Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
1
The Cross-species Mycobacterial Growth Inhibition Assay (MGIA) Project 2010-2014 1
2
Michael J. Brennan1+
, Rachel Tanner2*+
, Sheldon Morris3, Thomas J. Scriba
4, Jacqueline M. 3
Achkar5, Andrea Zelmer
6, David Hokey
1, Angelo Izzo
7, Sally Sharpe
8, Ann Williams
8, Adam 4
Penn-Nicholson4, Mzwandile Erasmus
4, Elena Stylianou
2, Daniel F. Hoft
9, Helen McShane
2, 5
Helen A. Fletcher6
6
7
1Aeras, Rockville, MD, 20850 USA;
2The Jenner Institute, University of Oxford, Oxford, OX3 8
7DQ UK; 3Center for Biologics, Evaluation and Research, Food and Drug Administration, 9
20892 USA; 4South African Tuberculosis Vaccine Initiative and Division of Immunology, 10
Department of Pathology, University of Cape Town, 7925 South Africa;
5Departments of 11
Medicine, Microbiology and Immunology, Albert Einstein College of Medicine, Bronx, NY, 12
10461 USA; 6Department of Immunology and Infection, London School of Hygiene & 13
Tropical Medicine, London, W1CE 7HT UK; 7Colorado State University, Fort Collins, 14
Colorado, 80523 USA; 8Public Health England, Porton Down, Salisbury, SP4 0JG UK; 15
9Saint Louis University, St. Louis, MO, 63104 USA. 16
17
*Corresponding author: Rachel Tanner. Email: [email protected] 18
+ denotes shared first authorship 19
Keywords: Mycobacterial Growth Inhibition Assay (MGIA); Tuberculosis; Correlates of 20
immunity; Vaccines 21
22
Abstract 23
The development of a functional biomarker assay in the tuberculosis (TB) field would be 24
widely recognized as a major advance in efforts to efficiently develop and test novel TB 25
CVI Accepted Manuscript Posted Online 12 July 2017Clin. Vaccine Immunol. doi:10.1128/CVI.00142-17Copyright © 2017 Brennan et al.This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license.
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
2
vaccine candidates. We present preliminary studies using mycobacterial growth inhibition 26
assays (MGIAs) to detect a BCG vaccine response across species, and extend this work to 27
determine if a standardized MGIA can be applied in characterizing new generation TB 28
vaccines. The comparative MGIA studies reviewed here aimed to evaluate robustness, 29
reproducibility, and ability to reflect in vivo responses. In doing so, they have laid the 30
foundation for the development of an MGIA that can be standardized and potentially 31
qualified. The major challenge ahead lies in better understanding the relationship between in 32
vivo protection, in vitro growth inhibition and the immune mechanisms involved. The final 33
outcome would be an MGIA that can be used in confidence in TB vaccine trials. We 34
summarize data arising from this project, present a strategy to meet the goals of developing a 35
functional assay for TB vaccine testing, and describe some of the challenges encountered in 36
performing and transferring such assays. 37
38
Introduction 39
The TB vaccine field is severely hampered by the lack of defined immune parameters that 40
correlate with vaccine efficacy in human studies. There has been recent progress, such as the 41
finding that antigen-specific IFN-γ ELISpot responses correlate with reduced risk of 42
developing disease in BCG-vaccinated South African infants[1]. However, heterogeneous 43
BCG vaccine responses are observed which may be associated with the immune environment 44
at the time of vaccination, including the proportion of monocytes and activated HLA-DR+ 45
CD4+ T cells[1, 2]. Development of a validated functional ‘biological response’ assay 46
capable of assessing the potential of a vaccine-induced immune response to protect against 47
Mycobacterium tuberculosis (M.tb) infection or TB disease would represent a major advance 48
in the effort to develop TB vaccines. Such an assay would circumvent the need to identify or 49
fully understand the specific aspects of the immune response contributing to this effect. 50
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
3
Mycobacterial growth inhibition assays (MGIAs) represent one such clinically relevant 51
approach. Using whole blood or peripheral blood mononuclear cells (PBMCs), MGIAs offer 52
an unbiased measure of the vaccine-induced antigen-specific immune response, taking into 53
account the ability of this response to function in its respective immune environment. They 54
also have the potential to shorten the time-frame and resources required to evaluate a new TB 55
vaccine in preclinical studies. 56
The MGIA project brought together an international group of scientific experts to work on 57
growth inhibition assays from 2010 to 2014. Participants included researchers from TB 58
endemic regions such as Cape Town, South Africa; as well as non-endemic areas such as the 59
UK and Washington DC, Saint Louis and Colorado in the US. 60
61
Goals of the MGIA project and assay selection 62
The MGIA project was initially supported by a World Health Organization (WHO) working 63
group, and was viewed as the follow-up to a whole blood ELISA assay that the WHO had 64
previously recommended[3]. The aim was to develop a functional assay that could be applied 65
in all clinical trials of TB vaccine candidates, thus aiding the identification of correlates of 66
immunity. Although several MGIAs were described in the literature previously (recently 67
reviewed in [4]), little work had been done on qualifying an assay that could be transferred 68
across laboratories using a standardized reproducible method for vaccine assessment. This 69
was a major goal of the project, funded by Aeras and the US Food and Drug Administration 70
(FDA). Prior to 2010, most MGIA development had focused on whole blood systems and had 71
shown promise[5-10]. However, especially given the logistical challenges associated with 72
real-time processing of fresh blood in clinical vaccine trials, in these studies, efforts were 73
made to assess the potential of using cryopreserved PBMC and to compare the outcomes with 74
those from whole blood. Other goals included determining the effect of M.tb infection status 75
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
4
on mycobacterial growth inhibition, comparing MGIA responses induced by wild-type BCG 76
and a novel recombinant BCG vaccine, and investigating the induction and functional role of 77
antibodies. The specific early goals are outlined in Table 1. 78
Although four previously-published MGIAs were considered for initial evaluation, a whole 79
blood assay based on the methods of Wallis et al. was selected as a focus for this project, 80
using the BACTEC MGIT system to quantitate mycobacterial growth[5] (henceforth referred 81
to as the ‘direct MGIA’). The rationale for this decision included the relative simplicity of the 82
assay and the use of standardized reagents and equipment that are readily available in many 83
clinical TB laboratories. Furthermore, the assay had previously been used to demonstrate 84
anti-mycobacterial activity of bactericidal drugs that correlated with sterilizing activity 85
observed during in vivo therapy[5, 11]. The BACTEC MGIT system has several advantages 86
over traditional colony-forming unit (CFU) based methods of mycobacterial quantification, 87
as it utilizes oxygen depletion as a detection method which is sensitive not only to the number 88
of viable bacteria, but also their rate of metabolism and growth. It is unaffected by clumping 89
and does not require serial dilutions, providing an accurate computer-generated read-out 90
based on validated technology. 91
It was also agreed that an adaptation of this assay using cryopreserved PBMC be developed 92
