Research M30 Neoepitope Expression in Epithelial Cancer ......Apoptosis detection To quantitatively evaluate spontaneous and drug-induced apoptosis in the cancer cell lines, four different

Imag

M30QuaTum

ElisabCarmAnton

Abst

Thequestistandmetastumortween

AuthoDepartm3IstitutoOncoloCROB,“SantaOncoloDepartm

Note: SResear

AuthorE. RosS. Indradata; R

Corresvia GatFax: 39

doi: 10

©2010

www.a

D

Published OnlineFirst October 26, 2010; DOI: 10.1158/1078-0432.CCR-10-1449

Clinical
Canceresearch
ing, Diagnosis, Prognosis

Neoepitope Expression in Epithelial Cancer:ntification of Apoptosis in Circulating

R

or Cells by CellSearch Analysis

etta Rossi1, Umberto Basso3, Romina Celadin1, Francesca Zilio1, Salvatore Pucciarelli2, Michele Aieta5,
en Barile6, Teodoro Sava7, Giorgio Bonciarelli4, Salvatore Tumolo8, Cristina Ghiotto3, Cristina Magro3, io Jirillo3, Stefano Indraccolo3, Alberto Amadori1,3, and Rita Zamarchi3
ractPur

quantExp

(CellScaspasand bRes

free foto 80%evaluaseriesnegatiby conCon

test h

cells (CT1 and 2

rs' Affiliatent of OnOncologic

gy, GeneraRionero inMaria dellagy, Universent, Osped

upplementach Online (h

contributionsi, R. Celadccolo, E. Ro. Zamarchi

ponding Autamelata, 6-049-80728

.1158/1078-

American A

acrjourna

ownloade

pose: This study aimed to detect the M30 neoepitope on circulating tumor cells (CTC) as a tool forifying apoptotic CTC throughout disease course and treatment.erimental Design: An automated sample preparation and analysis platform for computing CTCearch) was integrated with a monoclonal antibody (M30) targeting a neoepitope disclosed bye cleavage at cytokeratin 18 (CK18) in early apoptosis. The assay was validated using cell lineslood samples from healthy volunteers and patients with epithelial cancer.ults: M30-positive CTC could be detected in >70% of CTC-positive carcinoma patients, which werer both chemotherapy and radiologic treatments. The fraction of M30-positive CTC varied from 50%, depending on the histotype. To investigate the potential application of the M30 CTC assay for thetion of response in early phase trials, CTC and M30-positive CTC were enumerated in a small caseof breast cancer patients during treatment. Results indicate that changes in the balance of M30-ve/positive CTC may be used as a dynamic parameter indicating an active disease, as documentedsistent radiologic findings.clusions: M30 expression on CTC is detectable by immunofluorescence. The M30-integrated

as potential for monitoring dynamic changes in the quote of apoptotic CTC (in addition to CTC) to evaluate response in clinical trials of molecularly targeted anticancer therapeutics as well as
count
for translational research, in which there is a pressing need for informative circulating biomarkers.Clin Cancer Res; 16(21); 5233–43. ©2010 AACR.

modeapoptin brehand,

finding of tumor cells in peripheral blood raisesons as to their metastatic potential. In fact, notwith-ing that a single tumor cell was proved to sustaintasis in vivo (1), in humans the half-life of circulating

C) in peripheral blood is estimated at be-.4 hours, depending on the mathematical

countat anyindica(6). Nties ofare farshedshoulphenopressuCTC sAdd

velopCTC athe pdiseaslyzedan augood

ions: 1Oncology Section and 2Surgical Section,cology and Surgical Sciences, University of Padova,o Veneto (IOV-IRCCS), Padova, Italy;4Department ofl Hospital of Este-Monselice, Padova, Italy; 5IRCCS-Vulture (PZ), Italy; 6U.O.C Oncology - OspedaleMisericordia,” Rovigo, Italy; 7Department of Medicality of Verona, Verona, Italy; and 8Medical Oncologyale Santa Maria degli Angeli, Pordenone, Italy

ry data for this article are available at Clinical Cancerttp://clincancerres.aacrjournals.org/).

s: R. Zamarchi designed the research; R. Zamarchi,in, and F. Zilio carried out research; R. Zamarchi,ssi, U. Basso, S. Pucciarelli, and A. Amadori analyzedwrote the article.

thor:RitaZamarchi, IstitutoOncologicoVeneto - IRCCS,4 - 35128 Padova - Italy. Phone: 39-049-8215800;54; E-mail: [email protected].

0432.CCR-10-1449

ssociation for Cancer Research.

ls.org

Research. on January 9, 2clincancerres.aacrjournals.org d from

l of the extrapolated curve (2) and on the fact thatotic cells significantly contribute to the CTC fractionast (3) and prostate cancer (4) patients. On the othera strict correlation was established between CTCand prognosis (5), and elevated numbers of CTCtime during therapy was reported to be an accuratetion of subsequent rapid progression and mortalityevertheless, the phenotypic and biological proper-the CTC that are necessary for the metastatic processfrom clear. Theoretically, CTC should be adapted to

into peripheral blood and at least some of the CTCd be live cells. Moreover, although the metastatictype may be later acquired as a result of selectivere exerted at secondary sites, at least some of thehould be able to self-renew (7).ressing the role and mechanism of CTC in the de-ment of metastasis, we investigated whether or notre live cells, considering mainly when and how oftenercentage of apoptotic CTC changes throughoute course and treatment. For this purpose, we ana-CTC in our patient cohort by the CellSearch system,
tomated platform that permits serial testing withsensitivity and reproducibility (5). CTC assay was
5233

021. © 2010 American Association for Cancer

http://clincancerres.aacrjournals.org/

integrfor recthat bapoptM30with Aa stabThe

cancenumbtumortool f

Mate

PatienPeri

cancer29 me23 matientsbeforpatienof thestaticor colwererenalted pa(IOV-from30-60samp

patienfor stuview bseriallstatuson theby apCriterradiol

Cell lThe

cell lipurchand w

ApopTo

inducmethoanti-Mtransfsay, anpolymturer's

CTC aThe

the Cetion a

PhenTo

CTC w

Translational Relevance

The absolute number of circulating tumor cells(CTC) has proved to be a robust predictor of poorprognosis in metastatic breast, colorectal, and prostatecancer. Moreover, in the absence of tumor biopsiesCTC provide a “surrogate” index for monitoring re-sponse to treatment. However, the CTC biological sig-nificance is as yet undefined: why did the medianoverall survival not further decrease when >5 CTC(poor-prognosis threshold, very few CTC indeed) weredetected in 7.5 mL of blood?

