Clinical Chemistry 60:1 214–221 (2014)
Cancer Diagnostics
Prognostic Relevance of Viable Circulating Tumor Cells Detected by EPISPOT in Metastatic Breast Cancer Patients Jean-Marie Ramirez,1 Tanja Fehm,2 Mattea Orsini,3 Laure Cayrefourcq,1 Thierry Maudelonde,4 Klaus Pantel,5† and Catherine Alix-Panabie`res1,3†*
BACKGROUND: Detection of circulating tumor cells (CTC) in breast cancer patients is currently performed in many clinical trials, using different technologies, in particular the EpCAM-dependent CellSearch® system. The purpose of this study was to investigate the incidence and prognostic relevance of viable CTC in a large cohort of metastatic breast cancer (MBC) patients. METHODS: A total of 254 MBC patients were enrolled in a prospective multicenter study at first diagnosis of metastatic disease or disease progression (before the start of a new treatment regimen). After EpCAMindependent enrichment, viable CTC releasing cytokeratin-19 as an epithelial cell marker were detected in the peripheral blood by an EPISPOT assay, and the Food and Drug Administration cleared CellSearch was used as the reference method. RESULTS: Using the EPISPOT assay, CTC were detected in 59% of MBC patients. The overall survival (OS) was linked with the CTC status measured by EPISPOT (P ⫽ 0.0191), which allowed stratification of MBC patients in low- and high-risk groups. This stratification could be improved by addition of the CTC status assessed by the CellSearch system. In multivariate Cox proportionalhazards regression analysis, the 3 methods used to determine the level of CTC (EPISPOT, CellSearch, and combination of EPISPOT/CellSearch) were compared by the Bayesian information criterion method. Interestingly, the combination of the EPISPOT and CellSearch assays was the strongest predictor of OS (hazard ratio, 22.6; 95% CI, 2.8 –184.08). CONCLUSIONS: This is the first study in which CTC detection using the EPISPOT assay was evaluated on a
1
University Medical Centre, Saint-Eloi Hospital, Institute of Research in Biotherapy, Department of Cellular and Tissular Biopathology of tumors, Laboratory of Rare Human Circulating Cells, Montpellier, France; 2 Department of Obstetrics and Gynecology, University Medical Center, Duesseldorf, Germany; 3 University Institute of Clinical Research UM1—EA2415—Epidemiology, Biostatistics & Public Health, Montpellier, France; 4 University Medical Centre, Laboratory of Cellular and Hormonal Biology, Department of Cellular and Tissular Biopathology of Tumors, Arnaud de Villeneuve Hospital, Montpellier, France; 5 Department of Tumor Biology, University Medical, Center HamburgEppendorf, Hamburg, Germany. † Klaus Pantel and Catherine Alix-Panabie`res contributed equally to the work, and both should be considered as first authors.
214
large cohort of MBC patients, showing prognostic relevance of the presence of viable CTC. © 2013 American Association for Clinical Chemistry
The WHO has estimated that more than 1 million women develop breast cancer every year (1 ). Despite new forms of treatment, such as Herceptin® (trastuzumab), an anti– human epidermal growth factor receptor 2 (HER2)6 monoclonal antibody, or Tykerb® (lapatinib), a tyrosine kinase inhibitor targeting HER2, breast cancer remains an incurable disease, but the current overall survival (OS) rate at 10 years for breast cancer in general is approximately 75% in Western Europe. Circulating tumor cell (CTC) detection is a new approach with utility for (a) cancer prognosis, (b) personalized treatment depending on the CTC profile, (c) monitoring disease progression, and finally (d) better understanding of the genesis and evolution of breast cancer with the aim of developing new treatments. Detection of CTC currently is performed by use of the CellSearch® system, which was cleared by the Food and Drug Administration (FDA) in 2004 for CTC enumeration in the peripheral blood of patients with metastatic breast, prostate, and colon cancer (2– 4 ). Other innovative technologies for CTC detection have been developed, such as quantitative reverse-transcription PCR assays, CTC microchips, and filtration devices (5, 6 ). However, none of these devices are specific for living CTC, which is a serious limitation because many CTC are apoptotic in cancer patients (7, 8 ). We have developed a new technique in our laboratory, named the EPISPOT assay, which specifically
* Address correspondence to this author at: Laboratory of Rare Human Circulating Cells, Institute of Research in Biotherapy, Saint-Eloi Hospital, University Medical Centre, 80, Ave. Augustin Fliche, Montpellier, France. Fax ⫹33-(0)467-33-01-13; e-mail [email protected]. Received August 22, 2013; accepted October 21, 2013. Previously published online at DOI: 10.1373/clinchem.2013.215079 6 Nonstandard abbreviations: HER2, human epidermal growth factor receptor 2; OS, overall survival; CTC, circulating tumor cells; FDA, Food and Drug Administration; MBC, metastatic breast cancer; DTC, disseminated tumor cells; LCCRH, Laboratory of Rare Circulating Human Cells; CK19, cytokeratin-19; PFS, progression-free survival; BIC, Bayesian information criterion; EMT, epithelial– mesenchymal transition.
Viable Circulating Tumor Cells in Metastatic Breast Cancer Patients
detects living CTC and disseminated tumor cells (DTC) in cancer patients (9 ). This technique is based on the detection of proteins secreted by functional CTC or DTC cells combined with a negative enrichment (leukocyte depletion). This new method allows enumeration of only living CTC. Recently, it has been demonstrated that the majority of circulating breast cancer cells undergo apoptosis (7, 8 ). However, an efficient metastasis process requires living CTC (10 ). Thus, in contrast to other CTC technologies, the EPISPOT assay focuses on CTC with the potential to colonize secondary organs such as liver, lung, or bone. Moreover, the EPISPOT assay is a promising tool to identify proteins secreted by CTC or DTC, thereby providing an opportunity to better characterize their intrinsic properties responsible for survival, migration, and colonization of distant organs. Here, we report the results of the first prospective multicenter study to evaluate the prognostic impact of CTC detection using the EPISPOT assay in parallel with the CellSearch system (gold standard) in metastatic breast cancer (MBC) patients. Materials and Methods PATIENTS
A total of 254 MBC patients from 9 German university breast cancer centers [Du¨sseldorf (n ⫽ 4), Erlangen (30), Essen (46), Freiburg (9), Hamburg (79), Heidelberg (18), Munich (16), Regensburg (2), and Tuebingen (50)] were enrolled in this prospective open study from December 2007 until April 2009. The median (range) age of the patients was 60 (49 – 68) years, and 50% of the patients were over 60 years old. Briefly, 67% of the patients had metastasis in more than 1 site and 14% had metastasis localized in the bone, 39% had visceral metastatic sites, and 47% had both sites (visceral and bone). The criteria used for patient enrollment were (a) epithelial invasive carcinoma of the breast with metastasis to distant organs (M1), (b) age at least 18 years, (c) first diagnosis of metastatic disease or disease progression (before the start of a new line of therapy), (d) written informed consent, and (e) no secondary primary malignancy. A criterion for exclusion was the presence of a secondary primary malignancy (except in situ carcinoma of the cervix or adequately treated basal cell carcinoma of the skin). Patients were treated according to international guidelines, as indicated in the previous publications of the DETECT (Prognostic impact of circulating tumor cells assessed with the CellSearch System™ and AdnaTest Breast™ in metastatic breast cancer patients) study group (11, 12 ). All HER2⫹ patients received HER2targeted therapy (usually trastuzumab) that may in-
deed have influenced the prognosis of HER2⫹ patients. Hormone receptor–positive patients received endocrine therapy (eventually plus chemotherapy), and hormone receptor–negative patients received chemotherapy. Blood was collected before the start of a new treatment regimen. Participants gave their consent for the use of their blood samples. Data management was carried out from our databank (11 ). The local ethics committee approved this analysis (2007/B01). Blood was drawn before the start of a new line of therapy. All patients gave informed consent for the use of their blood samples. A web-based databank was designed for data management and online-documentation (www.detectstudy.de). By the use of this interface, clinical investigators were blinded for test results and the CTC test sites were blinded for the clinical data of the patients. The local institutional review boards approved the conduct of this study (2007/B01). METHODS
