Ovarian cancer is currently the fifth leading cause of cancer death of women in Europe and the United States.1 The prognosis of patients with ovarian cancer is limited owing to the lack of early symptoms and a high rate of recurrence. The disease is diagnosed in most patients at an advanced stage, and although the initial response to chemotherapy is generally good, a significant proportion of patients will have a relapse despite optimal treatment based on current guidelines.2 The identification of new biomarkers, reflecting current disease status and tumor activity, could optimize prediction and monitoring of oncologic treatment and provide insights into the biology of ovarian cancer. In this regard, recent studies have increasingly focused on disseminated tumor cells (DTCs) in the bone marrow (BM).
The presence of DTCs is a common phenomenon in solid tumors of epithelial origin. For breast cancer, DTC detection has been demonstrated to be a strong independent prognostic factor (level I evidence).3 Available data support the notion that hematogenous tumor cell dissemination may be clinically relevant in ovarian cancer as well.4–8 Detection rates of DTC, as a surrogate parameter for occult hematogenous spread, vary between 30% and 50% of patients with primary ovarian cancer. Interestingly, ovarian metastases to the bone are only rarely observed.5,9,10 Possibly, BM acts as a secondary “homing site” in these patients where tumor cells are able to persist and may subsequently cause a relapse.
The aim of the present 3-center study was to prospectively evaluate the impact of BM status on clinical outcome in a group of 495 patients treated for ovarian cancer at 3 comprehensive cancer centers (University Hospital Tuebingen, Germany; University Hospital Essen, Germany; and University Hospital Munich, Germany).
MATERIALS AND METHODS
This 3-center analysis was performed at the Department of Obstetrics and Gynecology, University Hospital Tuebingen, Germany; the Department of Gynecology and Obstetrics, University Hospital Essen, Germany; and the Department of Gynecology and Obstetrics, University Hospital Munich, Germany. A total of 495 patients with ovarian cancer patients (International Federation of Gynecology and Obstetrics [FIGO] stages I–IV) treated between January 1994 to November 2010 in these centers were included in the analyses (Tuebingen, 229 patients [46%]; Essen, 148 patients [30%]; and Munich, 118 patients [24%]). The patients’ characteristics at the time of diagnosis are shown in Table 1. All specimens were obtained after written informed consent. Tissue sampling and analysis of data were approved by the local ethic committee (114/2006A, 05-2870). Treatment of ovarian cancer and follow-up examinations were performed according to current treatment guidelines. Relapse was confirmed by physical examination, computed tomographic scan, x-ray, ultrasound, tumor marker, and/or relaparotomy depending on localization of recurrence.
Preparation of Slides
Detection of DTCs was performed as described in detail previously for all 3 centers.4 Ten to 20-mL BM was aspirated intraoperatively from the iliac crest of both sides into syringes containing heparin anticoagulant under general anesthesia using the Jamshidi technique. Bone marrow samples were processed within 24 hours. Tumor cell isolation and detection were performed based on the recommendations for standardized tumor cell detection.11 Briefly, samples were separated by density centrifugation using Ficoll (density, 1077 g/mL; Biochrom, Germany). Mononuclear cells (MNCs) were collected from the interphase layer, spun down onto a glass slide (106 MNCs per spot; Hettich cytocentrifuge, Tuttlingen, Germany) and air-dried overnight at room temperature.