By exploiting a M30-integrated CTC assay, we showhere that CTC are a heterogeneous cell population,which includes both apoptotic and viable cells: exceed-ingly high numbers of live CTC were associated withradiologic recurrence of disease, and also when a switchunder the threshold of poor prognosis was observedduring the therapy. Our data offer a rationale to theoption that a CTC subpopulation not expressing M30may be associated with decreased chances of survival.

Rossi et al.

Clin C5234

D


ated with a monoclonal antibody (mAb), anti-M30,ognizing (8) a neoepitope in cytokeratin 18 (CK18)ecomes available at a caspase cleavage event duringosis and is not detectable in vital epithelial cells; theneoepitope appears early in the apoptotic cascade,nnexin V reactivity, and it is generally regarded asle biomarker, specific for epithelial cell apoptosis.results obtained in breast, renal, and colorectal

r patients are presented, indicating that variableers of apoptotic CTC can be detected in all solid
s. The M30-integrated assay seems to be a feasible mAb,
BiochanalyzresultM30-p

StatisDat

(versioindicaand mnonpquanFisherwhere

Resu

QuandifferSev

ponen

or monitoring apoptotic CTC.

rial and Methods

tspheral blood was consecutively drawn from 34 breastpatients (33 female and 1 male, ages 39-83 years),tastatic renal cancer (mRCC) patients (6 female andle, ages 26-87 years), and 59 colorectal cancer pa-(21 female and 38male, ages 30-87 years) at baseline,e starting treatment. Breast and colorectal cancerts were enrolled in this study regardless of type or linerapy. Prior adjuvant treatment, treatment of meta-disease, or both were permitted in the case of breastorectal metastatic cancer, whereas mRCC patientsconsecutively enrolled in a pilot study, “Metastaticcancer: CTC determination in first-line Sunitinib trea-tients,” conducted at Istituto Oncologico VenetoIRCCS), Padova, Italy. Whole blood was also drawnhealthy control subjects (4 female and 4 male, ages
years) who had neither known illness at the time ofling nor history of malignant disease. All enrolled
apoptactivit

ancer Res; 16(21) November 1, 2010

Research. on January 9, 2clincancerres.aacrjournals.org ownloaded from

ts and healthy subjects gave their informed consentdy inclusion and were enrolled using institutional re-oard–approved protocols. After baseline evaluation,y monitored (1-10 months) reevaluations of diseasein the breast cancer patients were conducted dependingtype and schedule of treatment. Tumor measurements

propriate scans were done using Response Evaluationia in Solid Tumors guidelines without independentogy review, with no knowledge of the levels of CTC.

inesbreast cancer cell line MCF7, the prostate cancerne PC3, and the colon cancer cell line LoVo wereased from the American Type Culture Collectionere grown as described (9).

tosis detectionquantitatively evaluate spontaneous and drug-ed apoptosis in the cancer cell lines, four differentdologies were compared: Annexin V apoptosis assay,30 immunostaining, terminal deoxynucleotidyl

erase–mediated dUTP nick end labeling (TUNEL) as-dWestern blood analysis (WB) for poly(ADP-ribose)erase (PARP) cleavage, according to the manufac-instruction as detailed in Supplementary Materials.

ssayenumeration of CTC in whole blood was done byllSearch System according to manufacturer's instruc-s described (5).

otypic profiling of CTCquantify the fraction of apoptotic CTC, M30-positiveere detected integrating CTC assay with a specificM30 CytoDEATH Fluorescein (ALX-804-590, Alexisemicals), recognizing the M30 neoepitope of CK18,ed with the fourth filter of the CellSearch System;s were expressed as the total number of CTC andositive CTC per 7.5 mL of blood.

tical analysisa were analyzed utilizing the StatGraphics softwaren 2.6), as previously reported (10). Unless otherwiseted, all results are expressed as mean values ± 1 SD,ean values of three experiments are shown. The

arametric Mann-Whitney test was used to comparetitative variables. Frequencies were compared by' exact test (two tails) or χ2 test with Yates' correctionappropriate.

lts

titative comparison of apoptosis byent methodseral apoptosis assays devised to detect different com-ts of the apoptosis signaling cascade or specific
otic features, including DNA fragmentation, caspasey, membrane alterations, andmitochondrial changes,
Clinical Cancer Research



are custrongand smoston sawe firimmucologicytomThe

presenapoptby An15.3%specti17.2%specticleavaand inTo

singleassaywith c

discrifragmFig. 1tic ce(Fig. 1tion(Fig. 1ing thnecrotent wM30ptotictures

M30 CThe

specifThe iline tthe aearly

Fig. 1.apoptodifferenand druwas evfor AnnexpressPARP ccell lineof apop*, signifin cispl(responmedia oP < 0.0the intespontancultureswere noB, doubwith M3Cytogra(X-axis)in contrconditio36 houapoptoacquirean EPICreprese