and evaluated for comparison with whole blood. It is widely accepted that the cellular 93
immune response is central to protection against M.tb, and although innate whole blood 94
components such as neutrophils are known to play a role in the host response, they may be 95
less relevant when considering a vaccine-induced effect on adaptive immunity. Furthermore, 96
use of cryopreserved cells could aid in transferability of the assay to different trial sites, 97
eliminating the need to evaluate samples in real-time using fresh material. Rather, cells could 98
be stored and run in the future which is not only logistically simpler, but would negate the 99
need for a BACTEC MGIT machine at every site. Finally, the ability to use cryopreserved 100
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
5
cells would enable additional retrospective study of samples from historical clinical vaccine 101
trials for validation and exploratory work. A schematic and description of the method for the 102
direct MGIA assay is provided in Figure 1. 103
The studies involved in this project are presented chronologically; work began using samples 104
from clinical studies (as the ultimate goal was development of a human assay), but the 105
challenges encountered resulted in a decision to focus on preclinical adaptations in the latter 106
stages. One of the greatest impediments to initial MGIA developmental efforts using human 107
samples was the lack of availability of sufficiently large single whole blood or PBMC 108
samples from recently BCG vaccinated volunteers. This was required for assay optimization 109
and assessment of reproducibility between laboratories. Using animal samples such as mouse 110
splenocytes, however, it was possible to overcome this issue as 50-100 million splenocytes 111
may be isolated from each mouse. Furthermore, the ability to select age- and sex-matched 112
naïve inbred animals removes a number of potential sources of variability found in human 113
clinical trials. Animal models also provide the opportunity for biological validation by 114
correlating assay outcome with protection from in vivo challenge with pathogenic M.tb (see 115
below). 116
117
Early optimization work 118
BCG was selected as a surrogate for M.tb in the majority of these studies to avoid the need 119
for Biosafety level 3 (BSL3) facilities. It has previously been established that measures of 120
growth inhibition of BCG correlate with those of M.tb H37Rv in a murine MGIA[12], and 121
this was confirmed in the direct MGIA (Figure 2A). Early experiments using the direct 122
PBMC MGIA assay indicated high intra-assay variability between replicate cultures. This 123
was addressed by considering various aspects of assay preparation, the culture period and 96-124
hour processing. It was determined that the standard antibiotic supplement (penicillin and 125
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
6
streptomycin) in culture medium should not be used at any stage of sample processing after 126
thawing, due to the effect of artificially reducing the inoculum[13]. However, this altered the 127
multiplicity of infection (MOI) such that the vaccine effect was overwhelmed. A smaller 128
inoculum resulted in increased intra-assay variability. To maintain the same initial inoculum 129
volume, a lower MOI was achieved by using greater cell concentrations, and although this 130
did increase reproducibility, a balance had to be drawn given the limiting factor of cell 131
number in many clinical trials. Experiments altering the MOI in the direct splenocyte MGIA 132
have been described[14]. Once an MOI was identified, variability in the inoculum itself was 133
addressed by comparing different de-clumping methods such as vortexing, filtering and 134
sonicating. Vortexing with glass beads resulted in low intra-assay variability while retaining a 135
high recovery of CFU. Other variables examined included BCG strain[14], culture rotation 136
methods and lysis agents. Variability between sites in whole blood MGIA performance has 137
yet to be defined, but could result from factors such as differences in blood anticoagulants 138
and inoculum strain. 139
140
Human MGIA Studies 141
1) BCG vaccination and revaccination 142
A clinical trial of BCG vaccination and revaccination in UK adults was conducted at the 143
University of Oxford[13]. The direct PBMC MGIA detected a significant improvement in 144
mycobacterial growth inhibition following primary BCG vaccination, consistent with 145
epidemiological data on the efficacy of BCG in this population. No improvement in 146
mycobacterial growth inhibition was observed following BCG revaccination using either the 147
whole blood or PBMC assay[13]. This is in contrast to a previous study in US volunteers, in 148
which the direct whole blood MGIA, as well as the primary lymphocyte MGIA (effector 149
lymphocytes added to infected autologous monocytes[9]) and secondary lymphocyte MGIA 150
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
7
(using antigen expanded T cells[6]), detected enhanced mycobacterial growth inhibition after 151
BCG revaccination[7]. 152
In the Oxford study, the direct PBMC MGIA demonstrated a stronger primary vaccine effect 153
and greater reproducibility over repeated baseline bleeds compared with whole blood[13]; 154
most likely due to the processing of longitudinal PBMC samples in one batch. This is not an 155
option for fresh blood which may suffer from batch-effects including those associated with 156
different vials of mycobacterial inoculum. However, this was not the case in other studies, 157
such as the AERAS-422 vaccine trial at Saint Louis University described below, in which the 158
direct whole blood MGIA proved less variable among subjects included in each vaccinated 159
group compared to primary, secondary or titrated PBMC MGIAs[15]. 160
A study was also conducted in South Africa, comparing 10 week old infants who were BCG 161
vaccinated at birth with older adults. Infants demonstrated enhanced ability to inhibit growth 162
of BCG SSI (but not M.tb H37Rv) in the direct whole blood MGIA compared with adults 163
(Figure 2B). This data suggest that BCG vaccination may induce preferential growth 164
inhibition of BCG even in newborns. 165
2) Immune correlates of risk study 166
Recently, Fletcher et al. have reported the findings of an immune correlates of risk study in 167
BCG vaccinated infants in South Africa[1]. In this population, BCG-specific IFN- ELISpot 168
responses and levels of Ag85A-specific IgG antibodies measured at 4-6 months of age were 169
associated with a reduced risk of developing TB disease over the next 3 years of life[1]. 170
However, ability to control mycobacterial growth in the direct PBMC MGIA was not 171
associated with reduced risk of TB disease. This could be due to the very low frequency of 172
BCG specific IFN- secreting T-cells (median less than 50 SFC/million PBMC) and the fact 173
that a small number of PBMC were used in the assay (it has since been established that the 174
direct MGIA is more robust when greater numbers of PBMC are used). Furthermore, 175
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
8
autologous serum was not used in the assay and it would therefore not have measured the 176
contribution of Ag85A specific IgG, or any other antibody, to growth inhibition[1]. This was 177
also prior to subsequent further optimization of the assay. 178
3) Comparing wild-type BCG and a new recombinant BCG vaccine 179
Hoft et al. performed the direct whole blood MGIA as part of a clinical trial of AERAS-422; 180
a live recombinant BCG expressing perfringolysin and three M.tb antigens: 85A, 85B, and 181
