Circulating Tumor Cells: A Novel Prognostic Factor for Newly Diagnosed Metastatic Breast Cancer
http://www.100md.com
《临床肿瘤学》
The University of Texas M.D. Anderson Cancer Center, Houston, TX
The University of Michigan Comprehensive Cancer Center, Ann Arbor, MI
The Cleveland Clinic, Cleveland, OH
Washington University, St Louis, MO
University of Arizona, Phoenix, AZ
Immunicon Corporation, Huntingdon Valley, PA
ABSTRACT
PATIENTS AND METHODS: One hundred seventy-seven patients with measurable MBC were enrolled onto a prospective study. Eighty-three of the 177 patients were entering first-line treatment, and these patients are the focus of this analysis. CTCs from 7.5 mL of whole blood drawn before treatment initiation (baseline) and monthly thereafter for up to 6 months were isolated and enumerated using immunomagnetics.
RESULTS: The mean (± standard deviation) follow-up time was 11.1 ± 4.4 months (median, 12.2 months). Forty-three patients (52%) had ≥ five CTCs at baseline. The median PFS was 7.2 months (95% CI, 4.9 to 9.4 months), and the median OS was more than 18 months. Patients with ≥ five CTCs at baseline and at first follow-up (4 weeks) had a worse prognosis than patients with less than five CTCs (baseline: median PFS, 4.9 v 9.5 months, respectively; log-rank, P = .0014; median OS, 14.2 v > 18 months, respectively; log-rank, P = .0048; first follow-up: median PFS, 2.1 v 8.9 months, respectively; log-rank, P = .0070; median OS, 11.1 v > 18 months, respectively; log-rank, P = .0029). CTCs before and after the initiation of therapy were strong, independent prognostic factors.
CONCLUSION: Detection of CTCs before initiation of first-line therapy in patients with MBC is highly predictive of PFS and OS. This technology can aid in appropriate patient stratification and design of tailored treatments.
INTRODUCTION
Despite years of clinical research, the odds of achieving complete response for patients with metastatic breast cancer (MBC) are extremely low.1,2 Only a few patients who achieve a complete response after chemotherapy remain in this state for prolonged periods of time, with some patients remaining in remission beyond 20 years.2 These long-term survivors are usually young, have an excellent performance status, and, more importantly, have limited metastatic disease.3,4 The majority of patients with metastatic disease respond transiently to conventional treatments and develop evidence of progressive disease within 12 to 24 months of initiating treatment.1 For these patients, systemic treatment has not translated into a significant improvement in survival but has substantially improved their quality of life. Several clinical factors have been proposed but never prospectively validated that would help in the prediction of long-term outcome and efficacy of treatments.5,6
Circulating tumor cells (CTCs) can be detected in blood from patients with metastatic and primary carcinomas.7-11 Over the past several years, the development of immunomagnetic platforms has permitted accurate enumeration of CTCs at extremely low frequencies.11 In several case reports, the presence of CTCs has been associated with shortened survival times.12-15 We conducted a prospective, multicenter, double-blind study to determine the clinical significance of CTCs in patients with measurable MBC starting a new course of systemic therapy. The overall results of this study have been published elsewhere.16 This article represents the detailed analysis of only the cohort of patients with newly diagnosed MBC about to start first-line therapy. We believe this is an important analysis because patients who are newly diagnosed and about to start first-line therapy will derive the most benefit from appropriate risk stratification and treatment planning. Furthermore, an early prediction of treatment efficacy could have an impact in their quality of life.
PATIENTS AND METHODS
Before the initiation of therapy, patients had imaging evaluation (including computed tomography scans of chest and abdomen) of their metastatic sites and a baseline blood draw for enumeration of CTCs. Serial blood specimens were collected at roughly monthly intervals for a period of up to 6 months. Reassessments of disease status by the same modalities used at baseline were conducted every 9 to 12 weeks, depending on treatment type and schedule. Disease status was assessed using WHO criteria without knowledge of the patients’ CTC results.