Apoptosis-Specific M30 Neoepitope Expression on CTC

www.a

D


rrently available (11). Usingmore than onemethod isly recommended because of the limited sensitivitypecificity of current assays; moreover, choosing theappropriate apoptotic assay should also be basedmple type (tissues or cellular effusions). Therefore,stly evaluated the sensitivity and specificity of M30nostaining, comparing spontaneous and pharma-cally induced apoptosis in cancer cell lines by flowetry and WB.MCF7, LoVo, and PC3 cell lines were cultured in thece or the absence of cisplatin for 24 hours, raisingotic events in drug-treated cells, as shown (Fig. 1A)nexin V immunostaining (15.25% ± 4.8 MCF7-,± 0.3 LoVo-, and 13.6% ± 2.3 PC3-positive cells, re-

vely) and M30 immunostaining (15.2% ± 8 MCF7-,± 8.3 LoVo-, and 12.6% ± 1.3 PC3-positive cells, re-

vely) and confirmed by WB, showing strong PARPge in MCF7 (64.2% ± 0.5), in LoVo (55.2% ± 6.6),PC3 (34.4% ± 9.8).

quantitatively compare M30 immunostaining at-cell level, double fluorescence was done with TUNEL
in the cell lines 36 hours after apoptosis inductionisplatin. In the drug-treated cultu
cells,

ntative of three experiments.

acrjournals.org


minates the initial apoptotic phase before DNAentation (M30-positive TUNEL-negative cells;B, top left quadrant in right plot) from the apopto-lls, which are TUNEL-positive and M30-positiveB, top right quadrant in right plot). Again, a frac-

of cells were M30-negative and TUNEL-positiveB, bottom right quadrant in right plot), represent-e end phase of the process, when the cells becometic and the M30 epitope is lost. The data are consis-ith previous observations that the exposure of theneoepitope occurs at the initial phase of the apo-cascade, before the appearance of apoptotic fea-

in the nuclei (8).

TC assay developmentM30 mAb was integrated into the CTC assay to

ically quantify apoptosis of spreading tumor cells.ntegrated test was developed using the MCF7 cellhat was maintained in culture for 24 hours inbsence or the presence of paclitaxel, raising bothand late apoptotic events in drug-treated MCF7
as shown by flow cytometry (Fig. 2A). Untreated
res double staining controls and drug-conditioned cells were then spiked

Detection of epithelialsis. A, comparison oft methods. Spontaneous (−)g-induced apoptosis (+)aluated by flow cytometryexin V (Ann) or M30ion or by WB to detectleavage in three cancers 24 hours after inductiontosis with cisplatin.icant increases measuredatin-treated culturesses exceeding 2 SD thef untreated control,5 in every case). Due torexperimental variability ofeous apoptosis in bulk, PARP differencest significant in PC3 cells.le staining of MCF7 cells0 mAb and TUNEL method.ms of TUNEL reactivityand M30 binding (Y-axis)ol (left plot) and drug-ned culture (right plot)rs after induction ofsis with cisplatin. Data wered for 20,000 events usingS-XL device. Data are

Clin Cancer Res; 16(21) November 1, 2010 5235



into wobser7.5 mthe CThe

positiMCF7(6 ± 4by thefromCTC acytomtest dbeforecenceclearlyand e(Fig.morpvisiblunifowhenprogrecentraand lphotometricthe pdrug-cresemas detthe wBec

sibilitbe anthe paddreuntreainto tSearchadaptdid nM30-pThe

bloodpatienan onplemapproproceAPC adurinCom

patiendiscrithe Cis freqdisru

Fig. 2.CellSeaCytograbindingafter inquote ohigh Anquadrain drugB, quanobtainehistograCellSeamorpho*, signiftreatedcontrolMCF7 urow, CKsecondBasedarrows)Fragmeas a CTC (intact intermediate filament network progressively replacedby cytokeratin inclusions, cytoplasm/nuclear area overlying <50%).

Rossi et al.

Clin Cancer Res; 16(21) November 1, 20105236

Research. on January 9, 2clincancerres.aacrjournals.org Downloaded from


hole blood samples, at numbers similar to thoseved in vivo in cancer patients (200-1,000 cells/L peripheral blood) to be finally processed byellSearch system.CTC assay revealed a significant increase of M30-ve CTC (71 ± 16, 11.6%) in the drug-conditioned-spiked samples compared with untreated controls, 1%; P < 0.05; Fig. 2B). Because the values obtainedCTC assay and flow cytometry are generated startingdifferent pools of events (only nucleated cells in thessay versus both cells and nude nuclei in the flowetry), the fraction of apoptotic CTC by the integratediffers from the flow cytometric results obtainedthe spiking. Figure 2C shows an immunofluores-

image that stresses this point: M30 immunostainingdiscriminates intact CTC, which are M30 negative,

arly apoptotic CTC, which appeared M30 positive2C, left photo series); both these cells satisfy thehologic features required to be defined as CTC (cleare nucleus, cytoplasm/nuclear area overlying >50%,rmly immunostaining of cytoplasm). Conversely,the intact intermediate filament network is beingssively replaced by cytokeratin inclusions (Fig. 2C,l photo series), followed by chromatin condensationoss, all characteristic of apoptosis (Fig. 2C, rightseries), the events, which still entered in flow cyto-analysis, cannot be assigned as CTC. Remarkably,

ercentage of M30-positive CTC measured in theonditioned MCF7-spiked samples (11.6%) closelybles the quote of early apoptosis (14% in Fig. 1A)ermined by flow cytometry before the spiking intohole blood.ause CTC are considered “fragile” cells (12), the pos-y that an event assigned as M30-positive CTC couldartifact, due to apoptotic death occurring during

rocedure of enrichment and immunostaining, wasssed. To this purpose, blood samples spiked withted MCF7 cells were treated with paclitaxel directlyhe CellSave tube, 12 to 24 hours before the Cell-processing. Although we acknowledge that lab-

ed MCF7 cells could be less fragile than CTC, weot disclose in this case relevant changes of theositive fraction (1.5 ± 1, 0.2%).M30-integrated CTC assay was fully developed insamples obtained from healthy donors and cancerts. To use the test in follow-up studies, we generated-line staining procedure that is detailed in Sup-entary Materials. The obvious advantage of thisach is that antibodies of interest were added andssed simultaneously with the CK-PE and CD45-ntibodies, minimizing cell loss or disruption (13)g permeabilization and staining steps.pared with the MCF7-spiked samples, analysis oft's samples showed that the integrated assay allowsmination of the heterogeneous staining profile ofTC: an irregular CK staining (Fig. 3, event 519) that

