The Clinical Significance of HLA antibodies detected by ... · transplant rates of highly sensitized recipients with virtual crossmatching in kidney paired donation. Transplantation

The Clinical Significance of HLA antibodies

detected by Solid Phase Assay in Renal

Transplant Recipients

Samantha Jane Fidler

MSc (Med), BSc (Hons)

A thesis submitted for the degree of PhD

2016

School of Pathology and Laboratory Medicine

2

Statement of candidate Contribution

All work was performed by the author unless otherwise stated in the

thesis

Samantha Fidler

Student

Lloyd D’Orsogna

Coordinating Supervisor

Ashley Irish

Supervisor

3

The data presented in Chapter 3 has been published. The bibliographic

details and percentage contribution of each author to the work are set out

below.

Fidler SJ, Irish AB, Lim W, Ferrari P, Witt C, Christiansen FT. Pre-transplant

donor specific anti-HLA antibody is associated with antibody-mediated

rejection, progressive graft dysfunction and patient death. Transplant Immunology 2013; 28(4): 148-153

Author contribution:

Fidler S J (80%)

Irish AB (10%)

Lim W (2%)

Ferrari P (2%)

Witt C (4%)

Christiansen FT (2%)

Lloyd D’Orsogna


4



below.

Fidler S, Swaminathan R, Lim W, Ferrari P, Witt C, Christiansen FT,

D’Orsogna L, Irish AB. Peri-operative third party red blood cell transfusion in

renal transplantation and the risk of antibody-mediated rejection and graft

loss. Transplant Immunology 2013; 29: 22-27


Fidler S J (75%)

Swaminathan R (2%)

Lim W (2%)

Ferrari P (2%)

Witt C (2%)


D’Orsogna L (2%)

Irish AB (13%)

Lloyd D’Orsogna


5



below.

Ferrari, P., Fidler, S., Wright, J., Woodroffe, C., Slater, P., Van Althuis-

Jones, A., Holdsworth, R., Christiansen, F.T. Virtual Crossmatch Approach to Maximise Matching in Paired Kidney Donation. American Journal of Transplantation 2011: 11; 272-278


Ferrari P (40%)

Fidler S J (40%)

Wright J (10%)

Woodroffe C (1%)

Slater P (2%)

Van Althuis-Jones A (2%)

Holdsworth R (2%)


Lloyd D’Orsogna


6



below.

Ferrari P, Fidler S, Holdsworth R, Woodroffe C, Tassone G, Watson N,

Cantwell L, Bennett G, Thornton A, Christiansen FT, D’Orsogna L. High transplant rates of highly sensitized recipients with virtual crossmatching in kidney paired donation. Transplantation 2012; 94: 744-749


Ferrari P (40%)

Fidler S J (45%)

Holdsworth R (1%)

Woodroffe C (1%)

Tassone G (1%)

Watson N (2%)

Cantwell L (2%)

Bennett G (2%)

Thornton A (2%)


D’Orsogna L (1%)

Lloyd D’Orsogna


7



below.

Böhmig G, Fidler S, Christiansen FT, Fischer G, Ferrari P. Transnational

validation of the Australian algorithm for virtual crossmatch allocation in kidney paired donation. Human Immunology 2013; 74: 500-505


Böhmig G (15%)

Fidler S J (55%)


Fischer G (10%)

Ferrari P (10%)

Lloyd D’Orsogna


8

I Contents

i Contents..........................................................................................8

ii List of Tables..................................................................................13

iii List of Figures................................................................................16

iv Summary......................................................................................19

v Thesis overview..............................................................................20

vi Dedication....................................................................................22

vii Acknowledgments.........................................................................23

viii Abbreviations...............................................................................24

ix Publications...................................................................................26

1 Chapter 1 Literature Review...................................................27

1.1 The Human Leukocyte Antigen (HLA) system...............................28

1.1.1 HLA haplotypes and linkage disequilibrium..........................30

1.1.2 Expression of HLA...........................................................31

1.1.3 HLA polymorphism..........................................................31

1.2 Antigen presentation................................................................32

1.2.1 Endogenous antigen presentation......................................32

1.2.2 Exogenous antigen presentation........................................33

1.2.3 The MHC and allorecognition in renal transplantation...........34

1.2.4 Direct allorecognition.......................................................34

1.2.5 Indirect allorecognition....................................................35

1.3 The significance of HLA in transplantation....................................36

1.4 Humoral and cellular mechanisms of rejection..............................37

1.4.1 Hyperacute rejection........................................................37

1.4.2 Acute cellular rejection.....................................................38

1.4.3 Acute humoral rejection...................................................38

9

1.4.4 Chronic rejection.............................................................39

1.5 Production of antibody by B cells................................................39

1.5.1 T cell independent B cell activation....................................39

1.5.2 T cell dependent B cell activation.......................................40

1.6 Pre-transplant crossmatching and detection of HLA antibodies.......41

1.6.1 Crossmatching strategies.................................................41

1.6.2 HLA antibody detection by ELISA.......................................44

1.6.3 HLA antibody detection by Luminex bead assay...................46

1.7 Post-transplant HLA antibodies...................................................48

1.8 Therapies for the treatment of antibody mediated rejection............49

1.9 Allograft accommodation...........................................................51

1.10 Strategies for transplanting the sensitised patient.........................52

1.10.1 Desensitisation...............................................................52

1.10.2 Acceptable mismatch program..........................................53

1.10.3 Paired Kidney Exchange...................................................53

1.11 Aims of this thesis....................................................................56

2 Chapter 2 Materials and Methods...................................57

2.1 HLA antigen typing (HLA-A, -B, -C, -DRB1, -DR345, -DQB1, -DPB1)

by Sanger sequencing...............................................................58

2.1.1 DNA extraction...............................................................58

2.1.2 DNA quantitation using the Nanodrop spectrophotometer....58

2.1.3 PCR amplification of target DNA........................................58

2.1.4 Agarose gel electrophoresis..............................................59

2.1.5 Purification of PCR products for sequencing.........................60

2.1.6 Sequencing of PCR amplified template...............................61

2.1.7 Purification and concentration of Big Dye Terminator

sequencing products using ethanol precipitation..................62

2.1.8 Electrophoresis of sequencing products on 3730

sequencer......................................................................64

2.2 LABType® SSO for HLA-DQA1 typing...........................................65

10

2.2.1 DNA amplification............................................................65

2.2.2 Agarose gel electrophoresis..............................................66

2.2.3 Denaturation / neutralisation............................................66

2.2.4 Hybridisation..................................................................67

2.2.5 Labelling........................................................................68

2.2.6 Data analysis – OneLambda HLA Fusion™...........................68

2.3 Serological HLA typing..............................................................70

2.3.1 Isolation of T and B lymphocytes.......................................70

2.3.2 HLA antigen typing using commercial typing trays...............70

2.4 Crossmatching by complement-dependent lymphocytotoxicity

(CDC).....................................................................................71

2.4.1 Standard (NIH) CDC crossmatch.......................................71

2.4.2 Interpretation of standard crossmatch results.....................71

2.4.3 DTT crossmatch..............................................................72

2.4.4 Interpretation of DTT crossmatch results............................72

2.5 Luminex antibody testing..........................................................73

2.5.1 Characterisation of antibodies by Single Antigen bead

assay............................................................................73

2.5.2 Data analysis – OneLambda HLA Fusion™...........................74

2.6 Statistical analysis....................................................................75

3 Chapter 3 Pre-transplant donor specific anti-HLA antibody is

associated with antibody-mediated rejection,

progressive graft dysfunction and patient

death…………………………………………………………………..76

Fidler SJ, Irish AB, Lim W, Ferrari P, Witt C,

Christiansen FT. Transplant Immunology 2013;

28(4): 148-153

11

4 Chapter 4 Peri-operative third party red blood cell

transfusion in renal transplantation and the risk of

antibody-mediated rejection and graft

loss……………………………………………………………………..98

Fidler S, Swaminathan R, Lim W, Ferrari P, Witt C,

Christiansen FT, D’Orsogna L, Irish AB. Transplant

Immunology 2013; 29: 22-27

5 Chapter 5 Virtual crossmatch approach to maximise

matching in paired kidney donation................118

Ferrari, P., Fidler, S., Wright, J., Woodroffe, C., Slater, P., Van Althuis-Jones, A., Holdsworth, R.,

Christiansen, F.T. American Journal of Transplantation 2011: 11; 272-278

6 Chapter 6 High transplant rates of highly sensitized

recipients with virtual crossmatching in kidney

paired donation...........................................137

Ferrari P, Fidler S, Holdsworth R, Woodroffe C,

Tassone G, Watson N, Cantwell L, Bennett G, Thornton A, Christiansen FT, D’Orsogna L.

Transplantation 2012; 94: 744-749

7 Chapter 7 Transnational validation of the Australian

algorithm for virtual crossmatch allocation in

kidney paired donation.................................159

Böhmig G, Fidler S, Christiansen FT, Fischer G, Ferrari P. Human Immunology 2013; 74: 500-505

12

8 Chapter 8 Discussion and conclusions...........................176

8.1 The significance of preformed anti-HLA antibodies detected by

Luminex bead assay...............................................................177

8.2 Peri-operative blood transfusion is associated with AMR and renal

allograft loss..........................................................................180

8.3 Virtual crossmatching to facilitate transplantation......................181

8.4 Further research.....................................................................182

8.4.1 Post-transplant antibody testing......................................182

8.4.2 Rejection in the absence of HLA antibodies......................182

8.4.3 HLA DSA in the absence of rejection.................................183

8.4.4 Characterisation of HLA epitopes in kidney

transplantation..............................................................183

8.4.5 HLA specific B cells........................................................184

8.4.6 Complement-fixing HLA specific antibodies.......................185

8.5 Conclusions...........................................................................187

9 References.................................................................................188

10 Appendix...................................................................................210

13

ii List of Tables

Chapter 1

Table 1. A summary of studies investigating the association of graft

outcome with pre-transplant class I and II antibodies detected

by ELISA .......................................................................45


outcome with pre-transplant class I and II antibodies detected

by Luminex bead assay....................................................47


outcome with post-transplant class I and II antibodies detected

by Luminex bead assay....................................................49

Chapter 2

Table 1. PCR mastermix volumes per sample..................................59

Table 2. Expected number of PCR products and respective sizes for each

locus.............................................................................60

Table 3. Sequencing primers provided for use with each locus...........62

Table 4. 3730 instrument settings for HLA sequencing......................64

Table 5. PCR amplification mix volumes..........................................65

Table 6. Post-PCR reagent volumes................................................67

Table 7. SAPE volume requirements per sample..............................68

Chapter 3

Table 1. Demographic and clinical features at the time of transplant

according to entry antibody status.....................................84

Table 2. Longitudinal mixed-effects models of predictors of eGFR

(coefficient and 95% CI = mls/min) across the observed

follow-up (median 68 months) incorporating factors identified

as univariate predictors of eGFR........................................90

14

Chapter 4

Table 1. Demographic and clinical features at time of transplant by

entry HLA-antibody........................................................105

Table 2. Patient demographics, HLA antibody, rejection and outcomes

by transfusion status......................................................106

Table 3. Univariate and multivariate adjusted Cox Proportional Hazards

models for the effect of Delayed Graft Function (DGF), entry

HLA antibody, Antibody and Non-Antibody Mediated Rejection

(AMR and Non-AMR) and Transfusion status on death censored

and combined patient and graft loss.................................106

Chapter 5

Table 1. ABO blood group of donors and recipients in the Western

Australian paired kidney donation program.......................125

Table 2. Comparison of 3 test runs using different antibody resolution

and strength to excluded recipients with antibodies from

matching to donors in a pool of 32 incompatible donor-

recipient pairs...............................................................127

Table 3. Match results by blood type excluding recipients with donor

specific antibody at >2000 mean fluorescence intensity

(OneLambda) and using high-resolution antibody

definition......................................................................129

Chapter 6

Table 1. Match run conditions for the quarterly match runs of the

Australian KPD program.................................................142

Table 2. ABO blood group distribution of donors and recipients in the 4

initial match runs of the AKX program for KPD...................143

Table 3. Aggregated results of 149 CDC allogeneic crossmatches

performed between 37 matched donors and matched or

unmatched recipients, whose sera were available for

testing.........................................................................146

15

Chapter 7

Table 1. Baseline characteristics of donors and recipients in the 16 pairs

of the Vienna transplant center.......................................164

Table 2. CDC and flow cytometry crossmatch results from 112 donor-

recipient pair combinations where crossmatch data were

available in relation to the presence or absence of any donor-

specific antibodies (DSA) at >2000MFI.............................165

Table 3. Sensitivity and specificity of donor-specific antibody (DSA)

positive computer matching to predict a positive crossmatch

and negative predictive value of DSA negative matching to

identify crossmatch negative results using virtual

crossmatching in kidney paired donation...........................167

Chapter 8

Table 1. Summary of studies investigating the clinical significance of

HLA antibodies detected pre-transplant by Luminex bead

assay...........................................................................178

Table 2. Summary of studies investigating the clinical significance of

complement-fixing DSA..................................................186

Appendix

Table 1. Conexio Genomics HLA Sequencing Based Typing kit

details..........................................................................211

16

iii List of figures

Chapter 1

Figure 1. The human MHC map......................................................29

Figure 2. Structure of the HLA class I and II molecules......................30

Figure 3. Growth of the IMGT / HLA database...................................32

Figure 4. Exogenous and endogenous antigen processing...................34

Figure 5. Direct and indirect allorecognition......................................35

Figure 6. Death censored graft survival rates in relation to the number of

HLA-A, -B and –DR mismatches in two eras – 1985-1994 and

1995-2004.....................................................................36

Figure 7. Number of rejection episodes in the first year of transplant in

patients with a functioning graft at 1 year..........................37

Figure 8. Simplified representation of donor antigen recognition and the

induction of humoral immune response - T cell-independent

and T cell-dependent B cell activation................................41

Figure 9. A schematic representation of antibody mediated rejection

pathophysiology..............................................................51

Figure 10. Paired kidney exchange – simple two-way exchange...........54

Figure 11. Paired kidney exchange – non-directed donor chain.............55

Chapter 3

Figure 1. a. Time to first biopsy proven acute rejection (BPAR) excluding

24 patients with AMR by entry DSA....................................85

b. Time to antibody mediated rejection by entry DSA

status............................................................................86

17

c. Time to antibody mediated rejection by class of DSA at

entry.............................................................................86

Figure 2. a. Strength of antibody determined by the MFI of the highest

bead in the 37 patients with DSA according to rejection

category (no BPAR, non-AMR and AMR)..............................87

b. Time to AMR by category of the MFI of the highest DSA bead

compared with No-Antibody..............................................88

Figure 3. a. Time to patient or graft loss by entry DSA.......................89

b. Time to patient or graft loss by entry class of DSA...........89

Figure 4. Longitudinal mixed model of mean observed eGFR by entry

DSA status over time.......................................................91

Chapter 4

Figure 1. Kaplan-Meier plot of the time to AMR by transfusion status in

all patients...................................................................107

Figure 2. a. Time to Antibody Mediated Rejection (AMR) by antibody

status at the time of transplant in the combined 195 patients

receiving no-RBCT, Pre-RCBT and Post-RCBT....................108

b. Time to Antibody Mediated Rejection (AMR) by antibody

status at the time of transplant in the 65 patients with Pre +

Post-RBCT....................................................................108

Figure 3. Time to death censored graft loss by transfusion status.....110

18

Chapter 5

Figure 1. Matching diagram of test run 3 using 32 pairs, where donors

were excluded from matching to recipients with moderate and

strong (>2000 MFI) donor-specific antibodies defined at the 4-

digit level.....................................................................128

Chapter 6

Figure 1. Match probability by cPRA (-A, -B, -DR and –DQ) in 27

crossmatch-negative matched recipients among 53 registered

patients.......................................................................144

Figure 2. Pattern of cPRA (-A, -B, -DR, and –DQ) in 53 registered

recipients and in 20 recipients who underwent

transplantation..............................................................145

Figure 3. DSA strength and specificity in crossmatch–positive assays

from 112 crossmatches performed on unmatched panel

recipients in match runs 1 and 2 against available donor

cells.............................................................................147

Chapter 7

Figure 1. Probability of a positive crossmatch in relation to the

cumulative strength of donor-specific antibodies (DSA) in mean

fluorescence intensity (MFI) for class I and class II

DSA.............................................................................168

Chapter 8

19

Figure 1. Transplant activity by calculated panel reactive antibody

(cPRA) cohort...............................................................181

iv Summary

For patients with end stage renal failure there are only two options for renal replacement

therapy: dialysis and transplantation. Complications such as anaemia, hypertension and

infection along with the patient’s own co-morbidities contribute to the relatively short life

expectancy for dialysis patients of around seven years. Kidney transplantation remains the

treatment of choice, and can extend life expectancy by an additional five to ten years when

compared to staying on dialysis. One of the major immunological hurdles in transplantation is

the presence of pre-formed antibodies in the recipient directed against donor HLA antigens.

Whilst current immunosuppressive regimens have improved graft outcomes, the presence of

donor specific anti-HLA antibodies is associated with allograft rejection and poor long term

graft outcome. In recent years, the detection techniques for anti-HLA antibodies has evolved

from the relatively insensitive lymphocytotoxic crossmatch to the introduction of more

sensitive solid phase assays including ELISA and Luminex bead assays.

In this thesis I investigate the clinical significance of anti-HLA antibodies detected by Luminex

single antigen bead assay on renal transplant outcome.

20

v Thesis overview

Chapter 1 is a review of the human leukocyte antigen (HLA) system. It describes the role of

HLA in renal graft rejection and reviews pre-transplant crossmatching and detection of anti-

HLA antibodies and their role in transplant outcome. It also describes the protocols available

for transplanting highly sensitised kidney patients.

Chapter 2 describes the materials and methods used to perform the studies in this thesis. In

addition to HLA typing, crossmatching and Luminex antibody assays, the methods for statistical

testing performed in data analysis are recorded in this chapter.

The study described in chapter 3 investigates the clinical significance of pre-transplant HLA

antibodies detected by Luminex single antigen bead assay in the presence of a negative

lymphocytotoxic crossmatch on renal transplant outcome. The results confirm that pre-

transplant antibodies directed against donor HLA are associated with antibody mediated

rejection (AMR) and poor graft outcome in our cohort.

In chapter 4 we report the novel finding that the risk of AMR in patients with pre-transplant

donor specific HLA antibodies is greatly increased by peri-operative blood transfusion. Peri-

operative blood transfusion also increases the risk of graft loss in this group.

Chapter 5 describes the design of a paired kidney exchange program in which donors were

excluded from matching to recipients with donor specific antibodies detected by single antigen

bead assay. Modelling of the program using two different antibody cut-offs by mean

fluorescence intensity (MFI) was performed. In the simulation, using a conservative MFI cut-off

of 2000 resulted in greater than 50% of pairs having a match with a predicted negative

crossmatch (virtual crossmatch) whilst reducing the risk of AMR and poor long term graft

function.

In chapter 6 we review the first year of the Australian kidney exchange program. There were

concerns that using a conservative antibody assignment to avoid breakdown of identified

matches could potentially disadvantage sensitised patients by excluding crossmatch

compatible donors. However, we were able to demonstrate that 35% of the patients

transplanted in the first year of the program were highly sensitised.

21

In chapter 7 we were able to show using independent donor / recipient pair data from Austria

that virtual crossmatching in the Australian paired kidney exchange program reliably predicted

negative crossmatches with a sensitivity of 98.6% and specificity of 69.2%. Furthermore, the

number of compatible crossmatches which were not matched with the Australian criteria was

around 10%. Adjusting antibody cut-offs for highly sensitised patients may facilitate

transplantation in individuals who are extremely difficult to match.

Chapter 8 summarises the key findings of the studies reported in Chapters 3 to 7 and their

significance in context of contemporary studies. Further areas for additional research is also

addressed.

References are cited in chapter 9.

22

vi Dedication

This thesis is dedicated to two people who have influenced my life. Firstly, my late grandfather,

Flight Lieutenant Richard Colbourne. Secondly, my late friend and colleague, Linda Karen

Smith. I miss you both.

23

vii Acknowledgments

There are many people to thank who have helped and supported me during my studies.

Firstly I would like to thank my colleagues in the Nephrology departments of Fiona Stanley,

Royal Perth and Sir Charles Gairdner hospitals. A special thanks to Dr Ashley Irish, Dr Wai Lim,

Dr Suda Swaminathan and CNS Anne Warger who have given me much encouragement and

moral support.

Many thanks to all my collaborators, a special mention to the staff of the Australian Tissue

Typing laboratories in Brisbane, Sydney, Melbourne and Adelaide for all our useful discussions.

Thankyou to the patients without whose permission studies such as these would not be

possible.

I would like to thank the staff, past and present, of the Department of Clinical Immunology for

all their help along the way. In particular I would like to thank Alana Chin, Brittany King, Mary

Despot, Gabriella Tassone and Jacqueline Dey.

Finally, to my family, thankyou for all the support. It seems to have taken forever to get here.

24

viii Abbreviations

ADCC Antibody Dependent Cell Cytotoxicity

AKX Australian Kidney Exchange

AMR Antibody Mediated Rejection

APC Antigen presenting cell

BPAR Biopsy Proven Acute Rejection

Β2m Beta2 microglobulin

BXM B cell crossmatch

CAN Chronic Allograft Nephropathy

CLIP Class II-associated Ii peptide

CREG Cross-Reactive epitope group

CTLs Cytotoxic T lymphocytes

DGF Delayed graft function

DSA Donor Specific Antibody

DTH Delayed typed hypersensitivity

DTT Dithiothreitol

ELISA Enzyme-Linked ImmunoSorbant Assay

ER Endoplasmic reticulum

ERAPs ER amino peptidases

FITC Fluorescein Isothiocyanate

HIFCS McCoys 5A tissue Culture Medium + 5% Heat

inactivated Foetal Calf Serum

HLA Human Leukocyte Antigen

HR Hazard Ratio

HRP Horseradish Peroxidase

IFN-γ Interferon-gamma

25

Ig Immunoglobulin

IL Interleukin

IM Infectious Mononucleosis

IMGT International Immunogenetics information system

IVIG Intravenous immunoglobulin

KPD Kidney Paired Donation

LD Linkage Disequilibrium

MFI Mean Fluorescence Intensity

MHC Major Histocompatibility Complex

MIC MHC-Class I Related gene

NOMS National Organ Matching System

PBS Phosphate Buffered saline

PKE Paired Kidney Exchange

PRA Panel reactive antibody

PTC Peritubullar Capillary

PTLD Post Transplant Lymphoproliferative disorder

SAPE Streptavidin Phycoerythrin

TAP Transporter associated with antigen presentation

TCR T cell receptor

TG Transplant glomerulopathy

Th T helper

TMB Tetramethyl-benzidine

TNF Tumour necrosis factor

TXM T cell crossmatch

UNOS United Network Organ Sharing

XM Crossmatch

26

ix Publications

Publications completed during my candidature but not presented in this

thesis

Smith LK and Fidler S. HLA typing technologies. Chapter 10 in The HLA Complex in Biology and

Medicine – A resource book, pages 175-187. Ed N Mehra. Jaypee Brothers Medical Publishers

Ltd, New Delhi, India ISBN 978-81-8448-870-8 (2010)

D Trajanoski, SJ Fidler. HLA Typing Using Bead Based Methods. Chapter 4 in: Immunogenetics –

Methods and Applications in Clinical Practice. Ed: FT Christiansen, BD Tait. Humana Press (part

of the Springer publishing group), New York, USA.

G Tassone, S Fidler. Separation and Cryopreservation of Lymphocytes from Spleen and Lymph

Node. Chapter 20 in: Immunogenetics – Methods and Applications in Clinical Practice. Ed: FT

Christiansen, BD Tait. Humana Press (part of the Springer publishing group), New York, USA.

S Fidler. Crossmatching by Complement-Dependent Lymphocytotoxicity (CDC). Chapter 21 in:

Immunogenetics – Methods and Applications in Clinical Practice. Ed: FT Christiansen, BD Tait.

Humana Press (part of the Springer publishing group), New York, USA.

Ferrari, P., Fidler, S., Woodroffe, C., Tassone, G., D'Orsogna, L.J. Comparison of time on the

deceased donor kidney waitlist versus time on the kidney paired donation registry in the

Australian program. Transplant International 2012: 25; 1026-1031

Ferrari P, Hughes P, Cohney S, Woodroffe C, Fidler S, D’Orsogna L. ABO-incompatible matching

significantly enhances transplantation rates in Kidney Paired donation. Transplantation 2013;

96 (9): 821-826

Do Nguyen H, Wong G, Howard K, Claas F, Craig J, Fidler S, D’Orsogna L, Chapman J, Irish A,

Ferrari P, Christiansen F, Lim W. Modeling the benefits and costs of integrating an acceptable

HLA mismatch allocation model for highly sensitized patients. Transplantation 2014; 97 (7):

769-774)

27

Chapter 1. Literature Review

28

This literature review will cover literature published prior to the commencement

of this thesis (2008). Literature published after this time will be discussed in the

relevant chapters and in the final discussion chapter.

1.1 The Human Leukocyte Antigen (HLA) system

The genes of the HLA system are located within the Major Histocompatibility Complex

(MHC) on the short arm of chromosome 6. There are over 400 genes in the human MHC

region (Figure 1)1.

The class I region contains the classical HLA-A, -B and –C genes that encode the heavy

chains (α chain) of class I molecules. The α chain is associated with β2-microglobulin

encoded on chromosome 15. The α-chain consists of two peptide binding domains, an

immunoglobulin-like domain and a transmembrane region with a cytoplasmic tail (Figure

2).

The class II region consists of a series of subregions each containing A and B genes

encoding the α and β chains respectively. The α and β-chains consist of a peptide binding

domain, an immunoglobulin-like domain and a transmembrane region with a cytoplasmic

tail (Figure 2).

The class III region does not encode HLA molecules but does contain other

immunomodulatory genes such as Complement components and Tumour Necrosis Factor

(TNF)2.

29

Figure 1. The human MHC map. The MHC is divided into class I, class III and class II subregions

from telomere to centromere. The white boxes denote expressed genes. The black, striped

and grey boxes show pseudogenes, non-coding genes and gene candidates respectively.

From: T Shiina et al. Journal of Human Genetics 2009; 54: 15-391

30

Figure 2. Structure of the HLA class I and II molecules.

From: http://what-when-how.com/acp-medicine/adaptiveimmunity

1.1.1 HLA haplotypes and Linkage Disequilibrium

The HLA genes are closely linked on chromosome 6 and the pattern of inheritance is

determined by Mendelian genetics. HLA alleles are not in random recombination

because of the genes close proximity on the chromosome, a phenomenon known as

linkage disequilibrium3,4. HLA-B and –C alleles are in linkage disequilibrium in the class I

region and HLA-DRB1, -DRB345, -DQB1 and –DQA1 are in linkage disequilibrium in the

class II region. A haplotype is defined as a combination of HLA genes carried on one

parental chromosome. Individuals inherit one haplotype from each parent.

Recombination of HLA genes is not common, 1-3% compared to the whole genome. The

distribution and frequency of HLA alleles differs between ethnic groups. Some alleles are

found in all populations but with varying frequencies. However, there are some alleles

which are found only in a specific population for example DRB1*04:12 has only been

identified in Australian Aborigines5.

http://what-when-how.com/acp-medicine/adaptiveimmunity

31

1.1.2 Expression of HLA

HLA class I antigens are present on most nucleated cells. The level of expression varies,

with the lymphoid and myeloid cells having high expression6,7. HLA class II antigens are

present on professional antigen presenting cells (B cells, macrophages, dendritic cells,

Langerhans cells), thymic epithelium and activated T cells8. Cytokines such as interferon-

gamma (IFN-γ) can induce or upregulate the expression of HLA class II under certain

conditions9, 10. In the transplanted kidney, damage due to ischemia and reperfusion,

mechanical damage or cellular rejection for example, may induce a non-specific

inflammatory response. Antigen presentation to T cells is increased, and activated T cells

release cytokines which act on endothelial cells to upregulate class II HLA expression. The

increase in HLA class II expression can lead to a further increase in antigen presentation to

T cells amplifying the cellular immune response and also increase the antigen density of

donor (non-self) HLA target for an antibody response.

1.1.3 HLA polymorphism

The HLA system is highly polymorphic, the most polymorphic system in humans.

Polymorphism is not evenly spread throughout the HLA molecule but rather is

concentrated in the peptide-binding groove11. Polymorphisms resulting in amino acid

changes can modify the shape of the peptide binding groove and therefore alter the

peptide binding specificity of HLA molecules. The number of HLA alleles reported to the

international ImMunoGeneTics information system (IMGT) database increases by

approximately 500 every three months. The number of new alleles identified has

exponentially increased over the past several years, this may be a result of new HLA

typing technologies identifying polymorphisms outside regions previously sequenced

(Figure 3). There have been >10000 class I and 3500 class II alleles reported to the IMGT

HLA database in January 201612.

32

Figure 3. Growth of the IMGT / HLA Database. The number of new class I alleles reported

to the IMGT/HLA Database are shown in green whilst the number of class II alleles are

shown in black.

From: J Robinson, JA Halliwell et al. Nucleic Acids Research 2015; 43: D423-43113

1.2 Antigen Presentation

The major function of MHC molecules is to present unique peptides on the surface of cells

in an arrangement that permits their recognition by the T cell receptor (TCR) on T

lymphocytes14.

1.2.1 Endogenous Antigen presentation pathway

Endogenous peptides for HLA class I presentation are generated in the cytosol primarily by

the proteosome. Short peptides are transported to the endothelial reticulum (ER) by

Transporter associated with antigen presentation (TAP) where they are further processed

by ER amino peptidases (ERAPs) to produce peptides around 8 to 10 amino acids in length.

Assembly of the class I α chain with β2m and the peptides is coordinated by the peptide –

loading complex (tapasin and ERp57, calreticulin and TAP) (Figure 4). Some exogenous

proteins that are internalised by dendritic cells and macrophage subsets via phagosomes

33

or endosomes can be retranslocated into the cytosol where they are processed by this

pathway14, 15.

Class I restricted T cells recognise endogenous antigens synthesised within the target cell.

The class I molecule-peptide complex on the cell surface is recognised by the TCR of CD8+

lymphocytes.

1.2.2 Exogenous Antigen presentation pathway

Exogenous proteins are primarily presented by class II HLA. Antigens are internalised by

phagocytosis, macropinocytosis or endocytosis and traffic to the MHC class II

compartment (MIIC). Peptides of around 10 to 16 amino acids in length are generated in

the MIIC. HLA class I is assembled in the ER with invariant chain (Ii) which protects against

premature interaction with peptide in pre-lysosomal compartments. The complex traffics

to the MIIC where Ii is subjected to sequential proteolysis (Figure 4). The final cleavage

product, class II-associated Ii peptide (CLIP), occupies the peptide-binding groove.

Exchange of CLIP with peptide is facilitated by the chaperone-like molecule HLA-DM. Class

II restricted T cells recognise exogenously derived antigens. The class II molecule-peptide

complex is transported to the cell surface and recognised by the TCR of CD4+

lymphocytes14.

34

Figure 4. Exogenous and endogenous antigen processing.

From: JS Blum, PA Wearsch, P Cresswell. Annu Rev Immunol 2013; 31: 443-47314

1.2.3 The MHC and Allorecognition in renal transplantation

Allorecognition is initiated by CD4+ and CD8+ T cells which recognise either intact

allogeneic MHC molecules on the surface of donor APCs (direct allorecognition) or donor

peptides presented by class II HLA molecules after they have been processed and

presented by recipient APCs (indirect allorecognition)16, 17.

1.2.4 Direct allorecognition

Donor derived dendritic cells (passenger leucocytes) migrate from the transplanted kidney

to the recipient lymphoid organs where they stimulate the host immune response by

presenting host and donor derived peptides via intact donor MHC to recipient T cells

35

(Figure 5a)18. Direct allorecognition is believed to be the driving force behind early acute

graft rejection. The direct alloresponse diminishes as donor passenger leucocytes are

depleted because the graft parenchymal cells cannot drive T cell activation in the absence

of co-stimulatory molecules19, 20.

1.2.5 Indirect allorecognition

Recipient dendritic cells which infiltrate the transplanted kidney present peptides derived

from donor MHC. Following migration to recipient lymphoid organs, MHC-peptide

complexes are presented to recipient T cells (Figure 5b)21. The indirect alloresponse

persists causing chronic rejection via a delayed type hypersensitivity (DTH) mechanism and

the production of alloantibodies.

Figure 5. Direct and indirect allorecognition. A) direct allorecognition: T cells recognise

intact allogeneic MHC molecules on the surface of donor APCs; b) indirect

allorecognition: Alloantigens are recognised as peptides presented by class II HLA

molecules after they have been processed and presented by recipient APCs.

From: Nature reviews Immunology 2002; 2: 859-871. JA Bradley, EM Bolton, RA

Pedersen22

36

1.3 The significance of HLA in renal transplantation

Improvement in immunosuppressive therapies has been associated with better patient

and graft survival rates principally due to reduction in early acute T-cell mediated

rejection. However, despite the impressive reductions in early rejection, chronic allograft

rejection remains a leading cause of graft dysfunction and graft loss. Large collaborative

studies demonstrate that the number of HLA mismatches at HLA-A, -B and –DR loci have a

significant influence on graft survival23, 24. When stratified by two decades encompassing

improvements in immunosuppressive drugs, the effect of mismatching is still significant

(Figure 6). In addition, the need for rejection treatment is also associated with the number

of HLA mismatches in both eras (Figure 7).

Allograft rejection may occur at any time post-transplant and is categorised as early

(hyperacute), short-term (acute) or long-term (chronic).

Figure 6. Death censored graft survival rates in relation to the number of HLA-A, -B and –

DR mismatches in two eras- 1985-1994 and 1995-2004. P<0.001 in both decades.

From: G Opelz, B Dohler. Transplantation 2007; 84(2): 137-14324

37

Figure 7. Number of rejection episodes in the first year of transplant in patients with a

functioning graft at 1 year. P<0.001 in both decades

From: G Opelz, B Dohler. Transplantation 2007; 84(2): 137-14324

1.4 Humoral and cellular mechanisms of rejection

1.4.1 Hyperacute rejection

In hyperacute rejection the transplanted graft is rapidly rejected within minutes to hours

due to vascular injury via antibody-dependent cytotoxicity25, 26. Previous sensitisation

against HLA antigens through transfusion, pregnancy or previous transplant, or blood

group incompatibility allows pre-formed antibody binding to the vascular endothelium in

the presence of antigen and initiation of the complement cascade with the release of

inflammatory mediators and extensive microvascular thrombosis.

Hyperacute rejection has been virtually eliminated by the introduction of the pre-

transplant complement dependent lymphocytotoxicity (CDC) crossmatch27.

38

1.4.2 Acute cellular rejection

Acute cellular rejection can occur early, within days of transplant, but usually occurs within

weeks to months. Rejection is mediated by CD4+ and CD8+ lymphocytes that have been

activated against donor antigens28, 29. The stimulation of alloreactive T cells occurs in the

lymphoid tissues of the recipient by donor dendritic cells (passenger leucocytes) by direct

allorecognition. Donor passenger leucocytes stimulate HLA class II reactive CD4+ T cells,

which can stimulate the growth and differentiation of HLA class I reactive CD8+ cytotoxic T

cells. These alloreactive T cells enter the circulation and react with the allograft vascular

endothelium. The resulting cytokine release and inflammatory mediators leads to

migration of alloreactive T cells through the vascular wall. The infiltrating lymphocytes can

have a direct cytotoxic effect as well as mediating a DTH mechanism of graft rejection

involving recruitment of macrophages and release of cytokines30.

1.4.3 Acute humoral rejection

Acute humoral rejection is mediated by antibody and complement. Antibodies are either

pre-formed or produced de novo post-transplant and bind to donor antigens expressed on

the allograft which can activate the complement cascade. Damage to the endothelium can

be mediated either directly by formation of the membrane attack complex, or by the

infiltration of inflammatory cells recruited by soluble complement fragments or by

phagocytes that recognise complement fragments deposited on endothelial cells via a

complement receptor. There may also be complement independent antibody-dependent

cell cytotoxicity following antibody binding to endothelial cells. Endothelial damage can

initiate platelet activation and thrombosis, endothelial and smooth muscle cell

proliferation and direct graft damage from humoral and / or cellular infiltrates.

Mechanisms of acute humoral rejection are reviewed in 31-36.

39

1.4.4 Chronic rejection

Chronic rejection usually occurs years after transplantation and is characterised

functionally by a progressive loss of kidney function, which may be accompanied by

hypertension and proteinuria - both cellular and humoral mechanisms have been

implicated. Histologically, biopsies reveal vascular hypertrophy and hyalinosis with

interstitial fibrosis and chronic inflammation. Glomerular mesangial hypertrophy with

typical endothelial double contours and glomerulosclerosis are often associated with

chronic rejection and described as transplant glomerulopathy and are often refractory to

treatment37. Features of chronic humoral rejection include the histological features of

transplant glomerulopathy with evidence of chronic complement activation with binding of

C4d along the vascular compartment suggestive of humoral injury.

1.5 Production of antibody by B cells

1.5.1 T cell independent B cell activation

T cell independent B cell activation can be induced by direct interaction of antigen with the

B cell (Figure 8)38, 39, 40. Two mechanisms of independent activation are recognised: Type I

and Type II. In type I antigens, such as lipopolysaccharides, induce B cell activation through

Toll-like receptors directing inducing B cell division. At high antigen concentrations

polyclonal B cell activation occurs. Both immature and mature B cells can be activated. In

type II antigens, such as polysaccharides, induce an antigen specific B cell response via

interaction with the B cell immunoglobulin (Ig) receptor. Only mature B cells specific for

the antigen are activated by this process.

The antibodies produced by T cell independent activation are typically IgM and low affinity

compared to those derived using T cell help.

40

1.5.2 T cell dependent B cell activation

In T cell dependent activation, activation of B cells occurs following binding of antigen to

the B cell Ig receptor resulting in enhanced expression of co-stimulatory molecules on the

B cell surface (Figure 8)41. The antigen bound Ig receptor is internalised via receptor

mediated endocytosis. Antigen is processed internally and peptides are presented on the

B cell surface on MHC class II. As antigen processing occurs, B7-1 and B7-2 are expressed

by the B cell. CD4+ T helper (Th) cells which have been primed by previous encounter with

antigen presented by dendritic cells or macrophages recognise MHC-peptide complex on

the B cell and also bind B7 on the B cell surface via CD28 leading to Th cell proliferation.

Once activated, the T cells express CD40L which binds CD40 on the B cell, which in turn

leads to increased expression of B7 on the B cell surface and therefore more T cell

activation42. Antigen recognition by B cells enhances the expression of receptors for

cytokines released by the Th cells. CD4+ Th cell secreted cytokines IL-2, IL-4 and IL-5 induce

B cell proliferation and differentiation and determine the type of antibody produced by

promoting isotype class switching43-45.

41

Figure 8. Simplified representation of donor antigen recognition and the induction of

humoral immune response - T cell-independent and T cell-dependent B cell activation.

Upper panel – glycan antigens can be directly recognised by B cells and the cross-linking of the

B-cell receptors lead to IgM secretion through T cell-independent B cell activation. Lower panel

– Donor antigens can be endocytosed by antigen presenting cells and presented to helper T

cells leading to T cell activation and subsequent T cell-dependent B cell activation.

Adapted from S Julien, PA Videira, P Delannoy. Biomolecules 2012; 2: 435–46646

1.6 Pre-transplant crossmatching and detection of HLA antibodies

1.6.1 Crossmatching strategies

HLA antibodies can be formed following blood transfusion, pregnancy or transplant. In

1969 Patel and Terasaki published a landmark paper which showed that antibody directed

against donor class I HLA antigens, detected as a positive T cell crossmatch by the

complement-dependent lymphocytoxicity (CDC) assay, were associated with hyperacute

graft rejection27. Eighty percent (24/30) of grafts were lost to hyperacute rejection in

patients with HLA antibody. Following this report, the detection of pre-formed HLA class I

42

antibodies by the CDC crossmatch became the gold standard for pre-transplant testing and

has virtually eliminated hyperacute rejection. In the same year, Morris et al reported an

association of HLA antibodies detected post-transplant with renal graft failure47. HLA

antibodies were detected by CDC crossmatch in 38% (11/29) of patients whose grafts had

been rejected.

The CDC technique has historically been used to screen patients awaiting a transplant for

HLA antibodies, and an estimate of their sensitisation (Panel reactive antibody – PRA)

calculated as the percentage of donor lymphocytes with which a patient serum has a

positive reaction. The identification of donor-specific antibody detected in the screening

CDC assay can be used to predict a positive CDC crossmatch and is considered a

contraindication to transplantation.

The panel of donor lymphocytes was often variable from centre to centre and did not

represent the potential donor population. Therefore the same antibody profile could

result in a different PRA value between different centres. For this reason, calculation of

the PRA based on the frequency of HLA antigens in the donor population has helped to

standardise PRA48-51.