Rv3407[15]. Between November 2010 and August 2011, 24 volunteers were enrolled 182
(AERAS-422 high dose, n=8; AERAS-422 low dose, n=8; Tice BCG, n=8). High dose 183
AERAS-422 vaccination induced Ag85A- and Ag85B-specific lymphoproliferative responses 184
and marked anti-mycobacterial activity detectable in the direct whole blood MGIA[15]. A 185
systems biology approach using modular analysis identified positive correlations between 186
post-vaccination T cell gene expression and MGIA activity, and negative correlations 187
between post-vaccination monocyte gene expression modules and MGIA activity. 188
Unfortunately, the development of varicella-zoster virus reactivations occurred in two of 189
eight vaccine recipients given the high dose of AERAS-422, resulting in discontinuation of 190
AERAS-422 clinical development. However, these results demonstrated that the direct whole 191
blood MGIA can detect vaccine-induced increases in anti-mycobacterial immune 192
activity[15]. 193
4) Assessing M.tb infection status 194
As described, there was a significant correlation in outcome using BCG and M.tb in the direct 195
whole blood MGIA. However, no difference was detected in the magnitude of growth 196
inhibition between healthy M.tb infected and non-infected subjects using either mycobacterial 197
strain in the TB endemic region of South Africa (Baguma et al. submitted). This was in 198
contrast to a previous study in which growth of luminescent mycobacteria was significantly 199
lower in blood from tuberculin skin test (TST) positive subjects compared with TST negative 200
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
9
subjects from a non-endemic region[10]. The discordance between these findings may have 201
resulted from the cross-sectional designs in which infection with M.tb was defined using a 202
single IGRA or TST, and it is not known if volunteers from either study were recently or 203
historically exposed to M.tb. Furthermore, despite being QuantiFERON® negative, the South 204
African volunteers had readily detectable T cell responses to BCG stimulation, suggesting 205
high levels of immunological sensitization to mycobacteria (likely due to childhood BCG 206
vaccination and/or exposure to environmental mycobacteria). 207
208
Immune mechanisms of growth inhibition 209
One major goal of this project was to elucidate immune mechanisms underlying the 210
mycobacterial growth inhibition observed. The influence of IFN-γ producing T cells, 211
antibodies, monocyte:lymphocyte (ML) ratio and haemoglobin were assessed. 212
1) IFN-γ producing T cells 213
The magnitude of mycobacterial growth inhibition following BCG vaccination in the Oxford 214
study did not correlate with the antigen-specific IFN-γ ELISpot response, and a boosted IFN-215
γ ELISpot response observed following BCG revaccination was not reflected in enhanced 216
MGIA activity[13]. In the South African volunteers, despite QuantiFERON®-TB Gold 217
positive individuals having greater levels of ESAT-6 and CFP-10 specific IFN-γ secreting T 218
cells, this did not translate into improved MGIA responses. These findings are consistent with 219
previous observations that in vitro mycobacterial growth inhibition does not correlate with 220
IFN-γ production[7, 8]. 221
2) Antibodies 222
Chen et al. demonstrated that IgG antibody responses to arabinomannan (AM), a 223
mycobacterial envelope polysaccharide, increased significantly following BCG vaccination 224
using samples from the Oxford trial[16]. Their studies suggest that these antibodies, 225
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
10
particularly when targeting specific oligosaccharide epitopes, may play a functional role in 226
mycobacterial growth inhibition. Phagocytosis and intracellular growth inhibition were 227
significantly enhanced when BCG was opsonized with post-vaccination sera, and these 228
enhancements correlated with IgG titers to AM. Furthermore, increased phagolysosomal 229
fusion in M.tb infected macrophages was observed with post-vaccination serum but not pre-230
vaccination serum, indicating that intracellular growth reduction was FcR-mediated. 231
Interestingly, they also found that the antibody response at 4 weeks post BCG vaccination 232
correlated with mycobacterial growth inhibition in the direct PBMC MGIA[16]. A functional 233
role of antibodies to AM has since been further supported by murine immunization studies 234
with AM conjugate vaccines[17]. These data highlight the importance of assessing the 235
humoral compartment in MGIAs and comparing PBMC and whole blood based approaches 236
to better assess the potential contribution of antibody-mediated responses. 237
3) Monocyte-lymphocyte ratio 238
In blood samples from healthy volunteers, a greater proportion of monocytes to lymphocytes 239
(higher ML ratio) is associated with increased mycobacterial growth and an increased 240
probability of a Type I IFN transcriptional signature[18]. Altering the ML ratio in vitro 241
impacts the control of mycobacterial growth, with either very high or very low ML ratios 242
associated with poorer control in the direct PBMC MGIA[18]. This is consistent with the 243
observation that a profoundly altered ML ratio is associated with increased risk of developing 244
TB disease in non-HIV exposed infants, in HIV exposed uninfected infants, in HIV positive 245
adults starting antiretroviral therapy and in postpartum women[2, 19-21]. Accordingly, Hoft 246
et al. showed that anti-mycobacterial immune activity correlated positively with T-cell and 247
negatively with monocyte expression signatures in the Aeras 422 study[15]. Further studies 248
with animal models may better define the sub-populations of monocytes and lymphocytes 249
important for mycobacterial growth inhibition. 250
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
11
4) Haemoglobin 251
Using samples from the Oxford BCG vaccination study, a positive correlation was observed 252
between mean corpuscular haemoglobin (Hb) and mycobacterial growth in the direct whole 253
blood MGIA[22]. Experimental addition of Hb or ferric iron to PBMC from both humans and 254
non-human primates (NHPs) resulted in increased mycobacterial growth; an effect reversed 255
by addition of the iron chelator deferoxamine. Furthermore, expression of Hb-related genes 256
correlated significantly with mycobacterial growth in whole blood, and to a lesser extent, in 257
PBMC[22]. This association between Hb/iron levels and mycobacterial growth may in part 258
explain differences in outcome between whole blood and PBMC MGIAs and should be taken 259
into account when using these assays. 260
261
Animal MGIA Studies 262
Given that early clinical studies indicate the need for more investigations before a human 263
PBMC MGIA can be qualified, MGIA development using animal models of TB represents an 264
important ongoing strategy. In addition to the benefits of sample volume and reduced 265
heterogeneity described above, animal models have the important advantage that the MGIA 266
outcome can be correlated with protection from in vivo challenge with pathogenic M.tb. This 267
would then biologically validate the assay as capable of assessing effective anti-268
mycobacterial immunity. Where there is a need to test vaccine candidates for antigen dose, 269
adjuvant dose or antigen-adjuvant combinations, an MGIA could save time, animals and 270
funding. As the direct MGIA does not depend on any species-specific immune reagents, it 271
can be easily adapted for use across a range of animal species. 272
In an effort to improve inter-site reproducibility, the animal phase of the MGIA project 273