CTC detection. The detection of CTC was performed using both the EPISPOT assay and the CellSearch system (Veridex). CTC analysis by the CellSearch system was performed as described previously (11, 13 ). Blood samples for the EPISPOT assay were analyzed at the Laboratory of Rare Circulating Human Cells (LCCRH) at the Institute of Research in Biotherapy, UMC, Montpellier, France. The blood samples were sent at room temperature and processed within 24 h. EPISPOT assay. Preanalytical conditions were determined by spiking experiments with breast cancer cell lines. We observed a decrease in tumor cell viability and number after 24 h at room temperature. Storage at ⫹4 °C was less favorable. The EPISPOT assay was performed in only 194 of the 254 patients because Montpellier joined the DETECT study group after the start of the study and we therefore missed the first blood samples for the EPISPOT analysis and because of the exclusion of blood samples that failed to meet preanalytical quality criteria. These samples included those with less than the required 7.5 mL and those that were received more than 24 h following the blood draw. The EPISPOT assay was performed from a blood samples of 7.5 mL collected in EDTA tubes. Enrichment of CTC was performed using the RosetteSep human circulating epithelial tumor cell enrichment cocktail (StemCell Technology). The nitrocellulose membranes of the EPISPOT plate were precoated with a defined antibody to target a specific protein, anti– cytokeratin-19 (CK19) antibody in our study (9, 14, 15 ). The coating antibody CK19 (clone Ks 19.1, Progen BioClinical Chemistry 60:1 (2014) 215
technik) was diluted in PBS (6 g/mL) and was incubated at 4 °C for 12 h. After membrane blocking (100 L of 5% BSA/PBS at 37 °C for 2 h), CTC enriched by leukocyte depletion with the RosetteSep human circulating epithelial tumor cell enrichment cocktail (StemCell Technology) were incubated at 37 °C (5% CO2) for 18 h in cell line medium [DMEM ⫹ GlutaMax (Gibco) with 10% fetal calf serum and penicillin/streptomycin]. Membranes were then washed using PBS/0.1% Tween (6 times) and PBS (3 times). The CK19 proteins released during the cell culture were detected by a second Alexa555 conjugated anti-CK19 antibody (clone Ks 19.2, 3 g/mL; Progen Biotechnik GMBH). This antibody was diluted in PBS and incubated at 4 °C for 12 h. All single protein fingerprints generated CK19 immunospots that were observed and enumerated using video camera imaging and computer-assisted analysis (KS ELISPOT, Carl Zeiss Vision). CellSearch assay. Blood samples were collected into CellSave tubes (Veridex). The enrichment and enumeration of CTC were carried out using the CellSearch epithelial cell test (Veridex). Briefly, CTC were isolated from peripheral blood by an anti-EpCAM antibody– bearing ferrofluid and all DAPI(⫺)CD45(⫺)PanCK19(⫹) cells were identified as CTC (4 ). A blood sample containing at least 5 CTC was considered positive (16 ). STATISTICAL ANALYSIS
Associations between categorical variables were examined using 2 tests. Overall median survival was calculated by the Kaplan–Meier method. A time-dependent ROC curve was drawn and a MAX-STAT test was used to determine a CTC threshold for OS and progressionfree survival (PFS). PFS was calculated as described by Mu¨ller et al. (11 ). Log-rank tests were used to compare the survival curves by CTC detection groups. Univariate and multivariate Cox proportional hazards regression models were used to obtain unadjusted and fully adjusted hazard ratios and 95% CIs. The Bayesian information criterion (BIC) was used to compare Cox regression models. Results CTC DETECTION RATES
The specificity of the EPISPOT assay already had been determined in a group of 25 healthy controls, as reported in our previous publication (15 ). CTC expressing EpCAM and/or CK19 were detected in MBC patients using the CellSearch and the EPISPOT assays (Fig. 1). CTC counts determined by the EPISPOT assay were available for 194 of the 254 patients. Detection of CTC by the EPISPOT assay revealed that 59% (115/194) of patients were positive, 216 Clinical Chemistry 60:1 (2014)
Fig. 1. Viable CTCs identified in MBC patients using the EPISPOT assay. Analysis of viable CK19-releasing cells using the EPISPOT assay in a large cohort containing a total of 194 MBC patients. The SKBR3 cell line was used as positive control (left). Representative photo of single CK19 immunospots corresponding to viable CK19-releasing cells from a MBC patient (right).
whereas only 48.0% (122/254) were positive for CTC detected by the CellSearch assay. These data are summarized in Table 1. Comparison of the EPISPOT and CellSearch assays revealed a low agreement between the 2 methods as demonstrated by a low Cohen coefficient of 0.14 (P ⫽ 0.0416) (Table 2). Furthermore, we analyzed the combination of both methods in a total of 194 MBC patients. In 76.3% of cases, CTC were detected either by the EPISPOT assay alone (27.4%), or the CellSearch system alone (15.3%) or both methods (33.7%), whereas 23.7% of patients remained CTC negative. SURVIVAL ANALYSIS RELATED TO CTC RESULTS OBTAINED WITH EPISPOT ASSAY
To evaluate the EPISPOT assay as a predictor of survival, the OS according to the concentrations of CTC detected by the EPISPOT assay was estimated by the Kaplan–Meier method. A time-dependent ROC curve was drawn using the survival and CTC status (EPISPOT assay) of each MBC patient. ROC curve analysis showed that a cutoff value of 1 CTC has the highest sensitivity/specificity to predict OS. Of 194 patients, 115 had ⱖ1 CK19-releasing CTC (59%) per 7.5 mL of blood. These CTC-positive patients showed a significantly reduced OS compared to patients without CK19-releasing CTC (P ⫽ 0.0191, log-rank test) (Fig. 2A). The median OS of patients without CK19releasing CTC was significantly longer (5 years; 95% CI, 4.16 – 8.91) in comparison with CTC-positive
Viable Circulating Tumor Cells in Metastatic Breast Cancer Patients
Table 1. Prevalence of CTC: comparison between CellSearch and EPISPOT assays. CTC CellSearch, n (%) >1 (n ⴝ 142)
Variable
>5 (n ⴝ 97)
CTC EPISPOT, n (%)
>10 (n ⴝ 83)
>100 (n ⴝ 33)
>1 (n ⴝ 115)
>5 (n ⴝ 88)
>10 (n ⴝ 68)
>100 (n ⴝ 19)
Age class ⬍60 years
80 (56.3%) 53 (54.6%)
48 (57.8%) 22 (66.7%) 55 (47.8%) 42 (47.7%) 37 (54.4%) 13 (68.4%)
ⱖ60 years
62 (43.7%) 44 (45.4%)
35 (42.2%) 11 (33.3%) 60 (52.2%) 46 (52.3%) 31 (45.6%)
6 (31.6%)
Histology Ductal
112 (78.9%) 74 (76.3%)
62 (74.7%) 24 (72.7%) 87 (75.7%) 69 (78.4%) 54 (79.4%) 13 (68.4%)
Lobular
15 (10.6%) 11 (11.3%)
9 (10.8%)
Others
15 (10.6%) 12 (12.4%)
12 (14.5%)
Negative
36 (25.4%) 25 (25.8%)
Positive
106 (74.6%) 72 (74.2%)
6 (18.2%) 13 (11.3%)
9 (10.2%)
6 (8.8%)
3 (15.8%)
10 (11.4%)
8 (11.8%)
3 (15.8%)
24 (28.9%) 11 (33.3%) 29 (25.2%) 24 (27.3%) 16 (23.5%)
6 (31.6%)
3 (9.1%)
15 (13%)
Estrogen receptor status 59 (71.1%) 22 (66.7%) 86 (74.8%) 64 (72.7%) 52 (76.5%) 13 (68.4%)
Progesterone receptor status Negative
52 (36.6%) 36 (37.1%)
33 (39.8%) 14 (42.4%) 41 (35.7%) 36 (40.9%) 28 (41.2%)
Positive
90 (63.4%) 61 (62.9%)
50 (60.2%) 19 (57.6%) 74 (64.3%) 52 (59.1%) 40 (58.8%) 11 (57.9%)
8 (42.1%)
Missing
17 (12%)
10 (10.3%)
10 (12%)
Negative
88 (62%)
61 (62.9%)
49 (59%)
Positive
37 (26.1%) 26 (26.8%)
HER2 status 2 (6.1%)
15 (13%)
13 (14.8%)
8 (11.8%)
4 (21.1%)
22 (66.7%) 65 (56.5%) 47 (53.4%) 38 (55.9%) 10 (52.6%)
24 (28.9%)
9 (27.3%) 35 (30.4%) 28 (31.8%) 22 (32.4%)
5 (26.3%)
Metastatic sites Bone
19 (13.4%) 11 (11.3%)
Visceral
50 (35.2%) 32 (33%)
29 (34.9%)
9 (10.8%)
2 (6.1%)
12 (13.6%) 10 (14.7%)
3 (15.8%)
9 (27.3%) 45 (39.1%) 33 (37.5%) 25 (36.8%)
15 (13%)
4 (21.1%)
Multiple
73 (51.4%) 54 (55.7%)
45 (54.2%) 22 (66.7%) 55 (47.8%) 43 (48.9%) 33 (48.5%) 12 (63.2%)
41 (28.9%) 27 (27.8%)
24 (28.9%)
Number of metastatic sites One site Multiple sites
8 (24.2%) 36 (31.3%) 27 (30.7%) 20 (29.4%)
4 (21.1%)
101 (71.1%) 70 (72.2%) 59/83 (71.1%) 25 (75.8%) 79 (68.7%) 61 (69.3%) 48 (70.6%) 15 (78.9%)
Disease-free interval ⱕ12 months
36 (25.4%) 28 (28.9%) 22/83 (26.5%) 13 (39.4%) 25 (21.7%) 20 (22.7%) 17 (25%)
5 (26.3%)
⬎12 months
106 (74.6%) 69 (71.1%) 61/83 (73.5%) 20 (60.6%) 90 (78.3%) 68 (77.3%) 51 (75%)
14 (73.7%)
Therapeutic setting First line
58 (40.8%) 43 (44.3%) 36/83 (43.4%) 17 (51.5%) 45 (39.1%) 34 (38.6%) 27 (39.7%)
Second line
39 (27.5%) 22 (22.7%) 18/83 (21.7%)
Third line or more
45 (31.7%) 32 (33%)
18 (20.5%) 14 (20.6%)
4 (21.1%)
29/83 (34.9%) 11 (33.3%) 39 (33.9%) 36 (40.9%) 27 (39.7%)
6 (31.6%)
Table 2. Contingency table comparing the EPISPOT and CellSearch assays (N ⴝ 188).a EPISPOT
a
patients (2.8 years; 95% CI, 1.43–7.95). For PFS, no threshold of prognostically relevant CTC counts was identified by the Max-STAT and ROC curve analysis. SURVIVAL ANALYSIS RELATED TO CTC RESULTS OBTAINED WITH
CellSearch
CTC negative
CTC positive
CTC negative
45
52
CTC positive
29
62
Cohen coefficient ⫽ 0.14; P ⫽ 0.041.