Immediately prior to transplant, patient serum is crossmatched against potential donor

lymphocytes to detect pre-formed cytotoxic antibodies. Incubation of donor lymphocytes

with patient serum in the presence of complement, usually sourced from rabbits, results in

binding of antibody to target antigens. Antibody binding can induce conformational change

in the antibody molecule exposing the Fc (fragment crystallisable) segment that binds to

complement. The complement cascade is activated leading to membrane damage and cell

lysis. The standard CDC technique detects complement fixing IgG and IgM antibodies.

Several studies have shown that IgM antibodies are frequently non-HLA auto-reactive

antibodies and are not detrimental to graft outcome52-56. However, IgM antibodies can

undergo class switching under the influence of cytokines and co-stimulation of B cells on

contact with donor HLA antigen to which the patient has previously been senstitised45.

This may lead to the production of donor specific IgG antibodies which may be harmful to

the graft.

Crossmatches are performed on separated T and B cells due to the differential expression

of HLA class I and II antigens on these cells. A positive T cell crossmatch is considered an

43

absolute contraindication to transplantation; however a positive B cell crossmatch can

sometimes be acceptable.

Antibodies directed against class I HLA antigens can be detected by both T- and B-cell

crossmatches. However, only B cell crossmatches can detect antibodies directed against

HLA class II antigens. There are several reasons for a positive B cell crossmatch in the

presence of a negative T cell crossmatch: i) Low titre class I antibodies due to higher

expression of class I HLA on B cells57; ii) Non-HLA antibodies reacting with respective

antigens expressed on B cells; iii) Non-specific binding of IgG antibodies to Fc receptor on B

cells; iv) Class II HLA antibodies; v) Therapeutic antibody such as anti-CD20 (Rituximab).

The heterogeneity of the B cell crossmatch contributes to the uncertainty of the relevance

of a positive B cell crossmatch with concomitant negative T cell crossmatch in renal

transplantation38-44. Early studies reported no difference in graft outcome between

patients with a negative T and B cell crossmatch compared with those with a negative T

cell and positive B cell crossmatch58-62. More recent studies demonstrate that a positive B

cell crossmatch alone was not correlated with graft outcome, but when the cause of the

positive crossmatch was elucidated a difference in outcomes could be demonstrated.

Karrupan et al63 showed a B cell crossmatch effect only if class I antibodies were

demonstrated. Mahoney et al64 showed in a large cohort study that a positive B cell

crossmatch was predictive of poor outcome in sensitised patients. Additional studies have

also confirmed the importance of a positive B cell crossmatch where donor specific

antibodies can be demonstrated65-68.

Although a positive CDC crossmatch is highly predictive of early graft failure, an

unequivocally negative T cell CDC crossmatch does not necessarily result in long term graft

survival. Patel and Terasaki also reported that early graft loss occurred in some patients

despite a negative crossmatch. This is because weak or low-titre HLA antibodies are not

detected by the CDC crossmatch and despite their initial low level they may cause or

contribute to later graft loss. In addition, clinically relevant non-HLA antibodies are not

detected by the CDC crossmatch. The relative low sensitivity, and the inability to detect

HLA antibodies which are not able to fix complement lead to the introduction of more

sensitive assays such as the extended incubation time CDC crossmatch69, anti-human

globulin enhanced CDC (AHG-CDC) and flow crossmatching (FCXM). The AHG-CDC

44

crossmatch is more sensitive than the CDC assay because it detects low-titre antibodies

and antibodies which do not activate complement70, 71. Flow cytometry crossmatching has

greater sensitivity than the CDC assays because it allows the direct detection of recipient

antibody binding to donor lymphocytes72, 73. However, the significance of a positive

crossmatch using these more sensitive crossmatch assays is still unclear and has some

potential limitations74-78. It is possible that the assays may be over-sensitive and, in some

circumstances, the detection of low titre antibodies in otherwise low-risk patients might

exclude patients from proceeding to transplantation. Further investigation into the

relevance of a positive flow crossmatch, in particular a positive B cell flow crossmatch

alone is warranted.

1.6.2 HLA antibody detection by ELISA

Cellular assays have the major drawback in that the target cells, lymphocytes express non-

HLA antigens as well as HLA antigens on their cell surface. B cells express both class I and II

antigens, therefore a positive B cell crossmatch can be due to class I, class II, class I and II

or non-HLA antibodies including immune complexes and Fc receptor binding antibodies79.

To attempt to solve this problem, solid phase assays using isolated HLA antigens as targets

for antibody detection were developed. ELISA was the first type of solid phase assay to be

developed, using purified HLA antigens (pooled or single) bound to an ELISA plate.

Reactivity is determined by reading optical density using a spectrophotometer so removes

the subjectivity in antibody detection by CDC. The assay is semi-quantitative and specific

for all subclasses of IgG antibody80-82.

Several studies, including a large retrospective international study, have shown an

association between HLA antibodies detected in pre-transplant sera by ELISA with poor

graft outcome (Table 1)83-89. The majority of these studies considered the presence of any

HLA antibodies of relevance to the graft outcome and did not differentiate between

whether the HLA antibodies were donor specific, or non-specific. Not all studies reported

an association of pre-transplant antibodies detected by ELISA only with an adverse kidney

transplant outcome. In those studies which looked at the association of antibody with

graft loss, all found a correlation with the presence of class I antibodies whether they were

DSA89 or not83, 84. Of the studies investigating acute rejection however, only those looking

at DSA87, 88 rather than overall sensitisation85 found a correlation with the presence of class

45

I, II or I and II antibody. Patel et al87 were able to demonstrate this association even in

patients with a negative pre-transplant flow crossmatch. These results suggest that the

presence of DSA in pre-transplant sera is an important risk factor in acute rejection

episodes. However, detection of any HLA antibody which is associated with graft loss

might be due to the presence of DSA which have not been identified in the study, or a

predisposition to make HLA antibodies leading to production of DSA post-transplant.

Table 1. A summary of studies investigating the association of graft outcome with pre-

transplant class I and II antibodies detected by ELISA in renal transplants.

Authors and year of publication

Number in study / number of antibody positive

Antibody detected (any antibody or DSA)

Clinical end-point End-point correlation with antibody class

Class I Class II Class I and II

Monteiro et al. 1997

83

124 / 28 Any antibody

Graft loss Yes NI NI

Susal and Opelz. 2002

84


Graft loss Yes Yes NI

Slavcev et al. 2003

85


Acute rejection No No No

Susal and Opelz. 2004

86


Graft survival No Yes NI

Patel et al. 2007

87

330 / 20 DSA Acute humoral rejection

Yes Yes Yes

Lefaucheur et al. 2007

88

237 / 43 DSA Antibody mediated rejection

Yes Yes Yes

Susal et al. 2009

89


Graft failure Yes Yes Yes

NI = Not tested / investigated

The Lefaucheur group demonstrated that both current and peak (historical) sera positive

for HLA antibodies were associated with AMR, and could show a relationship between the

strength of the HLA antibody and graft outcome88. Furthermore, Wu et al showed that

class I antibodies which were detected on multiple occasions prior to transplant were

associated with poor graft survival, whereas donor-specific antibodies detected on only

one occasion were not detrimental to the allograft90.

46

1.6.3 HLA antibody detection by Luminex bead assay

Further sensitivity in HLA antibody detection by solid phase assay was achieved with the

development of the Luminex bead assay. In these assays, HLA antigens are attached to a

latex bead and then detected by flow cytometry91. The assay is more sensitive than ELISA92-

94, and the development of beads carrying a single recombinant HLA antigen (Single

Antigen Beads) has further assisted in the precise determination of antibody specificity

including antibodies directed against HLA-C, -DQ, -DP and –DR51/52/53. The assay can be

modified by changing the secondary detection antibody to detect different IgG subclasses

or different immunoglobulin isotypes95, 96. In addition the assay can be modified for

detection of complement-fixing antibodies which may be more pathogenic than antibodies

which are not complement fixing97, 98.

The clinical relevance of antibodies detected by Luminex bead assay in the absence of a

positive CDC crossmatch remains uncertain. At the time of commencement of this study

there were few publications investigating clinical outcomes from transplantation across

donor-specific antibodies detected by Luminex Single Antigen bead99-102. These papers are

summarised in Table 2. Although not always consistent, these studies indicate that both

class I and class II HLA antibodies are associated with a detrimental effect on long term

graft function due to high rates of rejection and delayed graft function.

47

Table 2. A summary of studies investigating the association of graft outcome with pre-

transplant class I and II DSA detected by Luminex bead assay in renal transplants.


Number in study / number of antibody (DSA) positive

Clinical end-point

Time to follow up

End-point correlation with antibody class


Gibney et al. 2006

99

136 / 20 DGF / Acute rejection / Graft loss

6 months Yes Yes Yes

Gupta et al. 2008

100

121 / 16 Acute rejection / DGF

1 and 5 years

No No No

Issa et al. 2008

101

598 / 103 Transplant glomerulopathy

Mean 54 months

No Yes Yes

Van den Berg-Loonen et al. 2008

102

34 / 13 Rejection episodes

180 days Yes Yes Yes

DGF = Delayed graft function

Gibney et al99 retrospectively studied 136 patients receiving a transplant between March

2003 and May 2004. Fifty-five patients had antibody detectable in pre-transplant sera by

Luminex PRA beads, of these 20 had demonstrable DSA. In the group of patients with DSA,

increased rates of primary non-function (10% vs 0%, P=0.06), delayed graft function (30%

vs 9%, P=0.04) and biopsy proven acute rejection (BPAR) (25% vs 3%, P=0.01) were

observed when compared to the group with no DSA. Graft survival at 6 months was also

significantly lower in those with DSA (75% vs 94%, P=0.04). Gupta et al100 could not confirm

these findings in their cohort of 121 patients transplanted between 1999 and 2001. HLA

antibodies were detected in 38 of the 121 patients, 16 patients had DSA. There was no

significant difference between 1 and 5 year patient and graft survival in patients with DSA,

non-DSA or no antibody. However, in a multivariate analysis, delayed graft function and

the presence of DSA were independent predictors of graft loss when using the non-

sensitised group as a reference. Issa et al101 studied a large cohort of 598 kidney recipients

transplanted between January 2000 and 2005. Of the 235 sensitised patients, 28% of the

patients with class I antibody had DSA and 55% of the patients with class II antibody had

DSA. In 100 patients with both class I and II antibodies, 31% had a class I DSA and 48% had

a class II DSA. The authors did not indicate the number of patients with both a class I and II

48

DSA. The group demonstrated that transplant glomerulopathy (TG) was associated with

the presence of class II antibodies (HR 1.89, 95%CI 1.42-2.52, p<0.0001) and that

association was significantly higher in patients with class II DSA (HR 3.524, 95%CI 1.67-7.44,

p=0.001). Furthermore, they showed graft survival in patients with TG to be significantly

lower than those without TG (62% versus 95%, p<0.001). In a smaller cohort of 34 highly

sensitised patients, van den Berg-Loonen et al102 showed a trend to earlier and higher

number of rejection episodes in 13 patients with DSA compared to those with no DSA

(p=0.08). They could not demonstrate a detrimental effect on graft function up to 8 years

post-transplant. These studies are all limited by including patients with a pre-transplant T

cell negative crossmatch which is likely to exclude any effect of high level class I antibody.

1.7 Post-transplant antibody monitoring

The presence of HLA antibodies detected at any time post-transplant has also been

associated with renal allograft rejection103-106. In addition, a number of studies have shown

the association of de novo HLA antibodies with acute and chronic rejection and poor graft

survival (Table 3)107-116. Terasaki et al108 showed that the production of any antibody post-

transplant (DSA or non-DSA) was associated with poor outcome. The group reported a 1

year survival rate after testing for HLA antibodies of 96% in those with no antibody versus

94% in those with any HLA antibody and 64% in those with DSA. The association of non-

DSA with poor outcome may be a marker of a general immune activation state rather than

HLA specific antibodies having a direct effect on the allograft, and may correspond to high

immunological responders. It has been hypothesized by some groups that DSA is ‘hidden’

and not detected in the serum by its binding to the graft. This concept is supported by the

finding that after nephrectomy of rejected kidneys, donor-specific antibodies are

subsequently detected in peripheral blood samples117-119. Eluates from kidney samples

have also shown donor-specific antibodies are bound to the endothelium120, 121.

49

Table 3. A summary of studies investigating the association of graft outcome with post-

transplant class I and II DSA detected by Luminex bead assay in renal transplants.


Number in study / number of antibody (DSA) positive

Pre-transplant antibody status

Clinical end-point

End-point correlation with antibody class


Mihaylova et al. 2006

107

72 / 16 No antibody Chronic rejection

Yes Yes Yes

Terasaki et al. 2007

108

1329 / 158 No antibody Graft failure Yes Yes Yes

Mao et al. 2007

109

54 / 32 No antibody DGF, Acute rejection, graft failure

Yes Yes Yes

Scornik et al. 2007

110

103 / 16 No antibody AMR Yes Yes Yes

Lee et al. 2009

111

50 / 37 No DSA AMR Yes Yes Yes

Haririan et al. 2009

112

297 / 93 Crossmatch negative

AMR No Yes Yes

Kedanis et al. 2009

113


AMR Yes Yes Yes

Stastny et al, 2009

114

137 /57 Negative T, neg / pos B flow crossmatch

Graft failure Yes Yes Yes

Cinti et al, 2009

115


Acute rejection Yes Yes Yes

Hidalgo et al. 2009

116

145 / 54 No DSA Microvascular pathology (subclinical rejection)

No Yes Yes

DGF = Delayed graft function; AMR = Antibody mediated rejection

1.8 Therapies for the treatment of antibody mediated rejection

Antibody mediated rejection is associated with a poorer prognosis than cellular rejection,

is frequently unresponsive to therapy and accounts for a large proportion of graft losses122-

126. Therapies used to treat AMR are targeted either against the alloantibody itself, or

against the antibody producing B cells (Figure 8)127. The most commonly used therapies

are i) plasmapheresis, a temporary removal of the antibody usually used in conjunction

with other therapies; ii) IVIg, which neutralises the alloantibody, inhibits complement

activation and modulates cell-mediated immunity128-130 ; and iii) Rituximab, an anti-CD20

50

monoclonal antibody which depletes B cells excluding plasma cells which do not express

CD20. These therapies are all temporary but can be effective in AMR treatment.

Lefaucheur et al published results of a controlled trial comparing a combination of

plasmapheresis, IVIg and Rituximab versus plasmapheresis alone and demonstrated a

91.7% graft survival at 36 months versus 50% respectively131.

More recently a therapy was introduced which not act directly on the antibody or antibody

producing B cells. Eculizumab is a humanised monoclonal antibody directed against the

complement protein C5 blocking the complement pathway at the terminus originally used

to treat paroxysmal nocturnal hemoglobulinuria (PNH)132 and recently found to be

effective in the treatment of AMR133. Stegall et all demonstrated that eculizumab was

effective in decreasing AMR in sensitised patients reducing the incidence from 41.2% in the

control group to 7.7% in those treated with the antibody134. A later study by the same

group indicated that eculizumab was not always successful with 2/26 patients treated

undergoing early AMR135.

Another potential therapy, Bortezomib, acts by inhibiting the 26s proteosome and

inducing apoptosis of rapidly dividing cells including bone marrow resident plasma cells136.

This drug is therefore able to deplete HLA antibody derived from long-lived plasma cells in

patients who are resistant to other therapeutic agents. Everly et al were able to show in a

short series of six treated patients improved graft function, a reduction in antibody levels

and rapid reversal of antibody-mediated rejection with no recurrence at 5 months137.

51

Figure 9. A schematic representation of antibody-mediated rejection pathophysiology.

Mechanisms of action of therapies are indicated. Therapies currently used are in red; drugs under

evaluation are in dashed orange. AMR, antibody-mediated rejection; Conv, convertase; IV,

intravenous; MAC, membrane attach complex; MHC, major Histocompatibility complex; mTOR,

mammalian target of rapamycin; NK, natural killer.

From: E Pouliquen, A Koenig et al. F1000Prime Rep 2015; 7: 51127

1.9 Allograft accommodation

The term accommodation was first used in the 1980’s in the context of blood group

incompatible transplants as a lack of immune response138-140. Anti-donor (blood group)

antibodies were removed prior to transplantation, but rejection was expected when these

transiently removed antibodies returned. On the contrary, these transplants survived and

functioned as well as blood group compatible transplants139, 140. The group of Bannett et al

were able to show this lack of response was not due to absence of blood group antigens

on the graft endothelium141. Others have shown that donor specific antibodies are

52

detectable and interact with the graft endothelium by C4d staining of peritubullar

capillaries142, 143. Accommodation is very common in blood group incompatible

transplantation. Haas et al reported over 80% of protocol biopsies in blood group

incompatible transplants showed C4d deposition with no effect on long term graft

outcome142.

Accommodation of HLA antibodies does not appear to be a common phenomenon and

many studies have demonstrated the presence of HLA antibodies prior to rejection

(reviewed in 1.6.2, 1.6.3, 1.7). It is clear, however, that this correlation is not absolute and

many patients have detectable DSA and no clinical or histopathological symptoms. It is

possible that accommodation of HLA antibodies is temporary, as the antibodies can be

detected many years before rejection as shown in long term studies such as those

reported by Mizutani et al144. However the frequency and mechanism by which

accommodation may occur in some patients with donor specific HLA antibodies is

unknown and warrants further research and elucidation.

1.10 Strategies for transplanting the sensitised patient

Approximately 30% of patients awaiting a kidney transplant in Australia are sensitised

against HLA antigens. The presence of anti-HLA antibodies often presents an impenetrable

barrier to transplantation and these patients have a longer wait on the transplant list, due

to donor incompatibility, and poorer survival on dialysis than their transplanted

counterparts145, 146. A further 5% of patients are highly sensitised, defined as a PRA of

greater than 80%. The transplant rates of highly sensitised patients is even lower, with

many of these patients remaining on the transplant list for greater than 5 years.

1.10.1 Desensitisation

The aim of desensitisation is to clear or decrease the presence of circulating alloantibodies

to a level which does not cause a positive crossmatch in order to prevent antibody

mediated rejection and to facilitate transplantation in highly sensitised patients. The two

most commonly used protocols involve the use of low dose intravenous immunoglobulin

(IVIG) with plasma exchange (Johns Hopkins protocol)147 or high dose IVIG alone (Cedars-

Sinai protocol)148. These protocols are often used in conjunction with Rituximab, an anti-

53

CD20 antibody, which depletes B lymphocytes but not plasma cells which do not express

CD20149. IVIG has multiple effects on different immune pathways150. It was initially thought

to have a neutralising effect on anti-HLA antibodies though anti-idiotypic antibodies128.

Later it was shown to inhibit complement activation129. IVIG also interacts with cells

involved in immune activation including macrophages, neutrophils and NK cells via the Fcϒ

receptor130. Stegall et al151 have shown that low dose IVIG with plasma exchange is more

effective than high dose IVIG alone. However, despite acceptable short-term patient and

graft survival in desensitised patients, a higher rejection rate and increased incidence of

AMR compared with unsensitised patients has been reported152, 153.

1.10.2 Acceptable mismatch program

The acceptable mismatch program is an initiative which originated in Eurotransplant

intended to facilitate the transplant of highly sensitised patients Europe-wide. The

program relies on two strategies to identify acceptable HLA mismatches for an individual: i)

identification and definition of HLA antibodies (unacceptable antigens) and subsequent

identification of HLA antigens towards which an individual has never formed an antibody

response154, 155; ii) definition of acceptable HLA antigens by identification of the repertoire

of epitopes shared between an individual’s HLA type and other HLA antigens using the

HLAMatchmaker program (www.matchmaker.net)156-158. In 2015, Eurotransplant reached

the 25th anniversary of the Acceptable Mismatch program and reported over 1000 highly

sensitised patients transplanted over this time who otherwise might not have found a

negative crossmatch159.

1.10.3 Paired kidney exchange

Paired kidney exchange was first proposed in 1986 by Rappaport160 as a simple exchange

between two incompatible donor and recipient pairs to create two compatible transplants

(a two way exchange). South Korea161 and Holland162, 163 were the first countries to

implement a paired kidney exchange program initially based on two-way exchanges

(Figure 10). To expand the number of potential matches, variations of paired donation

have been designed including closed chain matches with multiple pairs (3-way or higher)

and open chain matches starting with an altruistic donor and ending with a wait-list

54

recipient (Figure 11). These programs have a distinct advantage over desensitisation as

they avoid pre-formed HLA antibodies and therefore the risk of early AMR is much

reduced.

Figure 10. Paired kidney exchange – simple two-way exchange.

Green arrows denote compatible transplants, red dotted arrows denote incompatibility. Donor

1 is incompatible with recipient 1 and donor 2 is incompatible with recipient 2 due to HLA

sensitisation or blood group incompatibility. However, donor 1 is compatible with recipient 2

and donor 2 is compatible with recipient 1 and an exchange of donors is possible resulting in

two transplants.

55

Figure 11. Paired kidney exchange – non-directed donor chain.

Green arrows denote compatible transplants, red dotted arrows denote incompatibility. Donor

1 is incompatible with recipient 1, donor 2 is incompatible with recipient 2 and donor 3 is

incompatible with recipient 3 due to HLA sensitisation or blood group incompatibility. Recipient

1 does not have a compatible donor in the paired exchange pool, however, donor 1 is

compatible with recipient 2 and donor 2 is compatible with recipient 3. Introduction of an

altruistic donor who is compatible with recipient 1 forms the primary transplant and allows the

chain to proceed, with the end donor in the chain donating to the transplant wait list pool.

56

1.11 Aims of this thesis

One of the major immunological hurdles in transplantation is the presence of pre-formed

antibodies in the recipient directed against donor HLA antigens. Whilst current

immunosuppressive regimens have improved graft outcomes, the presence of donor

specific anti-HLA antibodies is associated with allograft rejection and poor long term graft

outcome. In recent years, the detection techniques for anti-HLA antibodies has evolved

from the relatively insensitive lymphocytotoxic crossmatch to the introduction of more

sensitive solid phase assays including ELISA and Luminex bead assays. The clinical

significance of donor specific HLA antibodies detected by Luminex single antigen bead

assay in pre-transplant sera with concurrent negative CDC crossmatch is uncertain. The

aims of this thesis are to retrospectively test pre-transplant sera from kidney transplant

recipients by Luminex single antigen bead assay and determine their clinical significance in

terms of rejection and patient and graft outcomes. We further aim to use this knowledge

to design virtual crossmatching strategies that allow transplantation of highly sensitised

patients in paired kidney exchange.

57

Chapter 2

Materials and Methods

58

2.1 HLA Antigen typing (HLA-A, -B, -C, -DRB1, -DRB345, -DQB1, -DPB1) by Sanger

sequencing

2.1.1 DNA Extraction

Recipient and donor Genomic DNA was extracted from peripheral blood samples

collected in ACD (Acid citrate dextrose) by commercial (QIAGEN) or in-house salting-

out method modified from the published method from Miller et al. (SA Miller, DD

Dykes, HF Polesky. A simple salting out procedure for extracting DNA from human

nucleated cells. Nucleic Acids Res 1988; 16(3): 1215)

2.1.2 DNA Quantitation using the Nanodrop spectrophotometer

1. Add 2ul of sterile deionised water to the Nanodrop probe to set a blank reference

2. Add 2µl of DNA sample to the Nanodrop probe to measure absorbance

3. Concentration (A260) and purity (A260 / A280 ratio) of the DNA sample is recorded.

The optimal purity is 1.8.

2.1.3 PCR amplification of Target DNA

1. For each locus set up one reaction for each sample being amplified. Include

appropriate positive controls of known genotype and at least one negative control for

each group of samples being amplified.

2. CG-HLA Sequence Based Typing Kits™ (Conexio Genomics) are supplied as separate

aliquots of primer mix and polymerase (Appendix Table I). The polymerase needs to be

added to the primer mix just prior to setting up the PCR reactions. Thaw the required

number of vials of the appropriate PCR Mix. Once thawed, vortex briefly to mix

components.

3. Calculate the number of samples to be tested and dispense the required amount of

PCR mix and DNA polymerase into a sterile tube (Table 1). Vortex the solution 3-4

times, for approximately 1 second each time.

59

Table 1. PCR mastermix volumes per sample

LOCUS HLA-A HLA-B HLA-C HLA-DRB1 HLA-DQB1

Locus-specific PCR mix

(e.g. CG-HLA-A mix)

16ul 16ul 16ul 16.7ul 16.7ul

DNA-Pol 1ul 1ul 1ul 0.3ul 0.3ul

4. Dispense 17µL of the primer/polymerase mix prepared in step 3 above into each

reaction well of either a 96 well plate, or 8 tube strips.

5. Add 3uL of sample DNA or appropriate positive control DNA to each reaction well.

Add 3uL of sterile water to the negative control reaction well.

6. Seal the reaction wells or tubes. Mix thoroughly but gently by vortexing, then

centrifuge briefly to ensure all reaction components are at the bottom of the wells (i.e.

turn centrifuge on, allow it to reach a speed of approximately 185g then turn off).

7. Place the 96 well plate or tube strips into a thermal cycler and amplify the target

sequence according to the thermal cycling conditions below.

Thermal Cycling Program

95°C 10 mins

96°C 20 secs

60°C 30 secs 33 cycles

72°C 3 mins

15°C hold

NB: Amplification takes approximately 2.5 hours to complete.

8. When the PCR is completed, remove the plate from the thermal cycler and either

proceed directly to gel electrophoresis or store at 4°C until required. Purification of

60

amplicons by ExoSap-IT® treatment should occur within 24 hours of completion of

PCR.

2.1.4 Agarose Gel Electrophoresis

1. Confirm successful amplification of the template DNA by agarose gel electrophoresis

with a 1% gel using 2µl of each PCR product combined with 5µl loading buffer.

2. Photograph the gel using a BIO-RAD GEL DOC 2000 Transilluminator. The ethidium

bromide labelled DNA bands are illuminated by the UV light.

3. The number and expected sizes of the resultant amplicons will vary according to the

locus and sample genotype. Expected PCR amplicon sizes are indicated in Table 2.

Band intensities should be reproducible within each assay if DNA quality is consistent.

A sample with a slightly weaker band can still be processed for sequencing using a

lower post- ExoSAP-IT® dilution volume. Very weak products should not be sequenced.

Table 2. Expected number of PCR products and respective sizes for each locus.

Locus

Expected band sizes

HLA-A

≈ 2 kb

HLA-B

≈ 2 kb

HLA-C

≈ 1 kb and 1.4 kb

HLA-DRB1

≈ 450bp - 650bp (Banding pattern will vary depending on which of the specific

allele groups are present, maximum of 2 bands expected)

HLA-DQB1

≈ 300bp and 350bp

2.1.5 Purification of PCR products for Sequencing

PCR products contain excess primers and unconsumed dNTP’s which require removal

before sequencing reactions can be performed. They may interfere with the efficiency of

the sequencing reaction, or can result in background “noise” in the sequence.

61

1. Prepare a mastermix consisting of 4µl of ExoSAP-IT® and 8uL of 2mM MgCl2 per

sample. Dispense 12µl of the mastermix into each PCR sample to be sequenced. Seal

the tubes or plates, vortex or place on shaker for 2 minutes and centrifuge to ensure

all reagents are at the bottom of the wells, before placing into the thermal cycler.

2. Run the thermal cycler according to the following protocol:

37°C - 30mins

80°C - 15mins

4°C - hold

3. Upon completion, dilute the purified product 1:8 with sterile water. This dilution

step will ensure that there is sufficient template to perform the sequencing reactions

and ensure that the concentration of the template is sufficient to produce good quality

sequence data. Note that weaker PCR products may require a lower dilution factor

than 1:8.

4. ExoSAP-IT® treated samples may be stored at 4°C until ready for use.

2.1.6 Sequencing of PCR Amplified Template

All sequencing reactions must be set up in a designated (“post-PCR”) area. This cannot be

the same area as used to set up the PCR reactions.

1. Allow the vial of BDT and other reagents to thaw at room temperature. Expose BDT

to light as little as possible. Once thawed, transfer reagents to ice. Reagents should be

kept on ice and the sequencing reactions should be prepared on a cooling block or on

an ice bath.

2. For each reaction prepare a fresh sequencing primer reaction mix as follows:

Reagent Volume Required

Sequencing Primer 2.0µl

BDT v3.1 Ready reaction mix 1.0µl

5 X Sequencing Buffer 3.5µl

Water 11.5µl

Sequencing primers are summarised in Table 3.

62

Table 3. Sequencing primers provided for use with each locus

LOCUS SEQUENCING PRIMERS

HLA-A AEX1F, AEX1R, AEX2F, AEX2R, AEX3F, AEX3R, AEX4F, AEX4R

HLA-B BEX1F, BEX2F, BEX2R, BEX3F, BEX3R, BEX4F, BEX4R

HLA-C CEX1F, CEX1R, CEX2F, CEX2R, CEX3F, CEX3R, CEX4F, CEX4R, CEX5F, CEX5R, CEX6F, CEX6R, CEX7F, CEX7R, CEX8F

HLA-DRB1 DRB1EX2F, DRB1EX3R, DRB1EX3R, RB-TG344-R

HLA-DQB1 DQB1EX2F, DQB1EX2R, DQB1EX3F, DQB1EX3R

3. Mix each sequencing primer reaction mix gently by pulse (2-3 seconds) vortexing.

4. Dispense 18μl of the sequencing reaction mix to each appropriate reaction tube or

well of a 96 well plate. If 8-tube strips or individual tubes are used, they should be

placed in a carrier so as to avoid mixing up the positions during the procedure.

5. Add 2μl of purified PCR product to each appropriate well.

6. Cap or seal wells or plate tightly, mix gently then centrifuge briefly to ensure all

reagents are at the bottom of the tubes.

7. Place the reaction tubes or plate into a thermal cycler and run according to the

following program:

96C for 10 seconds

50C for 5 seconds 25 cycles

60C for 2 minutes

then 4C indefinitely

8. Once the program is complete, remove the reaction tubes or plate from the thermal

cycler and either proceed directly to purification of the sequencing reaction products

or store at 4C until required. It is recommended that sequence reaction samples are

purified and run on the DNA sequencer within 24 hours.

63

2.1.7 Purification and Concentration of Big Dye Terminator sequencing products using

ethanol precipitation

1. Remove the reaction tubes or 96 well plate from the thermal cycler (or 4C storage)

and centrifuge briefly. If reusable lids/caps have been used during thermal cycling label

the lids/caps to avoid cross-contamination.

2. Each reaction requires 5µl of 125mM EDTA and 60µl Ethanol. Make a mix of the

following for a full 96 well plate:

550µl of 125mM EDTA

6.6ml of 100% Ethanol

Add 65µl of the above mix to each sample

The final ethanol concentration is 70%

3. Replace caps and invert tubes or plate 4 times to ensure thorough mixing.

4. Centrifuge briefly and leave at room temperature, protected from light, for 15

minutes to precipitate the extension products. Slightly longer precipitation times are

acceptable, preferably no longer than one hour.

5. Place the tubes or plate in a centrifuge and spin according to the following

guidelines:

Centrifuge speed 1400g – 2000 g: 45 minutes

Centrifuge speed 2000g to 3000g: 30 minutes

Important: Proceed to the next step immediately. If this is not possible, then

spin the tray for an additional 10 minutes immediately before performing the next

step.

6. Without disturbing the precipitates, remove the strip lids and discard the

supernatant, by inverting the tubes or plate onto absorbent tissues (e.g. facial tissues).

7. Place the inverted tubes or plate with the tissues into the centrifuge and spin briefly.

8. Remove the plate from the centrifuge, discard tissues, and then add 60µl of freshly

prepared 80% ethanol to each pellet. Cap the tubes, and then invert the tubes or plate

4 times to mix.

9. Place the tubes or plate in centrifuge and spin at approximately 1700g for 15

minutes.

10. Repeat steps 7 to 9, then remove from centrifuge, discard tissues. Vacuum dry for

15 minutes, or air dry for 1 hour, in the dark. Reseal plate if not proceeding

immediately to denaturation step.

64

11. Resolubilise pellets by adding 12µl of Hi-Di™ formamide. Thoroughly vortex tubes

or plate for approximately 15 seconds.

12. Denature samples by placing in a thermal cycler which is already at 98 degrees, for

5 minutes.

13. At the end of 5 minutes, remove from thermal cycler and immediately place on ice

for at least 5 minutes. Centrifuge briefly to bring down any condensation.

ENSURE THAT THERE ARE NO AIR BUBBLES IN THE WELLS. THESE CAN ENTER AND

DAMAGE THE CAPILLARY.

14. Load the reaction plate onto the automated sequencer and prepare the data

collection file according to the sequencer manufacturer specifications. Samples should

be run within 24-36 hours otherwise the Hi-Di™ formamide will begin to break down

to formic acid and there will be significant loss of resolution. If it is not possible to run

the plate within this time frame, it should be stored at 4 degrees.

2.1.8 Electrophoresis of sequencing products on 3730 sequencer

1. Run the sequencing product plates according to Table 4.

2. Use the instrument’s data collection software to process the raw collected data and

create the sequence files.

3. Analyse sequences using Assign ATF™ software.

Table 4. 3730 instrument settings for HLA sequencing.

PARAMETER SETTING

Dye set Z- BigDye V3

Mobility file KB_3730_POP7_BDTV3

Basecaller KB.bcp

Run Module Regular FastSeq50_POP7

Injection time 15 secs

Run Time 2.5 hours

65

2.2 LABType® SSO for HLA-DQA1 Typing

2.2.1 DNA Amplification

1. Turn on thermal cycler to keep heated lid warm.

2. Select the appropriate HLA typing kit (locus and resolution, OneLambda Inc) and

thaw out Amplification Primers, and D-Mix. Make sure to keep on ice until ready to

use.

3. Vortex D-mix and Amplification Primer for at least 15 seconds, and perform a quick

centrifuge (5 seconds) to spin down any liquid trapped under lid.

4. Create the amplification mixture by adding correct volumes of D-Mix and

Amplification Primers in a 1.5ml eppendorf tube as detailed in Table 5.

Table 5. PCR amplification mix volumes.

Number

of

Reactions

D-mix

(µl)

Amplification

Primer (µl)

AmpliTaq Polymerase (µl)

1 13.8 4 0.2

10 138.0 40 2.0

50 690.0 200 10.0

96 1491.0 432 21.6

5. Pipette 2µl of each DNA template into a 96 well PCR microplate.

6. Add the appropriate amount of AmpliTaq Polymerase to the amplification mixture

and vortex for 5 seconds.

7. Aliquot 18µl of Amplification mixture into each well containing DNA.

8. Cap or seal the plate and give a quick vortex to mix contents in plate.

66

9. Place PCR plate in thermal cycler and run according to the following program:

96°C for 3 minutes

96C for 20 minutes

60C for 20 minutes 5 cycles

72°C for 20 minutes


60°C for 15 minutes 30 cycles


72° for 1 minute

then 4C indefinitely

10. Remove PCR plate and store in a 4°C fridge.

2.2.2 Agarose Gel Electrophoresis

1. Confirm successful amplification of the template DNA by agarose gel electrophoresis

with a 1% gel using 5uL of each PCR product combined with 5uL loading buffer.

2. Photograph the gel using a BIO-RAD GEL DOC 2000 Transilluminator. The ethidium

bromide labelled DNA bands are illuminated by the UV light.

2.2.3 Denaturation/Neutralisation

1. Turn on Thermal cycler and run program to 60° C HOLD. Make sure lid is heated to

optimum temperature before use.

2. Remove reagents in Table 6, supplied in the SSO genotyping kit, from -80° C freezer

to room temperature to thaw and then prepare all reagent volumes in an

appropriately sized disposable tube as necessary. Return any unused reagents to 2° C

to 8°C (Do not re-freeze bead mixture after thawing).

67

Table 6. Post-PCR reagent volumes.

Number of

tests

Denaturation

Buffer (µl)

Neutralisation

Buffer (µl)

Hybridisation

Buffer (µl)

Wash Buffer

(µl)

Bead Mixture

(µl)

1 2.5 5 34 480 4

10 25 50 340 4800 40

20 50 100 680 9600 80

50 125 250 1700 24000 200

96 240 480 3264 46080 384

3. Prepare a crushed ice bath

4. Transfer 5µl of each amplified DNA sample into the corresponding well of a clean 96

well PCR plate.

5. Add 2.5µl Denaturation Buffer to each well of the PCR plate and mix thoroughly by

pipetting up and down. Incubate at room temperature (20° C to 24° C) for 10min.

6. Add 5µl Neutralisation Buffer with pipette to each well of the PCR plate, and mix

thoroughly.

7. Place PCR plate on ice bath.

2.2.4 Hybridisation

1. Add 38µl hybridisation mixture to each well.

2. Cover with tray seal and give the PCR plate a quick mix using the vortex.

3. Place PCR plate into the pre-heated (60° C) Thermal cycler, cover with PCR pad and

close and tighten lid. Incubate for 15 min.

4. Remove the PCR plate seal and quickly add 100µl Wash Buffer to each well.

5. Cover PCR plate with plate seal and centrifuge plate for 5min at 1000-1300g.

68

6. Remove the tray seal, then remove the wash buffer by dumping/flicking plate

upside down. Gently pat on absorbent paper 3 times to remove any residual liquid.

7. Repeat steps 4 to 6 two more times for a total of three wash steps.

8 Prepare the appropriate volume of 1x SAPE solution at the third centrifugation stage

(Table 7).

Table 7. SAPE volume requirements per sample.

Number of tests SAPE Stock Volume (µl) SAPE Buffer Volume (µl)

1 0.5 49.5

10 5.0 495.0

20 10.0 990.0

50 25.0 2475.0

96 48.0 4752.0

2.2.5 Labelling

1. Add 50µl of 1x SAPE solution to each well of the PCR plate.

2. Cover plate with seal and vortex thoroughly at low speed.

3. Place PCR plate into the pre-heated (60° C) Thermal cycler, cover with PCR pad and

close lid. Incubate for 5min.

4. Remove the PCR plate seal and quickly add 100µl Wash Buffer to each well.

5. Cover PCR plate with plate seal and centrifuge plate for 5min at 1000-1300g.

6. Remove wash buffer by dumping/flicking plate upside down.

7. Add 70µl Wash Buffer to each well and gently mix by pipetting.

8. Transfer contents of each well to a 96 well V-bottom microplate (reading plate)

using an 8-channel pipette. To avoid contamination use fresh pipette tips each time.

69

9. Cover reading plate with seal and aluminium foil. Keep in dark and at 4° C until ready

to read using LABScan™ Flow Analyser.

2.2.6 Data Analysis - OneLambda HLA Fusion™

Data analysis for the HLA assays are performed with the software package HLA

Fusion™. The LABType® analysis feature of the program analyses Luminex.csv output

files and is based on catalogue specifications provided with the software. Import the

data files as specified by the manufacturer.

The analysis software establishes the HLA type by firstly categorising each of the beads

via its internal fluorescence, then determining whether a probe is bound or not by the

strength of external fluorescence on that bead produced by the SAPE.

70

2.3 Serological HLA typing

2.3.1 Isolation of T and B lymphocytes

1. Centrifuge 9mls ACD blood samples at 1800rpm for 10 minutes with the brake off.

2. Remove the platelet rich plasma and replace with citrated phosphate buffered

saline (PBSC).

3. Add 50µl of class I Dynabeads® or 80µl of class II Dynabeads® and mix on a rotator

for 5 minutes.

4. Place on a Dynal® magnet for 2 minutes and remove supernatant.

5. Add 9mls PBSC and incubate at room temperature for 2 minutes.

6. Place on a Dynal® magnet for 2 minutes and remove supernatant.

7. Repeat steps 5 and 6.

8. Add McCoys 5A tissue Culture Medium + 5% Heat inactivated Foetal Calf Serum

(HIFCS).

9. Using a Hamilton syringe spot 1ul of cell suspension in triplicate onto an oiled

Terasaki tray.

10. Add 5ul Fluoroquench™ (Acridine orange / Ethidium bromide) stain to each well.

Allow cells to settle for approximately 20 minutes.

11. View the cells under the fluorescent microscope to determine cell concentration

and viability. Concentration is adjusted by addition or removal of 5% FCS/McCoys.

2.3.2 HLA antigen typing using commercial typing trays

Commercial HLA typing trays are prefilled Terasaki trays containing mineral oil, antisera

with well-documented specificity and complement.

1. Add 1µl prepared T cells to 72 well commercial class I typing trays (LM144A and

LM144B, OneLambda Inc) and 1µl B cells to commercial class II typing tray (MDR172,

OneLambda Inc).

2. Incubate trays at room temperature for 1 hour.

Add 5µl Fluoroquench™ staining solution to each well and allow cells to settle for

approximately 20 minutes.

71

3. View the cells under the fluorescent microscope and determine the percentage cell

death in each well.