Staining of Slides and DTC Identification
For detection of cytokeratin (CK)-positive tumor cells, slides were fixed in 4% neutral buffered formalin for 10 minutes and rinsed in phosphate-buffered saline. Automatic immunostaining was performed on the Dako Autostainer using the monoclonal mouse A45-B/B3 antibody (Micromet, Munich, Germany) and the DAKO-APAAP detection kit (DakoCytomation, Glostrup, Denmark) according to the manufacturer’s instructions. The A45-B/B3 antibody is directed against a common epitope of CK polypeptides, including the CK heterodimers 8/18 and 8/19. The A45-B/B3 antibody is the most extensively studied antibody for DTC detection in ovarian cancer to date.4,5,9,12 Reproducibility of DTC detection by A45-B/B3 antibody has been tested by using ovarian cancer cell line SKOV3.12 In our analysis, for reasons of feasibility in the clinical laboratory routine the malignant breast cancer cell line MCF-7 was used as a positive control. For each patient, 2 × 106 cells were analyzed on 2 slides. All slides were evaluated by 2 independent cytologists. In case nonconcordant results were obtained, the slides were evaluated by a third investigator to obtain consensus. Identification of tumor cells was performed according to the ISHAGE evaluation criteria and the DTC consensus statements.
The χ2 test and the Fisher exact test were used to evaluate the relationship between circulating tumor cells (CTCs) and clinicopathological factors. For the survival analysis, we considered in separate analyses the following primary end points: (1) death and (2) relapse, defined as distant or local disease recurrence, or both. Survival intervals were measured from the time of BM aspiration to the time of death or of the first clinical, histological, or radiographic diagnosis of relapse. We constructed Kaplan-Meier curves and used the log-rank test to assess the univariate significance of the parameters. The effects of multiple variables on survival were evaluated by a Cox proportional-hazards regression model. All reported P values are 2-sided. The initial model included BM status, age at diagnosis, FIGO stage, nodal status, resection status, and grading. Subjects with missing values were excluded from modeling. Statistical analysis was performed by SPSS version 17.0 (SPSS Inc, Chicago, IL). P < 0.05 was considered statistically significant.
Incidence of DTC
Bone marrow aspirates were obtained intraoperatively from 495 patients with primary ovarian cancer. Most (74%) of the patients were at FIGO stages III to IV, 19% at FIGO stage I, and 7% at FIGO stage II. The mean age was 58 years (range, 18–88 years). Disseminated tumor cells were detected in 134 (27%) of 495 BM aspirates. The number of CK-positive cells ranged from 1 to 42 per 2 × 106 mononuclear cells. Bone marrow status did not correlate with age (P = 0.116), FIGO stage (P = 0.459), lymph node status (P = 0.061), histopathologic grading (P = 0.107), or resection status (P = 0.632; Table 1). The lowest prevalence of positive BM status was observed in patients with low-grade serous cancer (7%), whereas 28% of patients with high-grade serous carcinoma presented with DTC (P = 0.011). The prevalence of DTC presence in other histologic subtypes is presented in Table 1.
BM Status and Progression-Free Survival
Patients were not considered for survival analysis if the follow-up period was less than 4 months. Thus, follow-up data were available in 456 cases. Only these patients were included in the survival analysis. Median follow-up was 46 months (range, 4–104 months). Data on clinical outcome are summarized in Table 2. One hundred ninety-eight patients (43%) had recurrence during follow-up. Relapse was diagnosed in 48% of DTC-positive and 42% of DTC-negative patients (P = 0.225). Median progression-free survival (PFS) was shorter in patients with DTC in the BM (29 months; 95% CI, 22–36 months vs 24 months; 95% CI, 18–30 months; P = 0.164, determined by log-rank test; Fig. 1). In the multivariate regression analysis, BM status (P = 0.030), FIGO stage (P < 0.001), and resection status (P = 0.003) were the only independent predictors of PFS (Table 3).
BM Status and Overall Survival
One hundred ninety-six patients (43%) died during follow-up. Patients with DTC were more likely to die than BM-negative patients (51% vs 40%; P = 0.029). The median overall survival (OS) was significantly shorter among DTC-positive patients compared with DTC-negative patients (51 months; 95% CI, 37–65 months vs 33 months; 95% CI, 23–43 months; P = 0.023, determined by log-rank test). Survival curves are presented in Figure 1. Multivariate Cox regression model confirmed BM status as a strong independent predictor of shorter OS (P = 0.001). Age at diagnosis (P = 0.004), FIGO stage (P < 0.001), nodal status (P = 0.022), and resection status (P < 0.001), but not the grading, also predicted reduced OS in the multivariate analysis (Table 3). In the subgroup analysis, BM status was a strong predictor of OS in patients with high-grade serous carcinoma (295 patients; P = 0.001; Fig. 2). The subgroup analysis was not performed for other histologic types owing to small sample sizes. The prognostic effect of BM status was assessed within FIGO stages; for FIGO stage III to stage IV tumors (n = 299), a significant association with OS was confirmed (P = 0.001; Fig. 3). No significant correlation with OS was observed in FIGO I (n = 84) and FIGO II (n = 31) tumors when analyzed separately.