5 (15.2%) 31 (27%)
9 (47.4%)
CELLSEARCH ASSAY
In the cohort of 254 MBC patients (DETECT study), the prognostic value of the CellSearch assay has been reported by Mu¨ller et al. (11 ). The median OS was 18.1 months in CTC-positive patients (95% CI 15.1–22.1). The PFS was not found to be correlated to the CTC Clinical Chemistry 60:1 (2014) 217
SURVIVAL ANALYSIS RELATED TO COMBINED EPISPOT AND CELLSEARCH RESULTS
We subsequently compared the prognostic power of the EPISPOT-based CTC detection with that of the CellSearch assay as the only FDA-cleared CTC assay (gold standard). OS was evaluated according to CTC status assessed by the EPISPOT assay, the CellSearch assay and by a combination of both methods. Interestingly, Kaplan-Meier analysis showed that the OS was significantly better in the patients negative for CTC using both methods of detection (Fig. 2B). Statistical analysis was carried out using the logrank test between CTC double-negative patients and the other 3 groups (EPISPOT positive, CellSearch positive, and double positive). We observed that the survival period was significantly shorter for the 3 groups with at least 1 test positive for CTC in comparison with the double-negative group (Both EPISPOT and CellSearch positive vs double negative, P ⫽ 0.0004, logrank test; only EPISPOT positive vs double negative, P ⫽ 0.0062, log-rank test; only CellSearch positive vs double negative, P ⫽ 0.0067, log-rank test). Thus, the combination of both CTC technologies was found to enhance the stratification of MBC patients into lowand high-risk groups. MULTIVARIATE COX ANALYSIS OF OS
Fig. 2. Overall survival (OS) of metastatic breast cancer (MBC) patient by EPISPOT assay and by the combination of the EPISPOT and CellSearch assays. Analysis (A), Kaplan-Meier plot estimating OS for 2 risk groups (0 CTC, ⱖ1 CTC per blood sample) obtained from the EPISPOT assay (P ⫽ 0.0191). (B), Kaplan-Meier plot estimating OS for 4 risk groups (CellSearch positive, EPISPOT positive, double negative, double positive) (P ⫽ 0.0021). Statistical analysis of OS between the 4 groups: EPISPOT ⫹ CellSearch vs double negative, P ⫽ 0.0004 log-rank test; EPISPOT vs double negative, P ⫽ 0.0062 log-rank test; CellSearch vs double negative, P ⫽ 0.0067 log-rank test.
The variables (CTC status, HER2, progesterone- and estrogen receptor status, therapeutic setting, and disease-free interval) found to be significant at a 20% threshold in a univariate Cox analysis for OS were included into the multivariate analysis (Table 3). Although clinical factors such as HER2 and estrogen receptor status and therapeutic setting (P ⫽ 0.0001, P ⫽ 0.0364, and P ⫽ 0.005, respectively) remained prognostically relevant, the CTC status (EPISPOT and CellSearch method) emerged as the second best independent predictor of OS (P ⫽ 0.0211) (Table 4). To determine the best predictor of OS among the 3 CTC detection approaches (i.e., EPISPOT assay, CellSearch system, or combined CellSearch and EPISPOT analyses), we compared them using the BIC method. We observed that the diagnostic approach using both the CellSearch and the EPISPOT results was the best predictor of OS (BIC, 317.650), followed by the EPISPOT (BIC, 346.203) and CellSearch methods (BIC, 435.1). Discussion
counts in this cohort receiving different types of treatments (P ⫽ 0.197). In multivariate analysis, we confirmed that the concentrations of CTC were an independent predictor for OS (data not shown). 218 Clinical Chemistry 60:1 (2014)
In this prospective multicenter study, the prognostic value of the EPISPOT assay was evaluated according to the concentrations of viable CK19-releasing CTC. We have previously shown that CK19 —although it is a cytoskeleton protein—is released from viable breast can-
Viable Circulating Tumor Cells in Metastatic Breast Cancer Patients
Table 3. Univariate analysis of overall survival. Variable
Age class Metastatic site
Number of metastatic sites Disease-free interval Therapeutic setting
CTC status (EPISPOT assay) CTC status (CellSearch assay) CTC status (CellSearch and EPISPOT assays)
Categories
Hazard ratio
ⱖ60 years
1
⬍60 years
1.138
Visceral
1 0.883
(0.526–1.481) (0.175–1.194)
1 Site
1
Multiple sites
1.166
⬎12 months
1
ⱕ12 months
0.584
0.0767 (0.322–1.059)
Third line or more
1 1.302
(0.74–2.291)
First line
0.456
(0.231–0.9)
⬎0 CTC/mL
1
0 CTC/mL
0.467
Positive
1
Negative
0.386
Double negative
1
0.0128
0.0191 (0.247–0.883) 0.0007 (0.223–0.668)
CellSearch
7.582
(1.646–34.923)
EPISPOT
6.175
(1.38–27.628) (2.053–37.132)
1
Postmenopausal
1.208
Histology
Others
1
Lobular Ductal Positive
1
Negative
2.204
Positive
1
Subtype
0.5883 (0.668–2.035)
Second line
8.732
HER2 status
0.279
0.457
Premenopausal
PR status
0.6033
Bone
CellSearch ⫹ EPISPOT
P
(0.699–1.853)
Visceral and bone
Menopausal status
ER status
95% CI
0.0308
(0.651–2.24)
0.5492
1.204
(0.421–3.445)
0.9339
1.141
(0.518–2.516) (1.338–3.63)
0.0019
(0.945–2.501)
0.0831
(0.94–3.034)
0.0794
Negative
1.537
Positive
1
Negative
1.689
Luminal
1
HER2 positive
0.993
(0.488–2.023)
0.9853
Triple negative
2.451
(1.088–5.519)
0.0304
cer cells and can be used as sensitive and specific marker in the EPISPOT assay (17 ). CK19 is also the most established and clinically validated marker for breast cancer patients in reverse-transcription PCR– based CTC assays, which detect minute amounts of CK19 mRNA (18 ). Using the EPISPOT assay, OS was correlated with the CTC status. In multivariate analysis, the presence of CTC as measured by the EPISPOT assay was not an independent predictor for OS. However, the combina-
tion of results from both the CellSearch and EPISPOT systems was a better predictor of OS. In the DETECT study previously published, no correlation was found between CTC detection and PFS using the CellSearch system. In the study we report here, performed in the same cohort of MBC patients, the CTC status determined by EPISPOT assay was not predictive of PFS either. As suggested by Mu¨ller et al., this lack of correlation could result from different definitions of disease progression Clinical Chemistry 60:1 (2014) 219
Table 4. Multivariate analysis of overall survival: CTC status (CellSearch and EPISPOT assays). Variable
Categories
Hazard ratio
95% CI
HER2 status
0.0364 Positive
1
Negative
2.241
Double negative
1
(1.052–4.772)
CTC status (CellSearch and EPISPOT assays)
0.0211 EPISPOT
14.479
(1.794–116.839)
CellSearch ⫹ EPISPOT
22.649
(2.787–184.081)
CellSearch
26.883
(3.168–228.143)
Therapeutic setting
0.005 First line
1
Second line
4.539
(1.636–12.594)
Third line or more
4.155
(1.652–10.454)
Positive
1
Negative
4.683
Estrogen receptor status
0.0001
based on institutional standards of the participating centers (11 ). The correlation between CTC detection and clinical outcome observed in our study suggests that the detected cells are tumor cells. However, circulation of normal epithelial cells under certain pathophysiological conditions may contribute to the pool of cells detected with the EPISPOT or CellSearch assays, as indicated by our previous publication (15 ). Further molecular characterization of CTC would be the only way to verify whether these circulating CK19-positive cells were of malignant or simply epithelial origin. However, in the EPISPOT assay, cells (including CTC) are washed off before the second antibody set revealed the immunospots. We are presently working on a new generation of EPISPOT in which the CK19-secreting cells can be collected for further molecular analysis. In peripheral blood of breast cancer patients, the majority of CTC were reported to be apoptotic (7, 8 ), and after mastectomy the half-life of CTC is estimated to be between 1 and 2.4 h (19 ). Although CTC undergo high levels of apoptosis in the peripheral blood, we observed that the majority of MBC patients present with viable CTC. However, this rate could be even higher. Due to the heterogeneity of CK expression in breast cancer (20, 21 ), we may miss viable CK19-negative CTC. Recent reports have related the high plasticity of CTC to an aggressive phenotype of tumor cells (22, 23 ). CTC can undergo epithelial–mesenchymal transition (EMT) and the reverse process of 220 Clinical Chemistry 60:1 (2014)
P
(2.168–10.114)
mesenchymal– epithelial transition. Epithelial CTC express E-cadherin, Epcam, and certain CKs. During the EMT process, these markers may be completely or partially downregulated in CTC, and expression of mesenchymal proteins such as vimentin and N-cadherin may become upregulated. In the present study, we used only epithelial markers and may have missed CTC with a mesenchymal phenotype. Thus, new markers that are not downregulated during the EMT process are urgently needed. For example, plastin-3, an actin bundling protein, is not repressed during EMT in colorectal cancer and has been recently used as a marker for prognostically relevant CTC of colorectal cancer (24 ). Plastin-3 is also expressed in other epithelial tumors (including breast cancer) and might, therefore, become a widespread marker for CTC assays. As an alternative, Yu et al. have used a combination of cell surface antigens to capture mesenchymal and epithelial CTC in MBC patients; however, CTC were detected in fewer than 50% of MBC patients (25 ). In summary, we found that the majority of MBC patients harbor viable CTC expressing and releasing CK19, and the presence of these tumor cells in the peripheral blood predicts an unfavorable clinical outcome, especially if combined with results of the CellSearch system. These findings highlight the need to develop more functional CTC assays and even to combine CTC assays based on different detection principles to address the problem of intra- and interpatient heterogeneity of CTC in MBC patients.