Isolation and Enumeration of CTCs
Blood samples were drawn into 10-mL EDTA Vacutainer tubes (Becton Dickinson, Franklin Lakes, NJ), to which a cell preservative was subsequently added. Samples were maintained at room temperature and processed within 72 hours after collection. All CTC evaluations were performed without knowledge of the patients’ clinical status in one of two central laboratories (Immunicon, Huntingdon Valley, PA, or IMPATH Predictive Oncology, Los Angeles, CA) or at selected participating centers. CTCs were immunomagnetically enriched from 7.5 mL of blood using ferrofluids coated with antibodies targeting the epithelial cell adhesion molecule.17,18 Isolated cells were fluorescently labeled with the nucleic acid dye 4',6-diamidino-2-phenylindole (DAPI) and monoclonal antibodies specific for leukocytes (CD45-allophycocyanin) and epithelial cells (cytokeratin 8,18,19–phycoerythrin) and analyzed by the CellSpotter Analyzer17,19 (Veridex LLC, Warren, NJ).
CTC Definition
The CellSpotter Analyzer presents a gallery of images of candidate CTCs present in a sample. To qualify as a CTC, an object must be round or oval, have a nucleus (as determined by positive 4',6-diamidino-2-phenylindole staining) contained within the cytoplasm (as determined by positive cytokeratin 8,18,19–phycoerythrin staining), and lack expression of CD45 (as determined by negative CD45-allophycocyanin staining). Results are always expressed as the number of cells per 7.5 mL of whole blood.17,19 Cell size ranges from 4 μm to more than 30 μm, and large heterogeneity in morphology is observed.17
Statistical Analysis
Fisher’s exact test was used to test for statistically significant differences in the proportions of patients with less than five or ≥ five CTCs at the blood draw time points of baseline, first follow-up visit (3 to 4 weeks), and first restaging visit after baseline (9 to 12 weeks). Time to disease progression or patient death was defined as the time elapsed between the date of the baseline blood draw and the date of clinical progression or death or, if neither progression nor death was observed during the follow-up period, the date of the last follow-up visit. Kaplan-Meier survival plots were generated based on CTC levels at baseline and follow-up blood collections, and the curves were compared using log-rank testing. Cox proportional hazards regression was used to determine univariate and multivariate hazard ratios for selected potential predictors of progression-free survival (PFS) and overall survival (OS). The cutoff of five CTCs to distinguish patients with favorable and unfavorable prognoses has been described previously.16 P < .05 was considered significant.
RESULTS
Prevalence of CTCs and Response to Therapy As Assessed by Imaging
We have previously reported that detection of circulating epithelial cells using CellSearch (Veridex) is rare in healthy women (mean ± SD, 0.1 ± 0.2 per 7.5 mL of blood) and patients with benign disease (mean ± SD, 0.1 ± 0.9 per 7.5 mL of blood).16 Patients who were started on a chemotherapy regimen (alone or in combination, n = 59) as their first-line treatment had a higher percentage of CTCs detected at baseline compared with patients receiving hormone therapy (alone or in combination, n = 24; P = .015; Table 2). Patients who were started on chemotherapy alone (n = 37) as their first-line treatment had a higher percentage of CTCs detected at baseline compared with patients receiving hormone therapy alone (n = 23), although this difference was not statistically significant (P = .064). There was no significant difference in CTC levels at any time point between patients with or without visceral metastasis (P ≥ .371) or between patients receiving hormone therapy alone (n = 23) or combined trastuzumab therapy (n = 17; P ≥ .108). However, patients who were started on chemotherapy alone (n = 37) had a higher percentage of CTCs detected at the first follow-up and first imaging visit blood draws compared with patients receiving combined trastuzumab therapy (n = 17; P ≤ .064).