This is the largest pooled analysis so far on the prognostic relevance of DTC in primary ovarian cancer. A total of 495 patients were included in this prospective 3-center analysis. Disseminated tumor cells, as a surrogate parameter for minimal residual disease, could be detected in 27% of the patients, irrespective of the stage of the disease.
Other studies reported DTC incidence ranging from 20% to 60% in primary ovarian cancer, depending on the methodology and patients’ collective.4–6,8,13–16 Interestingly, ovarian malignancies rarely cause bone metastases.10 This suggests that BM serves in these patients rather as a temporary “homing site” for isolated tumor cells, from where they are able to migrate and subsequently cause metastasis or locoregional recurrence. Because hematogenous dissemination of isolated tumor cells may already be observed in FIGO stage I, it may be hypothesized that contrary to the assumed natural history of ovarian cancer, single tumor cells acquire the potential to disseminate to extraperitoneal sites very early in the process of the disease.5 Because DTCs may spread by means of the blood stream, we cannot exclude that those might also be able to repopulate the peritoneal cavity, an environment that easily supports ovarian cancer growth.
Influence of DTC Detection on Clinical Outcome
For ovarian cancer, data on the prognostic value of DTCs are so far limited (Table 4). In the present study, presence of DTCs predicted significantly shortened OS (P = 0.001 in the multivariate survival analysis). This is in accordance with several smaller studies. Braun et al5 demonstrated impaired prognosis with regard to distant disease-free survival (DFS) in BM-positive patients at the time of diagnosis; in a subset of 64 optimally debulked patients, DTC presence remained a strong prognostic factor (P = 0.002), which highlights the role of DTC detection especially in patients who received successful surgical cytoreduction. We previously reported a significant correlation of positive BM status with reduced DFS in a group of 112 patients with FIGO stage I to stage III ovarian cancer.4 Interestingly, in some studies, the presence of isolated tumor cells in secondary sites such as BM and blood was also associated with higher risk for local recurrence as well.4,7 Therefore, it might be speculated that hematogenous tumor cell dissemination may serve as an indicator of a more aggressive phenotype of the primary disease that is likely to cause local relapse. In contrast, other authors reported no significant correlation between DTC detection and clinical outcome in ovarian cancer.13,17 This discrepancy might be due to differences in study protocols, for example, time point of BM sample collection (preoperative vs postoperative aspiration). Hypothetically, a transient increase in cancer cell dissemination from the primary tumor due to intraoperative manipulation could contribute to false-positive results and therefore affect further analysis.18
Interestingly, in the current analysis, BM status did not influence PFS. This may be due to inconsistent relapse diagnosis. Most patients were treated within a clinical trial; these patients received intensified follow-up with computed tomographic scans at regular intervals. In the group of patients treated outside clinical trials, follow-up care was carried out according to national guidelines and was based on clinical examination and patients’ symptoms. Possibly, relapse was diagnosed later in the group treated outside clinical trials.
Beyond the prognostic value of DTC detection, monitoring of minimal residual disease during and after treatment offers the opportunity to assess the residual risk of relapse. The presence of DTCs and/or CTCs may indicate persistent occult tumor load after treatment and thus an insufficient therapy response. Wimberger et al12 assessed changes in DTC levels before and after first-line chemotherapy; marked increase in DTC numbers predicted significantly shorter PFS. Whether the reevaluation of the DTC status after completion of therapy may contribute to selection of patients who might benefit from extended or intensified treatment remains to be investigated.