Viable Circulating Tumor Cells in Metastatic Breast Cancer Patients
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
Stock Ownership: None declared. Honoraria: None declared. Research Funding: Roche unrestricted grant; K. Pantel, European Research Council Investigator grant “DISSECT” (number 269081) and ERA-NET TRANSCAN grant “CTC-SCAN”; C. Alix-Panabie`res, INCa-DGOS-Inserm 6045 and ERA-NET TRANSCAN grant CTC-SCAN. Expert Testimony: None declared.
Authors’ Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest:
Patents: K. Pantel, patent number 07825055.2; C. Alix-Panabie`res, patent number IB2007/002549.
Employment or Leadership: L. Cayrefourcq, LCCRH. Consultant or Advisory Role: K. Pantel, Veridex, GILUPI, and Alere.
Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.
References 1. Coughlin SS, Ekwueme DU. Breast cancer as a global health concern. Cancer Epidemiol 2009;33: 315– 8. 2. Bidard FC, Vincent-Salomon A, Sigal-Zafrani B, Dieras V, Mathiot C, Mignot L, et al. Prognosis of women with stage IV breast cancer depends on detection of circulating tumor cells rather than disseminated tumor cells. Ann Oncol 2008;19: 496 –500. 3. Hayes DF, Cristofanilli M, Budd GT, Ellis MJ, Stopeck A, Miller MC, et al. Circulating tumor cells at each follow-up time point during therapy of metastatic breast cancer patients predict progression-free and overall survival. Clin Cancer Res 2006;12:4218 –24. 4. Riethdorf S, Fritsche H, Muller V, Rau T, Schindlbeck C, Rack B, et al. Detection of circulating tumor cells in peripheral blood of patients with metastatic breast cancer: a validation study of the CellSearch system. Clin Cancer Res 2007;13: 920 – 8. 5. Pantel K, Alix-Panabieres C, Riethdorf S. Cancer micrometastases. Nat Rev Clin Oncol 2009;6: 339 –51. 6. Pantel K, Brakenhoff RH, Brandt B. Detection, clinical relevance and specific biological properties of disseminating tumour cells. Nat Rev Cancer 2008;8:329 – 40. 7. Mehes G, Witt A, Kubista E, Ambros PF. Circulating breast cancer cells are frequently apoptotic. Am J Pathol 2001;159:17–20. 8. Rossi E, Basso U, Celadin R, Zilio F, Pucciarelli S, Aieta M, et al. M30 neoepitope expression in epithelial cancer: quantification of apoptosis in circulating tumor cells by CellSearch analysis. Clin Cancer Res 2010;16:5233– 43. 9. Alix-Panabieres C. EPISPOT assay: detection of viable DTCs/CTCs in solid tumor patients. Recent
Results Cancer Res 2012;195:69 –76. 10. Braun S, Naume B. Circulating and disseminated tumor cells. J Clin Oncol 2005;23:1623– 6. 11. Muller V, Riethdorf S, Rack B, Janni W, Fasching PA, Solomayer E, et al. Prognostic impact of circulating tumor cells assessed with the CellSearch System™ and AdnaTest Breast™ in metastatic breast cancer patients: the DETECT study. Breast Cancer Res 2012;14:R118. 12. Wallwiener M, Hartkopf AD, Baccelli I, Riethdorf S, Schott S, Pantel K, et al. The prognostic impact of circulating tumor cells in subtypes of metastatic breast cancer. Breast Cancer Res Treat 2013;137:503–10. 13. Riethdorf S, Muller V, Zhang L, Rau T, Loibl S, Komor M, et al. Detection and HER2 expression of circulating tumor cells: prospective monitoring in breast cancer patients treated in the neoadjuvant GeparQuattro trial. Clin Cancer Res 2010; 16:2634 – 45. 14. Deneve E, Riethdorf S, Ramos J, Nocca D, Coffy A, Daures JP, et al. Capture of viable circulating tumor cells in the liver of colorectal cancer patients. Clin Chem 2013;59:1384 –92. 15. Pantel K, Deneve E, Nocca D, Coffy A, Vendrell JP, Maudelonde T, et al. Circulating epithelial cells in patients with benign colon diseases. Clin Chem 2012;58:936 – 40. 16. Cristofanilli M, Budd GT, Ellis MJ, Stopeck A, Matera J, Miller MC, et al. Circulating tumor cells, disease progression, and survival in metastatic breast cancer. N Engl J Med 2004;351:781–91. 17. Alix-Panabieres C, Vendrell JP, Slijper M, Pelle O, Barbotte E, Mercier G, et al. Full-length cytokeratin-19 is released by human tumor cells: a potential role in metastatic progression of breast cancer. Breast Cancer Res 2009;11:R39. 18. Stathopoulou A, Gizi A, Perraki M, Apostolaki S,
19.
20.
21.
22.
23.
24.
25.
Malamos N, Mavroudis D, et al. Real-time quantification of CK-19 mRNA-positive cells in peripheral blood of breast cancer patients using the lightcycler system. Clin Cancer Res 2003;9:5145– 51. Meng S, Tripathy D, Frenkel EP, Shete S, Naftalis EZ, Huth JF, et al. Circulating tumor cells in patients with breast cancer dormancy. Clin Cancer Res 2004;10:8152– 62. Deng G, Herrler M, Burgess D, Manna E, Krag D, Burke JF. Enrichment with anti-cytokeratin alone or combined with anti-EpCAM antibodies significantly increases the sensitivity for circulating tumor cell detection in metastatic breast cancer patients. Breast Cancer Res 2008;10:R69. Willipinski-Stapelfeldt B, Riethdorf S, Assmann V, Woelfle U, Rau T, Sauter G, et al. Changes in cytoskeletal protein composition indicative of an epithelial-mesenchymal transition in human micrometastatic and primary breast carcinoma cells. Clin Cancer Res 2005;11:8006 –14. Alix-Panabieres C, Schwarzenbach H, Pantel K. Circulating tumor cells and circulating tumor DNA. Annu Rev Med 2012;63:199 –215. Bednarz-Knoll N, Alix-Panabieres C, Pantel K. Plasticity of disseminating cancer cells in patients with epithelial malignancies. Cancer Metastasis Rev 2012;31:673– 87. Yokobori T, Iinuma H, Shimamura T, Imoto S, Sugimachi K, Ishii H, et al. Plastin3 is a novel marker for circulating tumor cells undergoing the epithelial-mesenchymal transition and is associated with colorectal cancer prognosis. Cancer Res 2013;73:2059 – 69. Yu M, Bardia A, Wittner BS, Stott SL, Smas ME, Ting DT, et al. Circulating breast tumor cells exhibit dynamic changes in epithelial and mesenchymal composition. Science 2013;339:580 – 4.