4. Assign HLA type according to cell reactivity using antisera information.

2.4 Crossmatching by complement-dependent lymphocytotoxicity

2.4.1 Standard (NIH) CDC Crossmatch

1. Add 1µl recipient serum, with 1µl of appropriate controls (Positive and negative) to

an oiled Terasaki tray. Make trays in duplicate for both T and B cell crossmatches.

2. Add 1µl T or B cells using an automated cell dispenser to avoid cross-contamination.

Ensure all wells are thoroughly mixed.

3. Incubate the trays for 45 minutes at room temperature

4. Add 5µl of pooled rabbit complement to each well using an automated dispenser to

avoid cross-contamination.

5. Incubate the trays for 45 minutes at room temperature.

6. Add 5µl Fluoroquench™ stain to each well. Allow cells to settle for approximately 20

minutes.

7. Score the percentage cell death semi-quantitatively under the fluorescent

microscope.

2.4.2 Interpretation of standard crossmatch results

1. A positive or negative crossmatch is determined by comparing the crossmatch score

with the background reactivity i.e. the AB negative control serum.

2. Results are reviewed with other immunological data to identify level of risk.

2.4.3 DTT crossmatch

1. Add 1µl recipient serum, with 1µl appropriate controls (IgG and IgM positive

controls) in triplicate to an oiled Terasaki tray. Make trays in duplicate for both T and

B cell crossmatches.

2. Add 1µl 5mM Dithiothreitol (DTT) to one of the triplicate wells.

3. Add 1µl PBS to a second triplicate well to act as a dilution control.

72

4. Incubate at 37°C for 30 minutes.

5. Add 1µl L-Cystine (saturated solution) to the DTT and PBS wells.

6. Perform steps 2 to 7 of the standard CDC crossmatch as detailed in 2.3.1.

2.4.4 Interpretation of DTT crossmatch results

1. A positive or negative crossmatch is determined by comparing the crossmatch score

with the background reactivity i.e. the AB negative control serum.

2. A crossmatch is considered positive if the neat, DTT and PBS wells are positive.

3. A crossmatch is considered negative if the neat well is negative, or the neat and PBS

wells are positive and the DTT well is negative. Note: The PBS well should retain

reactivity for a valid result. If this well is negative, dilution cannot be excluded as a

cause.

73

2.5 Luminex Antibody Testing

2.5.1 Characterisation of antibodies by Single Antigen bead assay

1. Bring the antibody kit (OneLambda Inc) to room temperature.

2. Centrifuge recipient serum at 14,000 rpm for 10 minutes to remove any debris / fat.

3. Place 20µl serum into a 96 well V-bottomed microplate.

4. Add 2.5µl Class I or Class II Single Antigen beads to each well. Ensure the vial

containing the beads is well vortexed prior to use.

5. Cover with plate seal and incubate in the dark on a shaker for 30 minutes.

6. Make stock wash buffer by diluting the 10 x concentrate with sterile deionised

water.

7. After incubation, add 150µl wash buffer to each well using a 8-channel pipette.

8. Cover with fresh plate seal and centrifuge at 1400g for 5 minutes.

9. Remove wash buffer by dumping / flicking plate upside down.

10. Gently pat on absorbent tissue 5 times and dry vortex.

11. Add 200µl wash buffer to each well and cover with a fresh plate seal.

12. Centrifuge at 1400g for 5 minutes.

13. Prepare secondary conjugate by diluting 1µl FITC conjugated Goat anti-human IgG

in 99µl wash buffer for each sample.

14. Repeat steps 9 to 12. Remove wash buffer by dumping / flicking plate upside down.

15. Add 100µl diluted secondary conjugate to each well and cover with a fresh plate

seal.

16. Incubate in the dark on a shaker for 30 minutes.


18. Remove secondary conjugate by dumping / flicking plate upside down.


20. Add 200µl wash buffer to each well and cover with a fresh plate seal.


22. Remove wash buffer by dumping / flicking plate upside down.


24. Repeat steps 20 to 23.

25. Add 65µl cold PBS to each well.

74

26. Cover plate with seal and aluminium foil. Keep in dark and at 4° C until ready to

read using LABScan™ Flow Analyser.

2.5.2 Data Analysis - OneLambda HLA Fusion™

Data analysis for the HLA assays is performed with the software package HLA Fusion™.

The LABScreen® analysis feature of the program analyses Luminex.csv output files and

is based on catalogue specifications provided with the software. Import the data files

as specified by the manufacturer.

The analysis software establishes the antibody specificity by firstly categorising each of

the beads via its internal fluorescence, then determining whether a probe is bound or

not by the strength of external fluorescence on that bead produced by the FITC.

75

2.6 Statistical Analysis

Statistical analyses were performed by using SPSSv18 (SPSS Inc., Chicago IL, USA). For

categorical data Fisher's exact test or Pearson's chi-square tests were used. Parametric

data were compared by ANOVA or t-test, and for non parametric data Mann–Whitney

U test or Kruskal–Wallis one-way ANOVA was used. Comparisons of within group

differences by z-test were made with Bonferroni adjustment reported at the p < 0.05

level. Time to event of interest (AMR, graft and patient survival) was estimated by the

method of Kaplan–Meier and Cox proportional hazard regression analysis with the

predictor satisfying the proportional hazard assumption. Covariates examined were

HLA-antibody at entry, rejection, gender, re-transplantation, and delayed graft

function. Results were expressed as hazard ratios (HR) with 95% CI.

76

Chapter 3

Pre-transplant donor specific anti-HLA

antibody is associated with antibody-

mediated rejection, progressive graft

dysfunction and patient death.

77

Pre-transplant donor specific anti-HLA antibody is associated with antibody-

mediated rejection, progressive graft dysfunction and patient death.

Samantha J Fidler, Ashley B Irish, Wai Lim, Paolo Ferrari, Campbell S Witt, Frank T

Christiansen

The long-term clinical outcomes in patients with antibodies detected by Luminex single

antigen bead assay in the context of a negative CDC crossmatch are unclear. Previous

studies have shown varying results and significance. We sought to elucidate the clinical

significance of Luminex-defined antibodies in a cohort of patients transplanted with a

negative CDC T cell crossmatch in an era before Luminex antibodies were routinely used.

We retrospectively tested pre-transplant sera from a cohort of Western Australian kidney

transplant patients by Luminex single antigen bead and stratified patients by class of

antibody and looked at the long-term effects of such antibodies on clinical outcome.

These findings were published in:

Transplant Immunology 2013; 28: 148-153

Contributions:

SJF Performed all antibody assays, participated in study design,

participated in data analysis, prepared the manuscript

ABI Participated in study design, participated in data analysis, participated

in writing of the manuscript

WL, PF, CSW, FT Participated in study design, participated in writing of the manuscript

78

Pre-transplant donor specific anti-HLA antibody is associated

with antibody-mediated rejection, progressive graft

dysfunction and patient death

Samantha J. Fidler a, b, 1, Ashley B. Irish c, d, 1, Wai Lim e, 2, Paolo Ferrari f, 2,

Campbell S. Witt c, d, 3, Frank T. Christiansen c, d, 3

a Department of Clinical Immunology, PathWest, Royal Perth Hospital, University of Western Australia, Australia

b School of Pathology and Laboratory Medicine, University of Western Australia, Australia

c Department of Nephrology, Royal Perth Hospital, University of Western Australia, Australia

d School of Medicine and Pharmacology, University of Western Australia, Australia

e Department of Nephrology, Sir Charles Gairdner Hospital, University of Western Australia, Australia

f Department of Nephrology, Fremantle Hospital, University of Western Australia, Australia

a r t i c l e i n f o

Article history: Received 15 April 2013

Received in revised form 28 April 2013

Accepted 1 May 2013

Keywords: Donor-specific antibody (DSA) Kidney transplant

Abbreviations: AMR, Antibody-mediated Rejection; CDC, Complement Dependant Cytotoxicity; cPRA, Calculated

Panel Reactive Antibody; DSA, Donor Specific Antibodies; HB-MFI, Highest bead MFI; DGF, Delayed

Graft Function; BPAR, Biopsy Proven Acute Rejection; MFI, Mean Fluorescence Intensity; SAB,

Single Antigen Bead.

* Corresponding author at: Department of Clinical Immunology, PathWest, Royal Perth Hospital, Wellington Street,

Perth, WA 6000, Australia. Tel.: +61 8 9224 1144; fax: +61 8 9224 2920.

E-mail address: [email protected] (S.J. Fidler).

1 Participated in data analysis, performance of the research and writing of the paper.

2 Participated in performance of the research.

3 Participated in data analysis and writing of the paper.

mailto:[email protected]

79

a b s t r a c t

Background: The long term effect of donor specific antibodies (DSA) detected by

Luminex Single Antigen Bead (SAB) assay in the absence of a positive complement-

dependant cytotoxicity (CDC) crossmatch is unclear. DSA at the time of transplant

were determined retrospectively in 258 renal transplant recipients from 2003 to 2007

and their relationship with rejection and graft function prospectively evaluated. After

a median of 5.6 years follow-up 9% of patients had antibody mediated rejection (AMR)

(DSA 11/37(30%), DSA-Neg 13/221 (6%), HR 6.6, p < 0.001). Patients with anti-HLA

class II (HR 6.1) or both class I + II (HR 10.1) DSA had the greatest risk for AMR. The

Mean Fluorescent Intensity (MFI) of the DSA was significantly higher in patients with

AMR than those with no rejection (p = 0.006). Moreover, the strength of the antibody

was shown to be important, with the risk of AMR significantly greater in those

with DSA > 8000 MFI than those with DSA < 8000 MFI (HR 23, p < 0.001). eGFR

progressively declined in patients with DSA but was stable in those without DSA (35.7

± 20.4 mls/min vs 48.5 ± 22.7) and composite patient and graft survival was

significantly worse in those with class II (HR 2.9) or both class I + II (HR 3.7) but not

class I DSA. Class II DSA alone, or in combination with class I DSA had the strongest

association with graft loss and patient death. Patients with DSA not only have

increased rates of acute AMR, but also chronic graft dysfunction, graft loss and death.

Antibody burden quantified by SAB assay may identify patients at highest

immunological risk and therefore influence patient management and improve long-

term patient outcome.

Introduction

The use of the CDC T cell crossmatch for the detection of preformed anti-HLA class I

antibody to eliminate hyperacute rejection has allowed renal transplantation to

become the preferred method of management of end stage renal disease (ESRD) in

eligible patients. Advances in immunosuppressive therapies have also reduced acute

cellular rejection rates to very low levels [1]. However, despite these advances,

80

improving long-term graft survival remains problematic [2]. Increasingly the role of

antibody mediated graft injury, manifested either as acute humoral rejection or

chronic AMR (transplant glomerulopathy), has been recognised as an important,

potentially modifiable cause of graft attrition [3–6].

The introduction of sensitive and specific solid phase antibody detection assays allow

detection of lower levels of antibody than those associated with a positive CDC T- or B-

cell crossmatch. These assays can accurately determine the strength and specificity of

antibodies, thereby improving the ability to examine their role in graft rejection. We,

and others, have previously shown that the presence of anti-HLA antibodies detected

after transplant is associated with a significant risk of subsequent graft loss [4,7–9]

though it was not known whether these antibodies were present at the time of trans-

plant or arose de novo. However, there is contradictory data on the significance of

antibodies detected only by solid phase assay in the presence of a negative T-cell CDC

crossmatch at the time of transplant. Whilst some recent studies have shown that DSA

at the time of renal transplant are associated with increased risk of acute AMR [10–12]

when the crossmatch is positive and others have shown that even with a negative

crossmatch [9,12–15], pre-transplant DSA are associated with early AMR and

rejection, there is little information on the influence of pre-transplant DSA on long

term renal function.

Current literature is unclear in terms of the importance of class I versus class II versus

strength of DSA with respect to AMR and graft loss, and these differences may be

related to differences in transplant management and immunosuppressive protocols.

We initially set out to determine how our data sit with respect to other publications in

terms of AMR and graft loss so that graft function, as measured by eGFR, could be

viewed in the context of the rest of our data.

Results presented here validate the association between pre- transplant DSA and

AMR. We show that, like some others, classes I and II DSA are more important than

either class I or II alone for AMR. However, we further propose that the class of DSA

may be unimportant and the effects of DSA on graft rejection and outcome can be

explained by antibody strength. These findings may help to explain the differing

importance of classes I and II antibody between centres. We also show that DSA is

associated with non-AMR though not as strongly as with AMR.

81

Subjects and methods

Patients

We studied 267 patients with a negative T cell CDC crossmatch who underwent 270

consecutive renal transplants between June 2003 and October 2007. We excluded one

patient with a simultaneous liver–kidney transplant, and those patients who died (n =

3) or lost their graft within the first 30 days of transplantation (n = 8; 3 of whom were

subsequently re-transplanted within the period of observation), leaving 258 patients

and allografts. Of the 258 patients, 173 were also B cell crossmatched. 246 patients

received a kidney transplant in Perth, Western Australia and 12 patients received

simultaneous kidney–pancreas transplants at Westmead Hospital in New South Wales.

Patients were managed with varying immunosuppressive regimes but all patients

received a Calcineurin inhibitor (CNI) (Tacrolimus or Cyclosporine) at the time of

transplantation in combination with mycophenolate mofetil or mycophenolate sodium

and corticosteroids. Consistent with Australian practice, the IL2R Antibody Basiliximab

was commonly used for induction but the use of anti-T-cell preparations for induction

was rare (Table 1). The need for biopsy, diagnosis and management of rejection,

medication adjustments was determined by the caring clinician and was not protocol

driven.

Prospective testing

The Department of Clinical Immunology (DCI), Royal Perth Hospital performed all T

and B-cell CDC crossmatching against WA donors. For organs donated from other

Australian states the crossmatch was performed at the state of donor origin. Pre-

transplant HLA typing and HLA antibody testing of the kidney recipients was

performed by DCI. All deceased donors were typed for HLA-A, -B, -DR, -DR51, -DR52, -

DR53 and -DQ by serology using commercial monoclonal antibody trays (OneLambda

Inc). Living donors and all potential recipients were typed for HLA-A, -B, -DRB1 by DNA

sequencing based typing (In-house) [16].

Donor and recipient HLA-A, -B and -DR matching, patient sensitisation (PRA) and

waiting time were considered in recipient allocation. Antibodies directed against HLA-

A and -B were determined by CDC crossmatch or Luminex PRA bead (OneLambda Inc).

82

The presence of class I DSA in current or historical sera and previous donor class I or

class II HLA mismatches (HLA-A, -B and -DR only) were exclusions for organ allocation.

A negative pre-transplant T cell CDC crossmatch using current (within 3 months prior

to the transplant) and historical sera was mandatory for transplantation. B cell

crossmatching was performed for 67% of the patients, however a positive B cell

crossmatch was not considered an absolute contraindication to transplantation.

Retrospective testing

For this study, sera collected at the time of transplant were screened retrospectively

for anti-HLA class I and class II antibodies using the Luminex Mixed Screen assay and

those with a positive screen were characterised for HLA class I and/or class II antibody

specificity using Single Antigen beads (LABScreen Single Antigen beads, OneLambda

Inc). The assays were performed in accordance with the manufacturer's protocol.

Antibodies were considered to be positive if the MFI for a particular bead was greater

than 500. HLA antibodies with an MFI > 500 directed against a donor HLA antigen

(HLA-A, -B, -C, -DRB1, -DRB3,4,5, -DQB1, or -DPB1) were considered to be DSA.

Calculated PRA (cPRA) was determined as the frequency of antibody against HLA-A, -B,

-DRB1, DRB3,4,5 and -DQ antigen in a reference cohort of 200 Australian individuals.

Donors were typed at all HLA loci (HLA-A, -B, -C, -DRB1, -DRB3,4,5, -DQB1 and -DPB1)

by in-house sequence based typing [16].

Clinical outcome parameters

Delayed graft function (DGF) was defined as the need for dialysis within the first 48 h

of transplantation. Graft loss was defined as the return to dialysis. All rejection

episodes were proven by biopsy, undertaken when clinically indicated, whilst those

episodes clinically defined in the absence of biopsy confirmation were not considered

biopsy proven acute rejection (BPAR). Histological reporting of renal biopsies was

undertaken by the local histopathologists as part of routine clinical care and was made

without information as to the presence or absence of DSA. The biopsy findings were

graded according to the Banff Classification 2003 [17]. BPAR that did not fulfil the

83

criteria for AMR was termed Non-AMR. The first BPAR was used to construct time to

event analysis and where multiple rejections occurred, the highest reported grade was

recorded. Time to AMR was recorded as a separate event to allow analysis by rejection

type (AMR vs non-AMR). eGFR was calculated at 1, 2, 3, 6 and 12 months and then

annually according to the abbreviated MDRD formula.

Statistical analysis

Statistical analyses were performed by using SPSSv18 (SPSS Inc, Chicago IL, USA) and

Stata (Version 11, StataCorp, Texas, USA). For categorical data Fishers Exact Test or

Pearson's chi-square tests were used. Parametric data were compared by ANOVA or t-

test, and for non parametric data Mann–Whitney U test or Kruskall Wallis 1 way

ANOVA were used. Time to event was estimated by the method of Kaplan–Meier and

Cox Proportional Hazards was used to determine Hazard Ratio's. Predictors of time

dependent eGFR estimates were determined by a Longitudinal Mixed Model with time

being treated as a random effect with observations at multiple but irregular time

intervals. Mean eGFR was estimated over the months following transplant. Graft loss

was censored when no further eGFR estimates occurred due to return to dialysis or

death.

Results

Features of the patient groups

Patients were initially divided into 3 groups according to their antibody status at the

time of transplant: i) DSA group n = 37 (14%) — patients with HLA antibody with

MFI >500 against one or more donor HLA antigens, ii) Non-DSA group n = 28 (11%) —

patients with HLA antibodies (MFI >500) but not directed against donor antigens or iii)

No-antibody group n = 193 (75%) — patients with no HLA antibodies defined using the

MFI cut-off of <500 or with a negative antibody screen. Baseline clinical and

demographic data of these groups is shown in Table 1. In the DSA group 13 (35%)

patients had class I DSA only, 11 (30%) had class II DSA only and 13 (35%) patients had

both class I and II DSA. Of the 258 patients, 173 were also B cell CDC crossmatched

(BXM). Only 4 patients had a positive BXM and DSA, however the Australian algorithm

84

at the time did not exclude patients with positive BXM. Thirty patients had a positive

BXM with no DSA, and only one patient with DSA was not B cell crossmatched. As

expected, patients in the DSA and non-DSA groups were more commonly female (p =

0.003), were more likely to have received prior transfusion (p = 0.011) and more likely

to have undergone prior kidney transplant (p < 0.001).

Table 1. Demographic and clinical features at time of transplant according to entry

antibody status. Results shown as mean ± SD, median (IQR) or proportion.

Pre-transplant DSA is associated with AMR and non-AMR BPAR

Because the non-DSA group had similar rejection events (Table 1) to the No-antibody

group, these groups were combined as the DSA-Neg group (n=221) in subsequent analysis.

Compared with the DSA-Neg group, patients with DSA had a significantly increased risk of

non-AMR (Figure 1a) (HR 1.9, P=0.033) and especially AMR (Figure 1b). 30% of the DSA

group experienced AMR compared with only 6% of the DSA-Neg group (HR 6.6, P<0.001).

AMR showed a bi-modal distribution: AMR occurred early with a median time of around 1

month (39 days) in the DSA group compared to a median time of 49 months in the DSA-

Neg group (Figure 1b). In the 26 DSA patients who did not have AMR, 22 (85%) received at

least 1 ‘for cause’ biopsy but histological features consistent with AMR were not identified.

The risk of AMR varied according to the specificity of HLA antibody at the time of

transplantation (Fig. 1c). Compared with the DSA-Neg group, patients with only class I

85

or only class II DSA had a 4–6 fold increased risk of AMR and those with both class I +

II DSA had the highest risk (HR 10.1).

Figure 1a. Time to first biopsy-proven acute rejection (BPAR) excluding 24

patients with AMR by entry DSA (HR 1.9, 95%CI 1.1–1.3, p = 0.033).

86

Figure 1b. Time to Antibody Mediated Rejection by entry DSA status (HR 6.6,

95%CI 2.9– 14.7, p < 0.001).

Figure 1c. Time to Antibody Mediated rejection by class of DSA at entry (Class I

HR 4.4, 95%CI 1.3–15.5, p = 0.021; Class II HR 6.1, 95%CI 1.7–21.4, p = 0.005;

Class I + II HR 10.1, 95%CI 3.5–28.2, p < 0.001).

b

a

87

Strength of pre-transplant DSA is associated with AMR and non-AMR

In order to determine whether the strength of DSA could be correlated with rejection

we compared the MFI of the highest DSA bead amongst the various rejection

categories (Fig. 2a). There was no significant difference between the strength of HLA-

antibody defined as the highest MFI, or in the breadth of sensitisation defined as

cPRA, between the DSA group and non-DSA groups. However, in the patients with

DSA, the MFI of the highest DSA (HB-MFI) bead was significantly higher in those

with both class I + II DSA (Mean MFI 8070) than those with class I alone (3549) or class

II DSA alone (2540) (p = 0.004) In most cases this was due to a high class II antibody

(data not shown).

Amongst the 37 patients with DSA, the HB-MFI in the AMR group was significantly

higher than the HB-MFI in patients with no rejection (Fig. 2a) (p = 0.006). Additionally,

analysis of time to AMR by HB-MFI showed that the risk of AMR was significantly

greater when the HB-MFI was >8000 compared to those with an HB-MFI between 500

and 8000 (HR 23 95%, p < 0.001) (Fig. 2b). This same effect was observed when the

sum of the DSA beads was analysed. In our laboratory, a positive CDC crossmatch is

anticipated with an MFI of >8000 (data not shown).

Figure 2a. Strength of antibody determined by the MFI of the highest DSA bead

in the 37 patients with DSA according to rejection category (No BPAR, non-

AMR and AMR) (p = 0.006 AMR versus No BPAR).

88

Figure 2b. Time to AMR by category of the MFI of the highest DSA bead

compared with No-Antibody (p < 0.0001, HR MFI 500–8000 3.0 (0.9–9.0) p =

0.06 and HR >8000 = 23 (8.9–59)).

Pre-transplant DSA is associated with graft loss and recipient death

There was a trend for death-censored graft loss to be associated with DSA at the

time of transplant: 16% (6/37) of the DSA group lost their graft compared to 9%

(21/221) of the DSA-Neg group (p = 0.075). There was a significant increase in the

composite outcome of the combined patient and graft loss in the DSA group

compared to the DSA-Neg group (Fig. 3a). All 6 graft losses in the DSA group were

directly attributed to rejection episodes whilst in the DSA-Neg group 18/21 graft

losses were attributed to chronic allograft nephropathy, acute or chronic rejection

and 3/21 were due to recurrent disease (n = 2) or BK virus nephropathy (n = 1).

We next examined the effect of the HLA class of the DSA upon patient and graft

loss. In the DSA group, patients with both class I + II antibodies had a

statistically significant increase in both cumulative death-censored graft loss

(HR 4.9, p = 0.022) and combined patient or graft loss (HR 3.7, p = 0.003) (Fig.

3b). Actual graft survival in patients with class I DSA only, class II only, class I + II

DSA and DSA-Neg was 92% (12/13), 62% (7/11), 54% (7/13) and 83% (184/221)

respectively.

89

Figure 3a. Time to Patient or Graft Loss by entry DSA (DSA versus DSA negative

HR 2.1, 95%CI 1.1–4.0, p = 0.030).

Figure 3b. Time to patient or graft loss by entry class of DSA. There was a

significantly higher incidence of patient or graft loss in those with class II DSA

(HR 2.9, 95%CI 1.0–8.0, p = 0.047) and those with class I + II DSA (HR 3.7,

95%CI 1.6–8.9, p = 0.003) but not class I DSA.

90

Pre-transplant DSA is associated with chronic graft dysfunction

A longitudinal model incorporating established clinical predictors and additional

exploratory predictors having a p < 0.1 in univariate analyses, including DSA, was used

to determine independent longitudinal predictors of eGFR. There was a significant

continuous decrement in eGFR over time within the DSA group when compared with

the DSA-Neg group which was apparent as early as 1 year post transplant (Fig. 4). After

6 years of follow up the estimated mean eGFR of the DSA group was 35.7 ± 20.5

mls/min (n = 8) compared to 48.5 ± 22.7 (n = 104) for those in the DSA-Neg group.

Factors shown to be significantly and independently associated with loss of eGFR over

the time of follow-up were increased donor age, female gender, DGF, AMR, DSA at the

time of transplant and CNI use (Cyclosporine vs Tacrolimus) but not non-AMR (Table

2).

Table 2. Longitudinal mixed-effects models of predictors of eGFR (coefficient and

95% CI = mls/min) across the observed follow-up (median 68 months) incorporating

factors identified as univariate predictors of eGFR. (Random Effects of subject ID and

time since transplant were not shown). IR test of mixed model versus linear

regression χ2 = 1074, p < 0.0001.

91

Figure 4. Longitudinal mixed model of mean observed eGFR by entry DSA status

over time p < 0.0001. Observed mean eGFR at 72 months is 48.5 ± 22.7 (n =

108) in the no-DSA (negative) group versus 35.7 ± 20.4 (n = 8) in the DSA

(positive) group.

Discussion

This study confirms that the presence of pre-transplant anti-HLA antibody directed

against donor antigens detected by Luminex Single Antigen Bead assay with a negative

CDC T-cell crossmatch impacts on subsequent graft outcome. Such antibodies are

significantly associated with an increased risk of rejection, as previously reported by

others [9,12–15] and contrary to the findings of the Collaborative Transplant Study

group [18]. However results presented here further these findings and suggest that

DSA is also associated with poorer long term graft function as shown by declining

eGFR, graft loss and patient death. Importantly, the effect of these DSA on loss of

eGFR is independent of their effects on AMR. Moreover, the risk is graded and appears

greatest in those with both class I + II DSA who also have the highest antibody levels,

and intermediate in those with either class I only or class II only DSA. In addition, we

demonstrate that irrespective of class I or class II specificity, the MFI of the highest

DSA bead (> 8000 MFI) was predictive of AMR consistent with the concept that

antibody burden as quantified by SAB is associated with risk.

14% of patients had DSA detected in their pre-transplant serum and AMR occurred in

11/37 (30%) of the DSA group, compared with 6% of patients in the DSA-Neg group

92

(HR = 6.6). Only patients with DSA had early AMR, and in those without DSA at entry

AMR occurred a median of 4 years after transplant and was associated with the

development of de novo DSA (data not shown). BPAR in the absence of AMR (i.e. T cell

mediated rejection) was also significantly associated with the presence of DSA,

indicating that both cellular and antibody mediated rejection are associated with DSA,

and that DSA prospectively identifies patients at the highest risk of all types of

rejection.

Although both class I and class II DSA were associated with an increase in rejection,

the presence of both class I + II DSA together appeared to confer the greatest risk (HR

= 10). This finding is contrary to many of the previously published studies in which

class I DSA shows the greatest effect [19–21,23], but supports other studies [14,21,22]

showing pre-existing class II DSA to be associated with rejection and graft loss. The

exclusion of strong class I DSA by pre-transplant screening and the exclusion of those

with a positive prospective T cell CDC crossmatch may in part explain the lack of a

significant class I effect in our study.

We have examined the relevance of the absolute values of the MFI of HLA antibody.

Patients with DSA had similar average MFI when compared with patients with non-

DSA antibody. They also had similar antecedents of sensitisation (more females, higher

prior transfusion or re-transplants) yet the outcomes of patients with non-DSA were

similar to those without any antibody suggesting that pathogenicity of antibody

requires donor specificity. We have demonstrated that higher antibody strength

measured as MFI increases the risk of AMR. Patients with both class I + II DSA had

higher MFI and the highest risk of AMR. When considering the MFI of the highest DSA

bead, an MFI > 8000 (a value which in our laboratory is associated with a positive CDC

crossmatch) conferred a > 20 fold increased risk of AMR. These data suggest that the

risk of AMR is better associated with the absolute strength of DSA rather than HLA

class, and suggests a threshold of antibody burden that triggers an acute humoral

pattern of injury.

In our study we examined the effects of DSA and AMR on long- term graft function.

We have shown a significant and continuous decrease in graft function as measured

by eGFR in the DSA group. This loss is evident by 1 year post-transplant, and after 6

years of follow-up equates to a difference of 13 mls/min eGFR between the groups.

93

Furthermore, in the mixed-effect model we have shown that both DSA (-9mls/min)

and AMR (-3mls/min) are independently associated with loss of eGFR suggesting that

DSA have both acute and chronic humoral mechanisms of graft injury which translate

into loss of kidney function, reduced eGFR and premature graft loss. These incident

data are consistent with other prevalent studies [24–26] that identify chronic antibody

injury with graft impairment, proteinuria and early graft failure.

Most importantly DSA was associated with not only AMR but also increased graft loss

and patient death (HR = 2.1), and this was most marked in those with class II or class I

+ II DSA. Many patients died with a functioning graft suggesting that DSA is associated

with all cause mortality. Immunosuppressive therapy given to patients with DSA may

be associated with an increased risk of infection, vascular disease and/or malignancy

and therefore results presented here suggest that avoidance of pre-transplant DSA

may not only avoid acute AMR episodes but may actually improve patient morbidity

and mortality.

Detection of DSA in immediate pre-transplant serum in the absence of a positive T cell

CDC crossmatch is a significant risk factor for subsequent poorer graft outcome in

terms of both acute rejection episodes, long-term renal function and death. This risk

may be better predicted by the strength (MFI) of DSA and in our laboratory we now

consider an MFI >8000 regardless of crossmatch, to be a contraindication to

transplant. Alternative strategies including de-sensitisation, kidney-exchange

programmes and/or modification to Immunosuppressive regimens might be preferred

in patients with pre-existing high-level DSA in order to prolong the graft life and

reduce risk of rejection. Further studies to determine the utility of quantification of

the burden of antibody as measured by MFI in prediction of clinical risk are warranted.

94

Acknowledgements

The authors wish to acknowledge the clinicians and transplant nurses at Royal Perth

Hospital, Sir Charles Gairdner Hospital and Fremantle Hospital in Western Australia

involved in the collection of this data. We wish to thank the Department of Clinical

Immunology, Royal Perth Hospital for compatibility testing of the patients in this

study.

We also wish to thank Michael Phillips for statistical advice and help with statistical

analysis and A/Prof Lloyd D'Orsogna for critical reading of the manuscript.

Disclosure

This is an original work and the manuscript or parts of it have not been submitted

elsewhere for publication. All authors have read and approved submission of the

manuscript, and that material in the manuscript has not been published and is not

being considered for publication elsewhere in whole or part in any language except as

an abstract.

95

References

[1] Ekberg H, Tedesco-Silva H, Demirbas A, Vitko S, Nashan B, Gurkan A, et al. Reduced

exposure to calcineurin inhibitors in renal transplantation. N Engl J Med 2007;357:

2562–75.

[2] Lamb KE, Lodhi S, Meier-Kriesche HU. Long-term renal allograft survival in the United

States: a critical reappraisal. Am J Transplant 2011; 11: 450–62.

[3] Terasaki P, Mizutani K. Antibody mediated rejection: update 2006. Clin J Am Soc

Nephrol 2006; 1: 400–3.

[4] Langan LL, D'Orsogna L, Park LP, Hughes TL, Irish A, Luxton G, et al. HLA antibodies and

soluble CD30 are associated with poor renal graft outcome: updated results of a single-

center cross-sectional study. Clin Transpl 2006: 219–25.

[5] Kaposztas Z, Podder H, Mauiyyedi S, Illoh O, Kerman R, Reyes M, et al. Impact of

rituximab therapy for treatment of acute humoral rejection. Clin Transplant 2009; 23:

63–73.

[6] Einecke G, Sis B, Reeve J, Mengel M, Campbell PM, Hidalgo LG, et al. Antibody-mediated

microcirculation injury is the major cause of late kidney transplant failure. Am J

Transplant 2009; 9: 2520–31.

[7] Terasaki PI, Ozawa M. Predicting kidney graft failure by HLA antibodies: a

prospective trial. Am J Transplant 2004; 4: 438–43.

[8] Cooper JE, Gralla J, Cagle L, Goldberg R, Chan L, Wiseman AC. Inferior kidney allograft

outcomes in patients with de novo donor-specific antibodies are due to acute rejection

episodes. Transplantation 2011; 91: 1103–9.

[9] Gebel HM, Bray RA, Nickerson P. Pre-transplant assessment of donor-reactive, HLA-

specific antibodies in renal transplantation: contraindication vs. risk. Am J Transplant

2003; 3: 1488–500.

[10] Bachler K, Amico P, Honger G, Bielmann D, Hopfer H, Mihatsch MJ, et al. Efficacy of

induction therapy with ATG and intravenous immunoglobulins in patients with low-

level donor-specific HLA-antibodies. Am J Transplant 2010; 10: 1254–62.

http://refhub.elsevier.com/S0966-3274(13)00034-8/rf0005



































96

[11] Bentall A, Cornell LD, Gloor JM, Park WD, Gandhi MJ, Winters JL, et al. Five-year

outcomes in living donor kidney transplants with a positive crossmatch. Am J

Transplant 2013; 13: 76–85.

[12] Eng HS, Bennett G, Tsiopelas E, Lake M, Humphreys I, Chang SH, et al. Anti-HLA

donor-specific antibodies detected in positive B-cell crossmatches by Luminex

predict late graft loss. Am J Transplant 2008; 8: 2335–42.

[13] Honger G, Hopfer H, Arnold ML, Spriewald BM, Schaub S, Amico P. Pretransplant IgG

subclasses of donor-specific human leukocyte antigen antibodies and development of

antibody-mediated rejection. Transplantation 2011; 92: 41–7.

[14] Otten HG, Verhaar MC, Borst HP, Hene RJ, van Zuilen AD. Pretransplant donor-specific

HLA class-I and -II antibodies are associated with an increased risk for kidney graft failure.

Am J Transplant 2012; 12: 1618–23.

[15] Lefaucheur C, Suberbielle-Boissel C, Hill GS, Nochy D, Andrade J, Antoine C, et al. Clinical

relevance of preformed HLA donor-specific antibodies in kidney transplantation. Am J

Transplant 2008; 8: 324–31.

[16] Sayer D, Whidborne R, Brestovac B, Trimboli F, Witt C, Christiansen F. HLA-DRB1 DNA

sequencing based typing: an approach suitable for high throughput typing including

unrelated bone marrow registry donors. Tissue Antigens 2001; 57: 46–54.

[17] Racusen LC, Halloran PF, Solez K. Banff 2003 meeting report: new diagnostic insights

and standards. Am J Transplant 2004; 4: 1562–6.

[18] Susal C, Ovens J, Mahmoud K, Dohler B, Scherer S, Ruhenstroth A, et al. No association

of kidney graft loss with human leukocyte antigen antibodies detected exclusively by

sensitive Luminex single-antigen testing: a Collaborative Transplant Study report.

Transplantation 2011; 91: 883–7.

[19] Rodriguez LM, Paris SC, Arbelaez M, Cotes JM, Susal C, Torres Y, et al. Kidney graft

recipients with pretransplantation HLA CLASS I antibodies and high soluble CD30 are at

high risk for graft loss. Hum Immunol 2007; 68: 652–60.

[20] Susal C, Opelz G. Kidney graft failure and presensitization against HLA class I and class II

antigens. Transplantation 2002; 73: 1269–73.





































97

[21] Susal C, Opelz G. Good kidney transplant outcome in recipients with presensitization

against HLA class II but not HLA class I. Hum Immunol 2004; 65: 810–6.

[22] Song EY, Lee YJ, Hyun J, Kim YS, Ahn C, Ha J, et al. Clinical relevance of pretransplant

HLA class II donor-specific antibodies in renal transplantation patients with negative T-

cell cytotoxicity crossmatches. Ann Lab Med 2012; 32: 139–44.

[23] Lefaucheur C, Loupy A, Hill GS, Andrade J, Nochy D, Antoine C, et al. Preexisting donor-

specific HLA antibodies predict outcome in kidney transplantation. J Am Soc Nephrol

2010; 21: 1398–406.

[24] Hohage H, Kleyer U, Bruckner D, August C, Zidek W, Spieker C. Influence of proteinuria on

long-term transplant survival in kidney transplant recipients. Nephron 1997; 75: 160–5.

[25] Amer H, Fidler ME, Myslak M, Morales P, Kremers WK, Larson TS, et al. Proteinuria after

kidney transplantation, relationship to allograft histology and survival. Am J Transplant

2007; 7: 2748–56.

[26] Cherukuri A, Welberry-Smith MP, Tattersall JE, Ahmad N, Newstead CG, Lewington AJ,

et al. The clinical significance of early proteinuria after renal transplantation.






























98

Chapter 4

Peri-operative third party red blood cell

transfusion in renal transplantation and

the risk of antibody-mediated rejection

and graft loss

99

Peri-operative third party red blood cell transfusion in renal transplantation and

the risk of antibody-mediated rejection and graft loss

Samantha Fidler, Ramyasuda Swaminathan, Wai Lim, Paolo Ferrari, Campbell Witt, Frank

T Christiansen, Lloyd J D’Orsogna, Ashley B Irish

In our previous publication we were able to demonstrate the association of pre-

transplant donor specific antibodies (DSA) detected by Luminex single antigen bead assay

with antibody mediated rejection (AMR) and progressive graft dysfunction. There was a

trend towards graft loss although this did not reach significance. However, a number of

patients with pre-transplant DSA did not develop AMR and we sought to elucidate

reasons for this. A coincidental finding in our preliminary study showed an association of

blood transfusions with AMR. We investigated in this paper the relationship of pre- and

post-transplant blood transfusions in patients with or without DSA.


Transplant Immunology 2013; 29: 22-27

Contributions:

SJF Performed all antibody assays, participated in study design,

participated in data collection, participated in data analysis, prepared

the manuscript

ABI Participated in study design, participated in data collection,

participated in data analysis, participated in writing of the manuscript

RS Participated in data collection and analysis

WL, PF Participated in data collection, participated in writing of the

manuscript

CSW, FTC Participated in study design, participated in writing of the manuscript

LD’O Participated in writing of the manuscript

100

Peri-operative third party red blood cell transfusion in renal

transplantation and the risk of antibody-mediated rejection and graft

loss

Samantha Fidler a,b,1, Ramyasuda Swaminathan c,d,2, Wai Lim e,3, Paolo Ferrari f,4,

Campbell Witt a,b,5, Frank T. Christiansen a,b,6, Lloyd J. D'Orsogna a,b,7, Ashley B. Irish c,d,⁎,8

a Department of Clinical Immunology, PathWest, Royal Perth Hospital, Western Australia, Australia b School of Pathology and Laboratory Medicine, University of Western Australia, Australia c Medical Renal Transplant Unit, Royal Perth Hospital, Western Australia, Australia d School of Medicine and Pharmacology, University of Western Australia, Australia e Department of Nephrology Sir Charles Gairdner Hospital, Perth, Western Australia, Australia f Department of Nephrology Fremantle Hospital, Fremantle, Western Australia, Australia


Article history: Received 12 August 2013

Received in revised form 23 September 2013

Accepted 23 September 2013

Keywords: Donor-specific antibody (DSA) Peri-operative transfusion Kidney transplantation Antibody mediated rejection

Abbreviations: AMR, antibody-mediated rejection; RBCT, red blood cell transfusion; cPRA, calculated panel reactive antibody; DSA, donor

specific antibodies; DGF, delayed graft function; BPAR, biopsy proven acute rejection; MFI, mean fluorescence intensity; SAB,

single antigen bead.

* Corresponding author at: Medical Renal Transplant Unit, Royal Perth Hospital, GPO Box X2213, Perth, Western Australia 6847, Australia.

Tel.: +61 8 9224 2162; fax: +61 89224 2160. E-mail address: [email protected] (A.B. Irish).

1 Participated in the performance of research laboratory methods, research design and writing the paper. 2 Participated with data collection and analysis. 3 Participated with data collection and writing the paper. 4 Participated with data collection and writing the paper. 5 Participated in research design, data analysis and writing the paper. 6 Participated in research design and writing the paper. 7 Participated in data analysis and writing the paper. 8 Participated in research design, data collection and analysis and writing the paper.


101

1. Introduction

The presence of pre-transplant anti-HLA antibody directed against the donor antigens

(DSA) in the presence of a negative CDC crossmatch is associated with increased risk of

antibody mediated rejection (AMR) and graft failure [1–3]. HLA antibodies are formed

as a consequence of prior transplantation, pregnancy and blood transfusion due to

exposure to foreign HLA antigens [4–9]. However blood transfusion prior to transplant

is immunomodulatory and appears to reduce the risk of acute allograft rejection and

graft loss despite an increased risk of sensitisation [10–12]. Historically it had been

observed that large volumes of third-party red blood cell transfusion (RBCT) (up to 20

units) over a prolonged period are required to induce enduring antibodies, especially in

males or nulliparous females [4,13–15]. However in the presence of another immune

stimulating process such as pregnancy or transplantation, co-administration of third

party RBCT results in broad HLA antibody production which is more potent and

enduring [6,16,17]. In the current transplant era, transfusion in patients with end stage

kidney disease is less frequent due to the widespread use of epoetins. However during

acute illness or surgery patients may still be exposed to blood products, although

specifically transfusing patients for immunological benefit is no longer routine [18–20].