Detection of tumor cells expressing the M30 neoepitope byrch analysis. A, flow cytometry analysis of apoptotic MCF7 cells.ms of Annexin V reactivity (X-axis) and propidium iodide (PI)(Y-axis) in treated culture (right dot plot) and control (left dot plot)duction of apoptosis with paclitaxel for 24 hours. An increasingf early apoptotic cells (lower right quadrant in all plots) withnexin and low PI staining, and late apoptotic cells (upper rightnt in all plots) with high Annexin and high PI staining was disclosed-treated MCF7 cells. Data are representative of three experiments.tification of apoptotic MCF7 by integrated CTC assay. MCF7d from treated culture (white histograms) and control (blackms) were spiked in blood of healthy volunteers and processed byrch platform. Unassigned events are those that do not satisfylogic features required for a CTC or contaminating leukocytes.icant increase of M30-positive cells measured in paclitaxel-cultures (responses exceeding 2 SD the media of untreated, P < 0.05). C, morphology of M30-positive and M30-negativesing an Analyzer II device. Computer-elaborated images: firstand 4’, 6-diamidino-2-phenylindole (DAPI) staining profile;row, anti-M30 staining; third row, anti-CD45 specificity control.

on CK, DAPI, and/or not-M30 staining profile, apoptotic (red openand intact cells (red closed arrow) can be clearly classified.nts (white open arrows) are events that cannot be assigned

uently observed in vivo and may closely resembleption of the filamentous network can be clearly




distin524),tion o

M30-pTo

frompatiencer, 2934 (5tientsnumbto 44tively.tal caM30-and Mand 1were17 ofdetectrange(mediThe

numb

breasAs suM30-any spwithbaselitrendover,distannode,grouped wiIn advatedtrend

SerialTo

therapsequethis pmultienrol

Fig. 3.HorizonCK PE only, and DAPI only. Red squares, positively stained cells: events 519 and 526 are live CTC, exhibiting irregular and strong CK staining, respectively;based is clas


www.a

D


guished from a M30-positive CTC (Fig. 3, eventprospectively minimizing discretionary interpreta-f morphologic features.

ositive CTC in solid tumorstest whether apoptosis could be detected in CTCpatients with carcinoma, blood samples from 122ts were tested (34 breast cancer, 59 colorectal can-mRCC) before therapy. CTC were detected in 19 of

6%) breast cancer patients; in 15 of these 19 pa-(79%), M30-positive CTC were also detected. Theer of CTC and M30-positive CTC ranged from 1(median, 4) and 1 to 13 cells (median, 3), respec-CTC were also detected in 26 of 59 (44%) colorec-ncer patients; in 24 of these 26 patients (92.3%),positive CTC were detected. The number of CTC30-positive CTC ranged from 1 to 10 (median, 2)to 7 cells (median, 2), respectively. Finally, CTC

detected in 19 of 29 (66.5%) mRCC patients; inthese 19 patients (89.5%), M30-positive CTC wereed. The number of CTC and M30-positive CTCd from 1 to 141 (median, 3) and 1 to 67 cellsan, 3), respectively.

on M30 staining profile (sufficient signal relative to background) event 524

percentage of CTC-positive patients and total CTCers strictly resemble data previously reported in

moniThe re

acrjournals.org


t (5, 14, 15) and colorectal cancer patients (16).mmarized in Table 1, the presence of CTC andpositive CTC at diagnosis was not associated withecific clinicopathologic feature in epithelial tumors,remarkable exceptions. The presence of CTC atne seemed weakly associated with metastasis (P for= 0.075) in colorectal cancer (Table 1B). More-the presence of CTC at baseline was associated witht sites of metastasis (lung, mediastinal lymphliver, or bone, P for trend = 0.026) in mRCC; in this, M30-positive CTC at baseline was weakly associat-th clear cell tumor (P for trend = 0.08; Table 1C).dition, M30-positive CTC were associated with ele-grading in breast cancer patients (Table 1A, P for= 0.018).

M30-integrated CTC assay during chemotherapyinvestigate whether the integrated test may predicteutic response, CTC and M30-positive CTC werentially assessed in eight breast cancer patients. Tourpose, depending on their consensus to undergople CTC tests, the patients were consecutivelyled regardless of type or line of therapy and were

sified as apoptotic CTC.

M30 immunostaining of early apoptotic CTC. Analysis of three rare cells in a blood sample of a breast cancer patient using an Analyzer II device.tally, the photos are of the same cell stained for the combination (Comp) of CK (green) and DAPI (violet), CD45 APC only, M30 FITC only,

tored for a time in the range of 1 to 10 months.sults are summarized in Table 2.




Table

A)

Breas

All sub

Age at

Sex

T (n =

N (n =

M (n =

Gradin

Estrog

Proges

Her2 (

B)

Colore

All sub

Age at

Sex

T (n =

N (n =

M (n =

Rossi et al.