Clinical Chemistry 60:1 (2014) 221
Cancer Diagnostics
Prognostic Relevance of Viable Circulating Tumor Cells Detected by EPISPOT in Metastatic Breast Cancer Patients Jean-Marie Ramirez,1 Tanja Fehm,2 Mattea Orsini,3 Laure Cayrefourcq,1 Thierry Maudelonde,4 Klaus Pantel,5† and Catherine Alix-Panabie`res1,3†*
BACKGROUND: Detection of circulating tumor cells (CTC) in breast cancer patients is currently performed in many clinical trials, using different technologies, in particular the EpCAM-dependent CellSearch® system. The purpose of this study was to investigate the incidence and prognostic relevance of viable CTC in a large cohort of metastatic breast cancer (MBC) patients. METHODS: A total of 254 MBC patients were enrolled in a prospective multicenter study at first diagnosis of metastatic disease or disease progression (before the start of a new treatment regimen). After EpCAMindependent enrichment, viable CTC releasing cytokeratin-19 as an epithelial cell marker were detected in the peripheral blood by an EPISPOT assay, and the Food and Drug Administration cleared CellSearch was used as the reference method. RESULTS: Using the EPISPOT assay, CTC were detected in 59% of MBC patients. The overall survival (OS) was linked with the CTC status measured by EPISPOT (P ⫽ 0.0191), which allowed stratification of MBC patients in low- and high-risk groups. This stratification could be improved by addition of the CTC status assessed by the CellSearch system. In multivariate Cox proportionalhazards regression analysis, the 3 methods used to determine the level of CTC (EPISPOT, CellSearch, and combination of EPISPOT/CellSearch) were compared by the Bayesian information criterion method. Interestingly, the combination of the EPISPOT and CellSearch assays was the strongest predictor of OS (hazard ratio, 22.6; 95% CI, 2.8 –184.08). CONCLUSIONS: This is the first study in which CTC detection using the EPISPOT assay was evaluated on a
1
University Medical Centre, Saint-Eloi Hospital, Institute of Research in Biotherapy, Department of Cellular and Tissular Biopathology of tumors, Laboratory of Rare Human Circulating Cells, Montpellier, France; 2 Department of Obstetrics and Gynecology, University Medical Center, Duesseldorf, Germany; 3 University Institute of Clinical Research UM1—EA2415—Epidemiology, Biostatistics & Public Health, Montpellier, France; 4 University Medical Centre, Laboratory of Cellular and Hormonal Biology, Department of Cellular and Tissular Biopathology of Tumors, Arnaud de Villeneuve Hospital, Montpellier, France; 5 Department of Tumor Biology, University Medical, Center HamburgEppendorf, Hamburg, Germany. † Klaus Pantel and Catherine Alix-Panabie`res contributed equally to the work, and both should be considered as first authors.
214
large cohort of MBC patients, showing prognostic relevance of the presence of viable CTC. © 2013 American Association for Clinical Chemistry
The WHO has estimated that more than 1 million women develop breast cancer every year (1 ). Despite new forms of treatment, such as Herceptin® (trastuzumab), an anti– human epidermal growth factor receptor 2 (HER2)6 monoclonal antibody, or Tykerb® (lapatinib), a tyrosine kinase inhibitor targeting HER2, breast cancer remains an incurable disease, but the current overall survival (OS) rate at 10 years for breast cancer in general is approximately 75% in Western Europe. Circulating tumor cell (CTC) detection is a new approach with utility for (a) cancer prognosis, (b) personalized treatment depending on the CTC profile, (c) monitoring disease progression, and finally (d) better understanding of the genesis and evolution of breast cancer with the aim of developing new treatments. Detection of CTC currently is performed by use of the CellSearch® system, which was cleared by the Food and Drug Administration (FDA) in 2004 for CTC enumeration in the peripheral blood of patients with metastatic breast, prostate, and colon cancer (2– 4 ). Other innovative technologies for CTC detection have been developed, such as quantitative reverse-transcription PCR assays, CTC microchips, and filtration devices (5, 6 ). However, none of these devices are specific for living CTC, which is a serious limitation because many CTC are apoptotic in cancer patients (7, 8 ). We have developed a new technique in our laboratory, named the EPISPOT assay, which specifically
* Address correspondence to this author at: Laboratory of Rare Human Circulating Cells, Institute of Research in Biotherapy, Saint-Eloi Hospital, University Medical Centre, 80, Ave. Augustin Fliche, Montpellier, France. Fax ⫹33-(0)467-33-01-13; e-mail [email protected]. Received August 22, 2013; accepted October 21, 2013. Previously published online at DOI: 10.1373/clinchem.2013.215079 6 Nonstandard abbreviations: HER2, human epidermal growth factor receptor 2; OS, overall survival; CTC, circulating tumor cells; FDA, Food and Drug Administration; MBC, metastatic breast cancer; DTC, disseminated tumor cells; LCCRH, Laboratory of Rare Circulating Human Cells; CK19, cytokeratin-19; PFS, progression-free survival; BIC, Bayesian information criterion; EMT, epithelial– mesenchymal transition.
Viable Circulating Tumor Cells in Metastatic Breast Cancer Patients
detects living CTC and disseminated tumor cells (DTC) in cancer patients (9 ). This technique is based on the detection of proteins secreted by functional CTC or DTC cells combined with a negative enrichment (leukocyte depletion). This new method allows enumeration of only living CTC. Recently, it has been demonstrated that the majority of circulating breast cancer cells undergo apoptosis (7, 8 ). However, an efficient metastasis process requires living CTC (10 ). Thus, in contrast to other CTC technologies, the EPISPOT assay focuses on CTC with the potential to colonize secondary organs such as liver, lung, or bone. Moreover, the EPISPOT assay is a promising tool to identify proteins secreted by CTC or DTC, thereby providing an opportunity to better characterize their intrinsic properties responsible for survival, migration, and colonization of distant organs. Here, we report the results of the first prospective multicenter study to evaluate the prognostic impact of CTC detection using the EPISPOT assay in parallel with the CellSearch system (gold standard) in metastatic breast cancer (MBC) patients. Materials and Methods PATIENTS
A total of 254 MBC patients from 9 German university breast cancer centers [Du¨sseldorf (n ⫽ 4), Erlangen (30), Essen (46), Freiburg (9), Hamburg (79), Heidelberg (18), Munich (16), Regensburg (2), and Tuebingen (50)] were enrolled in this prospective open study from December 2007 until April 2009. The median (range) age of the patients was 60 (49 – 68) years, and 50% of the patients were over 60 years old. Briefly, 67% of the patients had metastasis in more than 1 site and 14% had metastasis localized in the bone, 39% had visceral metastatic sites, and 47% had both sites (visceral and bone). The criteria used for patient enrollment were (a) epithelial invasive carcinoma of the breast with metastasis to distant organs (M1), (b) age at least 18 years, (c) first diagnosis of metastatic disease or disease progression (before the start of a new line of therapy), (d) written informed consent, and (e) no secondary primary malignancy. A criterion for exclusion was the presence of a secondary primary malignancy (except in situ carcinoma of the cervix or adequately treated basal cell carcinoma of the skin). Patients were treated according to international guidelines, as indicated in the previous publications of the DETECT (Prognostic impact of circulating tumor cells assessed with the CellSearch System™ and AdnaTest Breast™ in metastatic breast cancer patients) study group (11, 12 ). All HER2⫹ patients received HER2targeted therapy (usually trastuzumab) that may in-
deed have influenced the prognosis of HER2⫹ patients. Hormone receptor–positive patients received endocrine therapy (eventually plus chemotherapy), and hormone receptor–negative patients received chemotherapy. Blood was collected before the start of a new treatment regimen. Participants gave their consent for the use of their blood samples. Data management was carried out from our databank (11 ). The local ethics committee approved this analysis (2007/B01). Blood was drawn before the start of a new line of therapy. All patients gave informed consent for the use of their blood samples. A web-based databank was designed for data management and online-documentation (www.detectstudy.de). By the use of this interface, clinical investigators were blinded for test results and the CTC test sites were blinded for the clinical data of the patients. The local institutional review boards approved the conduct of this study (2007/B01). METHODS