utilized a single shared batch of frozen mycobacteria. The BCG batch was prepared at Aeras 274
and tested for viability and reproducibility prior to distribution to project partners. 275
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
12
Investigators also agreed that, for each experiment, the colony forming units (CFUs) of 276
mycobacterial stocks would be obtained and standard curves would be used to convert Bactec 277
MGIT time to positivity (TTP) data, presented as the log10 CFU of the growth count for ease 278
of data presentation and interpretation. 279
1) Murine model 280
An MGIA assay using infected macrophages cultured separately from mouse splenocytes has 281
previously demonstrated reproducible growth inhibition data at seven days of incubation[23]. 282
Using this MGIA, immune splenocytes from mice vaccinated with five different TB vaccines 283
limited mycobacterial growth in vitro compared to naïve controls. Importantly, the vaccine-284
induced MGIA correlated at a group level with protective immune responses induced in 285
experimentally-matched animals using a mouse model of pulmonary TB[23]. 286
More recent studies have focused on simplifying the mouse MGIA model. Marsay et al. 287
showed that the direct MGIA using splenocytes has the ability to detect mycobacterial growth 288
inhibition following BCG vaccination[24]. Furthermore, investigators from CBER/FDA used 289
the direct splenocyte MGIA to demonstrate reproducible in vitro protective responses from 290
mice immunized with either BCG or a subunit vaccine, and showed that these responses 291
correlated on a group level with in vivo protection data from experimentally-matched 292
mice[25]. They also confirmed that mycobacterial quantification using the Bactec MGIT 293
system correlates strongly with plating of organ homogenates and counting CFU, and is faster 294
and easier to perform than the traditional CFU method[26]. 295
BCG gives a robust and reproducible protective effect against challenge with M.tb in 296
C57BL/6 mice and this model has therefore been used for further optimization of the mouse 297
direct MGIA. Zelmer et al. have sought to enhance the sensitivity of the direct splenocyte 298
MGIA, and found that detection of vaccine-induced inhibition could be improved by 299
lowering the MOI[27]. It was also shown that the capacity to detect mycobacterial growth 300
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
13
inhibition in BCG vaccinated mice is time-sensitive, with growth inhibition detected only at 301
the peak of the BCG immune response (approximately 6 weeks in C57BL/6 mice)[14]. As 302
there is no amplification of the antigen specific immune response in the direct MGIA, it is 303
only the ability of the cells resident in the spleen at the time of harvest that can be assessed 304
for anti-mycobacterial capacity[14]. Since BCG is a live replicating mycobacterium, there 305
can be variation in the development of the peak of the immune response and variation in the 306
persistence of BCG vaccine in lungs and spleen. This can impact the ability to reproducibly 307
select the optimum time point for measuring a vaccine response[28, 29]. However, selecting 308
the time point of the peak immune response may be less of a challenge when assessing a 309
subunit vaccine and the ability of MGIAs to assess the effectiveness of subunit vaccines has 310
shown early promise[23]. 311
2) Guinea pig model 312
Investigators at Colorado State University and Public Health England are in the process of 313
adapting the direct MGIA for use with the guinea pig model. Early indications are that 314
PBMCs provide more robust information than whole blood, and further studies with vaccine 315
candidates are in progress to examine the utility of PBMC MGIAs in the guinea pig model. 316
3) NHP model 317
The direct whole blood MGIA has also been adapted for use with non-human primate (NHP) 318
samples at Oxford, in collaboration with Sharpe et al. at Public Health England. In a study of 319
7 Rhesus macaques the assay demonstrated a strong enhancement of mycobacterial growth 320
inhibition following BCG vaccination, which was still significant following correction for 321
changes in haemoglobin[22]. Work is currently ongoing to optimize an NHP direct PBMC 322
MGIA and correlate the outcome with protection from in vivo BCG and M.tb challenge on an 323
individual animal basis. Reproduction of the MGIA in NHPs could help to inform the human 324
studies and provide biological validation of the assay. 325
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
14
Conclusions 326
The MGIA represents a functional assay for assessing TB vaccine activity that has the 327
potential to correlate with vaccine-induced immune protection. To date, the direct whole 328
blood MGIA has demonstrated the most consistent effect across a range of different clinical 329
vaccine studies in humans. However, because this assay must be performed within hours of 330
blood draw it requires laboratory infrastructure in close proximity to the clinical site, making 331
it difficult to assess in many vaccine trials, particularly in TB endemic countries. 332
Cryopreserved PBMCs are thus the preferred specimen for logistical reasons, and the direct 333
PBMC MGIA is currently undergoing further optimization and harmonization across 334
laboratories as part of the EURIPRED collaborative infrastructure program (H. McShane and 335
R. Tanner, personal communication). Use of MGIAs in animal models of TB provides an 336
opportunity for biological validation against protection from pathogenic in vivo challenge, 337
and could be more time- and cost-effective. Animal MGIAs could also reduce the number of 338
virulent M.tb challenge experiments required to identify promising vaccine candidates for 339
progression to clinical trials[30]. 340
The results presented indicate that the MGIA has promise as a functional assay for the 341
assessment of candidate TB vaccines. It also provides a tractable model for investigating the 342
mechanisms involved in anti-mycobacterial immunity. For example, findings from this 343
project have highlighted the importance of both lymphoid and myeloid cell mediated as well 344
as antibody-mediated immunity in these assays. In the future, it would be interesting to 345
consider how exposure of an individual to other mycobacterial species or infections (such as 346
helminths) affects ability to control mycobacterial growth in vitro, and work to this end is 347
underway. The influence of regulatory immune mechanisms and suppressors, which will be 348
represented in unbiased samples such as whole blood and PBMC, should also be explored. 349
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
15
In this report, we have summarized the numerous accomplishments of this project. Many 350
parameters of a standardized MGIA that can be routinely performed in a laboratory with 351
access to BACTEC MGIT equipment have been established. However, MGIAs are 352
technically demanding and it is clear that further work is required, particularly to reduce the 353
variability intrinsic to functional assays. Use of cryopreserved PBMCs remains a challenge, 354
as does the goal of qualifying this assay for use in human studies in particular. Utilization of 355
the direct MGIA in animal models (especially mice and NHPs) for assessing new TB 356
vaccines seems more valuable at this time. A number of explorations into MGIAs as a 357
functional TB test should be performed in the future, and some of these goals are outlined in 358
Table 2. If validated and shown to reproducibly and consistently correlate with efficacy 359