Stem Cell Hypothesis
A provocative hypothesis has been introduced recently with respect to natural history and progression of ovarian cancer. While the “classical” stochastic model of cancer development holds that any cell may become a source of malignant transformation, emerging evidence supports the view that only a minor subpopulation of cancer cells has the potential to initiate cancer growth. These cells, called cancer stem cells (CSCs), have the ability to self-renew, propagate tumorigenesis, and are usually drug-resistant.19 Experimental studies on stem cell biology have given new impetus to the cancer stem cell theory. Cancer stem cells are assumed to play an important role in the development of various tumor entities, such as breast and gastrointestinal cancer, retinoblastoma, and ovarian cancer.20,21 Interestingly, ovarian cancer cell lines feature “side population” cells with potential to differentiate into cancers with different histologies, suggesting the pluripotent character of stem cells.22
One currently debated hypothesis is the theory that DTC/CTC, the surrogate marker for minimal residual disease and possibly precursor of systemic metastasis, represent in fact cancer stem cells. Most DTCs in breast cancer seem to express a putative CSC phenotype, such as ALDH1 positivity or presence of CD44 and absence of CD24, whereas CTCs were reported to exhibit stem cell and epithelial-mesenchymal transition markers.23,24 Several characteristics of ovarian cancer (eg, high recurrence rates and multidrug resistance) suggest that the disease might be initiated and maintained by a unique population of cells with stem cell–like properties.21 Advanced ovarian cancer generally responds to platinum-based combination therapy; however, this initial regression is often followed by emergence of therapy-resistant cell clones. One possible explanation for this phenomenon is the CSC-induced drug-resistance: standard therapies fail to target tumor-initiating cells.21
Despite advances in surgical and systemic therapy, ovarian cancer leads to relapse in 60% of patients within 5 years, resulting in poor OS. Currently, therapy efficacy is assessed by physical examinations, imaging, and evaluation of CA125 levels. New biomarkers are thus necessary for better prediction and prognostication.
Early hematogenous tumor cell dissemination is a common phenomenon observed in most solid tumors. Recent data support the clinical relevance of these cells; in the present report, we demonstrate significant impact of the presence of DTCs in the BM on survival. Whether these patients might benefit from extended or more aggressive therapy remains to be evaluated in future trials.
1. Ries LAG, Harkins D, Krapcho M, et al. SEER Cancer Statistics Review, 1975–2003
. Bethesda, MD: National Cancer Institute; 2006.
2. Cannistra SA. Cancer of the ovary. N Engl J Med
. 2004; 351: 2519–2529.
3. Braun S, Vogl FD, Naume B, et al. A pooled analysis of bone marrow
micrometastasis in breast cancer. N Engl J Med
. 2005; 353: 793–802.
4. Banys M, Solomayer EF, Becker S, et al. Disseminated tumor cells
in bone marrow
may affect prognosis of patients with gynecologic malignancies. Int J Gynecol Cancer
. 2009; 19: 948–952.
5. Braun S, Schindlbeck C, Hepp F, et al. Occult tumor cells in bone marrow
of patients with locoregionally restricted ovarian cancer
predict early distant metastatic relapse. J Clin Oncol
. 2001; 19: 368–375.
6. Aktas B, Kasimir-Bauer S, Heubner M, et al. Molecular profiling and prognostic relevance of circulating tumor cells in the blood of ovarian cancer
patients at primary diagnosis and after platinum-based chemotherapy. Int J Gynecol Cancer
. 2011; 21: 822–830.
7. Poveda A, Kaye SB, McCormack R, et al. Circulating tumor cells predict progression-free survival
and overall survival
in patients with relapsed/recurrent advanced ovarian cancer
. Gynecol Oncol
. 2011; 122: 567–572.