CTC detection. The detection of CTC was performed using both the EPISPOT assay and the CellSearch system (Veridex). CTC analysis by the CellSearch system was performed as described previously (11, 13 ). Blood samples for the EPISPOT assay were analyzed at the Laboratory of Rare Circulating Human Cells (LCCRH) at the Institute of Research in Biotherapy, UMC, Montpellier, France. The blood samples were sent at room temperature and processed within 24 h. EPISPOT assay. Preanalytical conditions were determined by spiking experiments with breast cancer cell lines. We observed a decrease in tumor cell viability and number after 24 h at room temperature. Storage at ⫹4 °C was less favorable. The EPISPOT assay was performed in only 194 of the 254 patients because Montpellier joined the DETECT study group after the start of the study and we therefore missed the first blood samples for the EPISPOT analysis and because of the exclusion of blood samples that failed to meet preanalytical quality criteria. These samples included those with less than the required 7.5 mL and those that were received more than 24 h following the blood draw. The EPISPOT assay was performed from a blood samples of 7.5 mL collected in EDTA tubes. Enrichment of CTC was performed using the RosetteSep human circulating epithelial tumor cell enrichment cocktail (StemCell Technology). The nitrocellulose membranes of the EPISPOT plate were precoated with a defined antibody to target a specific protein, anti– cytokeratin-19 (CK19) antibody in our study (9, 14, 15 ). The coating antibody CK19 (clone Ks 19.1, Progen BioClinical Chemistry 60:1 (2014) 215
technik) was diluted in PBS (6 g/mL) and was incubated at 4 °C for 12 h. After membrane blocking (100 L of 5% BSA/PBS at 37 °C for 2 h), CTC enriched by leukocyte depletion with the RosetteSep human circulating epithelial tumor cell enrichment cocktail (StemCell Technology) were incubated at 37 °C (5% CO2) for 18 h in cell line medium [DMEM ⫹ GlutaMax (Gibco) with 10% fetal calf serum and penicillin/streptomycin]. Membranes were then washed using PBS/0.1% Tween (6 times) and PBS (3 times). The CK19 proteins released during the cell culture were detected by a second Alexa555 conjugated anti-CK19 antibody (clone Ks 19.2, 3 g/mL; Progen Biotechnik GMBH). This antibody was diluted in PBS and incubated at 4 °C for 12 h. All single protein fingerprints generated CK19 immunospots that were observed and enumerated using video camera imaging and computer-assisted analysis (KS ELISPOT, Carl Zeiss Vision). CellSearch assay. Blood samples were collected into CellSave tubes (Veridex). The enrichment and enumeration of CTC were carried out using the CellSearch epithelial cell test (Veridex). Briefly, CTC were isolated from peripheral blood by an anti-EpCAM antibody– bearing ferrofluid and all DAPI(⫺)CD45(⫺)PanCK19(⫹) cells were identified as CTC (4 ). A blood sample containing at least 5 CTC was considered positive (16 ). STATISTICAL ANALYSIS
Associations between categorical variables were examined using 2 tests. Overall median survival was calculated by the Kaplan–Meier method. A time-dependent ROC curve was drawn and a MAX-STAT test was used to determine a CTC threshold for OS and progressionfree survival (PFS). PFS was calculated as described by Mu¨ller et al. (11 ). Log-rank tests were used to compare the survival curves by CTC detection groups. Univariate and multivariate Cox proportional hazards regression models were used to obtain unadjusted and fully adjusted hazard ratios and 95% CIs. The Bayesian information criterion (BIC) was used to compare Cox regression models. Results CTC DETECTION RATES
The specificity of the EPISPOT assay already had been determined in a group of 25 healthy controls, as reported in our previous publication (15 ). CTC expressing EpCAM and/or CK19 were detected in MBC patients using the CellSearch and the EPISPOT assays (Fig. 1). CTC counts determined by the EPISPOT assay were available for 194 of the 254 patients. Detection of CTC by the EPISPOT assay revealed that 59% (115/194) of patients were positive, 216 Clinical Chemistry 60:1 (2014)
Fig. 1. Viable CTCs identified in MBC patients using the EPISPOT assay. Analysis of viable CK19-releasing cells using the EPISPOT assay in a large cohort containing a total of 194 MBC patients. The SKBR3 cell line was used as positive control (left). Representative photo of single CK19 immunospots corresponding to viable CK19-releasing cells from a MBC patient (right).
whereas only 48.0% (122/254) were positive for CTC detected by the CellSearch assay. These data are summarized in Table 1. Comparison of the EPISPOT and CellSearch assays revealed a low agreement between the 2 methods as demonstrated by a low Cohen coefficient of 0.14 (P ⫽ 0.0416) (Table 2). Furthermore, we analyzed the combination of both methods in a total of 194 MBC patients. In 76.3% of cases, CTC were detected either by the EPISPOT assay alone (27.4%), or the CellSearch system alone (15.3%) or both methods (33.7%), whereas 23.7% of patients remained CTC negative. SURVIVAL ANALYSIS RELATED TO CTC RESULTS OBTAINED WITH EPISPOT ASSAY
To evaluate the EPISPOT assay as a predictor of survival, the OS according to the concentrations of CTC detected by the EPISPOT assay was estimated by the Kaplan–Meier method. A time-dependent ROC curve was drawn using the survival and CTC status (EPISPOT assay) of each MBC patient. ROC curve analysis showed that a cutoff value of 1 CTC has the highest sensitivity/specificity to predict OS. Of 194 patients, 115 had ⱖ1 CK19-releasing CTC (59%) per 7.5 mL of blood. These CTC-positive patients showed a significantly reduced OS compared to patients without CK19-releasing CTC (P ⫽ 0.0191, log-rank test) (Fig. 2A). The median OS of patients without CK19releasing CTC was significantly longer (5 years; 95% CI, 4.16 – 8.91) in comparison with CTC-positive
Viable Circulating Tumor Cells in Metastatic Breast Cancer Patients
Table 1. Prevalence of CTC: comparison between CellSearch and EPISPOT assays. CTC CellSearch, n (%) >1 (n ⴝ 142)
Variable
>5 (n ⴝ 97)
CTC EPISPOT, n (%)
>10 (n ⴝ 83)
>100 (n ⴝ 33)
>1 (n ⴝ 115)
>5 (n ⴝ 88)
>10 (n ⴝ 68)
>100 (n ⴝ 19)
Age class ⬍60 years
80 (56.3%) 53 (54.6%)
48 (57.8%) 22 (66.7%) 55 (47.8%) 42 (47.7%) 37 (54.4%) 13 (68.4%)
ⱖ60 years
62 (43.7%) 44 (45.4%)
35 (42.2%) 11 (33.3%) 60 (52.2%) 46 (52.3%) 31 (45.6%)
6 (31.6%)
Histology Ductal
112 (78.9%) 74 (76.3%)
62 (74.7%) 24 (72.7%) 87 (75.7%) 69 (78.4%) 54 (79.4%) 13 (68.4%)
Lobular
15 (10.6%) 11 (11.3%)
9 (10.8%)
Others
15 (10.6%) 12 (12.4%)
12 (14.5%)
Negative
36 (25.4%) 25 (25.8%)
Positive
106 (74.6%) 72 (74.2%)
6 (18.2%) 13 (11.3%)
9 (10.2%)
6 (8.8%)
3 (15.8%)
10 (11.4%)
8 (11.8%)
3 (15.8%)
24 (28.9%) 11 (33.3%) 29 (25.2%) 24 (27.3%) 16 (23.5%)
6 (31.6%)
3 (9.1%)
15 (13%)
Estrogen receptor status 59 (71.1%) 22 (66.7%) 86 (74.8%) 64 (72.7%) 52 (76.5%) 13 (68.4%)
Progesterone receptor status Negative
52 (36.6%) 36 (37.1%)
33 (39.8%) 14 (42.4%) 41 (35.7%) 36 (40.9%) 28 (41.2%)
Positive
90 (63.4%) 61 (62.9%)
50 (60.2%) 19 (57.6%) 74 (64.3%) 52 (59.1%) 40 (58.8%) 11 (57.9%)
8 (42.1%)
Missing
17 (12%)
10 (10.3%)
10 (12%)
Negative
88 (62%)
61 (62.9%)
49 (59%)
Positive
37 (26.1%) 26 (26.8%)
HER2 status 2 (6.1%)
15 (13%)
13 (14.8%)
8 (11.8%)
4 (21.1%)
22 (66.7%) 65 (56.5%) 47 (53.4%) 38 (55.9%) 10 (52.6%)
24 (28.9%)
9 (27.3%) 35 (30.4%) 28 (31.8%) 22 (32.4%)
5 (26.3%)
Metastatic sites Bone
19 (13.4%) 11 (11.3%)
Visceral
50 (35.2%) 32 (33%)
29 (34.9%)
9 (10.8%)
2 (6.1%)
12 (13.6%) 10 (14.7%)
3 (15.8%)
9 (27.3%) 45 (39.1%) 33 (37.5%) 25 (36.8%)
15 (13%)
4 (21.1%)
Multiple
73 (51.4%) 54 (55.7%)
45 (54.2%) 22 (66.7%) 55 (47.8%) 43 (48.9%) 33 (48.5%) 12 (63.2%)
41 (28.9%) 27 (27.8%)
24 (28.9%)
Number of metastatic sites One site Multiple sites
8 (24.2%) 36 (31.3%) 27 (30.7%) 20 (29.4%)
4 (21.1%)
101 (71.1%) 70 (72.2%) 59/83 (71.1%) 25 (75.8%) 79 (68.7%) 61 (69.3%) 48 (70.6%) 15 (78.9%)
Disease-free interval ⱕ12 months
36 (25.4%) 28 (28.9%) 22/83 (26.5%) 13 (39.4%) 25 (21.7%) 20 (22.7%) 17 (25%)
5 (26.3%)
⬎12 months
106 (74.6%) 69 (71.1%) 61/83 (73.5%) 20 (60.6%) 90 (78.3%) 68 (77.3%) 51 (75%)
14 (73.7%)
Therapeutic setting First line
58 (40.8%) 43 (44.3%) 36/83 (43.4%) 17 (51.5%) 45 (39.1%) 34 (38.6%) 27 (39.7%)
Second line
39 (27.5%) 22 (22.7%) 18/83 (21.7%)
Third line or more
45 (31.7%) 32 (33%)
18 (20.5%) 14 (20.6%)
4 (21.1%)
29/83 (34.9%) 11 (33.3%) 39 (33.9%) 36 (40.9%) 27 (39.7%)
6 (31.6%)
Table 2. Contingency table comparing the EPISPOT and CellSearch assays (N ⴝ 188).a EPISPOT
a
patients (2.8 years; 95% CI, 1.43–7.95). For PFS, no threshold of prognostically relevant CTC counts was identified by the Max-STAT and ROC curve analysis. SURVIVAL ANALYSIS RELATED TO CTC RESULTS OBTAINED WITH
CellSearch
CTC negative
CTC positive
CTC negative
45
52
CTC positive
29
62
Cohen coefficient ⫽ 0.14; P ⫽ 0.041.