Leucodepletion of blood products has also been shown not to prevent the risk of

allosensitisation associated with RBCT [14,21–23]. The majority of studies on the role of

blood transfusion were performed in the period before the use of sensitive and specific

solid phase antibody detection assays were available and cell-dependent cytotoxicity

assays were utilised. Although it is established that DSA detected at the time of

transplant is associated with an increased risk of AMR why some patients with DSA

develop AMR and others do not is unclear and may relate to variability in the antibody

subtype, complement binding ability, or the amount or breadth of antibody [1,24–26].

Transfusion in the peri-operative and early post- transplant period depends on

individualised patient management factors and is commonly thought not to be an

immunological stimulus because it is assumed that the concomitant use of

immunosuppression mitigates this risk. We hypothesised that post-transplant

transfusion in patients with preformed HLA antibody may provide additional

allostimulation or immunological recall and increase the risk of AMR. We therefore

investigated the relationship of pre-transplant and peri- operative transfusion in renal

102

transplant recipients with and without pre-transplant HLA antibody determined by

Luminex single antigen bead (SAB) assay.

2. Subject and methods

2.1. Patients

We studied 258 transplant recipients of which 246 patients received a kidney

transplant and 12 patients received a simultaneous pancreas– kidney transplant

between June 2003 and October 2007. Patients were transplanted at 3 tertiary centres

and peri-operative care and decision for transfusions was individualised, clinically

indicated and not mandated by protocol. No donor-specific transfusions occurred.

Leucocyte depleted packed red cells were used. All patients received a calcineurin

inhibitor (CNI) (tacrolimus or cyclosporine) at the time of transplantation in combination

with mycophenolate mofetil or mycophenolate sodium and corticosteroids and the

Interleukin-2 receptor antibody basiliximab was commonly used for induction. The need

for biopsy, medication adjustments and transfusion was determined by the caring

clinical teams and was not protocol driven. Transfusion history was obtained from the

West Australian Red Cross Blood Bank, the Westmead Hospital Transfusion Laboratory,

patient medical records and direct patient interrogation. Patient follow-up was a

median of 67 months (IQR 54–77). Patients provided written consent for participation

in this study.

2.2. Laboratory methods

These are reported in detail elsewhere however stored donor DNA was typed by

sequence based typing at HLA-A, -B, -C, -DRB1, DQB1, DPB1 loci and DRB3, 4, 5 and

DQA1 where required [27]. All recipients were transplanted with a negative T cell CDC

crossmatch. B cell crossmatching was performed for 80% of the patients; however a

positive B cell crossmatch was not considered an absolute contraindication to

transplantation. Sera collected at the time of transplant were screened retrospectively

for anti-HLA class I and/or class II antibodies using the Luminex Mixed Screen assay

(OneLambda Inc.) and those with a positive screen were characterised for HLA class I

103

and/or class II antibodies specificity using single antigen beads (LABScreen Single

Antigen beads, OneLambda Inc.). Antibodies were considered to be positive if the

normalised mean fluorescence intensity (MFI) value for a particular bead was greater

than 500. HLA antibodies with an MFI > 500 directed against a donor HLA antigen were

considered to be DSA.

2.3. Clinical outcome parameters

Transfusion history was recorded as never transfused (No-RBCT), transfused at any

time prior to renal transplant but not after renal transplant surgery (Pre-RBCT), not

transfused prior to transplant but transfused at the time of, or within 30 days of

transplant surgery (Post-RBCT) and transfused both prior to and within 30 days of the

transplant (Pre + Post-RBCT). Delayed graft function (DGF) was defined as the need

for dialysis within the first 48 h of transplantation. Graft loss was defined as the return

to dialysis (i.e. death-censored) unless otherwise indicated. All rejection episodes were

proven by biopsy (BPAR) and the first BPAR was used to construct time to event

analysis and where multiple rejections occurred, the highest reported grade was

recorded. Time to AMR was recorded as a separate event to allow analysis by rejection

type (AMR vs Non-AMR). Treatment of rejection was at the treating clinician's

discretion and was not mandated by protocol. Histological reporting of renal biopsies

was undertaken by the local histopathologists as part of routine clinical care and was

initially made without information as to the presence or absence of DSA (due to

varying laboratory testing and reporting changes over the period of study). The biopsy

findings were graded according to the Banff classification 2003. AMR was defined as

C4d positivity in PTC alone or in conjunction with transplant glomerulitis and/or peri-

tubular capillaritis and/or arteritis, and also in the absence of C4d when transplant

glomerulitis and peritubular capillaritis were detected.

104

2.4. Statistical analysis

Statistical analyses were performed by using SPSSv18 (SPSS Inc., Chicago IL, USA).

For categorical data Fisher's exact test or Pearson's chi-square tests were used.

Parametric data were compared by ANOVA or t-test, and for non parametric data

Mann–Whitney U test or Kruskal–Wallis one-way ANOVA was used. Comparisons of

within group differences by z-test were made with Bonferroni adjustment reported at

the p < 0.05 level. Time to event of interest (AMR, graft and patient survival) was

estimated by the method of Kaplan–Meier and Cox proportional hazard regression

analysis with the predictor satisfying the proportional hazard assumption. Covariates

examined were HLA-antibody at entry, rejection, gender, re-transplantation, and

delayed graft function. Results were expressed as hazard ratios (HR) with 95% CI.

3. Results

Sixty-five patients had pre-transplant HLA-antibody: DSA group n = 37 (14%) and

Non-DSA group n = 28 (11%) while the remaining 193 (75%) patients had no HLA

antibody defined using the MFI cut-off of <500 or with a negative antibody screen.

Baseline clinical and demographic data of these groups is reported in detail

elsewhere and summarised in Table 1. [27] As expected, patients with any HLA

antibody were more commonly female (41/65 vs 53/193, p = 0.003) and more

likely to have undergone prior kidney transplant (20/65 vs 7/193, p < 0.001) and to

have received Pre-RBCT (39/65 vs 70/193, p = 0.011). There was no difference in

haemoglobin between the groups either at time of transplant (DSA 124 ± 19, Non-

DSA 124 ± 18, No-Antibody 124 ± 15 g/L, p = 0.99) or at 30 days post transplant

(DSA 109 ± 17, Non-DSA 113 ± 13, No-Antibody 114 ± 17 g/L, p = 0.19). Patients

with pre-transplant DSA were significantly more likely to have been transfused

within the first 30 peri-operative days (DSA 70%) than those with Non-DSA (43%)

or no HLA antibody (38%, p < 0.001) although the amount of RBCT was not

different [DSA 4 (2–4), Non-DSA 2 (2–4) and No- Antibody 2 (2–4) units median

and IQR, p = 0.17] and > 90% of all post-transplant RBCT given within the first 2

peri-operative days.

105

Table 1. Demographic and clinical features at time of transplant by entry HLA-

antibody.

In order to explore further the relationship between transfusion and pre-

transplant DSA we divided the patients into four groups according to their

transfusion status — No- RBCT, Pre-RBCT, Post-RBCT and Pre + Post-RBCT groups

as previously defined (Table 2). Overall 109/258 (42%) received Pre-RBCT and

111/258 (43%) of patients received Post- RBCT. The prevalence of HLA antibody

amongst these groups varied significantly as expected. The No-RBCT group were

much more likely to have no HLA antibody (86%) than the other groups (p < 0.05).

Conversely however, the Pre + Post-RBCT group were more likely to have DSA (p <

0.05), receive a repeat transplant and less likely to receive a pre-emptive or living

donor transplant, although time on dialysis was similar to those with Pre- and

Post-RBCT. Patients with Pre-RBCT only were significantly less likely to have Non-

AMR rejection than all other groups (p < 0.05, Table 3). Patients in the Pre +

Post-RBCT group were more likely to have DGF than all other groups, and had a 4

fold increased risk of AMR (Table 3 and Fig. 1, p = 0.004) with a median time to

AMR of 2 months.

106

Table 2. Patient demographics, HLA antibody, rejection and outcomes by transfusion

status.

Table 3. Univariate and multivariate adjusted Cox Proportional Hazards models for

the effect of Delayed Graft Function (DGF), entry HLA antibody, Antibody and Non-

Antibody Mediated Rejection (AMR and Non-AMR) and Transfusion status on death

censored (n=27) and combined patient and graft loss (n=51).

107

Figure 1. Kaplan–Meier plot of time to AMR by transfusion status in all patients.

Patients with Pre + Post-RBCT had HR 4.1 (95%CI 1.6–10.8 p = 0.004) for AMR

compared with No-RBCT. Note Y-axis is truncated.

Given the association of Pre + Post RBCT with AMR we next examined the

importance of pre-transplant DSA, a known risk for AMR, in the various transfusion

groups. When stratified by the presence of DSA there was no difference in the risk

of AMR between those with or without DSA in either of the No-RBCT, Pre-RBCT or

Post-RBCT groups individually (data not shown) or when these 3 groups were

combined (Fig. 2a). In the Pre + Post RBCT group the risk of AMR was 13.9 times

greater in those patients with DSA than in patients with Non-DSA or No-Antibody

(Fig. 2b p = 0.001). Indeed all 13 episodes of AMR in the DSA group occurred

exclusively in patients who had received Pre + Post-RBCT. On the other hand, 0/6

patients with Non-DSA and 2/37 in the No- antibody group who had received Pre +

Post-RBCT developed AMR. The median time between post-operative transfusion

and AMR in the DSA patients was 25 days (IQR 5–761 days).

108

Figure 2a. Time to antibody mediated rejection (AMR) by antibody status at the

time of transplant in the combined 195 patients receiving No-RBCT, Pre-RBCT and

Post-RBCT (p = 0.56)

Figure 2b. Time to antibody mediated rejection (AMR) by antibody status at the

time of transplant in the 63 patients with Pre + Post-RBCT (p = 0.001, HR 13.9 DSA

vs No-antibody) Note different Y axis origin.

109

Univariate predictors of AMR were pre-transplant DSA (HR 6.6 95%CI 2.9–

14.7, p < 0.001), DGF (HR 2.6. 1.1–6.1, p = 0.039), re-transplant (HR 3.3 1.3–8.3 p =

0.024) and Pre + Post-RBCT (HR 4.1, 1.6–10.8 p = 0.005) but not females (HR 0.9

0.38–2.1). There was a significant interaction between Pre-RBCT and Post-RBCT

(HR 4.4 2.0–9.8 p = 0.001) and between DSA and Post-RBCT (HR 10.6 4.7–23.8 p <

0.0001) on the risk of AMR. In a multivariate Cox model incorporating the above

univariate factors and adding interaction terms, only the interaction between DSA

and Post-RBCT (HR 7.2, 2.9–18.0, p = 0.001) remained a significant predictor of

AMR.

Death-censored graft loss (Fig. 3 and Table 3) and combined patient and graft loss

(Table 3) was significantly increased in the Pre + Post-RBCT group (HR 7.1 p <

0.001) compared with all other transfusion groups. This difference persisted even

after exclusion of the 37 patients with preformed DSA (HR 4.9 95% CI 1.5–15.8 p =

0.007 and 3.9 1.7–7.8 p = 0.001 respectively). Neither gender nor retransplant

were associated with graft or patient loss. We used significant univariate

predictors of graft loss including DGF, HLA antibody (DSA and Non-DSA), AMR and

Non-AMR rejection and transfusion history in a multivariate Cox Proportional

Hazards model (Table 3). AMR, Non-AMR rejection and Pre + Post RBCT were

independently associated with death censored and all cause graft loss. However

DGF and DSA were no longer predictive by multivariate analysis.

110

Figure 3. Time to death censored graft loss by transfusion status HR 7.1 (2.4–21.3) p

< 0.001 for Pre + Post-RBCT compared with No-RBCT.

4. Discussion

We show that the risk of AMR associated with the presence of DSA at the time of

transplant is modulated by exposure to RBCT. In our cohort, AMR was predominantly

observed only in sensitised patients with DSA all of whom had received RBCT prior to

transplant and were then transfused within the first 30 days (usually 48 h). Pre-

transplant DSA and Pre + Post-RBCT were each independent predictors of AMR.

However, there was a strong interaction between pre-transplant DSA and Post-RBCT,

which eliminated all other predictors (DGF, re-transplant, gender), and conferred a 7.2

fold increase in the risk of AMR. In contrast, patients with DSA who received only Pre-

RBCT or Post-RBCT, or who received no transfusion did not experience AMR.

Overall, patients with DSA had the highest rate of post-operative transfusion. In our

centres, the decision to give peri-operative transfusion was solely determined by clinical

need, authorised by different clinical teams and, in all cases, made without knowledge

of the patient's anti-HLA antibody status. Our reported post-operative transfusion rate

of 43% is very similar to those reported elsewhere [28,29]. The reason for the higher

rate of peri-operative transfusion in patients with DSA compared with Non-DSA is

therefore uncertain. It may be due to the greater medical and surgical complexities of

111

this patient group and their greater waiting time on dialysis for example. However, it is

notable that there was no difference in gender, re-transplantation, deceased donors or

Pre-RBCT between the DSA and Non-DSA groups at time of surgery, or between the

haemoglobin at surgery or at 1 month post- surgery. Although residual confounding by

indication remains possible, it is not possible to either entirely adjust or explain and

this difference requires further testing.

The immunological interaction of blood transfusion and transplantation is complex. Pre-

RBCT is associated with better graft outcomes and less acute rejection, and this is

suggested to be due to immunomodulation with down-regulation of an immune response

and the induction of regulatory T-cells [11,30]. Indeed our study confirms the continued

benefit of Pre-RBCT alone with this group having the lowest rate of Non-AMR. We also

confirm that Pre-RBCT is still associated with an increased risk of HLA-antibody

sensitisation. Several recent reports [28,29] raise some concern that post-operative

transfusion is associated with poor graft outcome. However these studies did not consider

sensitisation or prior transfusion as potential modifiers and these factors may account for

their conflicting conclusions. Here we report that peri-operative blood transfusion is

associated with an increased risk of AMR, but only in recipients with pre-transplant DSA

detected using solid phase assays, all of whom had been previously exposed to RBCT and

other sensitising events. This effect of peri-operative transfusion was not found in

recipients without DSA, suggesting that the combination of DSA and peri-operative blood

transfusion may be particularly detrimental to the transplanted graft. Importantly, adverse

events after peri-operative blood transfusion included not only antibody mediated

rejection, but also poorer long term graft outcome and recipient death, independent of

the risk of AMR and Non-AMR, consistent with the findings of O'Brien et al. [28].

In light of our findings it is worth considering the immunological mechanisms whereby

blood transfusion could increase the pathogenicity of pre-existing DSA. This might be

through direct quantitative or qualitative alterations in antibody or indirectly via specific

transfusion factors. Scornik et al. [16] have previously identified that re-exposure to

blood in those with prior sensitising events such as transplant or pregnancy elicits a

broad antibody response; findings that are consistent with our study. Interestingly,

broad recall of an antibody response to previously exposed transplant antigens, rather

than to the transfusion antigens, was induced by third party transfusions. [31] Qian et al.

112

[32] have shown in a cardiac rodent model that pre-sensitisation with allogeneic RBCT

causes accelerated graft rejection in the presence of complement and antibody binding

to graft endothelium. Complement activation products and quantitative changes in

cytokines may be present within stored blood products [33,34]. Norda and colleagues

found that stored plasma components may differ significantly in the amount and

timing of complement activation products, particularly C3a, which could specifically

trigger pathological changes if pre-existing effector DSA is present. Theoretically

cytokines and other factors present in blood products could also induce non-

complement activating DSA to class switch to IgG1 and/or IgG3 complement activating

antibodies. Mice models of transfusion related acute lung injury suggest that MHC

class I specific antibody binding to nonhematopoietic cells drives complement activation

and production of reactive oxygen species [35]. T cell allorecognition of allogeneic HLA

molecules, present even in leucodepleted blood products, may be associated with

specific and non-specific immune activation including increased cytokine production and

cytotoxicity function. Whether exposure to RBCT in sensitised patients stimulates an

increase in the absolute amount or breadth of DSA and/or class switching and

complement binding was beyond the scope of this study. Serially monitoring these

changes would be informative however the frequency of sampling and interventions for

management of rejection which alter antibody measurement confound interpretation.

These putative adverse effects of peri-operative blood transfusion could be investigated

further using in-vivo animal models.

This data suggests that avoidance of peri-operative blood transfusion or, given the

impossibility of eliminating all transfusion risk, establishing a new paradigm for RBCT in

sensitised transplant recipients should be considered. It is established that

leucodepleted unmatched RBCT does not reduce sensitisation [23] and therefore

selecting HLA matched blood with its significantly reduced risk of HLA specific antibody

production may be particularly suited to patients with DSA [36]. The use of HLA

matched blood transfusion products for renal transplantation patients is feasible and

may be effective within a clinical setting [36,37]. Furthermore recipients with high levels

of sensitization may even benefit from pre-operative storage of autologous blood

products for use in the event of requiring a peri-operative blood transfusion [38]. The

beneficial effects of blood transfusion on graft survival have been reported to be

associated with HLA-DR matching, and while HLA typing of the blood donors was not

113

available for this study it is highly likely that the observed adverse effect is associated

with specific HLA mismatches between the blood donor and organ recipient [36].

In conclusion, we have shown that Pre-RBCT alone is still associated with a lower rate of

Non-AMR rejection and an increased risk of HLA antibody. However, peri-operative blood

transfusion in sensitised renal recipients with DSA and prior transfusion is associated

with AMR. Post- RBCT may therefore be an additional factor modifying the risk of AMR

in patients with HLA-antibody. We also confirm and expand upon the previous findings

that perioperative blood transfusion is associated with poorer graft and patient

survival, and show that this is most evident in those with previous exposure to RBCT,

independent of acute rejection episodes. These findings suggest that RBCT remains a

potent and complex modifiable immunomodulator of renal transplant outcomes and

additional studies to further define mechanisms for these effects are warranted.

Disclosure

This is an original work and the manuscript or parts of it have not been submitted

elsewhere for publication. All authors have read and approved submission of the

manuscript, and that material in the manuscript has not been published and is not

being considered for publication elsewhere in whole or part in any language except as

an abstract.

Acknowledgements

The authors wish to acknowledge the clinicians and transplant nurses at Royal Perth

Hospital, Sir Charles Gairdner Hospital and Fremantle Hospital in Western Australia

involved in the collection of this data. We wish to thank the Department of Clinical

Immunology, Royal Perth Hospital for compatibility testing of the patients in this

study.

114

References

[1] Lefaucheur C, Loupy A, Hill GS, Andrade J, Nochy D, Antoine C, et al. Preexisting

donor-specific HLA antibodies predict outcome in kidney transplantation. J Am Soc

Nephrol 2010; 21: 1398–406.

[2] Otten HG, Verhaar MC, Borst HP, Hene RJ, van Zuilen AD. Pretransplant donor-

specific HLA class-I and -II antibodies are associated with an increased risk for kidney

graft failure. Am J Transplant 2012; 12: 1618–23.

[3] Song EY, Lee YJ, Hyun J, Kim YS, Ahn C, Ha J, et al. Clinical relevance of pretransplant

HLA class II donor-specific antibodies in renal transplantation patients with negative T-cell

cytotoxicity crossmatches. Ann Lab Med 2012; 32: 139–44.

[4] Opelz G, Graver B, Mickey MR, Terasaki PI. Lymphocytotoxic antibody responses to

transfusions in potential kidney transplant recipients. Transplantation 1981; 32: 177–83.

[5] Sanfilippo F, Vaughn WK, Bollinger RR, Spees EK. The influence of pretransplant

transfusions, using different blood products, on patient sensitization and renal allograft

survival. Transplantation 1984; 37: 350–6.

[6] Scornik JC, Ireland JE, Howard RJ, Fennell III RS, Pfaff WW. Role of regular and

leukocyte-free blood transfusions in the generation of broad sensitization.


[7] Pfaff WW, Howard RJ, Scornik JC, Day C, Renderer J, Scott J, et al. Incidental and

purposeful random donor blood transfusion. Sensitization and transplantation.


[8] Chow MP, Hu HY, Lyou JY, Lin JS, Yung CH, Lee A, et al. Red cells, HLA and platelet

antibody formation in patients with multiple transfusions. Acta Haematol 1994; 92:

57–60.

[9] Novotny VM, van Doorn R, Witvliet MD, Claas FH, Brand A. Occurrence of allogeneic

HLA and non-HLA antibodies after transfusion of prestorage filtered platelets and red

blood cells: a prospective study. Blood 1995; 85: 1736–41.

[10] Blajchman MA. Immunomodulatory effects of allogeneic blood transfusions: clinical

manifestations and mechanisms. Vox Sang 1998; 74(Suppl. 2): 315–9.


































115

[11] Claas FH, Roelen DL, van Rood JJ, Brand A. Modulation of the alloimmune response by

blood transfusions. Transfus Clin Biol 2001; 8: 315–7.

[12] Opelz G, Graver B, Terasaki PI. Induction of high kidney graft survival rate by multiple

transfusion. Lancet 1981; 1: 1223–5.

[13] Scornik JC, Meier-Kriesche HU. Blood transfusions in organ transplant patients:

mechanisms of sensitization and implications for prevention. Am J Transplant 2011;

11: 1785–91.

[14] Karpinski M, Pochinco D, Dembinski I, Laidlaw W, Zacharias J, Nickerson P. Leuko-cyte

reduction of red blood cell transfusions does not decrease allosensitization rates in

potential kidney transplant candidates. J Am Soc Nephrol 2004; 15: 818–24.

[15] Scornik JC, Pfaff WW, Howard RJ, Fennell III RS, Ramos E, Peterson JC, et al. Increased

antibody responsiveness to blood transfusions in pediatric patients. Transplantation 1994;

58: 1361–5.

[16] Scornik JC, Ireland JE, Howard RJ, Pfaff WW. Assessment of the risk for broad

sensitization by blood transfusions. Transplantation 1984; 37: 249–53.

[17] Sanfilippo F, Vaughn WK, Bollinger RR, Spees EK. Comparative effects of pregnancy,

transfusion, and prior graft rejection on sensitization and renal transplant results.


[18] Fox KM, Yee J, Cong Z, Brooks JM, Petersen J, Lamerato L, et al. Transfusion burden in

non-dialysis chronic kidney disease patients with persistent anemia treated in routine

clinical practice: a retrospective observational study. BMC Nephrol 2012; 13: 5.

[19] Ibrahim HN, Skeans MA, Li Q, Ishani A, Snyder JJ. Blood transfusions in kidney

transplant candidates are common and associated with adverse outcomes. Clin Transpl

2011; 25: 653–9.

[20] Aalten J, Bemelman FJ, van den Berg-Loonen EM, Claas FH, Christiaans MH, de Fijter

JW, et al. Pre-kidney-transplant blood transfusions do not improve transplantation

outcome: a Dutch national study. Nephrology, dialysis, transplantation. Off Publ Eur

Dial Transplant Assoc Eur Ren Assoc 2009; 24: 2559–66.




































116

[21] Balasubramaniam GS, Morris M, Gupta A, Mesa IR, Thuraisingham R, Ashman N.

Allosensitization rate of male patients awaiting first kidney grafts after leukodepleted

blood transfusion. Transplantation 2012; 93: 418–22.

[22] Scornik JC. Leukoreduced blood and sensitization of organ transplant candidates.

Transfusion 2012; 52: 1838–9 [author reply 9].

[23] van de Watering L, Hermans J, Witvliet M, Versteegh M, Brand A. HLA and RBC

immunization after filtered and buffy coat-depleted blood transfusion in cardiac

surgery: a randomized controlled trial. Transfusion 2003; 43: 765–71.

[24] AlMahri A, Holgersson J, Alheim M. Detection of complement-fixing and non-fixing

antibodies specific for endothelial precursor cells and lymphocytes using flow

cytometry. Tissue Antigens 2012; 80: 404–15.

[25] Chen G, Sequeira F, Tyan DB. Novel C1q assay reveals a clinically relevant subset of

human

Leukocyte antigen antibodies independent of immunoglobulin G strength on single

antigen beads. Hum Immunol 2011; 72: 849–58.

[26] Mujtaba MA, Goggins W, Lobashevsky A, Sharfuddin AA, Yaqub MS, Mishler DP, et al.

The strength of donor-specific antibody is a more reliable predictor of antibody-

mediated rejection than flow cytometry crossmatch analysis in desensitized kidney

recipients. Clin Transplant 2011; 25: E96-102.

[27] Fidler SJ, Irish AB, Lim W, Ferrari P, Witt CS, Christiansen FT. Pre-transplant donor

specific anti-HLA antibody is associated with antibody-mediated rejection, progressive

graft dysfunction and patient death. Transpl Immunol Jun. 2013; 28(4): 148–53.

[28] O'Brien FJ, Lineen J, Kennedy CM, Phelan PJ, Kelly PO, Denton MD, et al. Effect of

perioperative blood transfusions on long term graft outcomes in renal transplant

patients. Clin Nephrol 2012; 77: 432–7.

[29] Scornik JC, Schold JD, Bucci M, Meier-Kriesche HU. Effects of blood transfusions

given after renal transplantation. Transplantation 2009; 87: 1381–6.

[30] Bushell A, Karim M, Kingsley CI, Wood KJ. Pretransplant blood transfusion without

additional immunotherapy generates CD25 + CD4+ regulatory T cells: a potential

explanation for the blood-transfusion effect. Transplantation 2003; 76: 449–55.

[31] Scornik JC, Ireland JE, Howard RJ, Pfaff WW, Fennell III RS. Sensitization by blood

transfusions in previously transplanted patients. Transplantation 1983; 35: 505–6.




































117

[32] Qian Z, Lee CY, Murata K, Liu J, Fox-Talbot K, Wasowska BA, et al. Antibody and

complement mediated injury in transplants following sensitization by allogeneic blood

transfusion. Transplantation 2006; 82: 857–64.

[33] Norda R, Schott U, Berseus O, Akerblom O, Nilsson B, Ekdahl KN, et al. Complement

activation products in liquid stored plasma and C3a kinetics after transfusion of

autologous plasma. Vox Sang 2012; 102: 125–33.

[34] Pandey P, Chaudhary R, Aggarwal A, Kumar R, Khetan D, Verma A. Transfusion-

associated immunomodulation: quantitative changes in cytokines as a measure of

immune responsiveness after one time blood transfusion in neurosurgery patients.

Asian J Transf Sci 2010; 4: 78–85.

[35] Strait RT, Hicks W, Barasa N, Mahler A, Khodoun M, Kohl J, et al. MHC class I-specific

antibody binding to nonhematopoietic cells drives complement activation to induce

transfusion-related acute lung injury in mice. J Exp Med 2011; 208: 2525–44.

[36] Magee BA, Martin J, Cole MP, Morris KG, Courtney AE. Effects of HLA-matched blood

transfusion for patients awaiting renal transplantation. Transplantation 2012; 94: 1111–6.

[37] Scornik JC. HLA-matched blood transfusions? Transplantation 2012; 94: 1085.

[38] Sasaki H, Chikaraishi T, Furuhata S, Tsutsumi H, Miyano S, Nakazawa R, et al. Autologous

blood transfusion for kidney transplant recipients. Transplant Proc 2008; 40: 1371–2.

























118

Chapter 5

Virtual crossmatch approach to

maximise matching in paired kidney

donation

119

Virtual crossmatch approach to maximise matching in paired kidney donation

Paolo Ferrari, Samantha Fidler, Jenni Wright, Claudia Woodroffe, Paul Slater, Aaron Van

Althuis-Jones, Rhonda Holdsworth, Frank T Christiansen

Paired kidney exchange may be an option where an individual awaiting kidney transplant

has a living donor who is medically suitable, but cannot donate due to blood group

incompatibility or the presence of donor specific HLA antibodies. Exchanges can be very

simple between two pairs, or more complex involving three or more pairs. Identification

of compatible pairs relies on the exclusion of incompatible pairs rather than HLA

matching. Sophisticated computer software was created for the Australian kidney

exchange (AKX) program utilising exclusion of clinically relevant HLA antibodies based on

our previous studies, with high resolution donor typing. The design and validation of this

computer algorithm is discussed in this chapter.


American Journal of Transplantation 2011: 11; 272-278

Contributions:

PF Prepared the manuscript, participated in study design, participated in

data analysis

SJF Performed all computer matching, participated in study design,

participated in data collection, participated in data analysis,

participated in writing the manuscript

JW, PS, AAA-J Programmed the computer module, participated in study design

CW Participated in data collection

RH, FTC Participated in study design, participated in writing of the

manuscript

120

Virtual Crossmatch Approach to Maximize Matching

in Paired Kidney Donation

P. Ferraria,b,c,∗, S. Fidlerc,d,e, J. Wrightf, C. Woodroffea,c, P. Slaterf, A. Van Althuis-

Jonesf, R. Holdsworthg and F. T. Christiansend,e

a Department of Nephrology, Fremantle Hospital, Perth, Western Australia

bSchool of Medicine and Pharmacology, University of Western Australia cAustralian Paired Kidney Exchange Program, Perth, Western Australia dDepartment of Immunology, PathWest, Royal Perth Hospital, Perth, Western Australia eSchool of Pathology and Laboratory Medicine, University of Western Australia, Perth, Western Australia fNational Organ Matching Service, Australian Red Cross Blood Service, Sydney, New South Wales gMedical, Transplantation and Quality Services Division, Australian Red Cross Blood Service, Melbourne, Victoria, Australia *Corresponding author: Paolo Ferrari, [email protected]

We developed and tested a new computer program to match maximal sets of

incompatible live donor/recipient pairs from a national paired kidney donation

(PKD) registry. Data of 32 incompatible pairs included ABO and 4 digit-high-

resolution donor and recipient HLA antigens and recipient’s HLA antibodies.

Three test runs were compared, in which donors were excluded from matching to

recipients with either donor-specific antibodies (DSA) >8000MFI (mean

fluorescent intensity) at low-resolution (Run 1) or >8000MFI at high-resolution

(Run 2) or >2000MFI and high-resolution (Run 3). Run 1 identified 22 703 possible

combinations, with 20 pairs in the top ranked, Run 2 identified 24 113

combinations, with 19 pairs in the top ranked and Run 3 identified 8843

combinations, with 17 pairs in the top ranked. Review of DSA in Run 1 revealed

that six recipients had DSA 2000–8000MFI causing a possible positive crossmatch

resulting in breakdown of two 3-way and three 2-way chains. In Run 2, four

recipients had DSA 2000–8000MFI, also potentially causing breakdown of three

2-way chains. The more prudent approach of excluding from matching recipients

with DSA with >2000MFI reduces the probability of matched pairs having a

positive crossmatch without significantly decreasing the number of possible

transplants.

Key words: ABO, computerization, HLA, incompatibility, kidney transplantation,

matching, paired kidney donation


121

Abbreviations: CDC, complement-dependent cytotoxicity; IPMM, ignoring

previous mismatched antigens; MP, match probability; NOMS, National Organ

Matching System; PKD, paired kidney donation; SAB, single antigen bead;

WAPKD, Western Australian PKD.

Received 20 July 2010, revised 22 August 2010 and accepted for publication 08 September 2010

Introduction

The shortage of kidneys from deceased organ donors and longer kidney transplant waiting

lists in many countries (1,2) have placed greater emphasis on living kidney donation to

meet the increasing demand for transplantation in patients with renal failure. Results of

living donor kidney transplantation, irrespective of tissue typing matches, are superior to

those achieved with deceased donor kidneys (3). Based on the prevalence of ABO blood

groups and recipients’ HLA sensitization it is estimated that approximately 30% of

recipients are biologically incompatible with their intended donor and do not proceed to

live donor kidney transplantation. Traditionally, these recipients are left to wait for a

deceased donor kidney for a variably long period of time. There is ample evidence that

waiting time on dialysis is the strongest modifiable risk factor for renal transplant

outcomes (4). To overcome immunological barriers of blood group or HLA sensitization,

new strategies to expand living kidney donations have been introduced, including paired

kidney donation (PKD) and altruistic donation, or HLA-desensitization and transplantation

across the blood type barrier. PKD is a strategy that helps patients finding a suitable donor

when their intended living donors are blood group or HLA incompatible (5). In a simple PKD,

two donor/recipient pairs surmount each other’s incompatibility problem by simply

exchanging donors (6,7). PKD can be arranged involving three or more pairs, which can be

matched optimally in large numbers only using sophisticated matching software.

The experience of the Western Australian PKD (WAPKD) program (8,9), has been used to

establish a national PKD program and to develop a matching protocol for PKD pairs in

Australia. There are 14 Australian transplantation centers with a living donation program,

but only 4 performing over 25 live donations yearly. The effectiveness of a PKD program

depends largely on the size of the donor/recipient pool and this has led to the development

of a national PKD program in which all Australian kidney transplantation centers

participate. To ensure an equitable and objective matching of participating pairs, matching

122

is performed by the National Organ Matching System (NOMS) using a newly developed

computer program. The software was based on the principles guiding the Dutch algorithm

(10), but taking advantage of advances in HLA technology to enable virtual crossmatching.

In this study, the allocation criteria are described and the outcomes of test runs obtained

with the NOMS matching software using pairs enrolled in the WAPKD program are

presented.

Materials and Methods

The WAPKD program was established in October 2007 as a new initiative of a single

transplantation service performing annually approximately 80 live and deceased donor

kidney transplants. As of December 2009, 32 incompatible donor/recipient pairs were

enrolled in the program. Before development of the NOMS PKD program, the software

used for matching was rudimentary, searching for recipients who were ABO compatible

with each of the registered donors and eliminating those recipients who had HLA

antibodies to an HLA antigen present in the donor. Loops of reciprocal compatible

donor/recipient pairs and combinations of those pairs had to be selected manually, a task

only achievable using 2-way chains with a low degree of complexity in a small pool of PKD

pairs.

HLA typing

Donor HLA-typing for the loci-A, -B, -Cw, -DRB1, -DPB1 and -DQB1 was performed by the

PathWest Department of Clinical Immunology, Royal Perth Hospital using direct DNA

sequencing according to published methods (11). HLA-alleles of each locus were entered in

the computer program at the 4- digit level.

Detection of anti-HLA antibodies

Recipients were tested for HLA class I and class II directed IgG antibody against HLA-A, -B, -

Cw, -DRB1, -DQB1 and -DPB1 in their sera, using single antigen bead (SAB) Luminex

technology (Lab Screen Single Antigen® , One Lambda, Inc., Canoga Park, CA, USA). Using the

vendor’s HLA-Fusion software, the mean fluorescence intensity (MFI) signal for each SAB

was normalized against the negative control to correct for nonspecific binding. Beads with

123

normalized MFIs of <500 were considered negative. HLA antibody specificities were

reported at the 4-digit level. According to the MFI of each SAB the strength of each

antibody was classified as strong (>8000), moderate (2000–8000) or weak (500–2000). HLA

antibody testing was performed every 2 months from the time of listing. Except for

individuals whose antibody strength declined over time, there was very little variation in

the range and strength of the HLA antibodies detected. For matching purposes, the highest

MFI for each HLA antibody was used.

Computer program matching rules and ranking

The NOMS developed a computer program to match participating pairs based on the

Dutch experience (7,10,12). The basic principle of the match algorithm is to ensure that

sensitized recipients with alloantibodies against HLA antigens in the donor pool have the

best chance to receive a kidney, ignoring any rules on mismatches between donor and

recipient HLA. Donors must have a recorded HLA typing at the 4-digit level for the loci HLA-

A, -B, -Cw, -DRB1, -DPB1 and -DQB1. Typing for loci HLA-DQA1, -DRB3, -DRB4 and -DRB5

may also be included. Recipients must be tested for both Class I and Class II HLA antibodies

by Luminex SAB assay and all antibodies with >500MFI are considered positive and entered

into the module.

The NOMS computer program matches each recipient’s HLA antibody against each of the

donor’s HLA antigens using a 3 steps process. First, matched pairs are found by comparing

the AB0 blood groups of each donor and each recipient, then by checking the donor

specific antibodies (DSAs) of recipients and comparing these in turn with the HLA-typing of

each ABO compatible donor. This is done for all potential recipients against all potential

donors. To maximize the number of O recipients being matched to an O donor, allocation

occurs first amongst ABO blood group identical pairs. When all possible ABO identical

matches are identified, the remaining recipients are matched against AB0-compatible

donors.

Second, the program looks for 2-way or 3-way chains of donor recipient pairs, if for

recipient X a match is possible with donor Y, the computer determines whether donor X is a

124

blood group compatible and HLA-acceptable donor for recipient Y. If recipient Y is indeed a

match to donor X, a new chain is formed.

Third, the program creates possible combinations using the following six ranking rules: (1)

prioritize combinations which maximize the number of potential transplants (i.e. favour

combinations allowing 10 possible transplants over combinations resulting only in 8

possible transplants), (2) maximize the number of ABO identical pairs (i.e. favour those

combinations with more O-to-O pairs over O-to-A or B or AB pairs), (3) minimize the

number of simultaneous transplants in a single center (i.e. favour combinations with an

even spread across centers for logistic reasons), 4) maximize the number of shortest chains

(i.e. favour combinations with shorter chains to minimize problems if a donor or recipient

drops out), (5) match probability (MP) of the recipients in the combination (i.e. favour

combinations with more patients with lower match probabilities) and (6) waiting time (i.e.

favour combinations with longer waiting patients). The MP is calculated as MP = (a/b),

where a = number of acceptable donors in the run (i.e. ABO compatible donors having no

unacceptable HLA antigens) and b = total number of ABO compatible donors in the run. The

MP range is 0 to 1, 0 indicating no compatible donors and 1 indicating all ABO compatible

donors are suitable match.

All mismatched HLA antigens from previous transplants are by default considered in the

allocation of deceased donor kidneys in Australia and result in exclusion of a potential

donor bearing any of the previous mismatched antigens. The NOMS PKD computer

program has an option that allows ignoring previous mismatched antigens (IPMM), in the

absence of specific DSAs in the recipient to the previous mismatch or because the

previously mismatched antigen may be considered an acceptable mismatch based on the

epitopes present (13). This results in a patient not being excluded for a donor bearing the

ignored previous mismatched HLA antigen allele. Where previous peak sera are no longer

available for SAB testing to demonstrate absence of the DSAs to mismatched antigens the

decision to assign IPMM antigens is made by the recipient’s transplant team.

125

Testing and validation of the new computer program

Three test runs were undertaken using the data from the 32 pairs identified in the WAPKD

program. Data included blood groups and high-resolution HLA antigens of donors and

recipients and recipients’ HLA antibody specificities (by One Lambda SAB assay). Donors

were excluded from matching to recipients with (a) strong (>8000MFI) DSA defined at the

2-digit level (Run 1), (b) strong (>8000MFI) DSA defined at the 4-digit level (Run 2) or (c)

moderate and strong (>2000MFI) DSA defined at the 4-digit level (Run 3), respectively.

Run 1 with low resolution and a cut-off of >8000MFI was chosen because it was considered

to best simulate the Dutch program. In the Netherlands antibodies are defined at 2-digit

level and are based on complement- dependent cytotoxicity (CDC) reactivity. The antibody

specificity and threshold for Run 2 was chosen because the higher specificity of 4-digit

typing is likely to result in a lower number of potential donors being excluded from

matching to sensitized recipients with multiple antibodies. The presence of DSA 2000–

8000MFI still has the potential of positive CDC or flow crossmatch (14). Run 3 was chosen

because DSA with values of <2000MFI are unlikely to have a positive CDC or flow

crossmatch (14), thus enabling a virtual crossmatch.

Table 1: ABO blood group of donors and recipients in the Western Australian paired

kidney donation program

Donor

A

Donor

AB

Donor

B

Donor

O

Total

Recipient A 3 – – 7 10

Recipient AB – – 1 – 1

Recipient B 2 – – – 2

Recipient O 11 1 2 5 19

Total 16 1 3 12 32

Results

Outcome of the WAPKD program

Between October 2007 and December 2009, 32 incompatible pairs and 2 altruistic donors

were entered into the WAPKD database. Of the 32 pairs, 50% participated because of ABO

incompatibility and 50% because of positive crossmatch, 59% recipients and 38% donors

126

were blood group O (Table 1). In sensitized recipients, HLA antibody distribution was 78%

anti-HLA-A, -B and -DRB1, 16% anti- HLA-Cw, -DPB1, -DQA1 and 6% anti-HLA-DQB1, -DRB3 or

-DRB5. Two patients were highly sensitized with a calculated PRA >85%. During the 2-year

period of the WAPKD program the maximum number of pairs per run was 14 and up to 135

possible matches were identified. The majority of these matches involved the same pairs

being present in multiple exchanges. After limiting possible exchanges to 2-way chains and

eliminating multiple matches, 11 PKD and 2 deceased donor list recipients (4×2-way +

1xaltruistic donor 3-way + 1x altruistic donor 2-way) were matched and transplanted

through the WAPKD program. All had a negative CDC T- and B-cell crossmatch. Of the 11

PKD recipients transplanted in the WAPKD program, 7 had anti- HLA antibodies and 4 were

ABO incompatible with their intended donors. Eight subjects were transplanted in the

presence of multiple ‘weak’ (500–2000MFI) DSAs. No flow crossmatch was performed for

these patients, but none had acute humoral rejection. Of the remaining PKD recipients, six

received a deceased donor kidney transplant, four were transplanted using an ABO

incompatible program, three left the transplant program and eight recipients remain on

the transplant waiting list.