demonstrated in clinical trials of candidate vaccines for TB, MGIAs hold great potential. 360
Much as neutralizing antibodies are currently used to assess the efficacy of certain other 361
vaccines, MGIAs could be applied for regulatory purposes as a surrogate marker of efficacy. 362
363
Acknowledgments 364
We would like to thank Robert S. Wallis for serving as a consultant on the MGIA project. We 365
thank Megan Fitzgerald, Veerabadran Dheenadhalayan and Nathalie Cadieux of Aeras for 366
production and distribution of the BCG standard for the MGIA. We also thank Lewis 367
Schrager and Barry Walker for critical comments on the manuscript. We are indebted to 368
Thomas Evans, former CEO of Aeras, for his support of the MGIA project. I thank Pere-Joan 369
Cardona for the portrait (MJB). 370
371
Ethics Statement 372
Participants in the BCG vaccination and revaccination study were recruited under a protocol 373
approved by the Oxfordshire Research Ethics Committee (OxREC A). Written informed 374
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
16
consent was obtained from all individuals prior to enrollment in the trial. Human studies 375
performed at the South African Tuberculosis Vaccine Initiative were approved by the Human 376
Research Ethics Committee of the University of Cape Town. All adult participants provided 377
written informed consent prior to participation in the studies. In the case of children, written 378
informed consent was provided by a parent or legal guardian. All of the work at Saint Louis 379
University was approved by the Saint Louis University IRB, and PBMC from independent 380
healthy BCG vaccinated subjects were collected under a protocol approved by the 381
Institutional Review Board of the Albert Einstein College of Medicine. All animal studies 382
were approved by the appropriate institution which used IACUC-approved protocols and 383
methods. At Public Health England, all animal procedures and study design were approved 384
by the Public Health England, Porton Down Ethical Review Committee, and authorized 385
under an appropriate UK Home Office project license. 386
387
Financial Disclosure: We thank the USFDA (FDA Grant #IU18FD004012-01) and Aeras 388
for financial support. The research leading to some of these results has also received funding 389
from the European Union’s Seventh Framework Program [FP7/2007-2013] under Grant 390
Agreement No. 312661: European Research Infrastructures for Poverty Related Diseases 391
(EURIPRED); the European Union’s Horizon2020 research and innovation program under 392
Grant Agreement No. 643381; the Universities Federation for Animal Welfare (UFAW); 393
Wellcome Trust; and the Bill and Melinda Gates Foundation. 394
395
Author Contributions: Conceived and designed the experiments: MJB, HAF, RT, SM, TJS, 396
DH, HM, JMA, AI, SS, AR and DFH. Performed the experiments: RT, TJS, DH, DFH, JMA, 397
AI, SS, AR, APN, ME and AZ. Wrote the paper: MJB, HAF, RT, SM, TJS, DH, HM, JMA, 398
AI and DFH. 399
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
17
400
Competing Interests: The authors have declared that there are no competing interests. 401
402
References 403
[1] Fletcher, H. A., Snowden, M. A., Landry, B., Rida, W., Satti, I., Harris, S. A., Matsumiya, M., Tanner, 404 R., O'Shea, M. K., Dheenadhayalan, V., Bogardus, L., Stockdale, L., Marsay, L., Chomka, A., 405 Harrington-Kandt, R., Manjaly-Thomas, Z. R., Naranbhai, V., Stylianou, E., Darboe, F., Penn-406 Nicholson, A., Nemes, E., Hatheril, M., Hussey, G., Mahomed, H., Tameris, M., McClain, J. B., Evans, 407 T. G., Hanekom, W. A., Scriba, T. J., and McShane, H., 2016, "T-cell activation is an immune correlate 408 of risk in BCG vaccinated infants," Nat Commun, 7, p. 11290. 409 [2] Fletcher, H. A., Filali-Mouhim, A., Nemes, E., Hawkridge, A., Keyser, A., Njikan, S., Hatherill, M., 410 Scriba, T. J., Abel, B., Kagina, B. M., Veldsman, A., Agudelo, N. M., Kaplan, G., Hussey, G. D., Sekaly, R. 411 P., Hanekom, W. A., and team, B. s., 2016, "Human newborn bacille Calmette-Guérin vaccination and 412 risk of tuberculosis disease: a case-control study," BMC Med, 14, p. 76. 413 [3] Hanekom, W. A., Dockrell, H. M., Ottenhoff, T. H., Doherty, T. M., Fletcher, H., McShane, H., 414 Weichold, F. F., Hoft, D. F., Parida, S. K., and Fruth, U. J., 2008, "Immunological outcomes of new 415 tuberculosis vaccine trials: WHO panel recommendations," PLoS Med, 5(7), p. e145. 416 [4] Tanner, R., O'Shea, M. K., Fletcher, H. A., and McShane, H., 2016, "In vitro mycobacterial growth 417 inhibition assays: A tool for the assessment of protective immunity and evaluation of tuberculosis 418 vaccine efficacy," Vaccine. 419 [5] Wallis, R. S., Palaci, M., Vinhas, S., Hise, A. G., Ribeiro, F. C., Landen, K., Cheon, S. H., Song, H. Y., 420 Phillips, M., Dietze, R., and Ellner, J. J., 2001, "A whole blood bactericidal assay for tuberculosis," J 421 Infect Dis, 183(8), pp. 1300-1303. 422 [6] Worku, S., and Hoft, D. F., 2000, "In vitro measurement of protective mycobacterial immunity: 423 antigen-specific expansion of T cells capable of inhibiting intracellular growth of bacille Calmette-424 Guérin," Clin Infect Dis, 30 Suppl 3, pp. S257-261. 425 [7] Hoft, D. F., Worku, S., Kampmann, B., Whalen, C. C., Ellner, J. J., Hirsch, C. S., Brown, R. B., Larkin, 426 R., Li, Q., Yun, H., and Silver, R. F., 2002, "Investigation of the relationships between immune-427 mediated inhibition of mycobacterial growth and other potential surrogate markers of protective 428 Mycobacterium tuberculosis immunity," J Infect Dis, 186(10), pp. 1448-1457. 429 [8] Kampmann, B., Tena, G. N., Mzazi, S., Eley, B., Young, D. B., and Levin, M., 2004, "Novel human in 430 vitro system for evaluating antimycobacterial vaccines," Infect Immun, 72(11), pp. 6401-6407. 431 [9] Silver, R. F., Li, Q., Boom, W. H., and Ellner, J. J., 1998, "Lymphocyte-dependent inhibition of 432 growth of virulent Mycobacterium tuberculosis H37Rv within human monocytes: requirement for 433 CD4+ T cells in purified protein derivative-positive, but not in purified protein derivative-negative 434 subjects," J Immunol, 160(5), pp. 2408-2417. 435 [10] Kampmann, B., Gaora, P. O., Snewin, V. A., Gares, M. P., Young, D. B., and Levin, M., 2000, 436 "Evaluation of human antimycobacterial immunity using recombinant reporter mycobacteria," J 437 Infect Dis, 182(3), pp. 895-901. 438 [11] Cheon, S. H., Kampmann, B., Hise, A. G., Phillips, M., Song, H. Y., Landen, K., Li, Q., Larkin, R., 439 Ellner, J. J., Silver, R. F., Hoft, D. F., and Wallis, R. S., 2002, "Bactericidal activity in whole blood as a 440 potential surrogate marker of immunity after vaccination against tuberculosis," Clin Diagn Lab 441 Immunol, 9(4), pp. 901-907. 442 [12] Kolibab, K., Parra, M., Yang, A. L., Perera, L. P., Derrick, S. C., and Morris, S. L., 2009, "A practical 443 in vitro growth inhibition assay for the evaluation of TB vaccines," Vaccine, 28(2), pp. 317-322. 444 [13] Fletcher, H. A., Tanner, R., Wallis, R. S., Meyer, J., Manjaly, Z. R., Harris, S., Satti, I., Silver, R. F., 445 Hoft, D., Kampmann, B., Walker, K. B., Dockrell, H. M., Fruth, U., Barker, L., Brennan, M. J., and 446
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
18