5 (15.2%) 31 (27%)
9 (47.4%)
CELLSEARCH ASSAY
In the cohort of 254 MBC patients (DETECT study), the prognostic value of the CellSearch assay has been reported by Mu¨ller et al. (11 ). The median OS was 18.1 months in CTC-positive patients (95% CI 15.1–22.1). The PFS was not found to be correlated to the CTC Clinical Chemistry 60:1 (2014) 217
SURVIVAL ANALYSIS RELATED TO COMBINED EPISPOT AND CELLSEARCH RESULTS
We subsequently compared the prognostic power of the EPISPOT-based CTC detection with that of the CellSearch assay as the only FDA-cleared CTC assay (gold standard). OS was evaluated according to CTC status assessed by the EPISPOT assay, the CellSearch assay and by a combination of both methods. Interestingly, Kaplan-Meier analysis showed that the OS was significantly better in the patients negative for CTC using both methods of detection (Fig. 2B). Statistical analysis was carried out using the logrank test between CTC double-negative patients and the other 3 groups (EPISPOT positive, CellSearch positive, and double positive). We observed that the survival period was significantly shorter for the 3 groups with at least 1 test positive for CTC in comparison with the double-negative group (Both EPISPOT and CellSearch positive vs double negative, P ⫽ 0.0004, logrank test; only EPISPOT positive vs double negative, P ⫽ 0.0062, log-rank test; only CellSearch positive vs double negative, P ⫽ 0.0067, log-rank test). Thus, the combination of both CTC technologies was found to enhance the stratification of MBC patients into lowand high-risk groups. MULTIVARIATE COX ANALYSIS OF OS
Fig. 2. Overall survival (OS) of metastatic breast cancer (MBC) patient by EPISPOT assay and by the combination of the EPISPOT and CellSearch assays. Analysis (A), Kaplan-Meier plot estimating OS for 2 risk groups (0 CTC, ⱖ1 CTC per blood sample) obtained from the EPISPOT assay (P ⫽ 0.0191). (B), Kaplan-Meier plot estimating OS for 4 risk groups (CellSearch positive, EPISPOT positive, double negative, double positive) (P ⫽ 0.0021). Statistical analysis of OS between the 4 groups: EPISPOT ⫹ CellSearch vs double negative, P ⫽ 0.0004 log-rank test; EPISPOT vs double negative, P ⫽ 0.0062 log-rank test; CellSearch vs double negative, P ⫽ 0.0067 log-rank test.
The variables (CTC status, HER2, progesterone- and estrogen receptor status, therapeutic setting, and disease-free interval) found to be significant at a 20% threshold in a univariate Cox analysis for OS were included into the multivariate analysis (Table 3). Although clinical factors such as HER2 and estrogen receptor status and therapeutic setting (P ⫽ 0.0001, P ⫽ 0.0364, and P ⫽ 0.005, respectively) remained prognostically relevant, the CTC status (EPISPOT and CellSearch method) emerged as the second best independent predictor of OS (P ⫽ 0.0211) (Table 4). To determine the best predictor of OS among the 3 CTC detection approaches (i.e., EPISPOT assay, CellSearch system, or combined CellSearch and EPISPOT analyses), we compared them using the BIC method. We observed that the diagnostic approach using both the CellSearch and the EPISPOT results was the best predictor of OS (BIC, 317.650), followed by the EPISPOT (BIC, 346.203) and CellSearch methods (BIC, 435.1). Discussion
counts in this cohort receiving different types of treatments (P ⫽ 0.197). In multivariate analysis, we confirmed that the concentrations of CTC were an independent predictor for OS (data not shown). 218 Clinical Chemistry 60:1 (2014)
In this prospective multicenter study, the prognostic value of the EPISPOT assay was evaluated according to the concentrations of viable CK19-releasing CTC. We have previously shown that CK19 —although it is a cytoskeleton protein—is released from viable breast can-
Viable Circulating Tumor Cells in Metastatic Breast Cancer Patients
Table 3. Univariate analysis of overall survival. Variable
Age class Metastatic site
Number of metastatic sites Disease-free interval Therapeutic setting
CTC status (EPISPOT assay) CTC status (CellSearch assay) CTC status (CellSearch and EPISPOT assays)
Categories
Hazard ratio
ⱖ60 years
1
⬍60 years
1.138
Visceral
1 0.883
(0.526–1.481) (0.175–1.194)
1 Site
1
Multiple sites
1.166
⬎12 months
1
ⱕ12 months
0.584
0.0767 (0.322–1.059)
Third line or more
1 1.302
(0.74–2.291)
First line
0.456
(0.231–0.9)
⬎0 CTC/mL
1
0 CTC/mL
0.467
Positive
1
Negative
0.386
Double negative
1
0.0128
0.0191 (0.247–0.883) 0.0007 (0.223–0.668)
CellSearch
7.582
(1.646–34.923)
EPISPOT
6.175
(1.38–27.628) (2.053–37.132)
1
Postmenopausal
1.208
Histology
Others
1
Lobular Ductal Positive
1
Negative
2.204
Positive
1
Subtype
0.5883 (0.668–2.035)
Second line
8.732
HER2 status
0.279
0.457
Premenopausal
PR status
0.6033
Bone
CellSearch ⫹ EPISPOT
P
(0.699–1.853)
Visceral and bone
Menopausal status
ER status
95% CI
0.0308
(0.651–2.24)
0.5492
1.204
(0.421–3.445)
0.9339
1.141
(0.518–2.516) (1.338–3.63)
0.0019
(0.945–2.501)
0.0831
(0.94–3.034)
0.0794
Negative
1.537
Positive
1
Negative
1.689
Luminal
1
HER2 positive
0.993
(0.488–2.023)
0.9853
Triple negative
2.451
(1.088–5.519)
0.0304
cer cells and can be used as sensitive and specific marker in the EPISPOT assay (17 ). CK19 is also the most established and clinically validated marker for breast cancer patients in reverse-transcription PCR– based CTC assays, which detect minute amounts of CK19 mRNA (18 ). Using the EPISPOT assay, OS was correlated with the CTC status. In multivariate analysis, the presence of CTC as measured by the EPISPOT assay was not an independent predictor for OS. However, the combina-
tion of results from both the CellSearch and EPISPOT systems was a better predictor of OS. In the DETECT study previously published, no correlation was found between CTC detection and PFS using the CellSearch system. In the study we report here, performed in the same cohort of MBC patients, the CTC status determined by EPISPOT assay was not predictive of PFS either. As suggested by Mu¨ller et al., this lack of correlation could result from different definitions of disease progression Clinical Chemistry 60:1 (2014) 219
Table 4. Multivariate analysis of overall survival: CTC status (CellSearch and EPISPOT assays). Variable
Categories
Hazard ratio
95% CI
HER2 status
0.0364 Positive
1
Negative
2.241
Double negative
1
(1.052–4.772)
CTC status (CellSearch and EPISPOT assays)
0.0211 EPISPOT
14.479
(1.794–116.839)
CellSearch ⫹ EPISPOT
22.649
(2.787–184.081)
CellSearch
26.883
(3.168–228.143)
Therapeutic setting
0.005 First line
1
Second line
4.539
(1.636–12.594)
Third line or more
4.155
(1.652–10.454)
Positive
1
Negative
4.683
Estrogen receptor status
0.0001
based on institutional standards of the participating centers (11 ). The correlation between CTC detection and clinical outcome observed in our study suggests that the detected cells are tumor cells. However, circulation of normal epithelial cells under certain pathophysiological conditions may contribute to the pool of cells detected with the EPISPOT or CellSearch assays, as indicated by our previous publication (15 ). Further molecular characterization of CTC would be the only way to verify whether these circulating CK19-positive cells were of malignant or simply epithelial origin. However, in the EPISPOT assay, cells (including CTC) are washed off before the second antibody set revealed the immunospots. We are presently working on a new generation of EPISPOT in which the CK19-secreting cells can be collected for further molecular analysis. In peripheral blood of breast cancer patients, the majority of CTC were reported to be apoptotic (7, 8 ), and after mastectomy the half-life of CTC is estimated to be between 1 and 2.4 h (19 ). Although CTC undergo high levels of apoptosis in the peripheral blood, we observed that the majority of MBC patients present with viable CTC. However, this rate could be even higher. Due to the heterogeneity of CK expression in breast cancer (20, 21 ), we may miss viable CK19-negative CTC. Recent reports have related the high plasticity of CTC to an aggressive phenotype of tumor cells (22, 23 ). CTC can undergo epithelial–mesenchymal transition (EMT) and the reverse process of 220 Clinical Chemistry 60:1 (2014)
P
(2.168–10.114)
mesenchymal– epithelial transition. Epithelial CTC express E-cadherin, Epcam, and certain CKs. During the EMT process, these markers may be completely or partially downregulated in CTC, and expression of mesenchymal proteins such as vimentin and N-cadherin may become upregulated. In the present study, we used only epithelial markers and may have missed CTC with a mesenchymal phenotype. Thus, new markers that are not downregulated during the EMT process are urgently needed. For example, plastin-3, an actin bundling protein, is not repressed during EMT in colorectal cancer and has been recently used as a marker for prognostically relevant CTC of colorectal cancer (24 ). Plastin-3 is also expressed in other epithelial tumors (including breast cancer) and might, therefore, become a widespread marker for CTC assays. As an alternative, Yu et al. have used a combination of cell surface antigens to capture mesenchymal and epithelial CTC in MBC patients; however, CTC were detected in fewer than 50% of MBC patients (25 ). In summary, we found that the majority of MBC patients harbor viable CTC expressing and releasing CK19, and the presence of these tumor cells in the peripheral blood predicts an unfavorable clinical outcome, especially if combined with results of the CellSearch system. These findings highlight the need to develop more functional CTC assays and even to combine CTC assays based on different detection principles to address the problem of intra- and interpatient heterogeneity of CTC in MBC patients.