Validation of the new computer software

In the first step, the NOMS computer program compares each donor in turn against all

recipients and determines whether they are a match. Only a few seconds processing time

are necessary to find matching pairs. For the 32 recipients, although 992 (= (N-1) x N)

combinations are possible, depending on the antibody resolution and HLA antigen

specificity, only 355 (Run 3) to 445 (Run 2) combinations were found. In the second step,

these separate matched pairs (donor with new recipients) are used to identify all possible 2-

way or 3-way chains. This process takes a few minutes and results in between 191 to 316

chains with a maximum chain size of 3 (Table 2). In the third step, the computer program

selects all possible combinations of chains in which each matched pair appears only once

in a combination and ranks all possible combinations according to the preset conditions. As

the program finds all possible combinations of chains, the number of results increases

exponentially leading to 8000–25 000 possibilities when 32 pairs are enrolled and takes up

to several hours computing time (Table 2).

127

Table 2: Comparison of 3 test runs using different antibody resolution and strength to

exclude recipients with antibodies from matching to donors in a pool of 32 incompatible

donor - recipient pairs.

Run 1 used 2-digit HLA typing, with antibodies for exclusion >8000 MFI; run 2 used 4-digit HLA typing with

antibodies for exclusion >8000 MFI; and run 3 used 4-digit HLA typing with antibodies for exclusion >2000

MFI.

Run 1 Run 2 Run 3

Low

Resolution

High

Resolution

High

Resolution MFI > 8000 MFI > 8000 MFI > 2000

Time to match 3h 45 min 3h 58 min 0h 50 min

No. of matched pairs 439 445 355 No. of chains 308 316 191 No. of combinations 22 703 24 113 8843 No. of patients in 1st combination 20 19 17 3-way chains in 1st combination 4 5 5 2-way chains in 1st combination 4 2 1 Recipients with DSA 2000–8000MFI 6 4 0 No. of patients in chains with predicted negative

crossmatch

8 10 17 Donor/Recipient age difference (years) 0.9 ± 14.2 1.2 ± 13.6 4.0 ± 14.7

(–24 to 24) (–25 to 29) (–24 to 29)

Run 1 was completed in 3h 45min and identified 22703 possible combinations, with the

highest ranking comprising 20 pairs in 4 × 3-way and 4 × 2-way exchanges, of whom 10

(50%) were O recipients (Table 2). Run 2 was completed in 3h 58min and identified 24113

combinations, with the highest ranking comprising 19 pairs in 5 × 3- way and 2 × 2-way

exchanges, of whom 9 (47%) were O recipients. Run 3 was completed in 50 min and

identified 8843 combinations, with the highest ranking comprising 17 pairs in 5 × 3-way and

1 × 2-way exchanges (Figure 1), of whom 6 (35%) were O recipients (Tables 2 and 3).

Comparison of donor and recipient HLA antigens at conventional HLA-A, -B and -DRB1 loci

for those pairs matched in test Run 3 revealed and average of 4.8 ± 1.2 mismatches (A-

/B- 3.2 ± 0.9, DRB1- 1.6 ± 0.5) between each matched donor and recipient. In test Run 1,

a review of each recipient’s DSA revealed that 6 recipients had DSA with MFI of between

2000 and 8000 and may have had a positive flow crossmatch, causing breakdown of chain

1 + 2 (3-way) and 3 + 4 + 6 (2-way) if a negative flow crossmatch is a prerequisite for

transplantation. This would have resulted in only eight PKD pairs with predicted negative

flow and CDC crossmatch. In test Run 2 DSA 2000–8000MFI were present in 4 recipients

resulting in only 10 pairs with predicted negative flow and CDC crossmatch suitable for

transplantation (Table 2).

128

Figure 1: Matching diagram of test run 3 using 32 pairs, where donors were

excluded from matching to recipients with moderate and strong (>2000MFI) donor-

specific antibodies defined at the 4-digit level.

Fine lines indicate all possible chains; bold lines with arrows indicate chains in the 1st

combination with the highest number of pairs matched at the lowest match probability with the

highest number of blood group identical matches.

Table 3: Match results by blood type excluding recipients with donor specific antibody at >2000 mean fluorescence intensity (OneLambda) and using high-resolution antibody definition

129

In the 16 patients enrolled because of positive crossmatch to their intended recipients,

median class I PRA was 36 (range 10–89) in the 11 matched recipients and 60 (range 20–

73) in the residual 5 unmatched recipients (p = 0.30 Mann-Whitney test). Median class II

PRA was 42 (range 10–99) in the 11 matched recipients and 23 (range 14–95) in the

residual 5 unmatched recipients (p = 0.76).

Discussion

Using the matching algorithm developed by the NOMS, simulations using different

definitions of antibody specificity and strength in sensitized recipients to match

donor/recipient pairs in a PKD pool showed that 53% of pairs could find a suitable match

with predicted negative virtual crossmatch even in a relatively small pool of 32

donor/recipient pairs. Outcomes of the WAPKD program used to provide testing data for

the NOMS computer program support the probability of a high success rate of matched

pairs progressing to actual kidney transplantation and support the view that the strategy

selected in Australia to achieve the maximum number of transplants with a given pool of

incompatible donors and recipients is highly effective. Algorithms used in other countries

select possible exchanges based on a points scoring system that includes the number of

HLA-mismatches, % PRA, distance between the exchange centers, donor age differences,

CMV status and other factors (15,16). These algorithms have some inherent disadvantages,

such as limiting the maximum number of potential matches, or a subsequent high rate of

positive CDC crossmatches after computer matching of pairs, with consequent breakdown

of chains. This has been demonstrated in a testing environment with the UK algorithm,

showing a simulated 25% match-rate with 30 pairs and 35–40% with 50 pairs (15) and one

of the US algorithms, showing 24% match-rate with 45 recipients (16). Johnson et al.

reported performing only 8 PKD transplants in the UK over a 16 months period despite

identification of 28 matches among 85 pairs (15).

A direct comparison of the performance of the NOMS and the Dutch programs was not

undertaken. However, a simulation was tested using a match run with low resolution HLA

antibody definition and a cut-off of >8000MFI, thereby only including strong antibodies likely

130

to have CDC reactivity. In the Dutch program antibody specificities are entered at low-

resolution and matching is based on antibodies defined by CDC reactivity, as those

antibodies are considered a contraindication for transplantation, whereas Luminex is

considered a risk factor, but not contraindication to transplantation. In test Run 1 although

20/32 (63%) of recipients had a matched donor, 6 (30%) of the patients matched by low

resolution and a cut-off of >8000MFI would have had a DSA with a 2000–8000MFI or

antibody against splits within the broad antigen specificity. These results are in keeping

with the reported incidence of 34% pairs having a positive crossmatch in the 183

computer-matched combinations from the Dutch experience (7). Although the attempt to

exclude all potentially incompatible pairs with high levels of DSA to replicate the Dutch

experience for purposes of comparison has the disadvantage that it may exclude

individuals without true DSA, our strategy indicates a prudent approach that minimizes the

risk of breakdown of chains. It is worth noting that in the WAPKD cohort used to validate

the new software none of the identified matches had a positive CDC crossmatch prior to

transplantation.

In recent years, solid phase assays using beads coated with HLA antigens have become the

preferred method of detecting HLA antibodies in transplant candidates (17). These assays

can identify the precise HLA-specificities of antibodies and donor cells are not required. It is

still unclear what strength of Luminex detected antibodies is considered equivalent to CDC

reactivity. In our experience the presence of DSA >8000MFI will often result in a positive

CDC crossmatch, whereas HLA antibodies with a DSA 2000–8000MFI may result in a positive

flow crossmatch. Our approach, where we exclude any donor against whom a recipient has

a HLA antibody at a strength of >2000MFI in the PKD pool, may be considered excessively

prudent because it does not take into account the possibility of noncomplement fixing

IgG2 or IgG4 antibodies. Noncomplement fixing anti-HLA alloantibodies are responsible for

discrepant results observed between solid phase assays and negative CDC reactivity.

Conventional solid phase assays cannot distinguish between complement fixing and

noncomplement fixing antibodies, therefore possibly leading to an overestimation of

sensitized patients than with the CDC assay (18,19). This brings into question whether it is

safe to transplant patients who may be CDC negative, but positive solid phase DSAs.

Recent studies in kidney transplant recipients indicate that even with a negative CDC

crossmatch, the presence of DSA detected by flow cytometry (20) or Luminex (19,21,22) is

131

associated with higher rates of graft failure compared to antibody negative patients.

Recently, C4d-fixing capability of DSAs has been reported and was found to be a negative

predictor of graft survival in cardiac and kidney transplantation (23,24). However, C4d-fixing

DSAs have generally significantly higher MFI values than non-C4d-fixing DSA (>10000MFI

73% vs.18%; ) (25). Introducing the additional step of in vitro detection of C4d-fixing

alloantibodies would therefore probably not significantly increase the specificity of

antibody used for the exclusion of matches in our PKD program. Our conservative

approach would result in only 3 less matched pairs than using high titer, low definition

antibody to identify DSA. However, this virtual crossmatch approach is most likely to result

in a negative CDC crossmatch (14) allowing all matched pairs a better chance to proceed to

transplantation.

Because HLA typing of donors and the definition of unacceptable donor HLA antigens for

recipients are the basic parameters used in the NOMS matching algorithm, donor HLA

typing and recipient antibody profile are registered in more detail than that used for the

regular allocation of deceased donor kidneys. The advantage of listing antigens and

antibodies at 4-digit level is that in recipients with HLA antibodies it increases the specificity

and reduces the number of donors excluded from matching. For example, a recipient with

an HLA-B∗44 antibody will be excluded from matching with any donor with a B∗44

antigen. However, if the antibody specificity of the recipient is B∗44:02, donors with the

B∗44:02 antigen, but not B∗44:03 will be excluded from matching. This explains the higher

number of combinations identified in Run 2 using high-resolution matching compared to

the low-resolution matching results in Run 1.

The Australian PKD program limits the number of chains to 3-way exchanges because this

restriction does not significantly reduce the maximum number of possible matches (26),

but minimizes the complexity of logistics required to coordinate 4 or more simultaneous

donor surgeries. Although donor–donor or donor–recipient age differences are not used as a

scoring parameter in ranking match combinations, the NOMS program will not accept

exchanges with a donor-donor age difference of >30 years. In the 3 test runs the average

donor/recipient age difference of matched pairs was between 1 and 4 years depending on

132

the matching rules employed and no recipient was matched with a donor ≥30 years older

than the recipient’s age. This should ensure an excellent graft survival and function

unaffected by donor age (27).

A limitation of the current study is that we did not perform CDC or flow crossmatches on

patients with DSA 2000– 8000MFIs to substantiate the use of the more conservative

approach. Nevertheless, there are strong arguments to support such an approach. First,

our experience shows that using a cut off of 2000MFI, >95% of DSA negative recipients will

have a negative flow and CDC crossmatch, whereas approximately 50% of DSA positive

recipients will have a positive flow or CDC crossmatch. The finding of a 34% rate of positive

crossmatches in the Dutch study (25) provides additional support for this conclusion.

Second, a survey among transplant centers participating in the Australian PKD program has

indicated that clinicians are willing to perform transplants in the setting of DSA <2000MFI

with a negative CDC crossmatch, but, in the presence of DSAs with a strength of 2000–

8000MFI the majority would request a flow crossmatch and if positive, not proceed to

transplantation. This is likely to lead to delays in the exchanges and the breakdown of

matched donor/recipient chains. Third, even if such cases have negative crossmatches, there

is increasing evidence that transplantation in such circumstances is associated with higher

rates of acute rejection and poorer graft survival (19,21,22). It would seem prudent to avoid

such transplants and our approach allows this without compromising the viability of the

exchange program.

In conclusion, our data demonstrate that using a computer matching algorithm based on

the high definition of recipient HLA antibody and donor HLA antigens to match PKD pairs,

the number of possible matches with predicted negative virtual crossmatch can be

maximized. As with other PKD programs sensitized patient who are not matched tend to

be more highly sensitized. Although the sensitivity of non anti-human globulin enhanced

cytotoxic crossmatches is clearly lower than a virtual crossmatch based on DSA antibody

defined by SAB and which is likely to exclude matches with a negative cytotoxic

crossmatch, our data suggest that the more prudent approach of excluding from matching

recipients with antibody MFI-values >2000 does not result in a significantly lower number

of possible transplants compared to exclusion of recipients with antibody MFI-values

133

>8000 (χ 2 = 0.57), a threshold that may result in a higher number of matched pairs having

a positive crossmatch using standard or enhanced CDC or flow crossmatch techniques.

Acknowledgments

The authors wish to acknowledge the support of the Australian Organ and Tissue Donation

and Transplantation Authority for the establishment and funding of the Australian Paired

Kidney Donation program.

Disclosure

The authors of this manuscript have no conflicts of interest to disclose as described by

the American Journal of Transplantation.

134

References

1. Mathew T, Faull R, Snelling P. The shortage of kidneys for transplantation in Australia.

Med J Aust 2005; 182: 204–205.

2. Harter TD. Overcoming the organ shortage: Failing means and radical reform. HEC Forum

2008; 20: 155–182.

3. Cecka JM. Kidney transplantation in the United States. Clin Transpl 2008; 22: 1–18.

4. Meier-Kriesche HU, Kaplan B. Waiting time on dialysis as the strongest modifiable risk

factor for renal transplant outcomes: A paired donor kidney analysis. Transplantation

2002; 74: 1377– 1381.

5. Rapaport FT. The case for a living emotionally related international kidney donor exchange

registry. Transplant Proc 1986; 18(3 Suppl 2): 5–9.

6. Park K, Moon JI, Kim SI, Kim YS. Exchange-donor program in kidney transplantation.

Transplant Proc 1999; 31: 356–357.

7. de Klerk M, Witvliet MD, Haase-Kromwijk BJ, Claas FH, Weimar W. Hurdles, barriers, and

successes of a national living donor kidney exchange program. Transplantation 2008; 86:

1749–1753.

8. Ferrari P, Woodroffe C, Christiansen FT. Paired kidney donations to expand the living donor

pool: The Western Australian experience. Med J Aust 2009; 190: 700–703.

9. Ferrari P, De Klerk M. Paired kidney donations to expand the living donor pool. J Nephrol

2009; 22: 699–707.

10. Keizer KM, de Klerk M, Haase-Kromwijk BJ, Weimar W. The Dutch algorithm for allocation

in living donor kidney exchange. Transplant Proc 2005; 37: 589–591.

11. Sayer D, Whidborne R, Brestovac B, Trimboli F, Witt C, Christiansen F. HLA-DRB1 DNA

sequencing based typing: An approach suitable for high throughput typing including

unrelated bone marrow registry donors. Tissue Antigens 2001; 57: 46–54.

12. de Klerk M, Witvliet MD, Haase-Kromwijk BJ, Claas FH, Weimar W. A highly efficient living

donor kidney exchange program for both blood type and crossmatch incompatible donor-

recipient combinations. Transplantation 2006; 82: 1616–1620.

13. Duquesnoy RJ. A structurally based approach to determine HLA compatibility at the

humoral immune level. Hum Immunol 2006; 67: 847–862.

14. Morris GP, Phelan DL, Jendrisak MD, Mohanakumar T. Virtual crossmatch by identification

of donor-specific anti-human leukocyte antigen antibodies by solid-phase immunoassay: A

135

30-month analysis in living donor kidney transplantation. Hum Immunol 2010; 71: 268–273.

15. Johnson RJ, Allen JE, Fuggle SV, Bradley JA, Rudge C. Early experience of paired living kidney

donation in the United Kingdom. Transplantation 2008; 86: 1672–1677.

16. Saidman SL, Roth AE, Sonmez T, Unver MU, Delmonico FL. In- creasing the opportunity of

live kidney donation by matching for two- and three-way exchanges. Transplantation

2006; 81: 773– 782.

17. Tait BD. Solid phase assays for HLA antibody detection in clinical transplantation. Curr Opin

Immunol 2009; 21: 573–577.

18. Tambur AR, Bray RA, Takemoto SK, Mancini M, Costanzo MR, Kobashigawa JA et al. Flow

cytometric detection of HLA-specific antibodies as a predictor of heart allograft rejection.


19. Gibney EM, Cagle LR, Freed B, Warnell SE, Chan L, Wiseman AC. Detection of donor-

specific antibodies using HLA-coated microspheres: Another tool for kidney transplant risk

stratification. Nephrol Dial Transplant 2006; 21: 2625–2629.

20. Bielmann D, Honger G, Lutz D, Mihatsch MJ, Steiger J, Schaub S. Pretransplant risk

assessment in renal allograft recipients using virtual crossmatching. Am J Transplant 2007;

7: 626–632.

21. Amico P, Honger G, Mayr M, Steiger J, Hopfer H, Schaub S. Clinical relevance of

pretransplant donor-specific HLA antibodies detected by single-antigen flow-beads.


22. Lefaucheur C, Loupy A, Hill GS, Andrade J, Nochy D, Antoine C et al. Preexisting donor-

specific HLA antibodies predict outcome in kidney transplantation. J Am Soc Nephrol

2010; 21: 1398– 1406.

23. Wahrmann M, Exner M, Schillinger M, Haidbauer B, Regele H, Kormoczi GF et al. Pivotal

role of complement-fixing HLA alloantibodies in presensitized kidney allograft recipients.

Am J Transplant 2006; 6(5 Pt 1): 1033–1041.

24. Smith JD, Hamour IM, Banner NR, Rose ML. C4d fixing, luminex binding antibodies – a

new tool for prediction of graft failure after heart transplantation. Am J Transplant 2007;

7: 2809– 2815.

25. Honger G, Wahrmann M, Amico P, Hopfer H, Bohmig GA, Schaub

S. C4d-fixing capability of low-level donor-specific HLA antibodies is not predictive for

early antibody-mediated rejection. Transplantation 2010; 89: 1471–1475.

136

26. de Klerk M, Van Der Deijl WM, Witvliet MD, Haase-Kromwijk BJ, Claas FH, Weimar W. The

optimal chain length for kidney paired exchanges: An analysis of the Dutch program.

Transpl Int 2010; 23: 1120–1125.

27. Ferrari P, Lim W, Dent H, McDonald SP. Effect of donor- recipient age difference on graft

function and survival in live-donor kidney transplantation. Nephrol Dial Transplant 2010.

137

Chapter 6

High transplant rates of highly sensitized

recipients with virtual crossmatching in

kidney paired donation

138

High transplant rates of highly sensitized recipients with virtual crossmatching

in kidney paired donation

Paolo Ferrari, Samantha Fidler, Rhonda Holdsworth, Claudia Woodroffe, Gabriella

Tassone, Narelle Watson, Linda Cantwell, Greg Bennett, Alycia Thornton, Frank T

Christiansen, Lloyd D’Orsogna

The Australian paired kidney exchange (AKX) program was designed on the basis of

exclusion of clinically relevant antibodies detected by Luminex single antigen bead assay.

The low threshold for antibody exclusion, a mean fluorescence intensity (MFI) of greater

than 2000, conservatively excludes patients with a predicted positive complement-

dependent crossmatch. A concern was that highly sensitised patients might be

disadvantaged by excluding crossmatch compatible donors. This paper reviews the

match and transplant rates in the first year of the AKX program with respect to

recipient and donor characteristics, including blood group distribution, level of

recipient’s sensitization, and crossmatches.


Transplantation 2012: 94; 744-749

Contributions:

PF Prepared the manuscript, participated in study design,

participated in data analysis

SJF Performed all computer matching, participated in study

design, participated in data collection, participated in data

analysis, participated in writing the manuscript

RH, FTC, LD’O Participated in study design

CW Participated in data collection

GT, NW, LC, GB, AT Participated in study design, participated in data collection

139

CLINICAL AND TRANSLATIONAL RESEARCH

High Transplant Rates of Highly Sensitized Recipients With Virtual Crossmatching in

Kidney Paired Donation

Paolo Ferrari,1,2,10

Samantha Fidler,3,4

Rhonda Holdsworth,5 Claudia Woodroffe,

1

Gabriella Tassone,3 Narelle Watson,6 Linda Cantwell,7 Greg Bennett,8 Alycia

Thornton,9Frank T. Christiansen,

3,4 and Lloyd D’Orsogna

3,4

Background. In kidney paired donation (KPD), flexibility in the allocation of

incompatible pairs is required if a critical mass of pairs to efficiently find matches

cannot be reached.

Methods. In the Australian KPD program, virtual crossmatch is used for the allocation

of suitable donors to registered recipients. Matching is based on acceptable

mismatches, and donors are excluded from matching to recipients with donor-specific

antibodies (DSAs) greater than 2000 mean fluorescence intensity (MFI). Match and

transplant rates in the first year of the program were reviewed with respect to

recipient and donor characteristics, including blood group distribution, level of

recipient’s sensitization, and post allocation crossmatches.

Results. Four quarterly match runs were performed, which included 53 pairs and 2

altruistic donors. Human leukocyte antigen incompatibility accounted for 90% of the

listed pairs. In the second run, the DSA threshold was increased to greater than 8000

MFI, because no matches were found with standard allocation. Optional ABO-

incompatible matching was introduced from run 3. Matches were identified in 37

(70%) patients, of whom 92% had a negative crossmatch with their matched donor.

Crossmatch positive results were found only in recipients with DSAs greater than 2000

MFI in the second run. In 4 cases immunological reasons and in 4 cases other reasons

resulted in breakdown of chains and 17 patients not progressing to transplantation.

Eventually, 20 (38%) patients received a KPD transplant, and 35% of these had a

calculated panel-reactive antibody greater than 90%.

Conclusions. KPD using virtual crossmatch is a valid and effective solution for patients

with immunologically incompatible donors even in the context of highly sensitized

recipients.

140

Keywords: Kidney paired donation, ABO incompatibility, HLA incompatibility, Matching,

Computerization.

(Transplantation 2012; 94: 744-749)

The study was supported by the Australian Organ and Tissue Donation and Transplantation

Authority through the establishment of the National Paired Kidney Donation program and

funding of part of the staff (P.F., S.F., and C.W.).

The authors declare no conflicts of interest.

1Department of Nephrology, Fremantle Hospital, Perth, Western Australia, Australia.

2School of Medicine and Pharmacology, University of Western Australia, Perth, Western Australia,

Australia.

3Department of Immunology, PathWest, Royal Perth Hospital, Perth, Western Australia, Australia.

4School of Pathology and Laboratory Medicine, University of Western Australia, Perth, Western

Australia, Australia.

5National Transplant Services, Australian Red Cross Blood Service, Melbourne, Victoria, Australia.

6New South Wales Transplantation and Immunogenetics Service, Australian Red Cross Blood

Service, Sydney, New South Wales, Australia.

7Victorian Transplantation and Immunogenetics Service, Australian Red Cross Blood Service,

Melbourne, Victoria, Australia.

8 South Australian Transplantation and Immunogenetics Service, Australian Red Cross Blood

Service, Adelaide, South Australia, Australia.

9 Tissue Typing Laboratory, Pathology Queensland, Royal Brisbane and Women’s Hospital,

Herston, Queensland, Australia.

10Address correspondence to: Paolo Ferrari, M.D., Australian Paired Kidney Exchange Program,

Department of Nephrology, Fremantle Hospital, School of Medicine and Pharmacology,

University of Western Australia, Perth WA 6160, Australia.

E-mail: [email protected]

P.F., S.F., C.W., and L.D. participated in data analysis, performance of the research, and writing

of the paper. R.H. participated in performance of the research and data analysis. C.W., G.T.,

N.W., L.C., G.B., and A.T. participated in performance of the research. F.T.C. participated in data

analysis and writing of the paper.

Received 5 April 2012. Revision requested 24 April 2012. Accepted 23 May 2012.


141

Despite the increase in the number of deceased organ donors in recent years, live

kidney donation still remains a major source of kidney transplantations in Australia.

Not infrequently, a willing and healthy living donor is deemed incompatible owing to

blood group incompatibility or unacceptable donor-specific antibodies (DSAs). Kidney

paired donation (KPD) provides a solution to this dilemma by pairing two or more

incompatible pairs together to facilitate the exchange of live-donor kidneys between

the willing donors (1-4). The success of KPD depends on several factors, such as the

number of incompatible pairs in the database, the proportion of blood group-

incompatible versus human leukocyte antigen (HLA)-incompatible pairs, the level of

sensitization of enrolled patients, and the rules for allocation of KPD donors to

recipients. The Netherlands (1, 5) and the United Kingdom (UK) (3) have well-

established national KPD programs, whereas in the United States, several regional (6) or

single-center (2, 7) programs have established successful programs in the past few

years.

In Australia a national KPD program was established in August 2010. The

implementation of the Australian Paired Kidney Exchange (AKX) program required that

all participating transplant centers agree on donor and recipient medical suitability

criteria and the variables and rules used for allocation. We previously reported on the

allocation rules agreed for the AKX program (4), which were based on simulations

using different algorithms and data from a previous single-center KPD experience (8).

Our proposed approach was to ignore any HLA matching rules and to use matching

based on acceptable mismatches by excluding donors from matching to recipients

with DSAs greater than 2000 mean fluorescence intensity (MFI) for any class I or II

antibodies by Luminex single-antigen bead (SAB). This conservative approach to

antibody assignment was chosen to avoid the breakdown of identified chains due to

positive crossmatches (9) and to reduce the risk of inferior long-term outcome in the

presence of strong DSAs in patients with negative crossmatch at transplantation (10).

Depending on the level of sensitization of patients enlisted in a KPD program, our

approach could potentially disadvantage sensitized patients by excluding crossmatch-

compatible donors. We therefore undertook the task to review the activity of the first

12 months of the AKX program.

142

RESULTS

A total of 53 donor-recipient pairs were included in the 4 match procedures

performed between August 2010 and August 2011. There were 24 couples who

participated in 1, 14 in 2, 8 in 3, and 7 in all 4 rounds. Match run conditions are

summarized in Table 1. Twenty pairs were included in the first, 25 in the second, 26

in the third, and 33 in the fourth match run. In match runs 2 and 3, an additional

altruistic donor each contributed to the donor pool.

Table 1. Match run conditions for the quarterly match runs of the Australian KPD

program

Type of Immunological Incompatibility in Enlisted Pairs

Only 10% of recipients were included because of ABO blood group incompatibility

without HLA antibody against their coregistered donor. Most recipients (90%) were

included because of HLA sensitization; 40% were HLA and ABO incompatible with their

coregistered donor, and 50% were HLA incompatible (but ABO compatible) with their

coregistered donor. Table 2 summarizes the blood type distribution of donors and

recipients; 30% of the donors were blood group O, and there were twice as many

blood group O recipients. In the 53 registered recipients, the median calculated panel-

reactive antibody (cPRA) for loci -A, -B, -DR, and -DQ was 81% (mean, 71 ± 32). cPRAs of

greater than 75% and greater than 90% were found in 58% and 42% of the registered

recipients, respectively.

143

Table 2. ABO blood group distribution of donors and recipients in the 4 initial match

runs of the AKX program for KPD

Computer Matching

The matching algorithm does not consider any HLA matching rules, and allocation is

only based on acceptable mismatches by excluding donors from matching to recipients

with DSAs greater than 2000 MFI for any class I or II antibodies against any of the

donor HLA loci (4). The computer program selects between competing match offers on

the basis of prespecified ranking rules, such as favoring 3-way over 2- way exchanges,

to maximize the number of patients receiving a transplant, and then favoring patients

with low versus high match probability, which primarily gives an advantage to a

recipient with high versus low PRA. In match run 2, and despite the inclusion of an

altruistic donor, no suitable matches were identified using the 2000-MFI cutoff for

unacceptable mismatches, and a subsequent run was performed using the 8000-MFI

cutoff (Table 1). From match run 3, we agreed with referring centers that ABO-

incompatible matching could be considered as an option for selected patients with low

blood group-specific titers. After limiting possible exchanges to 2-way and 3-way

chains, without limitation of chain length with altruistic donors and eliminating

multiple matches, 31 KPD and 2 cadaveric list recipients were matched through the

AKX program. Three recipients were matched with multiple donors in subsequent

match runs. The match probability by cPRA in the 27 crossmatch-negative matched

recipients compared with the 53 registered patients is shown in Figure 1.

144

Figure 1. Match probability by cPRA (-A, -B, -DR, and –DQ) in 27 crossmatch-negative

matched recipients among 53 registered patients. Whereas the probability of matching

was inversely correlated with the cPRA, a matching donor was found for most highly

sensitized recipients, and recipients with a cPRA greater than 90% still had a 43% match

probability

Calculated Panel-Reactive Antibody in Listed Versus Matched Recipients

There was an inverse relationship between cPRA and match probability. Of the 37

matched recipients, the matched rate was 86% if the cPRA was less than 25% but was

43% if the cPRA was 90% to 100% (Fig. 1). However, a matching donor in a chain

combination could still be identified in a significant proportion of highly sensitized

patients. Of the 53 listed recipients, 58% had a cPRA greater than 75%, and in 42% of

them the cPRA was greater than 90%. In patients who underwent transplantation, a

cPRA greater than 75% was present in 45%, and a cPRA greater than 90% was

present in 35% of the recipients (Fig. 2).

145

Figure 2. Pattern of cPRA (-A, -B, -DR, and -DQ) in 53 registered recipients and in 20

recipients who underwent transplantation.

Complement-Dependent Cytotoxic Crossmatches

In total, 37 complement-dependent cytotoxic (CDC) crossmatches were performed

between matched donors and recipients. All the KPD donors and the altruistic

donor re- called for CDC crossmatch testing in the first 2 rounds were also

crossmatched against all recipients on the trays. Thus, an additional 112 T-cell

crossmatches (TCXMs) and B-cell crossmatches (BCXMs) results were available for

analysis, in which 60% of recipients were tested against the donors despite the

presence of single or multiple DSAs. The aggregated results of all crossmatches are

reported in Table 3.

146

Table 3. Aggregated results of 149 CDC allogeneic crossmatches performed between 37

matched donors and matched or unmatched recipients, whose sera were available for

testing.

Crossmatch Results of Matched Recipients

All 27 recipients who were matched using the 2000-MFI cutoff for unacceptable

mismatches had a negative CDC TCXM and BCXM. In match run 2, a less stringent

exclusion of unacceptable mismatches setting the cutoff at 8000 MFI resulted in 10

patients finding a possible donor. In 7 patients, 1 or more DSAs at strengths ranging

from 2254 to 6753 MFI were present; in 4 patients with a low-level single DSA

(2254-3373 MFI), both TCXM and BCXM were negative. However, 3 patients with 1 or

more DSAs at a strength greater than 6000 MFI had a positive BCXM, but negative

TCXM, against their matched donor and were not accepted for transplantation

resulting in the breakdown of chains.

Three patients were matched to an ABO-incompatible donor because an initial blood

group titer indicated that they were suitable for ABO-incompatible transplantation.

How- ever, in 2 patients, additional testing against donor red blood cells revealed an

excessively high blood group antibody titer, and it was decided not to progress to

transplantation. Thus, after computer matching and CDC crossmatch testing, 27

recipients were booked for live-donor kidney transplantation.

Panel Crossmatch Results

There were 21 positive TCXM with positive BCXM and 32 with positive BCXM and

negative TCXM results. All patients with positive TCXM had 1 or more class I DSA at

greater than 8000 MFI. In two patients, the initial testing suggested a weak or

147

moderate DSA, but the strength of the DSA was found to be greater than 8000 MFI

when the Luminex assay was repeated in the immunoglobulin (Ig)M-depleted serum

sample. A positive BCXM was found in 47% of patients with DSAs at greater than 8000

MFI and in 38% with DSAs at 2000 to 8000 MFI (Fig. 3A). DSAs against C-locus were

responsible for 15% of positive TCXMs, and DSA against DQB1 accounted for 50% of

positive BCXMs (Fig. 3B).

Figure 3. DSA strength and specificity in crossmatch-positive assays from 112

crossmatches performed on unmatched panel recipients in match runs 1 and 2 against

available donor cells. A, Strength of DSA. B, Specificity of DSA in those with

crossmatch-positive assays.

148

Kidney Paired Donation Transplants and Outcomes

In the first year of the AKX program, 20 recipients have undergone successful kidney

transplantation. One 3-way chain and two 2-way chains did not progress to

transplantation. One offer was refused by the transplant surgeon because of concerns

in relation to the vascular anatomy of the donor kidney, 1 because of a donor

withdrawing consent, and 1 because of an acute illness of a matched recipient. The

breakdown of these chains resulted in 7 patients not undergoing transplantation. Of

the patients who underwent transplantation, 18 were KPD recipients, and 2 were

transplant wait-list recipients who received the kidney from the last KPD donor in an

altruistic donor chain. This represents 34% transplant rate of all registered KPD

recipients, 49% of all matched recipients, and 74% of crossmatch-negative matches.

The mean follow-up period was 256 - 131 days, and 95% of the patients remained

rejection-free throughout this period. Fourteen patients underwent a total of 18

kidney biopsies (3 per cause, 15 protocol), and 6 patients did not have a kidney biopsy

either per cause or protocol. One patient with delayed graft function not requiring

dialysis demonstrated mild acute tubular necrosis on day 3 and showed BK virus

nephropathy at 3 months. One recipient with a cPRA of 99% who received a kidney

from an ABO-incompatible donor showed abundant C4d deposition and possible

antibody-mediated rejection with mild peritubular capillaritis. Records on renal

allograft function available to date revealed that serum creatinine levels were 108 ± 27

Kmol/L at discharge, 94 ± 22 Kmol/L at 3 months (n=20), 96 ± 29 Kmol/L at 6 months

(n=15), and 88 ± 11 Kmol/L at 1 year (n=5).

DISCUSSION

The first-year experience of the Australian KPD program demonstrates that by

using a stringent virtual crossmatch approach, it is possible to achieve match rates

of greater than 50% even with a small donor-recipient pool where most recipients

are highly sensitized. Thus, using a conservative approach to antibody assignment

to avoid positive crossmatches does not seem to significantly disadvantage

sensitized recipients.

These results are of particular importance in view of the relatively small number of

unsensitized ABO-incompatible recipients enrolled in the program, reducing the ability

to easily match donors to multiple recipients. In many instances, the barrier of blood

group incompatibility can be easily overcome without substantial changes in the

immunosuppressive regime (11, 12), and this has become standard practice in many

149

transplant centers in Australia, where the costs for apheresis for ABO-incompatible

transplants are not capped. This has contributed to the small number of ABO-

incompatible patients without HLA antibodies being listed in AKX.

The options for patients with high-level HLA sensitization are more limited, and there

are still concerns with regard to expense, increased morbidity, and inferior long-term

outcomes associated with desensitization techniques (10, 13, 14), although they have

been shown to provide a significant survival benefit for HLA-incompatible patients, as

compared with waiting for a compatible organ (15). If directed donation is not possible

because DSAs are not amenable to desensitization and conventional HLA matching in

the cadaveric program results in exceedingly long waiting time, the only hope for

highly sensitized recipients could be KPD. With the use of our virtual crossmatch

approach to KPD allocation (4), sensitized patients, with the exception of those with a

cPRA greater than 90%, do not seem to be overtly disadvantaged in favor of the

unsensitized patients. Moreover, that 35% of recipients who underwent

transplantation had a cPRA greater than 90% is a remarkable outcome. To perform

transplantation on a larger number of these more highly sensitized patients, a less

conservative approach to antibody assignment may be required. A restrictive policy to

reduce the increased risk for antibody-mediated rejection and the increased long-term

graft dysfunction will result in such transplantations being avoided in patients who

have no other options. Match run 2 demonstrates that a higher threshold for exclusion

from matching in the presence of DSAs of up to 8000 MFI resulted in 3 recipients with

negative TCXM and BCXM despite DSAs greater than 2000 MFI undergoing

transplantation without evidence of antibody-mediated rejection on protocol

biopsies after up to 12 months of follow-up. On the other hand, the only instances of

positive BCXM in matched recipients were found with this level of antibody assignment,

leading to breakdown of chains. Moreover, the results of the crossmatch testing

performed against the panel of unmatched recipients demonstrate that all positive

TCXM and 85% of positive BCXM were found in the presence of DSAs at greater than

2000 MFI, which is in line with previously reported data (9). It is worth noting that a

significant proportion of recipients had HLA-C or DQ antibodies. This underlines the

importance of using SAB assay for better definition of the exact specificities in view of

the higher rates of acute rejection episodes especially antibody-mediated acute

rejections and graft losses for immunologic reasons among recipients with

pretransplant DSAs against HLA-C, -DP, and -DQ (14, 16, 17). We do not routinely

perform flow crossmatches because CDC crossmatching and SAB assay are well-

established techniques for transplantation organ assignment, and flow crossmatching is a

150

very sensitive technique that could unnecessarily disadvantage highly sensitized

recipients with a negative CDC crossmatch and low-level DSAs. However, in some of the

patients with DSAs 2000 to 4000 MFI, flow crossmatches were performed and found

to be positive and would therefore suggest that our approach is also a good

predictor of flow crossmatch results.

The threshold of 2000 MFI to exclude donors from matching to recipients with a DSA

seems to be an excellent predictor of negative CDC crossmatch for class I antigens.

This prediction is dependent on the accuracy of HLA antibody testing. Initially, 1

sample was found to be crossmatch positive despite absence of DSAs greater than

2000 MFI. One possible explanation for this result is the presence of blocking IgM

antibodies (18), and, indeed, after IgM depletion, the index serum demonstrated a

significantly higher strength of the DSAs. This observation highlights the need to have

exactly agreed laboratory standards for correct antibody assignment. An alternative

explanation would be the prozone effect (19), where steric hindrance from a high-level

HLA antibody could compete for binding by anti-IgG detection antibody on the SAB.

However, although we do not routinely perform testing to unmask prozone in our

laboratories, this problem seems to be uncommon, because we rarely encounter

positive crossmatches without apparent DSA.

An alternative consideration to allow highly sensitized recipients the chance of live-

donor kidney transplantation is to leave the decision about whether and how to

proceed to the discretion of the transplant center, when low levels of DSA are

amenable to early therapeutic intervention. A review of all crossmatch data available

indicates that approximately 20% of donor-recipient combinations that were excluded

by virtual crossmatching because of DSAs 2000 to 8000 MFI had a negative TCXM and

BCXM, and in most instances, there were only 1 or 2 DSAs at a strength of greater than

4000 MFI. The risk of this approach is that identified matches have a greater risk of

breakdown, because the initial acceptance by the transplant center may later be

withdrawn. The result will be that the other immunologically suitable pairs that were

in the same chain will miss out on the potential organ exchange. Because donors have

to be re-called to provide fresh samples for crossmatches, the psychological and

emotional implications of informing them of a potential match but to be told later that

the match had been undone should not be discounted. In our program, there were 7

pairs (two 2-way and one 3-way chains) who were scheduled for exchange surgeries

and who did not progress to transplantation. Anecdotal reports by clinicians caring for

these patients would indicate significant emotional distress to those pairs who did not

151

progress to transplantation after the initial promise was not fulfilled.

When compared with available data from the first-year activity of other programs in

the Netherlands (1), the UK (3), and Texas (7), the transplant rates of the AKX program

(34% of enrolled pairs) seem to be excellent. Compared with the cumulative number

of enrolled recipients, the transplant rate in the first year was 40% in the Dutch

program (1), 9% in the UK (3), and 14% in the Austin (7). It is worth noting that these

programs differ significantly in the proportion of recipients included because of pure

ABO incompatibility with their coregistered recipients without the presence of

significant HLA sensitizations. Recipients who were ABO incompatible with their

intended donor were 56% in the Dutch program (1), 44% in the UK (3), and 36% in the

Austin (7), compared with only 10% in Australia. On the other hand, in contrast to our

program, neither the Dutch nor the UK program included altruistic donors in their first

year. The inclusion of altruistic donors is an effective approach to facilitate matching in a

KPD program (20). The inclusion of 2 altruistic donors in AKX resulted in a 4-way chain (3

KPD recipients and 1 wait-list recipient) and a 3-way chain (2 KPD recipients and 1 wait-

list recipient). If the altruistic donors were removed, only 2 KPD recipients could have

been matched in a 2-way chain. Thus, the total number of transplants purely by KPD

matching would have been 15, resulting in a transplant rate of 28%.