Clin Can5238

Dow


1. Patients and primary tumor characteristics by CTC and M30-positive CTC count

jects

19)

jects

47)

47)

M0 41

(Continued on the

cer Res; 16(21) November 1, 2010

Reseaon Januclincancerres.aacrjournals.org nloaded from

25

fol

rch.ary

16

lowing pag

9, 2021. ©

0.075

e)

2010 Ame

15

rican

1

Clinical C

Associatio

84

ancer

n for C

1

Researc

ancer

t cancer patients
n CTCgat ive
CTCposit
ive
P*
M30 + P* M30+ % P†
ne

34 15
(44 %) 19 (56 %) 15 (79 %) 53.9
≤35 0
— — — — diagnosis, y 36-50 12 4 8 0.47 7 1 65.1
≥51 22
11 11 8 45.8
M 1
0 1 1 0 0.21 0
F 33
15 18 15 56.9 T1 14 9 5 ‡ 0.42 5 ‡ 0.18 69.5 0.54
T2 4
2 2 1 100
T3 1
0 1 1 100 T4 0 — — — —
20)
N0 12 7 5 1 4 1 74.4 1 N1-N3 8 4 4 4 68.8
17)
M0 4 2 2 1 2 1 100 0.14
M+ 13
8 5 4 52
G1 5
4 1 ‡ 0.40 1 ‡ 0.38 50 0.018 g (n = 18)
G2 5
2 3 2 32.5
G3 8
4 4 4 100
en receptors (n = 20)
- 7 4 3 1 2 1 66.7 1
+ 13
7 6 5 49.6
tin receptors (n = 20)
- 10 4 6 0.36 5 1 66.3 1
+ 10
7 3 2 33.3
n = 20) - 19 11 8 0.45 7 1 52.8 1

+ 1 0 1 1 50

ctal cancer patients
n CTCgat ive
CTCposit
ive
P*
M30 + P* M30+ % P†
ne

59 33
(56 %) 26 (44 %) 24 (92. 3%) 84
≤35 1
0 1 1 1 1 87.5 diagnosis, y
36-50 5
3 ]
2
]
2
]
100

≥51 53
30 23 21 82
M 38
22 16 1 16 0.138 87
F 21
11 10 8 79
Tis 10
6 4 ‡ 0.128 3 ‡ 0.347 75 0.671
T0 3
3 0 — — T1 4 2 2 2 100
T2 9
4 5 5 95
T3 17
11 6 6 92
T4 4
0 4 4 77
N0 37
22 15‡ 0.687 14 ‡ 0.936 88 0.303
N1 6
2 4 4 97
N2 2
1 1 1 20
Nx 2
1 1 1 100
47)

M1 6 1 5 5 98

h


Ovedecreaof eigIn t

43 [p[stablnumbthe thMBC;of obrespoonly iin theCTC fthat wenumfew CIn t

shedd(6 CT

CTCessentpointrelatedIn p

<5 CTresenttheseto accnegatfollowfor anof apoFor

by anegatin relof lon

Table 1. Patients and primary tumor characteristics by CTC and M30-positive CTC count (Cont'd)

C)

Renal cancer patients n CTCnegative

CTCpositive

P* M30+ P* M30+ % P†

All subjects 29 10 (34.5%) 19 (65.5%) 17 (89.5%) 78Age at diagnosis, y ≤35 3 2 1 1 1 0.11 67

36-50 3 0]

3]

2 83≥51 23 8 15 4 90

Sex M 23 9 14 0.84 12 1 84F 6 1 5 5 96

T (n = 19) T1 4 1 3‡ 0.57 3‡ 0.63 78 0.47T2 3 2 1 1 80T3 11 4 7 5 67T4 1 0 1 1 100

N (n = 19) N0 9 4 5 0.64 5 0.46 89 1N1-N3 10 3 7 5 62

M (n = 11) M0 5 2 3‡ 0.81 2‡ 0.45 56 0.45M1 2 1 1 1 100Mx 4 1 3 3 89

Fuhrman grading (n = 15) G1 — — — — —G2 4 1 3 1 3 1 89 1G3 11 5 6 5 74

Histology (n = 27) CC carcinoma 22 8 14 1 13 0.33 84 0.08Others 5 2 3 2 38

Sites of metastasisat blood draw (n = 25)

Contralateral kidney 4 4 0‡ 0.026 — —Lung, mediastinal

LN or liver12 3 9 7 0.68 64 0.28

Bone 9 3 6 6 94

Abbreviations: T, tumor; N, node; M, metastasis; LN, lymph node; CC carcinoma, clear cell carcinoma.*Fisher' exact test or χ2 (‡) test were employed where appropriate.†Median test.‡ χ2 test.


www.a

D


rall the number of total and M30-positive CTCsed during treatment in six and increased in twoht patients.he first group, in four cases, consisting of patientsrogressive disease (PD)], 55 (PD), 94 (PD), and 97e disease/partial response (SD/PR)], the total CTCer switched from values greater than or equal toreshold of poor prognosis (5 CTC/7.5 mL forref. 5) to values less than the threshold at the endservation time, indicating a pharmacodynamicnse that was related to overall disease progressionn patient 97; the M30-positive CTC were very fewse patients and the relative percentage of M30-positiveluctuated, creating a pattern of peaks and troughsere difficult to evaluate. Only one or two cells wereerated in patients 29 (PD) and 53 (PD), being tooTC to discriminate treatment effect.he second group, patient 84 (PD) showed a major
ing of CTC that went from all alive to all deadC at the time point 3); in patient 50 (SD/PR) the total
procechang

acrjournals.org


number increased over the follow-up period andially all cells were apoptotic (10 of 11 CTC at the time2). The switch from values <5 to values >5 CTC wasto overall disease progression only in patient 84.rinciple, both decreased total CTC numbers to valueC and increased fraction of apoptotic CTC may rep-response-related markers. However, the fact thatare both rare events may preclude the possibilityurately assess significant differences in the M30-ive/positive CTC numbers in any patient; only-up studies of adequate patient cohorts monitoredappropriate time can address the predictive relevanceptotic CTC.this purpose the observed variations were expressedsimpler parameter: the detected numbers of M30-ive and M30-positive CTC were separately plottedation to time, and the area under the curve (AUC)gitudinal graphs was calculated (Fig. 4), following a
dure commonly adopted to evaluate cumulativees of serologic tumor markers (17). The difference



betwewas cformu

ΔA

Rela

• PCa

• Na5

• Δa

Asciated0.036old o55, anassocCTC n

Discu

CTCvidingthrouobtaiand papproto othpubliof pelacksbreasciatedby thshoultheseFur

macoing pof con

Tab

Patno.