McShane, H., 2013, "Inhibition of mycobacterial growth in vitro following primary but not secondary 447 vaccination with Mycobacterium bovis BCG," Clin Vaccine Immunol, 20(11), pp. 1683-1689. 448 [14] Zelmer, A., Tanner, R., Stylianou, E., Damelang, T., Morris, S., Izzo, A., Williams, A., Sharpe, S., 449 Pepponi, I., Walker, B., Hokey, D. A., McShane, H., Brennan, M., and Fletcher, H., 2016, "A new tool 450 for tuberculosis vaccine screening: Ex vivo Mycobacterial Growth Inhibition Assay indicates BCG-451 mediated protection in a murine model of tuberculosis," BMC Infect Dis, 16, p. 412. 452 [15] Hoft, D. F., Blazevic, A., Selimovic, A., Turan, A., Tennant, J., Abate, G., Fulkerson, J., Zak, D. E., 453 Walker, R., McClain, B., Sadoff, J., Scott, J., Shepherd, B., Ishmukhamedov, J., Hokey, D. A., 454 Dheenadhayalan, V., Shankar, S., Amon, L., Navarro, G., Podyminogin, R., Aderem, A., Barker, L., 455 Brennan, M., Wallis, R. S., Gershon, A. A., Gershon, M. D., and Steinberg, S., 2016, "Safety and 456 Immunogenicity of the Recombinant BCG Vaccine AERAS-422 in Healthy BCG-naïve Adults: A 457 Randomized, Active-controlled, First-in-human Phase 1 Trial," EBioMedicine, 7, pp. 278-286. 458 [16] Chen, T., Blanc, C., Eder, A. Z., Prados-Rosales, R., Souza, A. C., Kim, R. S., Glatman-Freedman, A., 459 Joe, M., Bai, Y., Lowary, T. L., Tanner, R., Brennan, M. J., Fletcher, H. A., McShane, H., Casadevall, A., 460 and Achkar, J. M., 2016, "Association of Human Antibodies to Arabinomannan With Enhanced 461 Mycobacterial Opsonophagocytosis and Intracellular Growth Reduction," J Infect Dis. 462 [17] Prados-Rosales, R., Carreño, L., Cheng, T., Blanc, C., Weinrick, B., Malek, A., Lowary, T. L., Baena, 463 A., Joe, M., Bai, Y., Kalscheuer, R., Batista-Gonzalez, A., Saavedra, N. A., Sampedro, L., Tomás, J., 464 Anguita, J., Hung, S. C., Tripathi, A., Xu, J., Glatman-Freedman, A., Jacobs, W. R., Chan, J., Porcelli, S. 465 A., Achkar, J. M., and Casadevall, A., 2017, "Enhanced control of Mycobacterium tuberculosis 466 extrapulmonary dissemination in mice by an arabinomannan-protein conjugate vaccine," PLoS 467 Pathog, 13(3), p. e1006250. 468 [18] Naranbhai, V., Fletcher, H. A., Tanner, R., O'Shea, M. K., McShane, H., Fairfax, B. P., Knight, J. C., 469 and Hill, A. V., 2015, "Distinct Transcriptional and Anti-Mycobacterial Profiles of Peripheral Blood 470 Monocytes Dependent on the Ratio of Monocytes: Lymphocytes," EBioMedicine, 2(11), pp. 1619-471 1626. 472 [19] Naranbhai, V., Moodley, D., Chipato, T., Stranix-Chibanda, L., Nakabaiito, C., Kamateeka, M., 473 Musoke, P., Manji, K., George, K., Emel, L. M., Richardson, P., Andrew, P., Fowler, M., Fletcher, H., 474 McShane, H., Coovadia, H. M., Hill, A. V., and Team, H. P., 2014, "The association between the ratio 475 of monocytes: lymphocytes and risk of tuberculosis among HIV-infected postpartum women," J 476 Acquir Immune Defic Syndr, 67(5), pp. 573-575. 477 [20] Naranbhai, V., Kim, S., Fletcher, H., Cotton, M. F., Violari, A., Mitchell, C., Nachman, S., 478 McSherry, G., McShane, H., Hill, A. V., and Madhi, S. A., 2014, "The association between the ratio of 479 monocytes:lymphocytes at age 3 months and risk of tuberculosis (TB) in the first two years of life," 480 BMC Med, 12, p. 120. 481 [21] Naranbhai, V., Hill, A. V., Abdool Karim, S. S., Naidoo, K., Abdool Karim, Q., Warimwe, G. M., 482 McShane, H., and Fletcher, H., 2014, "Ratio of monocytes to lymphocytes in peripheral blood 483 identifies adults at risk of incident tuberculosis among HIV-infected adults initiating antiretroviral 484 therapy," J Infect Dis, 209(4), pp. 500-509. 485 [22] Tanner, R., O'Shea, M. K., White, A. D., Müller, J., Harrington-Kandt, R., Matsumiya, M., Dennis, 486 M. J., Parizotto, E. A., Harris, S., Stylianou, E., Naranbhai, V., Bettencourt, P., Drakesmith, H., Sharpe, 487 S., Fletcher, H. A., and McShane, H., 2017, "The influence of haemoglobin and iron on in vitro 488 mycobacterial growth inhibition assays," Sci Rep, 7, p. 43478. 489 [23] Parra, M., Yang, A. L., Lim, J., Kolibab, K., Derrick, S., Cadieux, N., Perera, L. P., Jacobs, W. R., 490 Brennan, M., and Morris, S. L., 2009, "Development of a murine mycobacterial growth inhibition 491 assay for evaluating vaccines against Mycobacterium tuberculosis," Clin Vaccine Immunol, 16(7), pp. 492 1025-1032. 493 [24] Marsay, L., Matsumiya, M., Tanner, R., Poyntz, H., Griffiths, K. L., Stylianou, E., Marsh, P. D., 494 Williams, A., Sharpe, S., Fletcher, H., and McShane, H., 2013, "Mycobacterial growth inhibition in 495 murine splenocytes as a surrogate for protection against Mycobacterium tuberculosis (M. tb)," 496 Tuberculosis (Edinb), 93(5), pp. 551-557. 497
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
19
[25] Yang, A. L., Schmidt, T. E., Stibitz, S., Derrick, S. C., Morris, S. L., and Parra, M., 2016, "A 498 simplified mycobacterial growth inhibition assay (MGIA) using direct infection of mouse splenocytes 499 and the MGIT system," J Microbiol Methods, 131, pp. 7-9. 500 [26] Kolibab, K., Yang, A., Parra, M., Derrick, S. C., and Morris, S. L., 2014, "Time to detection of 501 Mycobacterium tuberculosis using the MGIT 320 system correlates with colony counting in 502 preclinical testing of new vaccines," Clin Vaccine Immunol, 21(3), pp. 453-455. 503 [27] Zelmer, A., Tanner, R., Stylianou, E., Morris, S., Izzo, A., Williams, A., Sharpe, S., Pepponi, I., 504 Walker, B., Hokey, D., McShane, H., Brennan, M., and Fletcher, H., 2015, "Ex vivo mycobacterial 505 growth inhibition assay (MGIA) for tuberculosis vaccine testing - a protocol for mouse 506 splenocytes,"BioRxiv. 507 [28] Kaveh, D. A., Carmen Garcia-Pelayo, M., and Hogarth, P. J., 2014, "Persistent BCG bacilli 508 perpetuate CD4 T effector memory and optimal protection against tuberculosis," Vaccine, 32(51), 509 pp. 6911-6918. 510 [29] Olsen, A. W., Brandt, L., Agger, E. M., van Pinxteren, L. A., and Andersen, P., 2004, "The 511 influence of remaining live BCG organisms in vaccinated mice on the maintenance of immunity to 512 tuberculosis," Scand J Immunol, 60(3), pp. 273-277. 513 [30] Tanner, R., and McShane, H., 2017, "Replacing, reducing and refining the use of animals in 514 tuberculosis vaccine research," ALTEX, 34(1), pp. 157-166. 515 516
517
Figure Legends 518
519
Figure 1: Schematic of the MGIT mycobacterial growth inhibition assay method. Whole 520
blood or cell culture medium containing the appropriate concentration of PBMC (or mouse 521
splenocytes in murine studies) is inoculated with an equal volume of mycobacteria at a low 522
multiplicity of infection (MOI) (~1 CFU:10,000 PBMC). Cultures are incubated in 2ml tubes 523
at 37°C for 96 hours with 360° rotation. Cells are then lysed to remove red blood cells and/or 524
release intracellularised mycobacteria. The lysate is inoculated into BACTEC mycobacteria 525
indicator tubes (MGITs). Tubes are placed on the BACTEC 960 machine and incubated at 526
37°C until detection of positivity by fluorescence. On day 0, direct-to-MGIT viability 527
controls are set up by directly inoculating MGIT tubes with the same volume of mycobacteria 528
as the samples. The time to positivity (TTP) is converted to a log10 CFU value using a 529
standard curve and the final value is expressed as an absolute CFU value or relative to the 530
control, indicating the amount of growth inhibition that has occurred during the culture 531
period. 532
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
20
Figure 2: Direct whole blood MGIA in South African infants. A) Whole blood samples 533
from 10 South African infants were assessed using the direct whole blood MGIA. There was 534
a significant correlation between growth of BCG SSI and M.tb H37Rv (Spearman’s r = 0.73, 535
p = 0.02). B) Whole blood samples from 10 South African infants were compared with those 536
from 53 South African adults using the direct whole blood MGIA. There was a significant 537
reduction in growth of BCG SSI in infants compared with adults (A Mann-Whitney U test 538
was performed between groups where *represents a p-value of <0.05). 539
540
Tables 541