8. Schindlbeck C, Hantschmann P, Zerzer M, et al. Prognostic impact of KI67, p53, human epithelial growth factor receptor 2, topoisomerase IIalpha, epidermal growth factor receptor, and nm23 expression of ovarian carcinomas and disseminated tumor cells
in the bone marrow
. Int J Gynecol Cancer
. 2007; 17: 1047–1055.
9. Fehm T, Becker S, Bachmann C, et al. Detection of disseminated tumor cells
in patients with gynecological cancers. Gynecol Oncol
. 2006; 103: 942–947.
10. Dauplat J, Hacker NF, Nieberg RK, et al. Distant metastases in epithelial ovarian carcinoma. Cancer
. 1987; 60: 1561–1566.
11. Fehm T, Braun S, Muller V, et al. A concept for the standardized detection of disseminated tumor cells
in bone marrow
from patients with primary breast cancer and its clinical implementation. Cancer
. 2006; 107: 885–892.
12. Wimberger P, Heubner M, Otterbach F, et al. Influence of platinum-based chemotherapy on disseminated tumor cells
in blood and bone marrow
of patients with ovarian cancer
. Gynecol Oncol
. 2007; 107: 331–338.
13. Marth C, Kisic J, Kaern J, et al. Circulating tumor cells in the peripheral blood and bone marrow
of patients with ovarian carcinoma do not predict prognosis. Cancer
. 2002; 94: 707–712.
14. Wimberger P, Roth C, Pantel K, et al. Impact of platinum-based chemotherapy on circulating nucleic acid levels, protease activities in blood and disseminated tumor cells
in bone marrow
of ovarian cancer
patients. Int J Cancer
. 2011; 128: 2572–2580.
15. Fan T, Zhao Q, Chen JJ, et al. Clinical significance of circulating tumor cells detected by an invasion assay in peripheral blood of patients with ovarian cancer
. Gynecol Oncol
. 2009; 112: 185–191.
16. Heubner M, Wimberger P, Dahlmann B, et al. The prognostic impact of circulating proteasome concentrations in patients with epithelial ovarian cancer
. Gynecol Oncol
. 2011; 120: 233–238.
17. Cain JM, Ellis GK, Collins C, et al. Bone marrow
involvement in epithelial ovarian cancer
by immunocytochemical assessment. Gynecol Oncol
. 1990; 38: 442–445.
18. Camara O, Kavallaris A, Noschel H, et al. Seeding of epithelial cells into circulation during surgery for breast cancer: the fate of malignant and benign mobilized cells. World J Surg Oncol
. 2006; 4: 67.
19. Dyall S, Gayther SA, Dafou D. Cancer stem cells and epithelial ovarian cancer
. J Oncol
. 2010; 2010: 105269.
20. Setoguchi T, Taga T, Kondo T. Cancer stem cells persist in many cancer cell lines. Cell Cycle
. 2004; 3: 414–415.
21. Szotek PP, Pieretti-Vanmarcke R, Masiakos PT, et al. Ovarian cancer
side population defines cells with stem cell-like characteristics and Mullerian inhibiting substance responsiveness. Proc Natl Acad Sci U S A
. 2006; 103: 11154–11159.
22. Hosonuma S, Kobayashi Y, Kojo S, et al. Clinical significance of side population in ovarian cancer
cells. Hum Cell
. 2011; 24: 9–12.
23. Aktas B, Tewes M, Fehm T, et al. Stem cell and epithelial-mesenchymal transition markers are frequently overexpressed in circulating tumor cells of metastatic breast cancer patients. Breast Cancer Res
. 2009; 11: R46.
24. Balic M, Lin H, Young L, et al. Most early disseminated cancer cells detected in bone marrow
of breast cancer patients have a putative breast cancer stem cell phenotype. Clin Cancer Res
. 2006; 12: 5615–5621.
Keywords:Copyright © 2013 by IGCS and ESGO
Ovarian cancer; Disseminated tumor cells; Overall survival; Progression-free survival; Bone marrow