Viable Circulating Tumor Cells in Metastatic Breast Cancer Patients
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
Stock Ownership: None declared. Honoraria: None declared. Research Funding: Roche unrestricted grant; K. Pantel, European Research Council Investigator grant “DISSECT” (number 269081) and ERA-NET TRANSCAN grant “CTC-SCAN”; C. Alix-Panabie`res, INCa-DGOS-Inserm 6045 and ERA-NET TRANSCAN grant CTC-SCAN. Expert Testimony: None declared.
Authors’ Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest:
Patents: K. Pantel, patent number 07825055.2; C. Alix-Panabie`res, patent number IB2007/002549.
Employment or Leadership: L. Cayrefourcq, LCCRH. Consultant or Advisory Role: K. Pantel, Veridex, GILUPI, and Alere.
Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.
References 1. Coughlin SS, Ekwueme DU. Breast cancer as a global health concern. Cancer Epidemiol 2009;33: 315– 8. 2. Bidard FC, Vincent-Salomon A, Sigal-Zafrani B, Dieras V, Mathiot C, Mignot L, et al. Prognosis of women with stage IV breast cancer depends on detection of circulating tumor cells rather than disseminated tumor cells. Ann Oncol 2008;19: 496 –500. 3. Hayes DF, Cristofanilli M, Budd GT, Ellis MJ, Stopeck A, Miller MC, et al. Circulating tumor cells at each follow-up time point during therapy of metastatic breast cancer patients predict progression-free and overall survival. Clin Cancer Res 2006;12:4218 –24. 4. Riethdorf S, Fritsche H, Muller V, Rau T, Schindlbeck C, Rack B, et al. Detection of circulating tumor cells in peripheral blood of patients with metastatic breast cancer: a validation study of the CellSearch system. Clin Cancer Res 2007;13: 920 – 8. 5. Pantel K, Alix-Panabieres C, Riethdorf S. Cancer micrometastases. Nat Rev Clin Oncol 2009;6: 339 –51. 6. Pantel K, Brakenhoff RH, Brandt B. Detection, clinical relevance and specific biological properties of disseminating tumour cells. Nat Rev Cancer 2008;8:329 – 40. 7. Mehes G, Witt A, Kubista E, Ambros PF. Circulating breast cancer cells are frequently apoptotic. Am J Pathol 2001;159:17–20. 8. Rossi E, Basso U, Celadin R, Zilio F, Pucciarelli S, Aieta M, et al. M30 neoepitope expression in epithelial cancer: quantification of apoptosis in circulating tumor cells by CellSearch analysis. Clin Cancer Res 2010;16:5233– 43. 9. Alix-Panabieres C. EPISPOT assay: detection of viable DTCs/CTCs in solid tumor patients. Recent
Results Cancer Res 2012;195:69 –76. 10. Braun S, Naume B. Circulating and disseminated tumor cells. J Clin Oncol 2005;23:1623– 6. 11. Muller V, Riethdorf S, Rack B, Janni W, Fasching PA, Solomayer E, et al. Prognostic impact of circulating tumor cells assessed with the CellSearch System™ and AdnaTest Breast™ in metastatic breast cancer patients: the DETECT study. Breast Cancer Res 2012;14:R118. 12. Wallwiener M, Hartkopf AD, Baccelli I, Riethdorf S, Schott S, Pantel K, et al. The prognostic impact of circulating tumor cells in subtypes of metastatic breast cancer. Breast Cancer Res Treat 2013;137:503–10. 13. Riethdorf S, Muller V, Zhang L, Rau T, Loibl S, Komor M, et al. Detection and HER2 expression of circulating tumor cells: prospective monitoring in breast cancer patients treated in the neoadjuvant GeparQuattro trial. Clin Cancer Res 2010; 16:2634 – 45. 14. Deneve E, Riethdorf S, Ramos J, Nocca D, Coffy A, Daures JP, et al. Capture of viable circulating tumor cells in the liver of colorectal cancer patients. Clin Chem 2013;59:1384 –92. 15. Pantel K, Deneve E, Nocca D, Coffy A, Vendrell JP, Maudelonde T, et al. Circulating epithelial cells in patients with benign colon diseases. Clin Chem 2012;58:936 – 40. 16. Cristofanilli M, Budd GT, Ellis MJ, Stopeck A, Matera J, Miller MC, et al. Circulating tumor cells, disease progression, and survival in metastatic breast cancer. N Engl J Med 2004;351:781–91. 17. Alix-Panabieres C, Vendrell JP, Slijper M, Pelle O, Barbotte E, Mercier G, et al. Full-length cytokeratin-19 is released by human tumor cells: a potential role in metastatic progression of breast cancer. Breast Cancer Res 2009;11:R39. 18. Stathopoulou A, Gizi A, Perraki M, Apostolaki S,
19.
20.
21.
22.
23.
24.
25.
Malamos N, Mavroudis D, et al. Real-time quantification of CK-19 mRNA-positive cells in peripheral blood of breast cancer patients using the lightcycler system. Clin Cancer Res 2003;9:5145– 51. Meng S, Tripathy D, Frenkel EP, Shete S, Naftalis EZ, Huth JF, et al. Circulating tumor cells in patients with breast cancer dormancy. Clin Cancer Res 2004;10:8152– 62. Deng G, Herrler M, Burgess D, Manna E, Krag D, Burke JF. Enrichment with anti-cytokeratin alone or combined with anti-EpCAM antibodies significantly increases the sensitivity for circulating tumor cell detection in metastatic breast cancer patients. Breast Cancer Res 2008;10:R69. Willipinski-Stapelfeldt B, Riethdorf S, Assmann V, Woelfle U, Rau T, Sauter G, et al. Changes in cytoskeletal protein composition indicative of an epithelial-mesenchymal transition in human micrometastatic and primary breast carcinoma cells. Clin Cancer Res 2005;11:8006 –14. Alix-Panabieres C, Schwarzenbach H, Pantel K. Circulating tumor cells and circulating tumor DNA. Annu Rev Med 2012;63:199 –215. Bednarz-Knoll N, Alix-Panabieres C, Pantel K. Plasticity of disseminating cancer cells in patients with epithelial malignancies. Cancer Metastasis Rev 2012;31:673– 87. Yokobori T, Iinuma H, Shimamura T, Imoto S, Sugimachi K, Ishii H, et al. Plastin3 is a novel marker for circulating tumor cells undergoing the epithelial-mesenchymal transition and is associated with colorectal cancer prognosis. Cancer Res 2013;73:2059 – 69. Yu M, Bardia A, Wittner BS, Stott SL, Smas ME, Ting DT, et al. Circulating breast tumor cells exhibit dynamic changes in epithelial and mesenchymal composition. Science 2013;339:580 – 4.
Clinical Chemistry 60:1 (2014) 221