542
Table 1: Major early objectives of the MGIA project 543
544
Major Early Objectives 545
546
Assess the variability, reproducibility and transferability of four different MGIAs in 547
naïve, BCG vaccinated and BCG re-vaccinated subjects. 548
549
Compare PBMC and WB MGIA responses induced by wild type BCG and a new 550
recombinant BCG vaccine. 551
552
In a TB endemic population (South Africa), determine the effect of M.tb infection 553
status on inhibition of mycobacterial growth. 554
555
Test sera obtained from subjects enrolled in the enhanced BCG vaccine MGIA study 556
for antibodies to capsular and cell surface constituents including major polysaccharide 557
antigens and to compare the pre- to post-immunization antibody levels with MGIA 558
outcomes. 559
560
Explore other immune mechanisms underlying mycobacterial growth inhibition (eg. 561
by correlating to ELISPOT and ICS-Flow analysis on blood samples from the BCG 562
study cohorts). 563
564
565
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
21
Table 2: Future goals of MGIA development 566
567
Future goals of MGIA development 568
569
Demonstrate that the direct PBMC MGIA meets requirements as a qualified assay 570
571
Determine if there are significant differences when using BCG or M.tb strains for in 572
vitro infection 573
574
Evaluate differences when using MGIAs in endemic or non-endemic countries 575
576
Determine if MGIAs can be used in M.tb infected populations 577
578
Identify factors in PBMC and whole blood that mediate and influence growth 579
inhibition 580
581
Evaluate performance of PBMC-MGIT in trials of novel TB vaccine candidates 582
583
Determine if the test is reproducible at different sites 584
585
Determine if the direct whole blood MGIA can be further adapted or refined for use in 586
vaccine or TB studies 587
588
589
590
Biographical Text 591
592
Dr. Michael J. Brennan: I was the first one in my family to go to college. I owe a great deal 593
to my parents and sister who sacrificed much so that I could take the time needed to become a 594
scientist. To have experienced a career I love, I have been exceptionally fortunate. 595
Throughout, I have held different positions at the National Institutes of Health, at the World 596
Health Organization and I eventually retired from the U.S Food and Drug Administration. 597
Now near the end of my career, I am employed by the non-profit organization, Aeras, and 598
have learned many new things about the manufacturing and clinical investigation of vaccines, 599
especially those for Tuberculosis. All over, I have travelled to six continents, participated in 600
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
22
many valuable conferences, and had meaningful and insightful conversations with hundreds 601
of scientists. Perhaps, those conversations have been the best part of all. 602
Dr. Rachel Tanner is a post-doctoral scientist at the Jenner Institute, University of Oxford. 603
She obtained a BA (Hons) in Biological Sciences at University of Oxford and developed an 604
early interest in vaccine immunology, taking up a Research Assistant post at the Centre for 605
HIV/AIDS Vaccine Immunology (CHAVI). Dr. Tanner received her DPhil in Clinical 606
Medicine at the Jenner Institute, University of Oxford in 2015. Her PhD project was 607
concerned with the development and optimisation of an MGIA across multiple species, and 608
she continues to work on such functional assays as part of her post-doctoral research. Dr. 609
Tanner’s broader interests are in immune correlates and the host immune response to TB 610
vaccination. 611
Dr. Sheldon Morris is a retired scientist who worked in the Laboratory of Mycobacterial 612
Diseases and Cellular Immunology at CBER/FDA from 1986-2015. He served as Laboratory 613
Chief from 2001-2015. Dr. Morris’s activities included the regulatory evaluation of 614
tuberculosis and malaria vaccines and scientific research involving the development of new 615
tuberculosis vaccines and novel assays to support vaccine development. During his tenure at 616
CBER, Dr. Morris authored or co-authored more than 100 publications on mycobacterial 617
disease research and was invited to give 52 scientific presentations. In addition, he was a 618
member of the Editorial Board for Antimicrobial Agents and Chemotherapy and served as an 619
ad hoc reviewer for 13 other journals. Importantly, Dr. Morris was selected to participate on 620
11 National Institute of Health grant review panels, 3 Welcome Trust grant reviews, and was 621
a scientific advisor to the World Health Organization during 4 meetings. 622
Dr. Thomas J. Scriba trained in Biological Sciences at Stellenbosch University, South 623
Africa and in T cell Immunology at Oxford University, United Kingdom. He returned to 624
South Africa for postdoctoral training in Paediatric and Clinical Immunology in Tuberculosis 625
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
23
and Vaccinology at the Institute of Infectious Disease and Molecular Medicine (IDM), 626
University of Cape Town. Dr. Scriba is Associate Professor at the University of Cape Town, 627
South Africa and is Deputy Director, Immunology at the South African Tuberculosis Vaccine 628
Initiative. His research interests include clinical immunology of tuberculosis and 629
development of novel prevention strategies against tuberculosis. 630
Dr. Jacqueline M. Achkar, M.D., M.Sc., is a physician scientist and an Associate Professor 631
in the Departments of Medicine (Division of Infectious Diseases) and Microbiology & 632
Immunology at the Albert Einstein College of Medicine, Bronx, New York. She received her 633
Medical Diploma and Doctoral Degree from the Free University of Berlin, Germany, her 634
Master of Science Degree in Clinical Research Methods from the Albert Einstein College of 635
Medicine, and completed her Medical Residency and Infectious Disease Fellowship at the 636
New York University School of Medicine. Dr. Achkar leads a successful translational 637
research program in the field of tuberculosis serology and biomarker discovery which 638
includes the identification of human antibody targets and the study of antibody functions in 639
the defense against TB. 640
Dr. Andrea Zelmer received her degree in Biology in 2002 and PhD in Infection Biology in 641
2005 from the Technical University Braunschweig, Germany. She subsequently moved to 642
London, UK to start a post-doctoral position at the UCL School of Pharmacy to work on 643
treatment and pathogenesis of E. coli K1 infection in neonates. In 2008 Dr. Zelmer moved to 644
the London School of Hygiene and Tropical Medicine to develop imaging technologies and 645
mouse models for testing of new drugs and vaccines against tuberculosis. She is now an 646
Assistant Professor at LSHTM. Over the last three years Dr. Zelmer has been focusing on 647
developing clinically relevant animal models for vaccine testing and studying the immune 648
response to vaccination. 649
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
24
Dr. David Hokey is the Senior Director of Human Immunology at Aeras in Rockville, 650
Maryland. He received his Ph.D. in Immunology from the University of Pittsburgh in 2005 651
where he worked primarily on dendritic cell-based tumor vaccines. He then performed his 652
postdoctoral research under the guidance of Dr. David Weiner at the University of 653
Pennsylvania which focused on DNA vaccines for HIV, cancer, and influenza with an 654
emphasis on primate immunology and polychromatic flow cytometry. Dr. Hokey’s current 655
work at Aeras involves evaluation of immune responses to TB vaccine candidates in pre-656