Early outcomes of the 20 recipients who underwent transplantation were excellent; all

patients left the hospital within the center’s usual length of stay without any surgical

complications or early rejection episodes, with only 1 case of delayed graft function not

requiring dialysis, unrelated to the relatively short cold ischemia of 6 hours of the

shipped kidney. At this stage, there are not sufficient data to comment on long-term

allograft outcomes. However, the observation that with the currently available follow-

up data, a biopsy-proven rejection was found in only 1 recipient, who had a cPRA of

99% and received a kidney from an ABO-incompatible donor, of the 20 patients who

underwent transplantation indicates that avoiding moderate and strong DSAs against

all HLA loci is a valid strategy to minimize the risk of rejection. The major stumbling

block in the AKX program to date has been the failure to progress to transplantation

surgery in 25% of crossmatch-negative matched pairs for non-immunological reasons.

Strategies to minimize the rate of non-conversion of matches to transplantations have

been devised.

In conclusion, high transplant rates in a KPD program using virtual crossmatching can

be achieved even with a small pool consisting largely of highly sensitized recipients,

indicating that KPD is a valid and effective solution for patients with immunologically

152

incompatible donors even in the context of highly sensitized recipients.

MATERIALS AND METHODS

The AKX program was established in August 2010 as a new national initiative to

enhance live kidney donation. Within the first 12 months, 78 incompatible donor-

recipient pairs had been referred to the program. Allocation procedures to match

compatible combinations were scheduled every 3 months. Only pairs who satisfied all

agreed medical criteria for donor and recipient registration, molecular HLA typing, and

HLA alloantibody testing were included in match run procedures. Thus, 53 donor-

recipient pairs were included in any of the 4 match procedures performed between

October 2010 and August 2011. Crossmatches between the new donors and recipients

were performed in the histocompatibility laboratory of the matched donors’ state.

Human Leukocyte Antigen Typing and Detection of Anti-Human Leukocyte Antigen

Antibodies

Donor molecular HLA typing and recipient HLA antibody testing for the loci -A, -B, -C, -

DRB1, -DPB1, -DQB1, and -DRB3/4/5 were performed at the 5 state histocompatibility

laboratories as previously reported (4). Donor HLA typing was performed using direct

DNA sequencing. Recipients were tested for HLA class I and II directed IgG antibody in

their sera using SAB Luminex technology (Luminex, One lambda Inc., Conga Park, CA);

the MFI signal for each SAB was normalized against the negative control to correct for

nonspecific binding.

Calculated Panel-Reactive Antibody

The cPRA of enrolled recipients was based on HLA-A, -B, -DR, and -DQ antigen

frequencies derived from the phenotypes of 200 representative participants in the

Busselton Health Study, which were found to have similar frequencies for HLA-A, -B,

and -DR of donors in the National Organ Matching System (NOMS) registry and the

United Network for Organ Sharing (data not shown). The correlation between the

cPRA derived using phenotypes from our cohort and the cPRA obtained using the

OPTN calculator was R2=0.981. We compared the cPRA distribution for patients who

were registered, were matched, and underwent transplantation.

Computer Program Matching

Allocation of suitable live-donor matches was performed using a software module

developed by the NOMS as previously described (4). The software takes advantage of

advances in HLA technology to enable virtual crossmatching. HLA alleles of each locus

153

and HLA antibody specificities were entered in the computer program at the 4-digit

level. Briefly, the algorithm does not consider any HLA matching rules for allocation,

and matching is only based on acceptable mismatches by excluding donors from

matching to recipients with DSAs greater than 2000 MFI for any class I or II antibodies

against any of the donor HLA loci -A, -B, -C, -DRB1, -DPB1, -DQB1, -DQA1, and -

DRB3/4/5 (4). By default, the program considers exclusively ABO-compatible

matching, but the option of ABO-incompatible matching is available in selected cases.

Complement-Dependent Cytotoxic Crossmatches of Matched Pairs

After suitable matches in a chain were identified and the peak and historical antibody

record was reviewed, TCXM and BCXM between a matched donor and recipient were

performed in the histocompatibility laboratory of the matched donor’s state using

fresh donor cells in duplicate. If all links within a chain had negative CDC crossmatches,

donation procedures in the centers were arranged within 3 months of allocation and

performed simultaneously. Sera from all recipients were distributed 1 week before

match runs 1 and 2, but from match run 3 onward, only sera from matched recipients

were sent to the laboratory of the matched donor. Donors re-called for CDC

crossmatch in runs 1 and 2 were crossmatched against all recipients on the trays to test

the accuracy of the virtual crossmatch in predicting a negative crossmatch in relation to

SAB antibody strength.

ACKNOWLEDGMENTS

The authors thank the collaborators of all participating transplant centers, in particular,

those involved in exchange transplant surgeries at Westmead Hospital, Royal Prince Alfred

Hospital, John Hunter Hospital, Prince of Wales Hospital, Royal North Shore Hospital, and

The Children’s Hospital at Westmead, in New South Wales; Monash Medical Centre and Royal

Melbourne Hospital, in Victoria; and Sir Charles Gairdner Hospital and Royal Perth Hospital,

in Western Australia. The authors also thank Ms Jenni Wright and the NOMS team for their

support throughout the program and Dr Scott Campbell, chair of the National Renal

Transplant Advisory Committee, for his advice and support. Finally, the authors thank the

patients and their donors for agreeing to participate in this pioneering study.

154

REFERENCES

1. de Klerk M, Keizer KM, Claas FH, et al. The Dutch national living donor kidney exchange program.

Am J Transplant 2005; 5: 2302.

2. Montgomery RA, Zachary AA, Ratner LE, et al. Clinical results from transplanting

incompatible live kidney donor/recipient pairs using kidney paired donation. JAMA 2005;

294: 1655.

3. Johnson RJ, Allen JE, Fuggle SV, et al. Early experience of paired living kidney donation in the

United Kingdom. Transplantation 2008; 86: 1672.

4. Ferrari P, Fidler S, Wright J, et al. Virtual crossmatch approach to maximize matching in

paired kidney donation. Am J Transplant 2011; 11: 272.

5. de Klerk M, Witvliet MD, Haase-Kromwijk BJ, et al. Hurdles, barriers, and successes of a

national living donor kidney exchange program. Transplantation 2008; 86: 1749.

6. Hanto RL, Reitsma W, Delmonico FL. The development of a successful multiregional kidney

paired donation program. Transplantation 2008; 86: 1744.

7. Bingaman AW, Wright FH, Murphey CL. Kidney paired donation in live-donor kidney

transplantation. N Engl J Med 2010; 363: 1091.

8. Ferrari P, Woodroffe C, Christiansen FT. Paired kidney donations to expand the living donor

pool: the Western Australian experience. Med J Aust 2009; 190: 700.

9. Morris GP, Phelan DL, Jendrisak MD, et al. Virtual crossmatch by identification of donor-

specific anti-human leukocyte antigen antibodies by solid-phase immunoassay: a 30-month

analysis in living donor kidney transplantation. Hum Immunol 2010; 71: 268.

10. Lefaucheur C, Loupy A, Hill GS, et al. Preexisting donor-specific HLA antibodies predict outcome

in kidney transplantation. J Am Soc Nephrol 2010; 21: 1398.

11. Flint SM, Walker RG, Hogan C, et al. Successful ABO-incompatible kidney transplantation

with antibody removal and standard immunosuppression. Am J Transplant 2011; 11: 1016.

12. Montgomery JR, Berger JC, Warren DS, et al. Outcomes of ABO- incompatible kidney

transplantation in the United States. Transplantation 2012; 93: 603.

13. Gloor J, Stegall MD. Sensitized renal transplant recipients: current protocols and future

directions. Nat Rev Nephrol 2010; 6: 297.

14. Ntokou IS, Iniotaki AG, Kontou EN, et al. Long-term follow up for anti-HLA donor specific

antibodies postrenal transplantation: high immunogenicity of HLA class II graft molecules.

Transpl Int 2011; 24: 1084.

15. Montgomery RA, Lonze BE, King KE, et al. Desensitization in HLA- incompatible kidney recipients

and survival. N Engl J Med 2011; 365: 318.

16. Gilbert M, Paul S, Perrat G, et al. Impact of pretransplant human leukocyte antigen-C

and -DP antibodies on kidney graft outcome. Transpl Proc 2011; 43: 3412.

17. Tran TH, Dohler B, Heinold A, et al. Deleterious impact of mismatching for human leukocyte

155

antigen-C in presensitized recipients of kidney transplants. Transplantation 2011; 92: 419.

18. Kosmoliaptsis V, Bradley JA, Peacock S, et al. Detection of immunoglobulin G human

leukocyte antigen-specific alloantibodies in renal transplant patients using single-antigen-

beads is compromised by the presence of immunoglobulin M human leukocyte antigen-

specific alloantibodies. Transplantation 2009; 87: 813.

19. Schnaidt M, Weinstock C, Jurisic M, et al. HLA antibody specification using single-antigen

beads: a technical solution for the prozone effect. Transplantation 2011; 92: 510.

20. Roodnat JI, Zuidema W, van de Wetering J, et al. Altruistic donor triggered domino-paired

kidney donation for unsuccessful couples from the kidney-exchange program. Am J

Transplant 2010; 10: 821.

156

Chapter 7

Transnational validation of the

Australian algorithm for virtual

crossmatch allocation in kidney paired

donation

157

Transnational validation of the Australian algorithm for virtual crossmatch

allocation in kidney paired donation

Georg A Böhmig, Samantha Fidler, Frank T Christiansen, Gottfried Fischer, Paolo Ferrari

Validation of the negative predictive value of the Australian paired kidney exchange (AKX)

program was achieved by the crossmatching of identified pairs. An unexpected positive

crossmatch was rare in the first year of the program, however whether further matches

with negative crossmatches were being excluded by the low threshold for antibody

exclusion was unclear. The transplant unit in Vienna, Austria supplied HLA antibody and

typing data from sixteen pairs in their recently developed paired exchange program. The

Austrian program matches compatible pairs based on a negative complement-dependent

and flow crossmatch, with donor specific HLA antibody with a mean fluorescence

intensity (MFI) less than 2000. The recipient HLA antibody data and donor HLA typing of

the Austrian pairs was matched using the Australian kidney exchange (AKX) computer

algorithm in order to compare virtual and actual crossmatch results. This paper reviews

the match and virtual crossmatch results in the Austrian paired kidney exchange

cohort.


Human Immunology 2013: 74; 500-505

Contributions:

PF Prepared the manuscript, participated in data analysis

SJF Performed all computer matching, participated in study

design, participated in data collection, participated in data

analysis, participated in writing the manuscript

GB, GF Participated in study design, participated in data collection

FTC Participated in study design

158

Human Immunology 74 (2013) 500–505

Transnational validation of the Australian algorithm for virtual

crossmatch allocation in kidney paired donation

Georg A. Böhmig a, Samantha Fidler b,c, Frank T. Christiansen b,c, Gottfried Fischer d, Paolo Ferrari e,f,⇑ a Division of Nephrology and Dialysis, Medical University Vienna, Vienna, Austria

b Department of Immunology, PathWest, Royal Perth Hospital, Perth, Western Australia, Australia

c School of Pathology and Laboratory Medicine, University of Western Australia, Perth, Western Australia, Australia

d Department for Blood Group Serology, Medical University Vienna, Vienna, Austria

e Department of Nephrology, Fremantle Hospital, Perth, Western Australia, Australia

f School of Medicine and Pharmacology, University of Western Australia, Australia


Article history:

Received 1 November 2012

Accepted 24 January 2013

Available online 1 February 2013

Abbreviations: AKX, Australian paired Kidney eXchange; BCXM, B-cell crossmatch; PRA,

Panel-reactive antibody; DSA, Donor-specific antibodies; KPD, Kidney paired donation;

MFI, Mean fluorescence intensity; NOMS, National Organ Matching System; TCXM, CDC

T-cell crossmatch.

⇑ Corresponding author at: Department of Nephrology, Fremantle Hospital, Perth,

Western Australia 6160, Australia. Fax: +61 8 9431 3619.

E-mail address: [email protected] (P. Ferrari).

a b s t r a c t

An independent pool of 16 incompatible live donor–recipient pairs registered

at the Vienna transplant unit was applied to test whether virtual crossmatch

allocation used in the Australian kidney paired donation (KPD) program reliably

predicts negative crossmatches. High resolution HLA data were entered into

the computer-matching algorithm and allocation was performed excluding any

www.ashi-hla.org

journal homepage: www.

elsevier.com/locate/ humimm

http://dx.doi.org/10.1016/j.humimm.2013.01.029


http://www.ashi-hla.org/

http://www.elsevier.com/locate/humimm

http://www.elsevier.com/locate/humimm

159

DSA > 2000MFI. CDC and flow crossmatch data of recipients against any of the

donors were available for 112 crossmatch combinations. The computer

program identified 19 possible pairings in 2-way or 3-way chains in multiple

combinations. The top ranked combination included one 3-way and two 2-way

ABO-compatible chains. Where crossmatches were available all recipients

were CDC crossmatch negative with the computer-matched donor. Excluding

allocation of KPD donors in the presence of DSA > 2000MFI had a negative

predictive of 99.9% for CDC and 96.4% for flow crossmatch. In the 12 pairings

with ≥1 DSA against crossmatched donors there was a negative CDC and flow

crossmatch. These results show excellent correlation between matching using

virtual crossmatch and actual crossmatch results. Using the 2000MFI cut-off

the number of potentially unacceptable CDC and flow crossmatch positive

pairings identified by virtual crossmatching is low, but some potential

crossmatch negative pairings are missed.

1. Introduction

Live kidney donation is an important source of kidney transplantations in many

countries. When live donors are unable to donate to their intended recipients

because of blood group or HLA-incompatibility, kidney paired donation (KPD) can

help overcome this barrier by pairing two or more incompatible pairs together to

facilitate the exchange of live donor kidneys between the willing donors [1–4]. The

rules for allocation of suitable pairings are paramount for the success of a KPD

program. We previously reported on the allocation algorithm of the Australian

paired Kidney eXchange (AKX) program [4], which was based on simulations using

different algorithms and data from a previous single center KPD experience [5]. Our

proposed approach was to ignore any HLA antigen matching rules and to use

matching based on acceptable mismatches by excluding donors from matching to

recipients with donor-specific antibodies (DSA) > 2000 mean fluorescence

intensity (MFI) for any class I or II antibodies detected by Luminex single-antigen

bead (SAB), in order to avoid the breakdown of identified chains due to positive

crossmatches [6] and to reduce the risk of inferior long-term outcome in the

presence of strong DSA in patients with negative crossmatch at transplantation

[7]. The proposed approach has been criticized [8], on the basis that the ability of

solid phase HLA antibody assays to provide quantifiable data are controversial and

160

although we suggested that the flow cytometric crossmatches may have been

positive in some cases, no actual flow cytometric crossmatches were performed [4].

Further validation of the proposed virtual crossmatch allocation based on exclusion

of unacceptable HLA antibodies by solid phase is therefore required.

The Vienna renal transplant service is in the preliminary stages of introducing a

KPD program to the country including the establishment of efficient serological

allocation strategies to accurately predict crossmatch outcomes and select HLA

antibody-compatible 2- to 3-way chains. The aim of this study was to compare if

the virtual crossmatch approach using the Australian KPD computer allocation

would predict negative crossmatches using the Austrian pool and whether these

stringent criteria would exclude potential crossmatch negative pairings to be

considered for matching.

2. Subjects and methods

2.1. Study subjects

As of September 2011, 16 incompatible donor–recipient pairs were registered at

the Vienna transplant unit. Baseline characteristics of donors and recipients are

given in Table 1. Recipients’ panel-reactive antibody (PRA) values for HLA class I

specificities were directly calculated according to the results obtained by CDC

screening against a panel of 50 HLA-typed volunteers using peripheral blood

mononuclear cells. Because many recipients had class II HLA antibodies and the

occurrence of antibody against HLA-DQ was high in recipients with previous

transplant, recipients’ calculated PRA (cPRA) for HLA-A, -B, -DR and -DQ antigen

was derived based on antigen frequencies from the UNOS registry for Caucasians

[9].

161

2.2. Solid phase antibody detection

All laboratory evaluations were performed at the Vienna unit as described

elsewhere [10]. Donor molecular HLA-typing and recipient HLA antibody testing

for HLA-A, -B, -C, -DRB1, -DPB1, -DQB1 and DRB3/4/5 were performed in

accordance with the standards used in the AKX program as previously reported

[4]. Donor HLA- typing was performed using direct DNA sequencing. Recipients

were tested for HLA class I and class II directed IgG antibody in their sera

using single antigen bead (SAB) technology (One Lambda, Canoga Park, CA, USA)

on a Luminex platform performed on neat, non-IgM depleted samples. The MFI

signal for each SAB was normalized against the negative control to correct for non-

specific binding. Recipients without DSA (<500MFI) or with only weak DSA

(<2000MFI) were considered to have no HLA antibodies against matched donors

(Luminex negative).

2.3. Complement-dependent cytotoxic (CDC) and flow cytometry crossmatches

After what was considered a sufficiently large pool of blood group or HLA-

incompatible pairs who joined the Vienna program, all recipients underwent T-

and B-CDC (no enhancement) and T- and B-cell flow crossmatching against any of

the donors in the program. However, some of the patients were only typed and

screened because they joined the register after the initial series of crossmatches.

Thus, full crossmatch data (CDC and flow) of recipients against any of the donors,

excluding co-registered donor–recipient pairs, was available for a total of 112

crossmatch combinations. No crossmatch data were available for 4 recipients. The

crossmatch data were kept blinded to the Australian team running the computer

matching until the matrix and potential KPD chains were identified.

2.4. Computer program matching

Allocation of suitable live donor matches was performed using a software module

developed by the National Organ Matching System (NOMS) as previously

described [4]. HLA data of all Austrian donors and recipients was entered into the

NOMS computer module. HLA-alleles of each locus and HLA antibody specificities

were entered in the computer program at the 4-digit level. The computer program

162

enables virtual crossmatching by matching recipients to suitable donors based on

acceptable mismatches. The algorithm does not consider any HLA antigen

matching rules for allocation and donors are excluded from matching to

recipients with DSA > 2000MFI for any class I or II antibodies against any of the

do- nor HLA-A, -B, -C, -DRB1, -DPB1, -DQB1, -DQA1 and -DRB3/4/5 loci [4]. For KPD

allocation purposes only HLA antibody with strength >2000MFI were authorized

for exclusion from matching to donors with the target HLA antigen. A cumulative

approach of total DSA strength was not considered for KPA allocation. By default

the program now considers ABO-incompatible matching, but ABO compatible

matching only is available on demand. The computer program selects between

competing match offers based on prespecified ranking rules, such as favouring 3-

way over 2-way exchanges, in order to maximize the number of patients receiving

a transplant, and then favouring patients with low versus high match probability,

which primarily gives an advantage to recipients with a higher PRA. The computer

program also excludes previous donor HLA-mismatches or donor–donor age

difference of >30 years.

2.5. Comparison of virtual and actual crossmatch results

The probability of a positive crossmatch in relation to the strength of DSA for class I and

class II DSA was analysed for both T- and B-cell CDC and flow crossmatches. For donor-

recipient combinations where the recipient had ≥1 DSA the cumulative strength of DSA

was determined by adding the MFI for each DSA. The cumulative strength of DSA was

grouped in five categories, 0-2000 (DSA negative), 2000-4000, 4000-10000, 10000-20000

and >20000MFI. Because T cells do not constitutively express HLA class II, the results for

class II DSA were compared to the B-cell crossmatch results only. B cells express both HLA

class I and II so a positive B-cell crossmatch may be due to antibodies directed against

HLA class I or II or both. Therefore data for class I DSA was compared to the results

of both T- and B-cell crossmatches.

163

2.6. Statistical analysis

The sensitivity and specificity of predicting positive crossmatch results for pairings

matched in the presence of DSA and the predictive value for negative

crossmatches for pairings matched without DSA was estimated for two cut-off

threshold of 2000MFI and 4000MFI, respectively [11].

3. Results

3.1. Computer matching

Most pairs (69%) were included because of HLA-incompatibility, five pairs were

ABO-incompatible with a PRA of <5%, the proportion of blood group O donors and

recipients (31%) was identical (Table 1). Of the 240 possible one-way combinations

achievable with 16 pairs (i.e. (N-1) x N), the computer program identified 75

suitable one-way donor-recipient matches based on the virtual crossmatch

criteria. Six recipients did not have any match because they had 1 or more DSA

against any of the donors in the pool. The one-way matches resulted in 19 possible

pairings in 2-way or 3- way chains in multiple combinations. Without consideration

of crossmatch results the top ranked combination where each pair appeared only

once in any chain included one 3-way and two 2- way ABO-compatible chains

(chain 5, 9 and 28). The second ranked combination included again the same 3-

way chain and two different 2-way chains (chain 5, 15 and 22), although in one

case an acceptable match was found between an AB donor and an A recipient. In

all instances where crossmatch data were available, these recipients were

complement-dependent cytotoxic (CDC) and flow-crossmatch negative with the

computer-matched donor (P < 0.001).

164

Table 1. Baseline characteristics of donors and recipients in the 16 pairs of the

Vienna transplant center.

3.2. Analysis of available HLA antibodies and crossmatch data

All recipients had complete Luminex HLA data and 12 had complete or partial

crossmatch data, but no crossmatch data were available for four recipients

who were listed at a later stage. Excluding crossmatch results between co-

registered donors and recipients, 112 full CDC and flow crossmatch data were

available for analysis. Of these, 28 pairings were DSA negative and 84 had at

least 1 DSA against their crossmatched donors, 61 had both ≥1 class I and ≥1

class II DSA, 3 pairings had ≥1 class I DSA only, and 20 had ≥1 class II DSA only.

In the 28 DSA negative pairings there were 27 with both negative flow and CDC

crossmatches and 1 with positive flow, but negative CDC crossmatch (Table 2).

165

In the DSA positive pairings 41 had a positive and 43 had a negative CDC

crossmatch. Of those with positive CDC crossmatch 40 had both positive flow

and CDC crossmatches. In the 43 DSA positive and CDC negative, 30 had positive

flow and 13 had both negative CDC and flow crossmatches (Table 2). In these

cases 2 or more DSAs were present in only 18% of pairings, compared to ≥2 DSA

in 80% of those with positive crossmatch. Moderate DSA in the range of 2000–

4000MFI were present in 8 crossmatch negative Luminex positive pairings.

Interestingly, all these cases only had class II DSA with specificities to DP and DQ

antigens against the crossmatched donor.

In the pairings without crossmatch data, 68 had 1 or more DSA against any of

the donors in the pool and 60 were Luminex DSA negative. The total number of

DSA negative pairings with or without crossmatch data were 88, which is

higher than the total number of DSA negative pairings identified by the

computer matching algorithm (N = 75), because the computer program

excludes previous mismatches or donor–donor age difference of >30 years.

Table 2. CDC and flow cytometry crossmatch results from 112 donor-recipient

pair combinations where crossmatch data were available in relation to the

presence or absence of any donor-specific antibodies (DSA) at >2000MFI.

166

3.3. Value of DSA strength to predict crossmatch results

The probability of a positive T- or B-cell CDC or flow crossmatch in relation to DSA

strength by class I and class II antibodies is shown in Fig. 1. The sensitivity and

specificity of any DSA positive (class I and/or II) computer matching to predict a

positive (T- and/ or B-cell) crossmatch and the negative predictive value of DSA

negative matching to identify crossmatch negative results is reported in Table 3.

The cutoff of 2000MFI for DSA to exclude recipients from matching to donors with

unacceptable antigens had a sensitivity of 98.6% and a specificity of 69.2% for a

positive CDC and flow crossmatch result (Table 3). If only a positive CDC

crossmatch result was considered sensitivity was 97.6% and a specificity of 38.6%.

If the cut off for DSA to exclude recipients from matching to donors with

unacceptable antigens were increased to 4000MFI sensitivity would be 97.6% and

a specificity of 89.7% for a positive CDC and flow crossmatch result. The negative

predictive value for both negative CDC and flow crossmatches was 99.9% using the

2000MFI cut off. If the cut off were increased to 4000MFI negative predictive

value for negative CDC and flow crossmatch would be 94.4% (Table 3). The

relationship between number of DSA and crossmatch results was also analyzed.

There was however no ‘dose–response’ relationship between increasing number

of DSA and crossmatch positivity. For instance, for CDC T-cell crossmatch a positive

result was found in 50%, 55%, 46%, 67% if there were 1, 2, 3 or ≥4 DSA

respectively. If there were no DSA the crossmatch was negative in 99% of cases.

The MFI for each DSA in pairings with 1–2 DSA and a positive crossmatch were

generally high (>10000MFI), whereas some pairings with ≥4 DSA as a cause of

positive had multiple moderate DSA.

167

Table 3. Sensitivity and specificity of donor-specific antibody (DSA) positive

computer matching to predict a positive crossmatch and negative predictive

value of DSA negative matching to identify crossmatch negative results using

virtual crossmatching in kidney paired donation.

Estimations were calculated using two cut-offs of Luminex HLA antibodies to exclude

recipients from matching to incompatible donors. Results are in % (and confidence

interval).

Figure 1. Probability of a positive crossmatch in relation to the cumulative

strength of donor-specific antibodies (DSA) in mean fluorescence intensity (MFI)

for class I (panel A) and class II (panel B) DSA.

The top panels show results for CDC crossmatches, the bottom panels show data for flow

crossmatch results. The results for class II DSA are reported for B-cell crossmatches only,

because T cells do not constitutively express HLA class II antigens.

168

4. Discussion

This analysis provided a unique opportunity to evaluate a KPD allocation algorithm

that utilizes virtual crossmatch strategy based on the allocation of recipients to

suitable donors that takes into consideration acceptable mismatches based on Luminex

HLA antibody assignment in an attempt to address the concerns previously raised by

Bingaman et al. [8] with regard to the proposed virtual allocation for KPD [4]. The present

findings demonstrate that the Australian KPD allocation algorithm allows computer

matching of live donors to recipients that will result in nearly 99% of identified matches

having a negative CDC and flow crossmatch. On the other hand this high sensitivity comes

at a cost of a low specificity, with many Luminex positive patients with DSA above the

threshold used for exclusion of matching that could potential have a negative CDC and/or

flow crossmatch. Although a weakness of this study relates to the nature of the

population studied, which is small and with only a modest degree of sensitization

based on cytotoxic PRA for class I only, the immunological profile of our recipients

compares favourably with cohorts of other national programs [3,12,13]. Many

recipients from the Vienna center were sensitized against class II HLA antibody

and not surprisingly anti- DQ antibody were frequently observed in recipients with

previous transplant resulting in a cPRA of 66 ± 41% and half of the recipients having

a cPRA >90%. In the Austrian program 75% of recipients were enrolled because of

HLA incompatibility to their co-registered donor, which is similar to the rate

observed in the Australian program (85%) [13], and significantly higher than the

rate observed in the Dutch (44%) [12] and the UK (56%) programs [3].

In recent years several algorithms for allocation of KPD pairs have been developed

to select possible exchanges based on a points scoring system that includes the

number of HLA-mismatches, % PRA, distance between the exchange centers,

donor age differences, CMV status and other factors [3,14,15]. The sensitivity and

specificity of these algorithms has not been widely tested and most of the

information on expected matched rates in relation to the number of registered

donor–recipients pairs have been based on simulation [3,15]. Once the computer

finds a match, crossmatches between matched donors and recipients are still

required before proceeding to transplantation and not infrequently even a single

unpredicted positive crossmatch prevents two or more pairs progressing to kidney

transplantation [16]. Because CDC and flow crossmatches are both cost and labor

intensive, if a recipient has a positive crossmatch with a computer matched donor,

additional crossmatches for this recipient against any other donors whose cells

169

may be available are usually not performed and/or reported.

The advantage of the cohort we used in our study is that for a large number of

recipients, both CDC and flow crossmatches had been performed in advance without

consideration of any putative compatibility. This has allowed us to assess whether the

allocation rules adopted for the Australian KPD program, and that were mainly based

on simulated data, are appropriate. The Australian matching algorithm uses high

definition of recipient HLA antibodies and donor HLA antigens to match KPD pairs with

the objective to achieve the maximum number of possible matches with predicted

negative crossmatch [4]. The Luminex threshold of 2000MFI was chosen because in

our experience a CDC crossmatch between a donor and recipient with a weak DSA is

almost invariably negative. If no DSA > 2000MFI was present, the probability of a

positive CDC crossmatch was negligible and the probability of a positive flow

crossmatch was 4%. The specific reason for a positive flow in the absence of DSA was

not investigated. Clearly the sensitivity of flow cytometry is higher than CDC

crossmatching [17], and positive flow crossmatches are not unusual in the presence of

a weak DSA (500–2000MFI). An alternative explanation is that some donor HLA alleles

are not represented in the bead panels and our algorithm does not take into account

serotypes for missing alleles. There was a clear dose–response relationship between

the cumulative DSA strength and the probability of a positive B-cell crossmatch for

class II DSA. Although not so clear-cut, this relationship was also present for the

association between cumulative strength of class I DSA and T- or B-cell crossmatches.

Interestingly, the difference in sensitivity values between CDC + flow results and CDC

only was not statistically significant (Table 3), whereas the difference in specificity

values between CDC + flow results and CDC results only was highly significant. The

present data indicates that using a threshold of 2000MFI will predict a negative CDC

crossmatch in virtually all of cases and rising the threshold to 4000MFI it would still

provide a negative predictive value of 97.2%. Thus, this prudent approach seems to

meet the objective to minimize the number of positive CDC crossmatches.

Attempts to define a cut off for Luminex technology, allowing the prediction of a

positive cross-match using either CDC or flow crossmatch have been hampered by the

complexity of the problem and differences in experimental design [18–20]. Our

data, along with the recent findings reported by Moreno et al. [21], contribute in

improving our understanding of the clinical predictive value of HLA antibodies by

Luminex technology in the context of virtual crossmatch. One drawback of the

current allocation rules of the Australian algorithm is that it doesn’t use different

170

values for HLA-C, DRB3/4/5, DQ and DP compared to HLA-A, B, and DRB1.

However, based on correlation with crossmatch results and density of antigen on

the beads these should be treated differentially. The observation in our cohort

that eight crossmatch negative cases had moderate DSA in the range of 2000–

4000MFI and all these cases had exclusively class II DSA with specificities to DP and

DQ antigens corroborates this view. Indeed, if anti-HLA antibody directed against

the HLA-C, DRB3/4/5 and DP locus were ignored the specificity of DSA positive

computer matching to predict a positive crossmatch would increase, without

affecting the negative predictive value of DSA negative matching to identify

crossmatch negative results.

For patients with high-level HLA-sensitisation the option of HLA-incompatible live

kidney transplantation [7,22,23] is often limited by the presence of multiple and

high-titre DSA that are not amenable to desensitisation. These patients also have

exceedingly long waiting time on the deceased donor list. Since many programs do

transplant patients with DSA ranging from 2000 to 8000MFI when the actual

crossmatch is negative, our approach could disadvantage such patients by

excluding crossmatch compatible donors and inappropriately deny a patient

access to transplantation.

The task of assigning an antigen as unacceptable in a KPD program where multiple

centers with differing expertise in managing immunological complex patient are

involved is daunting. Cut-offs must be sufficiently stringent to prevent positive

crossmatches, but should not be so rigid that they disallow transplants that some

centers would consider compatible. One interesting observation is that in pairings

with DSA in the 2000–4000MFI range a significant proportion of CDC and also flow

crossmatches have a negative result. Thus, by increasing the allocation cut-off to

4000MFI in the present cohort, 65% of the Luminex positive, but crossmatch

negative pairings would have been matched in the KPD allocation, increasing the

pool size of one-way donor–recipient pairs to be linked in chains and combinations

by 25%. This increase in the cut-off level appears to be associated with an

acceptable predicted rate of positive CDC crossmatches (rv3%), but undoubtedly

higher rates of positive flow crossmatches (rv10%). While transplantation will

likely not proceed in the setting of a positive CDC in the presence of a DSA,

desensitisation to enhance transplant rates using donor exchange may be

171

considered in the setting of a single weak-to-moderate DSA and positive flow

crossmatch [24]. Cut-offs may need to be tailored to the level of HLA sensitisation

and blood group distribution of recipients registered in the program. Continuous

adjustment of the cut-off may be subject to regulatory control in a national or

regional program and thus may be difficult to achieve. Such decision must also

take account of the likelihood of long-term allograft outcome with increasing

evidence that solid phase assays, even in the absence of a positive crossmatch is

associated with poorer long-term outcome [7,25,26]. Even if our exact same

protocol was used it is likely that each laboratory will need to define their own

clinically relevant MFI threshold, because of issues such as batch-to-batch

variation in SABs, difference in antibody affinity, DSA definition using C1q-assay

and other sources of variation. Ultimately the main problem with using virtual

cross match in highly sensitized patients is that solid phase assays at present still

do not predict complement binding antibodies. This problem could be overcome

by newer assays that may detect complement [10,27], therefore improving the

utility of the virtual crossmatch. We recently demonstrated that C4d-fixation to

detect complement-activating DSA significantly improved predictive accuracy of

CDC crossmatch, with better performance for T-cell than B-cell crossmatch [10].

Avoiding antibody is important, but there may be a greater survival benefit for a

highly sensitized recipient of a living donor kidney who has a negative crossmatch,

but positive DSA. Thus, an alternative consideration to allow highly sensitized

recipients the chance of live donor kidney transplantation is to leave the decision

as to whether and how to proceed to the discretion of each transplant center,

when low levels of DSA are amenable to early therapeutic intervention, with the

goal of finding a better immunologic match rather than a completely compatible

pairing. The risk of this approach is that identified matches have a greater risk of

breakdown, because the initial acceptance by the transplant center may later be

withdrawn. However, the likelihood of a chain breakdown would still be lower

than the predicted rate of positive crossmatches if all pairings were allowed at a

higher MFI cut-off.

The aim of this study was to identify a strategy to optimize virtual crossmatch

allocation for KPD using solid phase assay technology. Clearly, the role of DSA in

the context of crossmatch results cannot be dissociated from antibody-mediated

172

rejection and long-term survival [7], and the limitation of the current analysis is

that it does not address these points because it is prediction without

transplantation. However, maximizing the number of crossmatch negative

matches, while minimizing the risk of breakdown of chains is an important

measure to reduce the emotional burden that KPD pairs are exposed when a

match is proposed, but later cancelled because of positive crossmatch results.

In conclusion, these results show excellent correlation between matching using

virtual crossmatch and actual crossmatch results. The number of potentially

acceptable CDC crossmatch negative pairings excluded by virtual crossmatching is

approximately 10%, suggesting that the Australian algorithm won’t disallow too

many potential compatible transplants.

Disclosure

The authors of this manuscript have no conflicts of interest to disclose as

described by the Journal.

Acknowledgments

The authors wish to acknowledge the support of the Australian Organ and Tissue

Donation and Transplantation Authority for the establishment of the National

Paired Kidney Donation program and funding of part of the staff (PF, SF). We wish

to thank Elisabeth Lehner for her valuable contribution to patient management

and data collection.

173

References

[1] de Klerk M, Keizer KM, Claas FH, Witvliet M, Haase-Kromwijk BJ, Weimar W. The

Dutch national living donor kidney exchange program. Am J Transplant 2005; 5: 2302.

[2] Montgomery RA, Zachary AA, Ratner LE, Segev DL, Hiller JM, Houp J, et al. Clinical

results from transplanting incompatible live kidney donor/recipient pairs using kidney

paired donation. JAMA 2005; 294: 1655.

[3] Johnson RJ, Allen JE, Fuggle SV, Bradley JA, Rudge C. Early experience of paired living

kidney donation in the United Kingdom. Transplantation 2008; 86: 1672.

[4] Ferrari P, Fidler S, Wright J, Woodroffe C, Slater P, Van Althuis-Jones A, et al. Virtual

crossmatch approach to maximize matching in paired kidney donation. Am J


[5] Ferrari P, Woodroffe C, Christiansen FT. Paired kidney donations to expand the living

donor pool: the Western Australian experience. Med J Aust 2009; 190: 700.

[6] Morris GP, Phelan DL, Jendrisak MD, Mohanakumar T. Virtual crossmatch by

identification of donor-specific anti-human leukocyte antigen antibodies by solid-

phase immunoassay: a 30-month analysis in living donor kidney transplantation. Hum

Immunol 2010; 71: 268.

[7] Lefaucheur C, Loupy A, Hill GS, Andrade J, Nochy D, Antoine C, et al. Preexisting

donor-specific HLA antibodies predict outcome in kidney transplantation. J Am Soc

Nephrol 2010; 21: 1398.

[8] Bingaman AW, Bray RA, Gebel HM, Newell KA. Paired kidney donation: a (virtual)

balancing act. Am J Transplant 2011; 11: 194.

[9] Cecka JM, Kucheryavaya AY, Reinsmoen NL, Leffell MS. Calculated PRA: initial results

show benefits for sensitized patients and a reduction in positive crossmatches. Am J


[10] Wahrmann M, Hlavin G, Fischer G, Marinova L, Schwaiger E, Hörl WH, et al.

Modified solid-phase alloantibody detection for improved crossmatch prediction.

Hum Immunol 2013; 74: 32.

[11] Mackinnon A. A spreadsheet for the calculation of comprehensive statistics for the

assessment of diagnostic tests and interrater agreement. Comput Biol Med 2000; 30:

127.

174

[12] de Klerk M, Witvliet MD, Haase-Kromwijk BJ, Claas FH, Weimar W. Hurdles, barriers,

and successes of a national living donor kidney exchange program. Transplantation

2008; 86: 1749.

[13] Ferrari P, Fidler S, Holdsworth R, Woodroffe C, Tassone G, Watson N, et al. High

transplant rates of highly-sensitised recipients with virtual crossmatching in

kidney paired donation. Transplantation 2012; 94: 744

[14] Keizer KM, de Klerk M, Haase-Kromwijk BJ, Weimar W. The Dutch algorithm for

allocation in living donor kidney exchange. Transplant Proc 2005; 37: 589.

[15] Saidman SL, Roth AE, Sonmez T, Unver MU, Delmonico FL. Increasing the

opportunity of live kidney donation by matching for two- and three-way exchanges.

Transplantation 2006; 81: 773.

[16] Miyata R, Shimazu M, Tanabe M, Kawachi S, Hoshino K, Wakabayashi G, et al.

Clinical characteristics of thrombotic microangiopathy following ABO incompatible

living donor liver transplantation. Liver Transplant 2007; 13: 1455.

[17] Vaidya S, Partlow D, Susskind B, Noor M, Barnes T, Gugliuzza K. Prediction of

crossmatch outcome of highly sensitized patients by single and/or multiple antigen

bead luminex assay. Transplantation 2006; 82: 1524.

[18] Setoguchi K, Ishida H, Shimmura H, Shimizu T, Shirakawa H, Omoto K, et al. Analysis

of renal transplant protocol biopsies in ABO-incompatible kidney transplantation. Am

J Transplant 2008; 8: 86.

[19] Tanabe K. Interinstitutional variation in the measurement of anti-A/B antibodies:

the Japanese ABO-incompatible transplantation committee survey. Transplantation

2007; 84: S13.

[20] Tanabe K. Japanese experience of ABO-incompatible living kidney transplantation.

Transplantation 2007; 84: S4.

[21] Roth AE, Sonmez T, Unver MU. Kidney paired donation with compatible pairs.

Am J Transplant 2008; 8: 463.

[22] Gloor J, Stegall MD. Sensitized renal transplant recipients: current protocols and

future directions. Nature Rev Nephrol 2010; 6: 297.

[23] Ntokou IS, Iniotaki AG, Kontou EN, Darema MN, Apostolaki MD, Kostakis AG, et

al. Long-term follow up for anti-HLA donor specific antibodies postrenal

transplantation: high immunogenicity of HLA class II graft molecules. Transplant

Int 2011; 24: 1084.

175

[24] Montgomery RA, Lonze BE, Jackson AM. Using donor exchange paradigms with

desensitization to enhance transplant rates among highly sensitized patients. Curr

Opin Organ Transplant 2011; 16: 439.

[25] Gebel HM, Moussa O, Eckels DD, Bray RA. Donor-reactive HLA antibodies in renal

allograft recipients: considerations, complications, and conundrums. Hum Immunol

2009; 70: 610.

[26] Gupta A, Sinnott P. Clinical relevance of pretransplant human leukocyte antigen

donor-specific antibodies in renal patients waiting for a transplant: a risk factor.

Hum Immunol 2009; 70: 618.

[27] Chen G, Sequeira F, Tyan DB. Novel C1q assay reveals a clinically relevant subset

of human leukocyte antigen antibodies independent of immunoglobulin G

strength on single antigen beads. Hum Immunol 2011; 72:849.