29

43

50

53

55

84

94

97

Abb*ΔA†As

Rossi et al.

Clin C5240

D


en live and apoptotic CTC concentration-time areaalculated in all patients according to the followingla:

UC ¼ M30−negative CTC AUC−M30−positive CTC AUC

tive numbers were obtained in the following way:

ositive ΔAUC value is the expression of extra liveTC over the follow-up period (e.g., patients 84nd 94; Fig. 4B);egative ΔAUC value is the expression of extrapoptotic CTC over the follow-up period (e.g., patients0 and 97; Fig. 4B);AUC = 0 derives from balanced numbers of livend apoptotic CTC.shown in Table 2, positive ΔAUC value was asso-with radiologic recurrence of disease (P for trend =), including cases where a switch under the thresh-f 5 CTC was observed during therapy (patients 43,d 94 in Table 2); conversely, negative ΔAUC was
iated with SD/PR also in patient 50, whose totalum
dereg

le i C d u u h

ien G in

N

determined by instrumental findings (computerized tomography or sci



ssion

can today be quantified in cancer patients, pro-a robust predictor of treatment efficacy and survival

ghout the continuum of the care (6, 18, 19). Dataned in metastatic breast (5), colorectal (16, 20),rostate cancer (4, 20, 21) by immunocytometricach strongly support extending these observationser solid tumor histotypes. Otherwise, it was recentlyshed (22) that the threshold of 5 CTC/7.5 mLripheral blood, firstly set up by Cristofanilli (5),prognostic significance in inflammatory metastatict cancer, where a value <5 CTC was not asso-with better prognosis than ≥5 CTC. As suggested

e authors, the biological characterization of CTCd be addressed to discover unknown properties ofcells.thermore, early clinical trials require validated phar-dynamic biomarkers (hopefully blood-based) show-roof of mechanism (drug hits target) and/or proofcept (tumor responds to drug). Apoptosis is often
ulated in cancer, and the induction of tumor cell
ber increased to value >5 CTC. death is a primary goal of many targeted therapies, directly

th
2. Ser al CT an M3 0-po sitive CT C co nt d ring c emo
1

ee8

25

e

ntigrafy

021. ©

erapy

L

) simultaneous

2010 Ameri

0

0

001

05

0

20

0

ly done w

Cli

can Asso

ith CTC co

nical Canc

ciation for

PD

SD/P

unt.

er Rese

Cancer

t Age/Sex

T
N M rad g ER PGR HER2 Testno.
Total7.5

CTC/mL

M30+7.5 m

/ M30+
% ΔAUC * Diseastatu
ses†

79/M
Neg Neg Neg 1 2 0 0 Pos 2 1 1 10
PD

3
1 0 0 83/F T N+ M Pos Pos Pos 1 8 4 5 Pos x 1
2
1 0 0 PD
3
1 1 10 49/F T1c N0 M0 G Neg Neg Neg 1 3 3 10 Neg SD/P R 3
2
1 10 9 64/F T1b N G Pos Neg Neg 1 2 0 0 Pos PD x 1
2
1 1 10 50/F N0 1 8 2 2 Pos PD
2
N g Neg 40/F T N M G Neg Neg Neg 1 N g Neg Pos 1c 0 0 3
2
5 1 18 3 PD
3
6 6 10 64/F T M Pos Pos Pos 1 2 0 0 Pos 2 1b 0
2
2 6 1 3 6 3 5 4 4 0 0
R
62/F T N M G Neg Neg Neg 1 5 4 8 Neg 4b 2 0 3
2
N g Neg 3 N g Neg e4 1 1 1005 2 1 50
reviations: ER, estrogen receptor; PGR, progestin receptor; HER2, human epidermal growth factor receptor 2.UC = M30-negative CTCAUC - M30-positive CTCAUC

arch


or indregulaa uniqsence

Incascadsons.disclocaspanetwoturesserumprognbothresultincludThe

patienthat thighaggregphase(4, 25also dtate cyondis thequantdetermhowecells isurpriFirs

prolifand aical febothfeaturly rela(25).were aSec

long-ssidereperiodshowity ofsuppo<0.1%fully wThi

an ovwere drare cthreshto expan evcollecexcludM30 n

Fig. 4.A, longfor patiwith mepoor prconcen(•, red aof the MCTC ov84 andCTC ovpatients


www.a

D


irectly hinting molecular components of apoptosistory pathways. Either way, apoptosis is regarded as

er the follow-up period (negative ΔAUC) were detected in50 and 97, which were SD/PR by imaging.

ue biomarker of treatment efficacy, and in the ab-of tumor biopsies CTC may offer a surrogate sample.

mentfor su

acrjournals.org


evaluating different components of the apoptotice we focused on the M30 neoepitope for three rea-First, the anti-M30 mAb defines an epitope of CK18sing early phases of apoptosis. In this phase, despitese cleavage, CK18 is still retained in a filamentousrk and tumor cells still satisfy the morphologic fea-of CTC. Second, a clinical correlate exists betweenlevels of these CK fragment and tumor load and

osis in breast (23) and colorectal cancer (24). Finally,anti-M30 and Annexin V provided us consistents in flow cytometry, altogether recommendinging M30 in the new test.integrated assay was validated in 122 cancerts before therapy at the first blood draw, disclosinghe M30 neoepitope is expressed on CTC at a veryfrequency. Based on the presence of cytokeratinates in their cytoplasm, apoptotic CTC in the lateof the process were previously described in prostate) and in lung cancer (26). M30-positive CTC wereocumented in metastatic castration-resistant pros-ancer patients (27), but their enumeration was be-the purposes of that study. To our knowledge, thisfirst report on CellSearch technology applied toify apoptotic CTC. Larger studies are warranted toine the prevalence of M30-positive CTC in vivo;

ver, although the detection of large numbers of theses counterintuitive, the data presented here are notsing.t, it is well known that higher grade and increasederation are often associated with tumor necrosispoptosis that may also be regarded as adverse biolog-atures (28). Moreover, it was recently reported thatintact CTC and granular CTC (whose morphologices strictly resemble early apoptotic cells) are inverse-ted to survival in castration-resistant prostate cancerIndeed, in our breast cancer series, M30-positive CTCssociated with higher grading (P for trend = 0.018).ond, apoptotic CTC were previously described inurviving breast cancer patients and have been con-d a sign of occult niches of proliferating tumor cells,ically shedding into the blood flow (2). Here wethat M30-positive CTC were detectable in the major-cancer patients at different disease stages, possiblyrting that apoptotic dying is the mechanism becauseof CTC released daily into the circulation success-ill settle in secondary organs (29).