clinical and clinical settings. His research interests include evaluation of novel vaccine 657
platforms and adjuvants for manipulation of immune responses as well as assay optimization 658
and he has worked in this area for 18 years. 659
Dr. Angelo Izzo received his B.Sc. (Hons) and PhD at the University of Adelaide, South 660
Australia. He then undertook post-graduate training at the Universita di Milano, Instituto di 661
Virologia, Milan, Italy and returned to the University of Adelaide as a Research Assistant. 662
Dr. Izzo later took up a Post-doctoral fellowship at the Trudeau Institute, Saranac Lake, New 663
York followed by the University of New Mexico School of Medicine, Albuquerque, New 664
Mexico. He then became Assistant Professor at Midwestern University Department of 665
Microbiology, Illinois and is now Assistant Professor at Colorado State University, Fort 666
Collins, Colorado. Dr. Izzo has over 25 years’ experience investigating the immune response 667
to pulmonary Mycobacterium tuberculosis infection in animal models. He has obtained 668
extensive experience in understanding vaccine induced immunology and 669
immunopathogenesis of pulmonary tuberculosis using the animal models. 670
Dr. Sally Sharpe is a Scientific Leader at Public Health England, Porton Down. She received 671
a B.Sc. from London University in 1984 then took up a Research Assistant post at The Centre 672
of Applied Microbiology and Research at Porton Down where she was trained in T cell 673
immunology and flow cytometry. She worked within a team developing non-human primate 674
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
25
models of HIV and was awarded a PhD (Open University, 1997) for research into T cell 675
responses to simian immunodeficiency virus. Since 2003, she has led work to develop refined 676
macaque models of Tuberculosis for the evaluation of new interventions at the Health 677
Protection Agency and subsequently Public Health England on the Porton Down site. 678
Dr. Ann Rawkins nee Williams obtained her B.Sc. at the University of Wales in Cardiff and 679
began her career as a Research Assistant studying bacterial pathogenesis at the Centre of 680
Applied Microbiology and Research at Porton Down in 1985. She conducted her post-681
graduate studies on the pathogenic mechanisms of Legionella pneumophila and was awarded 682
a PhD in 1995. After a brief period investigating Salmonella infection of poultry, she began 683
M. tuberculosis research, developing aerosol-challenge in-vivo models for pathogenesis 684
studies and subsequently for the evaluation of interventions, predominantly vaccines. The 685
vaccine evaluation team which she established is internationally renowned and collaborates 686
with the majority of TB vaccine developers and she has been involved in multiple EU-funded 687
consortia. She is currently a Scientific Leader providing scientific direction to the TB 688
research programme at the Public Health England laboratories at Porton Down, Salisbury, 689
UK. She has published widely in the name of Williams. 690
Dr. Adam Penn-Nicholson is a Research Officer/Scientist at the South African Tuberculosis 691
Vaccine Initiative, at University of Cape Town, South Africa. He previously worked as a 692
postdoctoral Fellow at the University of Cape Town. Dr. Penn-Nicholson received his PhD 693
at Case Western Reserve University, in Cleveland, Ohio, USA. Dr. Penn-Nicholson’s main 694
focus is the discovery of blood-based biomarkers that prospectively predict TB disease risk, 695
and understanding the biology involved in progression from latent M. tuberculosis infection 696
to active TB disease. He also provides scientific oversight of immunology for several clinical 697
TB vaccine trials currently being conducted at SATVI. 698
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
26
Mr Mzwandile Erasmus has a BSc from the University of Cape Town and joined the South 699
African Tuberculosis Vaccine Initiative (SATVI) shortly after graduating in 2007. He is 700
currently a Senior Laboratory Technologist has been involved in multiple clinical and 701
research studies conducted at SATVI. 702
Dr. Elena Stylianou has a BSc. and MSc. in Biochemistry from Imperial College, London 703
and a PhD in Immunology from St George’s, University of London. She has since been at the 704
Jenner Institute, University of Oxford, training as a post-doctoral immunologist. Dr. 705
Stylianou has always found the complexity of the interaction between Mycobacterium 706
tuberculosis (M.tb) and the human host fascinating and both her PhD and post-doctoral 707
studies focused on pre-clinical vaccine research against tuberculosis. The key areas of her 708
research include mucosal and systemic delivery of viral vectors, immune responses after 709
vaccination and evaluation of vaccine protective efficacy using M.tb infection models. 710
Dr. Dan F. Hoft M.D., Ph.D. is the Director of the Division of Infectious Diseases, Allergy 711
& Immunology at Saint Louis University. Dr. Hoft was trained in Infectious Diseases at the 712
University of Iowa, where he received an NIH Physician Science Training Award (PSTA) 713
allowing him to complete a Ph.D. in microbiology/immunology. He came to Saint Louis 714
University in 1992 and has received continuous NIH funding since his arrival. In 2006, he 715
was named the Director of the new Division of Immunobiology. In 2010, the Divisions of 716
Immunobiology, Infectious Diseases and Allergy & Immunology were joined into the new 717
Division of Infectious Diseases, Allergy & Immunology with Dr. Hoft as Director. In 2014 718
he became the Principal Investigator of SLU’s Vaccine & Treatment Evaluation Unit 719
(VTEU). The capacity to induce protective immune memory by vaccination is a powerful 720
public health tool, and Dr. Hoft has been working in this area for 32 years. 721
Professor Helen McShane is currently a Professor of Vaccinology at Oxford University, 722
Wellcome Senior Clinical Research Fellow, Deputy Head of Department for Nuffield 723
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from
27
Department of Medicine and Honorary Consultant physician in HIV medicine. Professor 724
McShane obtained an intercalated BSc in 1988, followed by a degree in medicine in 1991 725
(both University of London). In 1997 She was awarded an MRC Clinical Training Fellowship 726
to undertake a PhD with Adrian Hill in Oxford, and was later awarded a PhD in 2001 727
(University of London). In 2001 she was awarded a Wellcome Clinician Scientist Fellowship, 728
allowing her to complete her clinical training and subsequently awarded a CCST in HIV and 729
GU Medicine in 2003. In 2005, she was awarded a Wellcome Senior Clinical Research 730
Fellowship; this Fellowship was renewed in 2010. Professor McShane has led the TB vaccine 731
research group at Oxford University for 16 years. 732
Dr. Helen A. Fletcher is an Associate Professor of Immunology and Deputy Director of the 733
TB Centre at the London School of Hygiene & Tropical Medicine (LSHTM). Dr. Fletcher 734
has worked in the field of TB immunology for 17 years at University College London, 735
University of Oxford and LSHTM. She has used transcriptomics and cellular immunology to 736
identify correlates of TB disease risk including increased monocyte to lymphocyte ratio and 737
increased CD4+ T cell activation in South African infants at risk of developing TB disease. 738
Dr. Fletcher has published more than 90 papers and her current work at LSHTM continues to 739
focus on immune correlates of risk and the host response to TB vaccination. 740
on January 30, 2020 by guesthttp://cvi.asm
.org/D
ownloaded from