176

Chapter 8

Discussion and conclusions

177

8 Discussion and conclusions

The aims of this thesis were to evaluate the clinical relevance of preformed antibodies

directed against HLA detected by Luminex bead assay in the absence of a positive CDC

crossmatch in renal allograft transplantation. Furthermore, whether this data could be

utilized prospectively in clinical practice to facilitate transplantation was studied. This

chapter summarises the key findings of the studies presented in this thesis and will

address further research in these areas.

8.1 The significance of preformed anti-HLA antibodies detected by Luminex bead assay

The Western Australian tissue typing laboratory has been performing pre-transplant

antibody detection and analysis by Luminex bead assay since 2004. This assay has several

key advantages such as increased sensitivity and the ability to accurately determine

specificity including, importantly, definition of antibodies against class II HLA antigens.

There is, however, some controversy over whether antibodies detected by the more

sensitive solid phase assays are clinically significant when found with a negative

crossmatch. Our study has shown a high risk (HR 6.6, 95%CI 2.9-14.7, p<0.001) of AMR in

patients with DSA detected by Luminex single antigen bead assay. A number of studies

have investigated the association of Luminex detected DSA with kidney graft outcome and

are summarised in chronological order in Table 1164-177. Consistent with our findings,

several groups have demonstrated a significant correlation between the presence of pre-

transplant DSA by Single Antigen bead assay and AMR164, 167,168, 170, 172-177. However, in

contrast, Phelan et al reported no AMR in their small cohort of 64 patients165 whilst Gupta

et al could not demonstrate a significant association of acute rejection with DSA at 1 and 5

years follow-up100.

We found the risk of AMR was increased substantially (HR 10.1, 95%CI 3.5-28.2, p<0.001)

in patients with both class I and class II antibodies. Otten et al reported similar findings

with a 30% graft survival at 10 years in those with both class I and II DSA compared to 72%

graft survival in patients with no DSA171. However, this group found no association of class

I or class II DSA alone, whilst we were able to demonstrate a significant association with

both (Class I HR 4.4, 95%CI 1.3-15.5, p=0.021; Class II HR 6.1, 95%CI 1.7-21.4, p=0.005).

178

Table 1. Summary of studies investigating the clinical significance of HLA antibodies

detected pre-transplant by Luminex bead assay.

Authors and year of

publication

Number in

study / DSA

positive

Incidence of

AMR

(No DSA vs DSA)

P value Graft survival (No DSA vs DSA)

P value

Vlad et al. 2009164

325 / 27 6.7% vs 45% P<0.0001 90% vs 81% (1yr) P=0.21

Gupta and Sinnott 2009

165

121 / 16 Not reported 92% vs 86% (1yr) 77% vs 62% (5yr)

NS NS

Phelan et al. 2009166

64 / 12 7.6% vs 8.3% NS 88% vs 83% (4yr) NS

Aubert et al. 2009167

113 / 11 0% vs 9% P<0.005 99% vs 100% (1yr) NS

Amico et al. 2009168

334 / 67 35% vs 71% P<0.0001 89% vs 68% #(5yr) P=0.002

Lefaucheur et al.

2010169

237 / 83 Not reported 84% vs 61% (8yr) P<0.001

Riethmüller et al.

2010170

155 / 20 6% vs 35% P=0.048 Not reported

Otten et al. 2012171

837 / 290 Not reported 72% vs 30% (10yr, class I+II DSA) 72% vs 65% (10yr,class I only ) 72% vs 69% (10yr, class II only)

<0.001 NS NS

Thammanichanoud

et al. 2012172

116 / 28 14.7% vs 25% P=0.252 96.6% vs 92.9% (1yr) 92.7% vs 92.9% (5yr)

NS NS

David-Neto et al.

2012173

94 / 16 0% vs 19% P=0.004 82% vs 68.7% (6yr) P=0.21

Pérez-Flores et al.

2013174

190 / 13 30% vs 77% P=0.03 100% vs 60% (1yr) P=0.01

Wu et al. 2013175

221 / 15 3.4% vs 26.7% P=0.0001 99% vs 87% (2yr) P=0.0009

Tsapepas et al. 2014

176

174 / 62 7.1% vs 32.3% P<0.001 Not reported

Malheiro et al.

2015177

462 / 40 0.9% vs 35% P<0.001 94.9% vs 84.8% (5yr)

P=0.006

# In DSA positive group with AMR

NS= Not significant

We reported that the MFI of the DSA was higher in those patients with AMR than those

without AMR. This finding was independently reported by two groups. Lefaucheur et al

reported the prevalence of AMR increased with increasing pre-transplant DSA169. The rate

of AMR in the DSA negative group was 0.9%, compared with 18.7% if the DSA was 466-

3000 MFI, 36.4% if the DSA was 3001-6000 MFI and 51.3% if the DSA was greater than

6000 MFI. Wu et al also reported the rate of AMR increased with increasing MFI127.

Others could not show a significant association of antibody strength with AMR172. These

contrary findings might be related to the difficulties in measuring the strength of HLA

antibodies using MFI in Luminex assays. It is now recognised that there may be an

179

underestimation of HLA antibodies due to weak or false negative reactions. Two potential

causes of weak reactivity have been described. The phenomenon of prozone has been

shown in Luminex assays178, which may mask strong antibodies if serial dilutions are not

performed. One of the major causes of prozone has been identified as complement

interference. The prozone effect has been found to be related to the amount of antibody-

triggered C1q, C4 or C3 split product deposition179-182. This problem can be resolved by

dilution, treatment with DTT, heat treatment or addition of EDTA to test sera183. A second

cause of false negative or weak reactivity in Luminex bead assays was identified in the

serial dilution of sera which showed IgG antibody binding appearing or increasing as the

sera were diluted. This can be due to the presence of IgM antibodies, which block IgG

antibody binding to the antigen coated bead184. This blocking can be eliminated by serial

dilution, heat treatment, treatment with DTT or hypotonic dialysis184, 185.

In addition to the problems of underestimation of antibody levels, overestimation can also

occur. It is recognised that the density of HLA antigen present on Luminex beads can vary

from bead to bead, from batch to batch, and from locus to locus. In particular, HLA-C,

HLA-DQ and HLA-DP antigens are over-represented on Luminex beads when compared to

cell density186, 187. This can lead to detection of antibodies against these antigens by in vitro

Luminex assay alone, which may not be harmful in vivo in the context of clinical

transplantation. Secondly, the process of manufacture of HLA specific beads may result in

the exposure of cryptic epitopes. Morales-Buenrostro et al found reactivity in the Luminex

antibody assay in 63% of non-sensitized males188. The reactivity was found to be against

epitopes on denatured antigen189, 190. Otten et al found that antibodies against intact but

not denatured HLA are associated with kidney graft function and loss, with 5 year serum

creatinine levels of 177µmol/l in patients with antibody against intact antigen versus

144µmol/l in patients with antibody directed against denatured antigen (P=0.02)191.

We also showed graft function, as measured by estimated glomerular filtration rate

(eGFR), progressively declined over time in the DSA group but was stable in patients

without DSA. Wu et al also demonstrated that a decline in eGFR was preceded by the

presence of DSA, either pre-existing or de novo post-transplant192. This suggests that

chronic antibody mediated injury exerts a longer term deleterious effect hastening graft

loss.

Although we could not show a significant association, there was a trend for graft loss to be

associated with the presence of pre-transplant DSA in our study (p=0.075). However,

when we stratified graft loss by antibody class we saw a significant effect of class II DSA

(HR 2.9, 995%CI 1.0-8.0, p=0.047) and class I and II DSA (HR 3.7, 95%CI 1.6-8.9, p=0.003)

180

but not class I DSA. Graft survival in the class I and II DSA group was 54% at 8 years

compared with 83% in the DSA negative group. Graft survival was better in patients with

class I DSA (92% at 8 years); this may be due to the differences in cause of graft loss. In the

DSA positive group, all losses were due to rejection, however in the DSA negative group

graft losses occurred due to rejection, disease recurrence and BK nephropathy. Otten et al

also showed graft survival was only significantly influenced by class I and II antibodies171

and other groups have shown a significant association between graft survival and pre-

transplant DSA168, 169, 174, 175, 177. Studies are not consistent however, as many groups have

not reported a significant association165-167, 172, 173.0

8.2 Peri-operative blood transfusion is associated with AMR and renal allograft loss

Blood transfusion pre-renal-transplant is associated with sensitization against HLA antigens

but is not associated with an adverse allograft outcome; indeed it can be beneficial by

reducing episodes of acute rejection193, 194. One of the reasons suggested for this

phenomenon is that patients who are susceptible to antibody formation become broadly

sensitised, and these antigens are subsequently avoided in transplantation. In our

preliminary study we found an association of DSA detected by Luminex bead assay, in

crossmatch negative patients, with AMR and graft dysfunction. It was noted that AMR was

more prevalent in patients receiving a blood transfusion at, or shortly after, the time of

transplant and this was further explored. We found in our second study that the risk of

AMR was 13.9 times greater in patients with DSA (p=0.001), than those with no DSA, if

they received a transfusion both pre- and post-transplant. We also showed that graft loss

was significantly increased in patients who received pre- and post-transplant transfusions

(HR 7.1, p<0.001), even in the group of patients without DSA. Patients with DSA who

received transfusions only pre- or post-transplant did not experience AMR. We postulate

that there is a broad recall of antibody response following re-exposure to blood as

described by Scornik et al195. These findings are novel and have not been reported

elsewhere, and may have important implications for clinical practice. As blood

transfusions are associated with AMR and graft loss, methods to avoid transfusion at the

time of kidney transplant such as iron infusions, erythropoietin or autologous blood

transfusion could potentially have a positive effect on transplant outcome and may

prevent or reduce rejection episodes and graft loss.

181

8.3 Virtual crossmatching to facilitate transplantation.

Chapters 5, 6 and 7 addressed the use of Luminex antibody data in virtual crossmatching

for paired kidney donation. The Australian kidney exchange (AKX) program was based on

the principles guiding the Dutch algorithm162, but using high resolution HLA typing of

donors and Luminex single antigen bead analysis of recipient antibodies to enable virtual

crossmatching. We aimed to use an antibody threshold which would exclude positive

crossmatches but not at the detriment to highly sensitised patients. We, and others, have

shown the utility of the virtual crossmatch in predicting crossmatch results in the clinical

setting196- 200. In the first 4 years of the Australian paired kidney exchange program 105

patients were transplanted in matches from 215 registered donor / recipient pairs

(49%)201. A further 4 patients on the deceased donor wait-list were transplanted from the

end of a non-directed altruistic donor chain. Matching using conservative antibody cut-offs

did not appear to be detrimental to highly sensitised patients (Figure 1) who had

acceptable transplant rates even in those with a PRA >90%. One year patient and graft

survival were excellent at 97% and 96% respectively.

% of programme registrants, % of registrants transplanted, % of cPRA group transplanted

Figure 1. Transplant activity by calculated panel reactive antibody (cPRA) cohort: cPRA

97% is the true threshold for diminished access; the 95-96% groups are transplanted at a

proportion equivalent to their proportion in the registry, comparable with any other

cPRA cohort.

182

From: L Cantwell, C Woodroffe, R Holdsworth, P Ferrari. Nephrology 2015; 20(3): 124-

131201

8.4 Further research

8.4.1 Post-transplant antibody testing

Our two studies investigating the significance of pre-transplant HLA antibodies in chapters

3 and 4 showed the association of donor specific antibodies with rejection and poor

transplant outcome. We hypothesise that those patients who did not have DSA pre-

transplant but did undergo AMR developed de novo DSA post-transplant. Other groups

have shown the association of DSA found at any time post-transplant with poor allograft

outcome as reviewed in section 1.7107-116. Our group has collected sera from all kidney

transplant patients at 0, 1, 3, 6 months and yearly since 2003. Antibody analysis of these

sera to determine the prevalence of de novo DSA in our cohort is currently being

performed. Preliminary results, not shown, confirm the association of de novo DSA with

graft rejection consistent with current literature.

8.4.2 Rejection in the absence of HLA antibodies

Despite the very sensitive techniques available for detecting HLA antibodies,

histopathological features of antibody mediated rejection can be found in the absence of

demonstrable DSA in the peripheral blood. One explanation for this is that circulating DSA

is sequestered onto the donor kidney endothelium causing damage, but at levels in the

circulation undetectable by current assays. This hypothesis is supported by studies

showing DSA detection following nephrectomy of rejected kidneys117-119 and elution of

antibody from kidney endothelium120, 121.

In addition to HLA antibodies and the risk of AMR, non-HLA antibodies such as antibodies

directed against MHC Class I related gene-A (MICA), vimentin or angiotensin type I

receptor (AT1R) may also play an important role in graft rejection and give a similar

histological appearance. The role of non-HLA antibodies in allograft rejection are

comprehensively reviewed in publications by Dragun et al202, 203 and Sumitran-

Holgersson204. Preliminary investigation of MICA antibodies by our group has not found an

association of presence of MICA antibodies with graft outcome. We believe this is due to

the low prevalence of MICA antibodies in our cohort, and the finding that all patients with

MICA antibody also had HLA DSA.

183

8.4.3 HLA DSA in the absence of rejection

Although detection of DSA prior to transplant has been associated with poor graft

outcome, not all patients with DSA undergo rejection or lose their grafts. A potential

explanation for this may be accommodation of the graft in some patients. It has been

shown in highly sensitised patients who have undergone desensitisation to temporarily

remove DSA prior to transplant, that they do not always experience accelerated rejection

despite having return or persistence of antibody and can have excellent long term

outcomes205, 206. More recently, in a study of 1025 kidney transplant recipients who

received an HLA incompatible transplant from a living donor following desensitisation,

graft survival at 8 years was better than in controls who either received a compatible

deceased donor transplant or remained on the wait list (76% versus 62.9% versus

43.9%)207. The frequency and mechanisms by which accommodation may occur in some

patients with donor specific HLA antibodies is unknown and warrants further research and

elucidation.

An additional parameter to consider that may impact upon the detection of antibody and

the initiation of immune injury is the antibody affinity and avidity. Whilst Luminex

antibody detection provides a sensitive means of detecting HLA antibodies, it provides no

data regarding the strength of interaction between the antibody and corresponding

epitope. Antibody binding may be influenced by the epitope orientation and accessibility

and the physiochemical properties of amino acids in the binding site. Kosmoliaptsis et al

have shown in experiments analysing surface electrostatic potentials that some changes in

amino acids do not result in a change in electrostatic motif, whereas other changes at

critical positions can affect antibody binding208,209. This is a potential area for further

research and may help us to understand the pathogenicity of HLA antibodies.

8.4.4 Characterisation of HLA epitopes in kidney transplantation

HLA matching between donor and recipient is a major component in kidney allocation as

the greater number of HLA mismatches is associated with increasing risk of allograft

rejection23, 24. The crossreactivity between HLA antigens was first recognised by serological

analysis of antisera210. The definition of HLA epitopes followed the elucidation of the

three-dimensional structure of the HLA molecule and the availability of amino acid

sequence data. Duquesnoy developed a computer-based algorithm, HLAMatchmaker,

based upon linear triplets of amino acids on the HLA molecule which are critical for

184

antibody recognition211. Subsequently the algorithm was updated to include non-linear

amino acid triplets, referred to as eplets212. Further refinement followed the recognition

that in some cases self-epitopes are required for antibody binding213, 214. El-Awar et al

showed in in vitro experiments evidence of epitopes by absorption / elution studies using

Luminex single antigen bead assays215. Furthermore, Duquesnoy and Marrari were able to

correlate class I and II epitopes determine by HLAMatchmaker with those identified by El-

Awar et al 216, 217.

Several studies have associated an increasing number of epitope mismatches with a

greater chance of producing HLA antibodies218-220. In addition it has been demonstrated

that patients matched at the epitope level, in broad antigen level mismatches, have similar

outcomes to broad antigen matched transplants156. HLAMatchmaker can be used

successfully as a tool to facilitate transplantation, particularly in highly sensitised patients.

The Eurotransplant Acceptable Mismatch program reports considerably shorter waiting

times for highly sensitised patients, whilst achieving allograft outcomes similar to that of

unsensitised patients157, 158. Validation of these findings in the Australian kidney transplant

setting is essential for the ongoing improvement of the allocation algorithm.

8.4.5 HLA specific B cells

Donor specific antibodies are associated with acute and chronic rejection, however DSA

are not always found in patients undergoing rejection even when using the sensitive

Luminex antibody assays. There are a number of suggestions why this may occur i) very

low titre antibodies; ii) antibodies are sequestered to the transplanted organ; iii) patient

unresponsiveness; or iv) timing of antibody testing. In all these situations it is unclear

whether the patient immune system is primed for a rapid response on re-exposure to

antigen.

It is possible to detect HLA-specific B cells which produce alloantibody in vitro. The first

assays utilised HLA tetramers to detect HLA antibody producing B cells221. The surface

immunoglobulin antigen receptor of B cells binds to HLA molecules on the tetramer with a

comparable specificity to that of secreted antibody. The tetramer staining can then be

quantified by flow cytometry. The next assays involved in vitro culture of B cells and testing

of supernatants by Luminex single antigen bead assay for HLA antibody. Han et al found

DSA in all B cell supernatants from patients who had rejected their organs (13/13) and no

DSA in unsensitised patients222. In four patients DSA was only found in the B cell

supernatant and not in peripheral blood, two patients had rejected kidneys in situ.

185

Heidt et al developed an ELISPOT assay, allowing a straightforward sensitive and specific

laboratory test to quantify HLA-specific B cells223. The group showed that patients with low

numbers of HLA-specific B cells could be re-transplanted, with good outcomes at 18

months follow-up, and hypothesised that there may be a level which can be interpreted as

below clinical significance. Lucia et al used a similar in-house ELISPOT to detect HLA-

specific B cells and were able to correlate detection of antibody with a high risk of humoral

rejection224. This group also found HLA specific B cells in the absence of HLA specific

antibodies in peripheral blood.

These experiments suggest that an increase in HLA-specific B cells post-transplant may be

a precursor to antibody formation, and therefore may be a target for adjusting

immunosuppression. In addition detection of HLA-specific B cells may reveal sensitisation

when HLA antibodies are not found in peripheral blood, and pre-transplant testing may

assist in evaluating risk.

8.4.6 Complement fixing HLA specific antibodies

The presence of DSA detected pre- or post-transplant is associated with AMR and graft

loss. However, not all patients with DSA develop rejection or lose their grafts. The ability

to activate complement has come under scrutiny in a bid to understand the pathogenicity

of antibodies due to the finding of depositions of the complement component C4d on the

kidney in rejection. The Luminex antibody assay can be modified, via the use of different

secondary detection antibodies to immunoglobulin isotypes, subclasses or complement

components, to assess the significance of complement-fixing versus non-fixing HLA

antibodies225. Studies investigating clinical outcomes in patients with DSA stratified by

their complement-fixing ability are summarised in Table 2.

186

Table 2. Summary of studies investigating the clinical significance of complement-fixing

DSA.

Authors and

year of

publication

Number in

study (DSA

positive) /

Number of

C’-fixing

DSA

Incidence of

AMR

(DSA vs C’-fixing

DSA)

P value Graft survival (DSA vs C’-fixing DSA)

P value

Wahrmann et

al. 2009226

46 / 21 4% vs 24% P<0.001 90% vs 76% P=0.01

Hönger et al.

2010227

64 / 11 53% vs 55% NS Not reported

Yabu et al.

2011228

31 / 12 Not reported 37% vs 21% P=0.026

Hönger et al.

201195

74 / 21 25% vs 57% P=0.23 100% vs 78% P=0.74

Sutherland et

al. 2012229

35 / 15 16.7% vs 60% P=0.04 85% vs 53.3% P=0.04

Lawrence et

al. 2013230

42 / 13 8% vs 24% P <0.01 91.7% vs 80% P=0.001#

Crespo et al.

2013231

28 / 15 30.8% vs 26.7% P=0.67 76.9% vs 80% P=0.8

Loupy et al.

2013232

239 / 77 16% vs 48% P<0.001 93% vs 54% P<0.001

C’ = complement; abs = antibodies, DSA = Donor specific antibodies; NS = Not significant

#AMR free graft survival

Several groups have demonstrated a significant association of pre-transplant complement-

fixing DSA with AMR when compared to non-complement-fixing DSA226, 230, 232. Others

could not find a significant association95, 227, 231. All groups reported an increased risk of

AMR with DSA when compared to patients with no DSA. Yabu et al228 and Sutherland et

al229 investigated de novo DSA in the post-transplant period and were able to demonstrate

a strong association of complement-fixing DSA with AMR. These results suggest that

187

complement-fixing ability of DSA might only be important in the post-transplant period.

Most groups reported a correlation between complement-fixing DSA and poor graft

survival226, 228-230, 232, and all groups reported lower graft survival in patients with DSA

compared to no DSA.

Zeevi et al studied complement-fixing antibodies in heart transplants and showed high

titre antibodies correlated with the ability to fix complement and were associated with

AMR (P=0.005)233. In our own publication, we were able to demonstrate a correlation

between strong DSA assessed by MFI and renal graft outcome (Chapter 3). It may be

possible to use MFI as a surrogate test for complement-fixing antibodies, and other groups

have suggested that pre-transplant the strength of DSA rather than complement-fixing

ability is important 231, 234, 235. Larger cohort or collaborative studies are required to

determine the clinical utility of testing for complement-fixing antibodies as a further tool

to stratify transplant risk.

In addition to the complement-activating ability of different subclasses of IgG antibodies, it

has been recognised the glycosylation of antibody may be important in modulating

complement fixation. The process of glycosylation in the context of transplantation is

reviewed by Thomas et al236 and appears to influence the affinity of the Fcϒ receptor,

shifting IgG activity from a pro-inflammatory to an anti-inflammatory pathway237.

8.5 Conclusions

The studies in this thesis have shown that pre-transplant donor specific HLA antibodies

detected by Luminex single antigen bead assay with a concurrent negative CDC crossmatch

are associated with increased risk of rejection episodes, AMR, chronic graft dysfunction

and graft loss. We have shown that the risks are increased with increasing strength of

antibody as measured by MFI. In addition we have described the novel finding that that

the risk of AMR was significantly increased in patients with DSA, than those with no DSA, if

they received a transfusion both pre- and post-transplant. These findings have important

implications for clinical practice.

We have also shown that use of antibodies detected by the sensitive Luminex method for

virtual crossmatching and transplant allocation is successful in the Australian paired kidney

exchange program with minimal adverse affects on the transplantation of highly sensitised

patients.

Further investigation into the reasons why some patients with DSA do not undergo graft

rejection is warranted.

188

References

189

References

1. T Shiina, K Hosomichi, H Inoko, JK Kulski. The HLA genomic loci map: Expression,

interaction, diversity and disease. J Hum Genet 2009; 54: 215-39

2. J Klein, A Sato. The HLA system. N Engl J Med 2000; 343: 702-709

3. JD Wall, JK Pritchard. Haplotype blocks and linkage disequilibrium in the human

genome. Nat Rev Genet 2003; 4: 587-597

4. AB Begovich, GR McClure, VC Suraj, RC Helmuth, N Fildes, TL Bugawan, HA Erlich, W

Klitz. Polymorphism, recombination, and linkage disequilibrium within the HLA class II

region. J Immunol 1992; 148: 249-258

5. X Gao, A Veale, SW Serjeantson. HLA class II diversity in Australian aborigines:

Unusual DRB1 alleles. Immunogenetics 1992; 36: 333-337

6. AS Daar, SV Fuggle, JW Fabre, A Ting, PJ Morris. The detailed distribution of HLA-A, B,

C antigens in normal human organs. Transplantation 1984; 38: 287-293.

7. J Klein, A Sato. The HLA system. First of two parts. NEJM 2000; 343: 702-709

8. P Cresswell. Regulation of HLA class I and class II antigen expression. Br Med Bull

1987; 43: 66-80

9. Y Watanabe, CO Jacob. Regulation of MHC class II antigen expression. Opposing

effects of tumor necrosis factor-alpha on IFN-gamma-induced HLA-DR and Ia

expression depends on the maturation and differentiation stage of the cell. J

Immunol 1991; 146: 899-905

10. FJ Salgado, J Lojo, CM Fernández-Alonso, JE Viñuela, OJ Cordero, M Nogueira.

Interleukin-dependent modulation of HLA-DR expression on CD4 and CD8 activated T

cells. Immunology and Cell Biology 2002; 80: 138-147

11. KR Prilliman, D Crawford, HD Hickman, KW Jackson, J Wang, WH Hildebrand. Alpha-2

domain polymorphism and HLA class I peptide loading. Tissue Antigens 1999; 54:

450-460

12. http://www.ebi.ac.uk/imgt/hla

13. J Robinson, JA Halliwell, JD Hayhurst, P Flicek, P Parham, SGE Marsh. The IPD and

IMGT / HLA database: Allele variant databases. Nucleic Acids Res 2015; 43: D423-431

14. JS Blum, PA Wearsch, P Cresswell. Pathways of antigen processing. Annu Rev

Immunol 2013; 31: 443-473

http://www.ebi.ac.uk/imgt/hla

190

15. P Cresswell, AL Ackerman, A Giodini, DR Peaper, PA Wearsch. Mechanisms of MHC

class-I-restricted antigen processing and cross-presentation. Immunol Rev 2005; 207:

145-157

16. NJ Rogers, RI Lechler. Allorecognition. Am J Transplant 2001; 1: 97-102

17. G Benichou, AW Thomson. Direct versus indirect allorecognition pathways: On the

right track. Am J Transplant 2009; 9: 655-656

18. AN Warrens, G Lombardi, RI Lechler. Presentation and recognition of major and

minor histocompatibility antigens. Transpl Immunol 1994; 2: 103-107

19. RI Lechler, JR Batchelor. Restoration of immunogenicity to passenger cell-depleted

kidney allografts by the addition of donor strain dendritic cells. J Exp Med 1982;

155(1): 31-41

20. DS Game, RI Lechler. Pathways of allorecognition: implications for transplantation

tolerance. Transpl Immunol 2002; 10: 101-108

21. DA Shoskes, KJ Wood. Indirect presentation of MHC antigens in transplantation.

Immunol Today 1994; 15: 32-38

22. JA Bradley, EM Bolton, RA Pederson. Stem cell medicine encounters the immune

system. Nat Rev Immunol 2002; 2: 859-871

23. L Martins, I Fonseca, S Sousa, C Matos, J Santos, L Dias, AC Henriques, AM Sarmento,

A Cabrita. The influence of HLA mismatches and immunosuppression on kidney graft

survival: An analysis of more than 1300 patients. Trans Proc 2007; 39: 2489-2493

24. G Opelz, B Döhler. Effect of human leukocyte antigen compatibility on kidney graft

survival: Comparative analysis of two decades. Transplantation 2007; 82: 137-143

25. P Gailiunas Jr, M Suthanthiran, GJ Busch, CB Carpenter, MR Garovoy. Role of humoral

presensitization in human renal transplant rejection. Kidney Int 1980; 17(5): 638-646

26. RB Colvin, RN Smith. Antibody-mediated organ-allograft rejection. Nat Rev Immunol

2005; 5: 807-817

27. R Patel, P Terasaki. Significance of the positive crossmatch test in kidney

transplantation. N Eng J Med 1969; 280: 735-739

28. AS Rosenberg, T Mizuochi, SO Sharrow, A Singer. Phenotype, specificity, and function

of T cell subsets and T cell interactions involved in skin allograft rejection. J Exp Med

1987; 165: 1296-1315

29. F Boisgérault, Y Liu, N Anosova, E Ehrlich, MR Dana, G Benichou. Role of CD4+ and

CD8+ T cells in allorecognition: lessons from corneal transplantation. J Immunol 2001;

167(4): 1891-1899

30. E Ingulli. Mechanism of cellular rejection. Pediatr Nephrol 2010; 25: 61-74

191

31. BA Wasowska. Mechanisms involved in antibody- and complement-mediated

allograft rejection. Immunol Res 2010; 47: 25-44

32. NM Valenzuela, EF Reed. Antibodies in transplantation: The effects of HLA and non-

HLA antibody binding and mechanisms of injury. Methods Mol Biol 2013: 1034

33. S Mauiyyedi, RB Colvin. Humoral rejection in kidney transplantation: New concepts in

diagnosis and treatment. Curr Opin Nephrol Hypertens 2002; 11: 609-618

34. S Moll, M Pascual. Humoral rejection of organ allografts. Am J Transplant 2005; 5:

2611-2618

35. RB Colvin, RN Smith. Antibody mediated organ-allograft rejection. Nat Rev Immunol

2005; 5: 807-817

36. RB Colvin. Antibody-mediated renal allograft rejection: Diagnosis and pathogenesis. J

Am Soc Nephrol 2007; 18: 1045-1056

37. H Regele, GA Böhmig, A Habicht, D Gollowitzer, M Schillinger, S Rockenschaub, B

Watschinger, D Kerjaschki, M Exner. Capillary deposition of complement split product

C4d in renal allografts is associated with basement membrane injury in peritubullar

and glomerular capillaries: A contribution of humoral immunity to chronic allograft

rejection. J Am Soc Nephrol 2002; 13: 2371-2380

38. TV Obukhanych, MC Nussenzweig. T-independent type II immune responses generate

memory B cells. J Exp Med 2006; 203(2): 305-310

39. JJ Mond, A Lees, CM Snapper. T cell independent antigens type 2. Annu Rev Immunol

1995; 13: 655-692

40. JJ Mond, Q Vos, A Lees, CM Snapper. T cell independent antigens. Curr Opin Immunol

1995; 7: 349-354

41. DC Parker. T cell-dependent B cell activation. Annu Rev Immunol 1993; 11: 331-360

42. Y Yang, JM Wilson. CD40 ligand-dependent T cell activation: Requirement of B7-CD28

signalling through CD40. Science 1996; 273: 1862-1864

43. M Croft, SL Swain. B cell responses to fresh and effector T helper cells. Role of

cognate T-B interactions and the cytokines IL-2, IL-4, and IL-6. J Immunol 1991; 146:

4055-4064

44. WE Paul, RA Seder. Lymphocyte responses and cytokines. Cell 1994; 76: 241-251

45. JM Purkerson, PC Isakson. Interleukin 5 (IL-5) provides a signal that is required in

addition to IL-4 for isotype switching to immunoglobulin (Ig) G1 and IgE. J Exp Med

1992; 175: 973-982

46. S Julien, PA Videira, P Delannoy. Sialyl-Tn in cancer: (How) did we miss the

target? Biomolecules 2012; 2: 435–466

192

47. P Morris, P Kincaid-Smith, V Marshall. Leukocyte antigens in cadaver transplantation.

Transplant Proc 1969; 1: 376-378

48. MS Leffell, AG Steinberg, WB Bias, CH Machan, AA Zachary. The distribution of HLA

antigens and phenotypes among donors and patients in the UNOS registry.

Transplantation 1994; 58(10): 1119-1130

49. AA Zachary, JT Sholander, JA Houp, MS Leffell. Using real data for a virtual

crossmatch. Hum Immunol 2009; 70: 574-579

50. JM Cecka. Calculated PRA (cPRA): the new measure of sensitization for transplant

candidates. Am J Transplant 2010; 10: 26-29

51. AR Tambur, J Leventhal, DB Kaufmann, J Friedwald, J Miller, MM Abecassis. Tailoring

antibody testing and how to use it in the calculated panel reactive antibody era: The

Northwestern University experience. Transplantation 2008; 86: 1052-1059

52. JR Chapman, CJ Taylor, A Ting, PJ Morris. Immunoglobulin class and specificity of

antibodies causing positive T cell crossmatches. Relationship to renal transplant

outcome. Transplantation 1986; 42: 608-612

53. CJ Taylor, JR Chapman, A Ting, PJ Morris. Characterization of lymphocytotoxic

antibodies causing a positive crossmatch in renal transplantation. Relationship to

primary and regraft outcome. Transplantation 1989; 48: 953-958

54. DE Cross, R Greiner, FC Whittier. Importance of the autocontrol crossmatch in human

renal transplantation. Transplantation 1976; 21: 307-311

55. Y Iwaki, M Lau, PI Terasaki. Successful transplants across T-warm positive

crossmatches due to IgM antibodies. Clin Transplant 1998; 2: 81-94

56. C Baron, M Pastural, P Lang, R Bentabet, N el Kassar, T Seror, D Dahmane, D Desvaux,

D Chopin, G Fruchard, P Remy, P Grimbert, E Lepage, P Bieling. Long-term kidney

graft survival across a positive historical but negative current sensitized cross-match.


57. MA Pellegrino, M Belverdere, AG Pellegrino, S Ferrone. B peripheral lymphocytes

express more HLA antigens than T peripheral lymphocytes. Transplantation 1978;

25(2): 93-95

58. T Mohanakumar, MA Gledlin, CL Rhodes. Relationship of B cell alloantibodies to renal

allograft survival. Transplantation 1979; 27(4): 273-8

59. G Ayoub, MS Park, PI Terasaki, Y Iwaki, G Opelz. Transplantation 1980; 29(3): 227-229

60. RB Ettinger, PI Terasaki, G Opelz, M Malekzadeh, C Uittenbogaart, C Pennisi, R Fine.

Successful renal allografts across a positive crossmatch for donor B-lymphocyte

alloantigens. Lancet 1976; 1: 56-58

193

61. M Jeannet, G Benzonana, I Arni. Donor-specific B and T lymphocyte antibodies and

kidney graft survival. Transplantation 1981; 31(3): 160-163

62. A Ting, PJ Morris. Renal transplantation and B-cell crossmatches with autoantibodies

and alloantibodies. Lancet 1977; 2: 1095-1097

63. SS Karuppan, A Lindholm, W Möller. Characterisation and significance of donor-

reactive B cell antibodies in current sera of kidney transplant patients.


64. RJ Mahoney, S Taranto, E Edwards. B-cell crossmatching and kidney allograft

outcome in 9031 United States transplant recipients. Human Immunology 2002;

63(4): 324-335

65. S Le Bas-Bernardet, M Hourmant, N Valentin, C Paitier, M Giral-Classe, S Curry, G

Follea, JP Soulillou, JD Bignon. Identification of the antibodies involved in B-cell

crossmatch positivity in renal transplantation. Transplantation 2003; 75(4): 477-482

66. A Goh, E Oei, A Vathsala. Successful transplantation across positive B-cell cross-match

in deceased donor renal transplants. Transplant Proc 2012; 44(1): 193-199

67. HS Hwang, HE Yoon, BS Choi, EJ Oh, JI Kim, IS Moon, YS Kim, CW Yang. B–cell

complement dependent cytotoxic crossmatch positivity is an independent risk factor

for long term renal allograft survival. J Korean Med Sci 2011; 26(4): 528-533

68. HS Eng, G Bennett, P Bardy, P Coghlan, GR Russ, PT Coates. Clinical significance of

anti-HLA antibodies detected by Luminex: Enhancing the interpretation of CDC-BXM

and important post-transplantation monitoring tools. Hum Immunol 2009; 70(8):

595-599

69. DE Cross, FC Whittier, P Weaver, J Foxworth. A comparison of the antiglobulin versus

extended incubation time crossmatch: Results in 223 renal transplants. Transplant

Proc 1977; 9: 1803-1806

70. TC Fuller, AB Cosimi, PS Russell. Use of an antiglobulin-ATG reagent for detection of

low levels of alloantibody – improvement of allograft survival in presensitized

recipients. Transplant Proc 1978; 10: 463-466

71. AH Johnson, RD Rossen, WT Butler. Detection of alloantibodies using a sensitive

antiglobulin microcytotoxicity test: Identification of low levels of preformed

antibodies in accelerated allograft rejection. Tissues Antigens 1972; 2: 215-226

72. Y Iwaki, M Lau, DJ Cook, S Takemoto, PI Terasaki. Crossmatching with B and T cells

and flow cytometry. Clin Transpl 1986; 277-284

73. M Karpinski, D Rush, J Jeffery, M Exner, H Regele, S Dancea, D Pochino, P Birk, P

Nickerson. Flow cytometric crossmatching in primary renal transplant recipients with

194

a negative anti-human globulin enhanced cytotoxicity crossmatch. J Am Soc Nephrol

2001; 12: 2807-2814

74. AA Zachary, L Klingman, N Thorne, AR Smerglia, GA Teresi. Variations of the

lymphocytotoxicity test: An evaluation of sensitivity and specificity. Transplantation

1995; 60: 498-503

75. MH Christiaans, R Overhof, A ten Haaft, F Niemann, JP van Hoof, EM van den Berg-

Loonen. No advantage of flow cytometry crossmatch over complement-dependent

cytotoxicity in immunologically well-documented renal allograft recipients.


76. RH Kerman, B Süsskind, I Buyse, P Pryzbylowski, J Ruth, S Warnell, SA Gruber, S Katz,

CT van Buren, BD Kahan. Flow cytometry-detected IgG is not a contraindication to

renal transplantation: IgM may be beneficial to outcome. Transplantation 1999; 68:

1855-1858

77. MC Bittencourt, JM Rebibou, Y Saint-Hillier, J Chaboud, I Dupont, JM Chalopin, P

Herve, P Tiberghien. Impaired renal graft survival after a positive B-cell flow-

cytometry crossmatch. Nephrol Dial Transplant 1998; 13: 2059-2064

78. HM Gebel, RA Bray. Sensitization and sensitivity: Defining the sensitized patient.


79. JC Scornik. Detection of alloantibodies by flow cytometry: Relevance to clinical

transplantation. Cytometry 1995; 22: 259-263

80. WW Alterman, B Seliger, S Sel, D Wendt, G Schlaf. Comparison of the established

standard complement-dependent cytotoxic and flow cytometric crossmatch assays

with a novel ELISA-based HLA crossmatch procedure. Histol histopathol 2006; 21:

1115-1124

81. AW Harmer, M Sutton, A Bayne, RW Vaughan, KI Welsh. A highly sensitive rapid

screening method for the detection of antibodies directed against HLA class I and

class II antigens. Transpl Int 1993; 6: 277-280

82. KJ Kao, JC Scornik, SJ Small. Enzyme-linked immunoassay for anti-HLA antibodies –

Alternative to panel studies by lymphocytotoxicity. Transplantation 1993; 55: 192-

196

83. F Monteiro, C Mineiro, H Rodrigues, J Kalil. Pretransplant and posttransplant-

monitoring of anti-HLA class I IgG1 antibodies by ELISA identifies patients at high risk

of graft loss. Transpl Proc 1997; 29: 1433-1434

84. C Susal, G Opelz. Kidney graft failure and presensitization against HLA class I and class

II antigens. Transplantation 2002; 73: 1269-1273

195

85. A Slavcev, J Lacha, E Honsova, H Sajdlova, A Lodererova, S Vitko, S Valhova, I Striz, E

Ivaskova. Clinical relevance of antibodies to HLA antigens detectable by the standard

complement-dependent cytotoxicity test. Transpl Int 2003; 16: 872-878

86. C Susal, G Opelz. Good kidney graft outcome in recipients with presensitization

against HLA class II but not HLA class I. Hum Immunol 2004; 65: 810-816

87. AM Patel, C Pancoska, S Mulgaonkar, FL Weng. Renal transplantation in patients with

pre-transplant donor-specific antibodies and negative flow cytometry crossmatches.

Am J Transplant 2007; 7: 2371-2377

88. C Lefaucheur, C Suberbielle-Boissel, GS Hill, D Nochy, J Andrade, C Antoine, C

Gautreau, D Charron, D Glotz. Clinical relevance of preformed HLA donor-specific

antibodies in kidney transplantation. Am J Transplant 2008; 8: 324-331

89. C Susal, B Döhler, G Opelz. Presensitized kidney graft recipients with HLA class I and II

antibodies are at increased risk for graft failure: A collaborative transplant study

report. Hum Immunol 2009; 70: 569-573

90. H-H Wu, Y-C Tien, C-I Huang, Y-J Chiang, S-H Chu, P-C Lai. HLA class I antibodies in

patients awaiting kidney transplantation and the association with graft survival.