rd, criticism was raised when, surprisingly, the medi-erall survival did not further decrease when >5 CTCetected in 7.5 mL of blood (30). In challenging withells, technical limits were evoked to account for aold of poor prognosis that is otherwise difficultlain, for example increasing mistakes in assigningent as a CTC when <5 CTC are detected or volumeted for the assay, but biological reasons cannot beed. Here we show that CTC frequently express aeoepitope, which may offer a rationale to the argu-

Changes in the number of CTC in breast cancer patients.itudinal graphs of CTC count over the indicated follow-up periodents 50, 84, 94, and 97. Patients 94 and 97 were diagnosedtastatic cancer at enrolling in this study; red lines, threshold ofognosis (5 CTC/7.5 mL for MBC; ref. 5). B, area under the bloodtration-time curve over the follow-up period of M30-negativerea) and M30-positive (○, green area) CTC; gray area, overlay30-positive and M30-negative longitudinal graphs. Extra liveer the follow-up (positive ΔAUC) were detected in patients94, which were PD by imaging; on the contrary, extra-apoptotic

that only the viable CTC cause the decreased chancesrvival (30).




OnCTCwhethmoniCTC rwhichmetasassaytreatmtrackincate tance aparamconsisabsolusponsoutcoassesstion. ItestingvideseffectsIn c

majorinduc

solidthis potherof res

Discl

No p

Ackn

We tartworassocia

Grant

GranTask 4cancer ”controfinalizz

Thepaymenadvertiindicate

Refe1. Fid

puof

2. Mepa81

3. Mecel

4. Laing46

5. CrdisJ M

6. HaeacpaRe

7. Amlatibre375

8. Leetioear

9. Fainahyp

10. AmCD

11. Vetic

12. Flocir

Rossi et al.

Clin C5242

D


the other hand, the high frequency of M30-positivein radio-chemo-free patients raises doubts as toer the integrated CTC test may be a useful tool totor drug-induced cell death. Our data show thatepresent a heterogeneous cell population, amongboth apoptotic cells and viable cells with possibletatic potential exist; we show that the M30-integratedmay be used to accurately quantify them duringent. In this setting, preliminary data obtained byg a small case series of breast cancer patients indi-

hat changes in the M30-negative/positive CTC bal-s expressed by ΔAUC may be used as a “dynamic”eter disclosing an active disease, as documented bytent radiologic findings. As in the case of the CTCte number, whether such an early assessment of re-e to treatment may result in an improved overallme or quality of life needs to be prospectivelyed in clinical trials designed to investigate this ques-n ongoing clinical trials at IOV-IRCCS, we are nowwhether assaying the quote of apoptotic CTC pro-

a more sensitive marker for rating pharmacodynamicin patients compared with total CTC counts.onclusion, although apoptosis is thought to play a
role in anticancer therapy, the clinical relevance oftion of apoptosis remains uncertain, particularly in
Recepublish

cell death. J Immunol Methods 2000;243:167–90.res LM, Kindelberger DW, Ligon AH, et al. Improving the yield ofculating tumour cells facilitates molecular characterisation and

recCa

13. WaγHtum

14. Kriresca

15. Satuman

16. SacoAn

17. deandru

18. BuimCa

19. Rieceval92

20. Smpro

21. DaanCli

22. Mesta

23. Olobioto



tumors. The proposed test might contribute to clarifyoint, and possibly provides a secondary end pointthan tumor size and tumor burden for evaluationponse in early phase trials.

osure of Potential Conflicts of Interest

otential conflicts of interest were disclosed.

owledgments

hank Ms. Colette Case for editing the manuscript and P. Gallo fork preparation. The CellSearch platform was sponsored by thetion “Il faro per lo IOV” of the ASCOM Padova.

Support

ts from the Italian Ministry of Health, (Oncology Program, Gender/“Characterization of circulating tumor cells in breast and ovary, R. Zamarchi); Banco Popolare di Verona (S. Indraccolo); Alleanzail cancro (ACC4; S. Indraccolo); Regione Veneto (Ricerca sanitariaata n.11/2008, A. Amadori); AIRC (A. Amadori).costs of publication of this article were defrayed in part by thet of page charges. This article must therefore be hereby markedsement in accordance with 18 U.S.C. Section 1734 solely tothis fact.

ived 05/31/2010; revised 08/16/2010; accepted 09/02/2010;ed OnlineFirst 10/26/2010.

rencesler IJ, Talmadge JE. Evidence that intravenously derived murinelmonary melanoma metastases can originate from the expansiona single tumor cell. Cancer Res 1986;46:5167–71.ng S, Tripathy D, Frenkel EP, et al. Circulating tumor cells intients with breast cancer dormancy. Clin Cancer Res 2004;10:52–62.hes G, Witt A, Kubista E, Ambros PF. Circulating breast cancerls are frequently apoptotic. Am J Pathol 2001;159:17–20.rson CJ, Moreno JG, Pienta KJ, et al. Apoptosis of circulat-tumor cells in prostate cancer patients. Cytometry A 2004;62:

–53.istofanilli M, Budd GT, Ellis MJ, et al. Circulating tumor cells,ease progression, and survival in metastatic breast cancer. N Engled 2004;351:781–91.

yes DF, Cristofanilli M, Budd GT, et al. Circulating tumor cells ath follow-up time point during therapy of metastatic breast cancertients predict progression-free and overall survival. Clin Cancers 2006;12:4218–24.adori A, Rossi E, Zamarchi R, Carli P, Pastorelli D, Jirillo A. Circu-ng and disseminated tumor cells in the clinical management ofast cancer patients: unanswered questions. Oncology 2009;76:–86.rs MP, Kolgen W, Bjorklund V, et al. Immunocytochemical detec-n and mapping of a cytokeratin 18 neo-epitope exposed duringly apoptosis. J Pathol 1999;187:567–72.varo E, Nardo G, Persano L, et al. Hypoxia inducible factor-1αctivation unveils a link between tumor cell metabolism andoxia-induced cell death. Am J Pathol 2008;173:1186–201.adori A, Zamarchi R, De Silvestro G, et al. Genetic control of the4/CD8 T-cell ratio in humans. Nat Med 1995;1:1279–83.rmes I, Haanen C, Reutelingsperger C. Flow cytometry of apopto-

ognition of discordant HER2 amplification in breast cancer. Br Jncer 2010;102:1495–502.ng LH, Pfister TD, Parchment RE, et al. Monitoring drug-induced2AX as a pharmacodynamic biomarker in individual circulatingor cells. Clin Cancer Res 2010;16:1073–84.

shnamurthy S, Cristofanilli M, Singh B, et al. Detection of minimalidual disease in blood and bone marrow in early stage breastncer. Cancer 2010;116:3330–7.ndri MT, Zorzino L, Cassatella MC, et al. Changes in circulatingor cell detection in patients with localized breast cancer before

d after surgery. Ann Surg Oncol 2010;17:1539–45.stre J, Maestro ML, Puente J, et al. Circulating tumor cells inlorectal cancer: correlation with clinical and pathological variables.n Oncol 2008;19:935–8.Haas EC, di Pietro A, Simpson KL, et al. Clinical evaluation of M30d M65 ELISA cell death assays as circulating biomarkers in ag-sensitive tumor, testicular cancer. Neoplasia 2008;10:1041–8.dd GT, Cristofanilli M, Ellis MJ, et al. Circulating tumor cells versusaging-predicting overall survival in metastatic breast cancer. Clinncer Res 2006;12:6403–9.thdorf S, Fritsche H, Muller V, et al. Detection of circulating tumorlls in peripheral blood of patients with metastatic breast cancer: aidation study of the CellSearch system. Clin Cancer Res 2007;13:0–8.irnov DA, Zweitzig DR, Foulk BW, et al. Global gene expressionfiling of circulating tumor cells. Cancer Res 2005;65:4993–7.nila DC, Heller G, Gignac GA, et al. Circulating tumor cell numberd prognosis in progressive castration-resistant prostate cancer.n Cancer Res 2007;13:7053–8.go M, De Giorgi U, Hsu L, et al. Circulating tumor cells in meta-tic inflammatory breast cancer. Ann Oncol 2009;20:1824–8.
fsson MH, Ueno T, Pan Y, et al. Cytokeratin-18 is a useful serummarker for early determination of response of breast carcinomaschemotherapy. Clin Cancer Res 2007;13:3198–206.



24. Aucyttum11

25. CoAllin

26. Hotumcan808

27. Sw

Chhyb

28. Ruapterwit

29. Daint20


www.a

D


sch C, Buxhofer-Ausch V, Olszewski U, et al. Caspase-cleavedokeratin 18 fragment (M30) as marker of postoperative residualor load in colon cancer patients. Eur J Surg Oncol 2009;35:

64–8.umans FA, Doggen CJ, Attard G, de Bono JS, Terstappen LW.circulating EpCAM+CK+CD45− objects predict overall survivalcastration-resistant prostate cancer. Ann Oncol 2010;21:1851–7.u JM, Greystoke A, Lancashire L, et al. Evaluation of circulatingor cells and serological cell death biomarkers in small cell lung
cer patients undergoing chemotherapy. Am J Pathol 2009;175:–16.ennenhuis JF, Tibbe AG, Levink R, Sipkema RC, Terstappen LW.
30. Tibfor15

acrjournals.org


aracterization of circulating tumor cells by fluorescence in situridization. Cytometry A 2009;75:520–7.pa JD, de Bruine AP, Gerbers AJ, et al. Simultaneous detection ofoptosis and proliferation in colorectal carcinoma by multiparame-flow cytometry allows separation of high and low-turnover tumorsh distinct clinical outcome. Cancer 2003;97:2404–11.wood S, Cristofanilli M. Integrating circulating tumor cell assayso the management of breast cancer. Curr Treat Options Oncol07;8:89–95.be AG, Miller MC, Terstappen LW. Statistical considerations
enumeration of circulating tumor cells. Cytometry A 2007;71:
4–62.




2010;16:5233-5243. Published OnlineFirst October 26, 2010.Clin Cancer Res Elisabetta Rossi, Umberto Basso, Romina Celadin, et al. CellSearch AnalysisQuantification of Apoptosis in Circulating Tumor Cells by M30 Neoepitope Expression in Epithelial Cancer:

Updated version

10.1158/1078-0432.CCR-10-1449doi:

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/16/21/5233.full#ref-list-1

This article cites 30 articles, 9 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/16/21/5233.full#related-urls

This article has been cited by 10 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/16/21/5233To request permission to re-use all or part of this article, use this link

Research. on January 9, 2021. © 2010 American Association for Cancerclincancerres.aacrjournals.org Downloaded from


http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-10-1449

http://clincancerres.aacrjournals.org/content/16/21/5233.full#ref-list-1

http://clincancerres.aacrjournals.org/content/16/21/5233.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/16/21/5233


Documents

Research M30 Neoepitope Expression in Epithelial Cancer ......Apoptosis detection To quantitatively evaluate spontaneous and drug-induced apoptosis in the cancer cell lines, four different