Transpl Proc 2008; 40: 2191-2194

91. R Pei, J-H Lee, T Chen, S Rojo, PI Terasaki. Flow cytometric detection of HLA

antibodies using a spectrum of microbeads. Hum Immunol 1999; 60: 1293-1302

92. W Herczyk. Luminex. ASHI Quarterly 2013: 104; PC Lee, M Ozawa, CJ Hung, YJ Lin, SS

Chang, TC Chou. Reappraisal of HLA antibody analysis and crossmatching in kidney

transplantation. Clin Transpl 2007: 219-226

93. N El-Awar, J Lee, PI Terasaki. HLA antibody identification with single antigen beads

compared to conventional methods. Hum Immunol 2005; 66: 989-997

94. S Jung, EJ Oh, CW Yang, WS Ahn, Y Kim, YJ Park, K Han. Comparative evaluation of

ELISA and Luminex panel reactive assay for HLA alloantibody screening. Korean J Lab

Med 2009; 29(5): 473-480

95. G Hönger, H Hopfer, ML Arnold, BM Spriewald, S Schaub, P Amico. Pretransplant IgG

subclasses of donor-specific human leukocyte antigen antibodies and development of

antibody-mediated rejection. Transplantation 2011; 92: 41-47

96. ML Arnold, FM Heinemann, P Horn, M Ziemann, N Lachmann, A Mulbacher, A Dick, A

Ender, D Thammanichanoud, GF Fischer, S Schaub, M Hallensleben, J Mytilineos, WE

Hitzler, C Seidl, II Doxiadis, BM Spriewald. 16(th) IHIW: Anti-HLA alloantibodies of the

IgA isotype in re-transplant candidates. Int J Immunogenet 2013; 40: 17-20

196

97. JD Smith, IM Hamour, NR Banner, ML Rose. C4d fixing, Luminex binding antibodies –

A new tool for prediction of graft failure after heart transplantation. Am J Transplant

2007; 7: 2809-2815

98. N Lachmann, K Todorova, H Schulze, C Schonemann. Systematic comparison of four

cell- and Luminex-based methods for assessment of complement-activating HLA

antibodies. Transplantation 2013; 95: 694-700

99. EM Gibney, LR Cagle, B Freed, SE Warnell, L Chan, AC Wiseman. Detection of donor-

specific antibodies using HLA-coated microspheres: Another tool for kidney

transplant risk stratification. Nephrol Dial Transplant 2006; 21: 2625-2629

100. A Gupta, V Iveson, M Varagunum, S Bodger, P Sinnott, RC Thuraisingham.

Pretransplant donor-specific antibodies in cytotoxic negative crossmatch kidney

transplants: Are they relevant? Transplantation 2008; 85: 1200-1204

101. N Issa, FG Cosio, JM Gloor, S Sethi, PG Dean, SB Moore, S DeGoey, MD Stegall.

Transplant glomerulopathy: Risk and prognosis related to anti-human leukocyte

antigen class II antibody levels. Transplantation 2008; 86: 681-685

102. EM van den Berg-Loonen, EVA Billen, CEM Voorter, LWE van Heurn, FHJ Claas,

JP van Hooff, MHL Christiaans. Clinical relevance of pretransplant donor-directed

antibodies detected by Single Antigen beads in highly sensitized renal transplant

patients. Transplantation 2008; 85: 1086-1090

103. N Lachmann, PI Terasaki, C Schonemann. Donor-specific HLA antibodies in

chronic renal allograft rejection: A prospective trial with four-year follow-up. Clin

Transpl 2006; 171-199

104. C Wiebe, IW Gibson, TD Blydt-Hansen, M Karpinski, J Ho, LJ Storsley, A

Goldberg, PE Birk, DN Rush, PW Nickerson. Evolution and clinical pathologic

correlations of de novo donor-specific HLA antibody post kidney transplant. Am J

Transplant 2012; 12: 1157-1167

105. MJ Everly, JJ Everly, LJ Arend, P Brailey, B Susskind, A Govil, A Rike, P Roy-

Chaudhury, G Mogilishetty, RR Alloway, A Tevar, ES Woodle. Reducing de novo

donor-specific antibody levels during acute rejection diminishes renal allograft loss.

Am J Transplant 2009; 9: 1063-1071

106. M Houmant, A Cesbron-Gautier, PI Terasaki, K Mizutani, A Moreau, A

Meurette, J Dantal, M Giral, G Blancho, D Cantarovich, G Karam, G Follea, JP Soulillou,

JD Bignon. Frequency and clinical implications of development of donor-specific and

non-donor specific HLA antibodies after kidney transplantation. J Am Soc Nephrol

2005; 16: 2804-2812

197

107. A Mihaylova, D Baltadjieva, P Boneva, M Ivanova, K Penkova, D Marinova, S

Mihailova, E Paskalev, P Simeonov, E Naumova. Clinical relevance of anti-HLA

antibodies detected by flow-cytometry bead-based assays – Single-center

experience. Hum Immunol 2006; 67: 787-794

108. PI Terasaki, M Ozawa, R Castro. Four-year follow-up of a prospective trial of

HLA and MICA antibodies on kidney graft survival. Am J Transplant 2007; 7: 408-415

109. Q Mao, PI Terasaki, J Cai, N El-Awar, L Rebellato. Analysis of HLA class I specific

antibodies in patients with failed allografts. Transplantation 2007; 83: 54-61

110. JC Scornik, G Guerra, JD Schold, TR Srinivas, D Dragun, H-U Meier-Krieshe.

Value of posttransplant antibody tests in the evaluation of patients with renal graft

dysfunction. Am J Transplant 2007; 7: 1808-1814

111. P-C Lee , L Zhu, PI Terasaki, MJ Everly. HLA-specific antibodies developed in the

first year posttransplant are predictive of chronic rejection and renal graft loss.


112. A Haririan, B Kiangkitiwan, D Kukuruga, M Cooper, H Hurley, C Drachenberg, D

Klassen. The impact of C4d pattern and donor-specific antibody on graft survival in

recipients requiring indication renal allograft biopsy. Am J Transplant 2009; 9: 2758-

2767

113. RL Kedainis, MJ Koch, DC Brennan, H Liapis. Focal C4d+ in renal allografts is

associated with the presence of donor-specific antibodies and decreased allograft

survival. Am J Transplant 2009; 9: 812-819

114. P Stastny, S Ring, C Lu, J Arenas, M Han, B Lavingia. Role of immunoglobulin

(Ig)-G and IgM antibodies against donor human leukocyte antigens in organ

transplant recipients. Hum Immunol 2009; 70: 600-604

115. P Cinti, R Pretagostini, Q Lai, ML Tamburro, M Rossi, L Poli, P Berloco.

Alloantibodies and outcomes of deceased donor kidney allografts. Hum Immunol

2009; 70: 651-654

116. LG Hidalgo, PM Campbell, B Sis, G Einecke, M Mengel, J Chang, J Sellares, J

Reeve, PF Halloran. De novo donor-specific antibody at the time of kidney transplant

biopsy associates with microvascular pathology and late graft failure. Am J Transplant

2009; 9: 2532-2541

117. EV Billen, MH Christiaans, J Lee, EM van den Berg-Loonen. Donor-directed HLA

antibodies before and after transplantectomy detected by the Luminex single antigen

assay. Transplantation 2009; 87: 563-569

198

118. M Marrari, RJ Duquesnoy. Detection of donor-specific HLA antibodies before

and after removal of a rejected kidney transplant. Transpl Immunol 2010; 22: 105-

109

119. D Tittelbach-Helmrich, P Pisarski, G Offermann, M Geyer, O Thomusch, UT

Hopt, O Drognitz. Impact of transplant nephrectomy on peak PRA levels and outcome

after kidney re-transplantation. World J Transplant 2014; 4: 141-147

120. L Martin, F Guignier, C Mousson, D Rageot, E Justrabo, G Rifle. Detection of

donor-specific anti-HLA antibodies with flow cytometry in eluates and sera from renal

transplant recipients with chronic allograft nephropathy. Transplantation 2013; 76:

395-400

121. A Girnita, E Portwood, P Brailey, M Cuffy, A Tica, F Paterno, T Diwan, S Shah, G

Mogilishetty, M Cardi, A Govil, RA Alloway, ES Woodle. Epitope cluster analysis for

donor-specific antibody identification in eluates from fine-needle allograft biopsy.

Hum Immunol 2015; 76: 125-126

122. SR Lederer, B Kluth-Pepper, H Schneeberger, E Albert, W Land, HE Feucht.

Impact of humoral alloreactivity early after transplantation on the long-term survival

of renal allografts. Kidney Int 2001; 59(1): 334-341

123. RS Gaston, JM Cecka, BL Kasiske, AM Fieberg, R Leduc, FC Cosio, S

Gourishankar, J Grande, P Halloran, L Hunsicker, R Mannon, D Rush, AJ Matas.

Evidence for antibody-mediated injury as a major determinant of late kidney allograft

failure. Transplantation 2010; 90(1): 68-74

124. S Mauiyyedi, M Crespo, AB Collins. Acute humoral rejection in kidney

transplantation. II. Morphology, immunopathology, and pathologic classification. J

Am Soc Nephrol 2002; 13: 779-789

125. G Einecke, B Sis, J Reeve, M Mengel, PM Campbell, LG Hidalgo, B Kaplan, PF

Halloran. Antibody-mediated microcirculation injury is the major cause of late kidney

transplant failure. Am J Transplant 2009; 9: 2520-2531

126. J Sellarés, DG de Freitas, M Mengel, J Reeve, G Einecke, B Sis, LG Hidalgo, K

Famulski, A Matas, PF Hallpran. Understanding the causes of kidney transplant

failure: The dominant role of antibody-mediated rejection and nonadherence. Am J

Transplant 2012; 12: 388-399

127. E Pouliquen, A Koenig, CC Chen, A Sicard, M Rabeyrin, E Morelon, V Dubois, O

Thaunat. Recent advances in renal transplantation: Antibody mediated rejection

takes center stage. F1000Prime Rep 2015; 7: 51

128. DB Tyan, VA Li, L Czer, A Trento, SC Jordan. Intravenous immunoglobulin

suppression of HLA alloantibody in highly sensitized transplant candidates and

199

transplantation with a histoincompatible organ. Transplantation 1994; 57 (4): 553-

562

129. J Watanabe, JC Scornik. IVIG and HLA antibodies. Evidence for inhibition of

complement activation but not for anti-idiotypic activity. Am J Transplant 2005; 5:

2786-2790

130. T Tha-In, J Bayry, HJ Metselaar, SV Kaveri, J Kwekkeboom. Modulation of the

cellular immune system by intravenous immunoglobulin. Trends Immunol 2008; 29:

608-615

131. C Lefaucheur, D Nochy, J Andrade, J Verine, C Gautreau, D Charron, GS Hill, D

Glotz, C Suberbielle-Boissel. Comparison of combination plasmapheresis / IVIg / anti-

CD20 versus high-dose IVIg in the treatment of antibody mediated rejection. Am J

Transplant 2009; 9: 1099-1107

132. P Hillmen, NS Young, J Schubert, RA Brodsky, G Socié, P Muus, A Röth, J Szer,

MO Elebute, R Nakamura, P Browne, AM Risitano, A Hill, H Schrezenmeier, C-L Fu, J

Maciejewski, SA Rollins, CF Mojcik, RP Rother, L Luzzatto. The complement inhibitor

Eculizumab in Paroxysmal Nocturnal Hemoglobinuria. N Engl J Med 2006; 355: 1233-

1243

133. A Barnett, R Nicholas, E Asgari, P Chowdhury, SH Sack, A Dorling, N Mamode.

The use of Eculizumab in renal transplantation. Clinical transplantation 2013; 27:

E216-229

134. MD Stegall, T Diwan, S Raghavaiah, LD Cornell, J Burns, PG Dean, FG Cosio, MJ

Gandhi, W Kremers, JM Gloor. Terminal complement inhibition decreases antibody-

mediated rejection in sensitized renal transplant recipients. Am J Transplant 2011;

11(11): 2405-2413

135. A Bentall, DB Tyan, F Sequeira, MJ Everly, MJ Gandhi, LD Cornell, H Li, NA

Henderson, S Raghavaiah, JL Winters, PG Dean, MD Stegall. Antibody-mediated

rejection despite inhibition of terminal complement. Transplant Int 2014; 27(12):

1235-1243

136. DK Perry, JM Burns, HS Pollinger, BP Amiot, JM Gloor, GJ Gores, MD Stegall.

Proteosome inhibition causes apoptosis of normal human plasma cells preventing

alloantibody production. Am J Transplant 2009; 9(1): 201-209

137. MJ Everly, JJ Everly, B Süsskind, P Brailey, LJ Arend, RR Alloway, P Roy-

Chaudhury, A Govil, G Mogilishetty, AH Rike, M Cardi, G Wadih, A Tevar, ES Woodle.

Bortezomib provides effective therapy for antibody- and cell-mediated acute

rejection. Transplantation 2008; 86(12): 1754-1761

200

138. AD Bannett, RF McAlack, R Raja, A Baquero, M Morris. Experiences with

known ABO-mismatched renal transplants. Transplant Proc 1987; 19: 4543-4546

139. MW Chopek, RL Simmons, JL Platt. ABO-incompatible renal transplantation:

Initial immunopathologic evaluation. Transplant Proc 1987; 19: 4553-4557

140. GPJ Alexandre, JP Squifflet, M De Bruyere, D Latinne, R Reding, P Gianello, M

Carlier, V Pirson. Present experiences in a series of 26 ABO-incompatible living donor

renal allografts. Transplant Proc 1987; 19: 4538-4542

141. AD Bannett, RF McAlack, M Morris, MW Chopek, JL Platt. ABO incompatible

renal transplantation: A qualitative analysis of native endothelial tissue ABO antigens

after transplant. Transplant Proc 1989; 21: 783-785

142. M Haas, MH Rahman, LC Racusen. C4d and C3d staining in biopsies of ABO-

and HLA-incompatible renal allografts: Correlation with histologic findings. Am J

Transplant 2006; 6: 1829-1840

143. M Mengel, J Bogers, JL Bosmans. Incidence of C4d stain in protocol biopsies

from renal allografts: Results from a multicenter trial. Am J Transplant 2005; 5: 1050-

1056

144. K Mizutani, P Terasaki, A Rosen, V Esquenazi, J Miller, RNJ Smith, R Pei, M

Ozawa, J Lee. Serial ten-year follow-up of HLA and MICA antibody production prior to

kidney graft failure. Am J Transplant 2005; 5(9): 2265-2272

145. FK Port, RA Wolfe, EA Mauger, DP Berling, K Jiang. Comparison of survival

probabilities for dialysis patients vs cadaveric renal transplant recipients. JAMA 1993;

270: 1339-1343

146. RA Wolfe, VB Ashby, EL Milford, AO Ojo, RE Ettenger, LY Agodoa, PJ Held, FK

Port. Comparison of mortality in all patients on dialysis, patients on dialysis awaiting

transplantation, and recipients of a first cadaveric transplant. N Engl J Med 1999; 341:

1725-1730

147. RA Montgomery, AA Zachary. Transplanting patients with a positive donor-

specific crossmatch: A single center’s perspective. Pediatr Transplant 2004; 8: 535-

542

148. SC Jordan, A Vo, S Bunnapradist, M Toyoda, A Peng, D Puliyanda, E Kamil, D

Tyan. Intravenous immune globulin treatment inhibits crossmatch positivity and

allows for successful transplantation of incompatible organs in living-donor and

cadaver recipients. Transplantation 2003; 76: 631-636

149. MD Pescovitz. Rituximab, an anti-CD20 monoclonal antibody: History and

mechanism of action. Am J Transplant 2006; 6: 859-866

201

150. H-P Hartnug. Advances in the understanding of the mechanisms of IVIG. J

Neurol 2008; 255: 3-6

151. MD Stegall, J Gloor, JL Winters, SB Moore, S Degoey. A comparison of

plasmapheresis versus high-dose IVIG desensitization in renal allograft recipients with

high levels of donor specific alloantibody. Am J Transplant 2006; 6: 346-351

152. K Marfo, A Lu, M Ling, E Akalin. Desensitization protocols and their outcome.

CJASN 2001; 6(4): 922-936

153. RA Montgomery, BE Lonze, KE King, ES Krause, LM Kucirka, JE Locke, DS

Warren, CE Simpkins, NN Dagher, AL Singer, AA Zachary, DL Segev. Desensitization in

HLA-incompatible kidney recipients and survival. N Engl J Med 2011; 365(4): 318-326

154. FH Claas, Y Gijbels, A von Veen, LP de Waal, J D’Amaro, GG Persijn, JJ van

Rood. Selection of cross-match negative HLA-A and/or –B mismatched donors for

highly sensitized patients. Transpl Proc 1989; 21: 665-667

155. FH Claas, J De Meester, MD Witvliet, JM Smits, GG Persijn, II Doxiadis.

Acceptable HLA mismatches for highly immunized patients. Rev Immunogenet 1999;

1(3): 351-8

156. RJ Duquesnoy, S Takemoto, P De Lange, IIN Doxiadis, GM Schreuder, GG

Persijn, FH Claas. HLAMatchmaker: a molecularly based algorithm for

histocompatibility determination. III. Effect of matching at the HLA-A, B amino acid

triplet level on kidney transplant survival. Transplantation 2003; 75(6): 884-889

157. FHJ Claas, MD Witvliet, RJ Duquesnoy, GG Persijn, IIN Doxiadis. The acceptable

mismatch program as a fast tool for highly sensitised patients awaiting a cadaveric

kidney transplantation: Short waiting time and excellent graft outcome.


158. FHJ Claas, A Rahmel, IIN Doxiadis. Enhanced kidney allocation to highly

sensitised patients by the acceptable mismatch program. Transplantation 2009;

88(4): 447-452

159. S Heidt, MD Witvliet, GW Haasnoot, FHJ Claas. The 25th anniversary of the

Eurotransplant Acceptable Mismatch program for highly sensitized patients. Transpl

Immunol 2015; 33(2): 51-57

160. FT Rappaport. The case for a living emotionally related international kidney

donor registry. Transplant Proc 1986; 18: 5-9

161. K Park, JI Moon, SI Kim, YS Kim. Exchange donor program in kidney

transplantation. Transplantation 1999; 67(2): 336-338

202

162. M de Klerk, KM Keizer, FH Claas, M Witvliet, BJ Haase-Kromwijk, W Weimar.

The Dutch national living donor kidney exchange program. Am J Transplant 2005;

5(9): 2302-2305

163. M de Klerk, MD Witvliet, BJ Haase-Kromwijk, FH Claas, W Weimar. A highly

efficient living donor kidney exchange program for both blood type and crossmatch

incompatible donor-recipient combinations. Transplantation 2006; 82: 1616–1620

164. G Vlad, EK Ho, ER Vasilescu, AI Colovai, MB Stokes, GS Markowitz, VD D’Agati,

DJ Cohen, LE Ratner, N Suciu-Foca. Relevance of different antibody detection

methods for the prediction of antibody-mediated rejection and deceased-donor

kidney allograft survival. Hum Immunol 2009; 70: 589-594

165. A Gupta, P Sinnott. Clinical relevance of pretransplant human leukocyte

antigen donor-specific antibodies in renal patients waiting for a transplant: A risk

factor. Hum Immunol 2009; 70: 618-622

166. D Phelan, T Mohanakumar, S Ramachandran, MD Jendrisak. Living donor renal

transplantation in the presence of donor-specific human leukocyte antigen antibody

detected by solid-phase assay. Hum Immunol 2009; 70: 584-588

167. V Aubert, J-P Venetz, G Pantaleo, M Pascual. Low levels of human leukocyte

antigen donor-specific antibodies detected by solid phase assay before

transplantation are frequently clinically irrelevant. Hum Immunol 2009; 70: 580-583

168. P Amico, G Honger, M Mayr, J Steiger, H Hopfer, S Schaub. Clinical relevance of

pretransplant donor-specific HLA antibodies detected by Single Antigen flow-beads.


169. C Lefaucheur, A Loupy, GS Hill, J Andrade, D Nochy, C Antoine, C Gautreau, D

Charron, D Glotz, C Suberbielle-Boissel. Preexisting donor-specific HLA antibodies

predict outcome in kidney transplantation. J Am Soc Nephrol 2010; 21: 1398-1406

170. S Riethmüller, S Ferrari-Lacraz, MK Müller, DA Raptis, K Hadaya, B Rüsi, G

Laube, G Schneiter, T Fehr, J Villard. Donor-specific antibody levels and three

generations of crossmatches to predict antibody-mediated rejection in kidney

transplantation. Transplantation 2010; 90: 160-167

171. HG Otten, MC Verhaar, HP Borst, RJ Hene, AD van Zuilen. Pretransplant donor-

specific HLA class-I and –II antibodies are associated with an increased risk for kidney

graft failure. Am J Transplant 2012; 12: 1618-1623

172. D Thammanichanoud, A Ingsathit, T Mongkolsuk, S Rattanasiri, S

Kantachuvesiri, C Sakhonrat, C Leenanupan, S Worawichawong, P Kitpoka. Pre-

transplant donor specific antibody and its clinical significance in kidney

transplantation. Asian Pac J Allergy Immunol 2012; 30: 48-54

203

173. E David-Neto, P Souza, N Panajotopoulos, H Rodrigues, CG Ventura, DSR David,

FBC Lemos, F Agena, WC Natias, JE Kalil, MCR Castro. The impact of pretransplant

donor-specific antibodies on graft outcome in renal transplantation: A six-year

follow-up study. Clinics 2012; 67: 355-361

174. I Pérez-Flores, JL Santiago, N Calvo-Romero, A Barnentos-Guzmán, AI

Sáanchez-Fructuoso. Pretransplant anti-HLA antibodies detected by Luminex in highly

sensitized renal transplanted patients. Biomed Res Int 2013: 738404

175. P Wu, J Jin, MJ Everly, C Lin, PI Terasaki, J Chen. Impact of alloantibody

strength in crossmatch negative DSA positive kidney transplantation. Clin Biochem

2013; 46: 1389-1393

176. DS Tsapepas, R Vasilescu, B Tanriover, Y Coppleson, Y Rekhtman, MA Hardy, G

Dube, RJ Crew, LE Ratner, DJ Cohen, S Mohan. Preformed donor-specific antibodies

and risk of antibody-mediated rejection in repeat renal transplantation.


177. J Malheiro, S Tafulo, L Dias, LS Martins, I Fonseca, I Beirão, A Castro-Henriques,

A Cabrita. Analysis of preformed donor-specific anti-HLA antibodies characteristics

for prediction of antibody-mediated rejection in kidney transplantation. Transpl

Immunol 2015; 32: 66-71

178. D Lowe, M Hathaway, D Briggs. The high dose hook effect in the detection and

monitoring of HLA specific antibody by Luminex assay. Int J Immunogenet 2007; 34:

288

179. C Weinstock, M Schnaidt. The complement-mediated prozone effect in the

Luminex single-antigen bead assay and its impact on HLA antibody determination in

patient sera. Int J Immunogenet 2013; 40(3): 171-177

180. E Schwaiger, M Wahrmann, G Bond, F Eskandry, GA Böhmig. Complement

component C3 activation: The leading cause of the prozone phenomenon affecting

HLA antibody detection on single-antigen beads. Transplantation 2014; 97(12): 1279-

1285

181. G Guidicelli, G Anies, T Bachelet, V Dubois, JF Moreau, P Merville, L Couzi, L

Taupin. The complement interference phenomenon as a cause for sharp fluctuations

of serum anti-HLA antibody strength in kidney transplant patients. Transpl Immunol

2013; 29: 17-21

182. J Visentin, M Vigata, S Daburon, C Contin-Bordes, V Fremeaux-Bacchi, C

Dromer, MA Billes, M Neau-Cransac, G Guidicelli, JL Taupin. Deciphering

complement interference in anti-human leukocyte antigen detection with flow beads

assays. Transplantation 2014; 98(6): 625-631

204

183. M Schnaidt, C Weinstock, M Jurisic, B Schmid-Horch, A Ender, D Wernet. HLA

antibody specification using single-antigen beads – A technical solution for the

prozone effect. Transplantation 2011; 92(5): 510-515

184. V Kosmoliaptsis, JA Bradley, S Peacock, AN Chaudhry, CJ Taylor. Detection of

immunoglobulin G human leukocyte antigen-specific alloantibodies in renal

transplant patients using single-antigen beads is compromised by the presence of

immunoglobulin M human leukocyte antigen-specific alloantibodies. Transplantation

2009; 87(6); 813-820

185. AA Zachary, DP Lucas, B Detrick, MS Leffell. Naturally occurring interference in

Luminex® assays for HLA-specific antibodies: Characteristics and resolution. Hum

Immunol 2009; 70(7): 496-501

186. PR Warner, D Youngs, KA Nelson. HLA density on microparticles versus

antibody detection sensitivity: Implications for crossmatch prediction. Hum Immunol

2007; 68(1): 528

187. S Hemmatpour, S Todd, D Sage. Variation of antigen density on HLA single

antigen beads – Implications for sensitised patients awaiting kidney transplantation.

Transpl Int 2011; 24(52): 354

188. LE Morales-Buenrostrom PI Terasaki, LA Marino-Vázquez, JH Lee, N El-Awar, J

Alberu. “Natural” human leukocyte antigen antibodies found in nonalloimmunized

healthy males. Transplantation 2008; 86(8): 1111-1115

189. N El-Awar, PI Terasaki, A Nguyen, N Sasaki, LE Morales-Buenrostro, H Saji, E

Maruya, F Poli. Epitopes of HLA antibodies found in sera of normal healthy males and

cord blood. Clin Transpl 2008: 199-214

190. N El-Awar, JH Lee, C Tarsitani, PI Terasaki. HLA class I epitopes: Recognition of

binding sites by mAbs or eluted alloantibodies confirmed with single recombinant

antigens. Hum Immunol 2007; 68(3): 170-180

191. HG Otten, MC Verhaar, HPE Borst, M van Eck, WGJ van Ginkel, RJ Hene, AD van

Zuilen. The significance of pretransplant donor-specific antibodies reactive with

intact or denatured human leucocyte antigen in kidney transplantation. Clin Exp

Immunol 2013; 173: 536-543

192. P Wu, MJ Everly, LM Rebellato, CE Haisch, KP Briley, P Bolin, WT Kendrick, C

Morgan, RC Harland, PI Terasaki. Trends and characteristics in early glomerular

filtration rate decline after post transplantation alloantibody appearance.


193. G Opelz, B Graver, PI Terasaki. Induction of high kidney graft survival rate by

multiple transfusion. Lancet 1981; 1: 1223-1225; MA Blajchman. Immunomodulatory

205

effects of allogeneic blood transfusions: Clinical manifestations and mechanisms. Vox

Sang 1998; 74: 315-319

194. FH Claas, DL Roelen, JJ van Rood, A Brand. Modulation of the alloimmune

response by blood transfusion. Transfus Clin Biol 2001; 8: 315-31

195. JC Scornik, JE Ireland, RJ Howard, WW Pfaff. Assessment of the risk for broad

sensitization by blood transfusions. Transplantation 1984; 37: 249-253

196. S Vaidya, D Partlow, B Susskind, M Noor, T Barnes, K Gugliuzza. Prediction of

crossmatch outcome of highly sensitized patients by single and / or multiple antigen

bead luminex assay. Transplantation 2006; 82: 1524-1528

197. AR Tambur, DS Ramon, DB Kaufman, J Friedwald, X Luo, B Ho, A Skaro, J

Caicedo, D Ladner, T Baker, J Fryer, L Gallon, J Miller, MM Abecassis, J Leventhal.

Perception versus reality?: Virtual crossmatch – How to overcome some of the

technical and logistic limitations. Am J Transplant 2009; 9: 1886-1893

198. GP Morris, DL Phelan, MD Jendrisak, T Mohanakumar. Virtual crossmatch by

identification of donor-specific anti-human leukocyte antigen antibodies by solid-

phase immunoassay: A 30-month analysis in living donor kidney transplantation. Hum

Immunol 2010; 71: 268-273

199. CM Worsley, ES Mayne, MS Suchard. Luminex-based virtual crossmatching for

renal transplantation in South Africa. S Afr Med J 2012; 102: 40-43; G Schlaf, B Pollok-

Kopp, WW Alterman. Solid phase-based cross-matching in order to avoid kidney

allografting against donor-specific anti-HLA antibodies: Long term experience with a

procedure allowing highly reliable diagnoses. Austin J Nephrol Hypertens 2014; 1:

1028

200. A Piazza, G Ozzella, E Poggi, D Caputo, A Manfreda, D Adorno. Virtual

crossmatch in kidney transplantation. Transplant Proc 2014; 46: 2915-2918

201. L Cantwell, C Woodroffe, R Holdsworth, P Ferrari. Four years of experience

with the Australian kidney paired donation programme. Nephol 2015; 20(30): 124-

131

202. D Dragun. Humoral responses directed against non-human leukocyte antigens

in solid organ transplantation. Transplantation 2008; 86(8): 1019-1025

203. D Dragun, A Phillippe, R Catar. Role of non-HLA antibodies in organ

transplantation. Curr Opin Organ Transplant 2012; 17(4): 440-445

204. S Sumitran-Holgersson. Relevance of MICA and other non-HLA antibodies in

clinical transplantation. Curr Opin Immunol 2008; 20(5): 607-613

205. D Glotz, C Antoine, P Julia, C Suberbielle-Boissel, S Boudjeltia, R Fraoui, C

Hacen, A Duboust, J Bariety. Desensitization and subsequent kidney transplantation

206

of patients using intravenous immunoglobulin (IVIg). Am J Transplant 2002; 2: 758-

760

206. RA Montgomery, BE Lonze, KE King, ES Kraus, LM Kucirka, JE Locke, DS Warren,

CE Simpkins, NN Dagher, AL Singer, AA Zachary, DL Segev. Desensitization in HLA-

incompatible kidney recipients and survival. N Engl J Med 2011; 365: 318-326

207. BJ Orandi, X Luo, AB Massie, JM Garonzik-Wang, BE Lonze, R Ahmed, KJ Van

Arendonk, MD Stegall, SC Jordan, J Oberholzer, TB Dunn, LE Ratner, S Kapur, RP

Pelletier, JP Roberts, ML Melcher, P Singh, DL Sudan, MP Posner, JM El-Amm, R

Shapiro, M Cooper, GS Lipkowitz, MA Rees, CL Marsh, BR Sankari, DA Gerber, PW

Nelson, J Wellen, A Bozorgzadeh, AO Gaber, RA Montgomery, DL Segev. Survival

benefit with kidney transplants from HLA-incompatible live donors. N Engl J Med

2016; 374: 940-950

208. V Kosmoliaptsis, LD Sharples, AN Chaudhry, DJ Halsall, JA Bradley, CJ Taylor.

Predicting HLA class II alloantigen immunogenicity from the number and

physiochemical properties of amino acid polymorphisms. Transplantation 2011; 91:

183-190

209. V Kosmoliaptsis, TR Dafforn, AN Chaudhry, DJ Halsall, JA Bradley, CJ Taylor.

High-resolution, three-dimensional modelling of human leukocyte antigen class I

structure and surface electrostatic potential reveals the molecular basis for

alloantibody binding epitopes. Hum Immunol 2011; 72(11): 1049-1059

210. VC Joysey, E Wolf. HLA-A, -B and –C antigens, their serology and cross-

reaction. Br Med Bull 1978; 34: 217-222

211. RJ Duquesnoy. HLAMatchmaker: A molecularly based algorithm for

histocompatibility determination: I. Description of the algorithm. Hum Immunol

2002; 63: 339-352

212. RJ Duquesnoy, M Askar. HLAMatchmaker: A molecularly based algorithm for

histocompatibility determination v. Eplet matching for HLA-DR, HLA-DQ and HLA-DP.

Hum Immunol 2007; 68(1): 12-25

213. RJ Duquesnoy. The antibody response to an HLA mismatch: a model for

nonself-self discrimination in relation to HLA epitope immunogenicity. Int J

Immunogenet 2012; 39(1): 1-9

214. M Resse, R Paolillo, BP Minucci, G Moccia, C Napoli. Antibody-reactive class I

epitopes defined by pairs of mismatched eplets and self-eplets. Tissue Antigens 2015;

86(5): 368-372

207

215. N El-Awar, JH Lee, C Tarsitani, PI Terasaki. HLA class I epitopes: Recognition of

binding sites by mAbs or eluted alloantibody confirmed with single recombinant

antigens. Hum Immunol 2007; 68: 170-180

216. RJ Duquesnoy, M Marrari. Correlations between Terasaki’s HLA class I epitopes

and HLAMatchmaker-defined eplets on HLA-A, -B and –C antigens. Tissue Antigens

2009; 74(2): 117-133

217. M Marrari, RJ Duquesnoy. Correlations between Terasaki’s HLA class II

epitopes and HLAMatchmaker-defined eplets on HLA-DR and -DQ antigens. Tissue

Antigens 2009; 74(2): 134-146

218. MKA Dankers, MD Witvliet, DL Roelen, P De Lange, N Korfage, GG Persijn, R

Duquesnoy, IIN Doxiadis, FH Claas. The number of amino acid triplet differences

between patient and donor is predictive for the antibody reactivity against

mismatched human leukocyte antigens. Transplantation 2004; 77(8): 1236-1239

219. RS Goodman, CJ Taylor, CM O’Rourke, A Lynch, JA Bradley, T Key. Utility of

HLAMatchmaker and single-antigen HLA-antibody detection beads for identification

of acceptable mismatches in highly sensitised patients awaiting kidney

transplantation. Transplantation 2006; 81(9): 1331—1336

220. V Kosmoliaptsis, JA Bradley, LD Sharples, A Chaudhry, T Key, RS Goodman, CJ

Taylor. Predicting the immunogenicity of human leukocyte antigen class I

alloantigens using structural epitope analysis determined by HLAMatchmaker.


221. AA Zachary, D Kopchaliiska, RA Montgomery, MS Leffell. HLA-specific B cells. i.

A method for their detection, quantification, and isolation using HLA tetramers.


222. M Han, JA Rogers, B Lavingia, P Stastny. Peripheral blood B cells producing

donor-specific HLA antibodies in vitro. Hum Immunol 2009; 70: 29-34

223. S Heidt, DL Roelen, YJH de Vaal, MGD Kester, C Eijsink, S Thomas, NM van

Besouw, HD Volk, W Weimer, FHJ Claas, A Mulder. A novel ELISPOT assay to quantify

HLA-specific B cells in HLA-immunized individuals. Am J Transplant 2012; 12: 1469-

1478

224. M Lucia, S Luque, E Crespo, E Melilli, JM Cruzado, J Martorell, M Jarque, S Gil-

Vernet, A Manonelles, JM Gringό, O Bestard. Preformed circulating HLA-specific

memory B cells predict high risk of humoral rejection in kidney transplantation.

Kidney International 2015; 88: 874-887

225. DB Tyan. New approaches for detecting complement-fixing antibodies. Curr

Opin Organ Transplant 2012; 17: 409-415

208

226. M Wahrmann, G Bartel, M Exner, H Regele, GF Körmöczi, GF Fischer, GA

Böhmig. Clinical relevance of preformed C4d-fixing and non-C4d-fixing HLA single

antigen reactivity in renal allograft recipients. Transpl Int 2009; 22(10): 982-989

227. G Hönger, M Wahrmann, P Amico, H Hopfer, GA Böhmig, S Schaub. C4d-fixing

capability of low-level donor-specific HLA antibodies is not predictive for early

antibody-mediated rejection. Transplantation 2010; 89(12): 1471-1475

228. JM Yabu, JP Higgins, G Chen, F Sequeira, S Busque, DB Tyan. C1q-fixing human

leukocyte antigen antibodies are specific for predicting transplant glomerulopathy

and late graft failure after kidney transplantation. Transplantation 2011; 91(3): 342-

347

229. SM Sutherland, G Chen, FA Sequeira, CD Lou, SR Alexander. Complement-fixing

donor-specific antibodies identified by a novel C1q assay are associated with allograft

loss. Pediatr Transplantation 2012; 16(1): 12-17

230. C Lawrence, M Willicombe, PA Brookes, E Santos-Nunez, R Bajaj, T Cook, C

Roufosse, D Taube, AN Warrens. Preformed complement-activating low-level donor-

specific antibody predicts early antibody-mediated rejection in renal allografts.


231. M Crespo, A Torio, V Mas, D Redondo, MJ Pérez-Sáez, M Mir, A Faura, R

Guerra, O Montes-Ares, MD Checa, J Pascual. Clinical relevance of pretransplant anti-

HLA donor-specific antibodies: Does C1q fixation matter? Transpl Immunol 2013; 29:

28-33

232. A Loupy, C Lefaucheur, D Vernerey, C Prugger, J-P Duong van Huyen, N

Mooney, C Suberbielle, V Frémeaux-Bacchi, A Méjean, F Desgrandchamps, D

Anglicheau, D Nochy, D Charron, J-P Empana, M Delahousse, C Legendre, D Glotz, GS

Hill, A Zeeva, X Jouven. Complement-binding anti-HLA antibodies and kidney-allograft

survival. N Engl J Med 2013; 369: 1215-1226

233. A Zeevi, J Lunz, B Feingold, M Shullo, C Bermudez, J Teuteberg, S Webber.

Persistent strong anti-HLA antibody at high titer is complement binding and

associated with increased risk of antibody-mediated rejection in heart transplant

recipients. J Heart Lung Transplant 2013; 32(1): 98-105

234. M Yell, BL Muth, DB Kaufman, A Djarnali, TM Ellis. C1q binding activity of de

novo donor-specific HLA antibodies in renal transplant recipients with and without

antibody-mediated rejection. Transplantation 2015; 99: 1151-1155

235. D Thammanichanond, P Wiwattanathum, T Mongkolsuk, S Kantachuvesiri, S

Worawichawong, SA Vallipakorn, P Kitpoka. Role of pretransplant complement-fixing

209

donor-specific antibodies identified by C1q assay in kidney transplantation.

Transplant Proc 2016; 48(3): 756-760

236. KA Thomas, NM Valenzuela, EF Reed. The perfect storm: HLA antibodies,

complement, FcϒRs and endothelium in transplant rejection. Trends Mol Med 2015;

21(5): 319-329

237. RM Anthony, F Nimmerjahn. The role of differential IgG glycosylation in the

interaction of antibodies with FcγRs in vivo. Curr Opin Organ Transplant 2011; 16(1):

7-14

210

Appendix

211

Appendix

Table I. Conexio Genomics HLA Sequencing Based Typing kit details. The PRE-PCR

component of each kit consists of a vial/s of a locus-specific PCR mix (e.g. CG-HLA-A mix )

consisting of PCR buffer, dNTPs, MgCl2, locus specific PCR primers, and a single vial of

DNA polymerase (DNA Pol ). The POST-PCR kit contains sequencing primers (e.g. AEX1F).

Kit Catalogue

No

No Tests Pre-PCR contents Post PCR contents

Reagent Volume Primer

Name

Volume

HLA-A XH-PD1.1-2(50) XH-PD1.1-

2(100)

50 tests 100 tests

DNA Pol

CG-HLA-A mix

DNA Pol

CG-HLA-A mix

1 x 55ul

1x 880ul

1 x 110ul

2 x 880ul

AEX1F,

AEX1R,

AEX2F,

AEX2R,

AEX3F,

AEX3R,

AEX4F,

AEX4R

AEX1F,

AEX1R,

AEX2F,

AEX2R,

AEX3F,

AEX3R,

AEX4F,

AEX4R

1 x 110ul of each

1 x 220ul of each

HLA-B BS-PD2.1-2(50) BS-PD2.1-

2(100)

50 tests

100 tests

DNA Pol

CG-HLA-B mix

DNA Pol

1 x 55ul

1x 880ul

1 x 110ul

BEX1F,

BEX2F,

BEX2R,

BEX3F,

BEX3R,

BEX4F,

BEX4R

BEX1F,

BEX2F,

1 x 110ul of each

1 x 220ul of each

212

CG-HLA-B mix

2 x 880ul

BEX2R,

BEX3F,

BEX3R,

BEX4F,

BEX4R

HLA-C HH-PD 3.2-2(50) HH-PD

3.2-2(100)

50 tests

100 tests

DNA Pol

CG-HLA-C mix

DNA Pol

CG-HLA-C mix

1 x 55ul

1x 880ul

1 x 110ul

2 x 880ul

CEX1F,

CEX1R,

CEX2F,

CEX2R,

CEX3F,

CEX3R,

CEX4F,

CEX4R,

CEX5F,

CEX5R,

CEX6F,

CEX6R,

CEX7F,

CEX7R,

CEX8F

CEX1F,

CEX1R,

CEX2F,

CEX2R,

CEX3F,

CEX3R,

CEX4F,

CEX4R,

CEX5F,

CEX5R,

CEX6F,

CEX6R,

CEX7F,

CEX7R,

CEX8F

1 x 110ul of each

1 x 220ul of each

DRB1 HH-PD5.2-4(50) HH-PD5.2-4(100)

50 tests

100 tests

DNA Pol

CG-HLA-DRB1

mix

DNA Pol

1 x 18ul

1 x 920ul

1 x 36ul

DRB1EX2F,

DRB1EX3R,

DRB1EX3R,

RB-TG344-R

DRB1EX2F,

DRB1EX3R,

DRB1EX3R,

RB-TG344-R

1 x 110ul of each

1 x 220ul each

213

CG-HLA-DRB1

mix

2 x 920ul

DQB1 PQ-PD6.2-1(50) PQ-PD6.2-

1(100)

50 tests

100 tests

DNA Pol

CG-HLA-DQB1

mix

DNA Pol

CG-HLA-DQB1

mix

1 x 18ul

1 x 920ul

1 x 36ul

2 x 920ul

DQB1EX2F,

DQB1EX2R,

DQB1EX3F.

DQB1EX3R

DQB1EX2F,

DQB1EX2R,

DQB1EX3F.

DQB1EX3R

1 x 110ul of each

1 x 220ul of each

Documents

The Clinical Significance of HLA antibodies detected by ... · transplant rates of highly sensitized recipients with virtual crossmatching in kidney paired donation. Transplantation