Patients treated with chemotherapy (alone or in combination, n = 59) showed a significant decrease in CTCs from baseline to the first follow-up (61% to 27%, P = .003) and first imaging visit blood draws (61% to 24%, P = .004). Patients treated with hormone therapy (alone or in combination with trastuzumab therapy, n = 24) also showed a significant decrease in CTCs from baseline to the first follow-up (29% to 19%, P = .018) and first imaging visit blood draws (29% to 17%, P = .003). Patients treated with chemotherapy alone (n = 37) showed a significant decrease in CTCs from baseline to the first follow-up (57% to 37%, P = .001) and first imaging visit blood draws (57% to 31%, P = .004). The largest decrease in CTCs from baseline to the first follow-up (59% to 7%, P = .600) and first imaging visit blood draws (59% to 0%, P = not applicable) was observed in patients receiving trastuzumab-combined regimens.
CTCs and Imaging to Assess Response to Therapy
Four of the 83 patients died before the first follow-up imaging visit; all of these patients had detectable CTCs, and their counts were 11, 24, 456, and 4,648. Objective evaluation of response to therapy was obtained by a centralized review of the imaging studies for 70 of the 79 patients with a follow-up disease status assessment (Table 3). The response to therapy for the other nine patients was determined by the clinical site. At the first follow-up imaging visit, 20 (25%) of the 79 patients were classified as having a partial response, with 11 of these patients (55%) having ≥ five CTCs before the initiation of therapy but only one patient (5%) having ≥ five CTCs at the first follow-up imaging visit. In contrast, of the 20 patients (25%) classified as having progressive disease, 14 patients (70%) had ≥ five CTCs before the initiation of therapy, and similar results were observed at the first follow-up blood draw and imaging visit. The remaining 39 patients (49%) were classified as having stable disease; 14 of them (36%) had ≥ five CTCs at baseline, and only three patients (8%) had persistent CTCs at restaging, possibly suggesting a therapeutic benefit despite the lack of criteria for objective remission.
At the first follow-up blood draw, the median PFS for patients with less than five CTCs was 9.5 months v 2.1 months for patients with ≥ five CTCs (P = .0057), and the median OS for patients with less than five CTCs was more than 18 months v 11.1 months for patients with ≥ five CTCs (P = .0012; Fig 1). At the first follow-up imaging visit, the median PFS for patients with less than five CTCs was 8.9 months v 1.8 months for patients with ≥ five CTCs (P = .0001), and the median OS for patients with less than five CTCs was 18 months v 11.1 months for patients with ≥ five CTCs (P = .0001; Fig 1). In summary, detection of CTCs 3 to 4 weeks after the initiation of therapy predicts treatment efficacy as determined by traditional imaging modalities 5 to 6 weeks later. The strong correlation between the presence of CTCs and poor prospects for survival suggests that the prognostic value of CTCs may be more relevant than the information obtained by imaging.
Prediction of PFS and OS Before Initiation of Therapy
In all 83 patients, the median PFS time was approximately 7.2 months (95% CI, 4.9 to 9.4 months), and the median OS time was more than 18 months. Figure 1 shows the Kaplan-Meier plots for prediction of PFS and OS using the baseline CTC counts (Figs 1A and 1D, n = 83 patients), the first follow-up CTC counts (Figs 1B and 1E, n = 60 patients), and the CTC counts at restaging (Figs 1C and 1F, n = 79 patients). Forty-three of the patients (52%) had ≥ five CTCs per 7.5 mL of blood at baseline. These patients had a significantly shorter median PFS (~ 4.9 months; 95% CI, 2.0 to 8.1 months) and median OS (~ 14.2 months; 95% CI, 11.1 to > 18 months) compared with patients with less than five CTCs per 7.5 mL of blood (median PFS, ~ 9.5 months; 95% CI, 6.1 to > 18 months; P = .0014; median OS, > 18 months; P < .0048). We further analyzed PFS and OS by patient characteristics (Table 4). Baseline CTC count was almost significantly associated with PFS in patients whose first-line therapy was chemotherapy (alone or with hormone therapy or trastuzumab therapy; P = .0464; Table 4). In patients with visceral disease, hormone receptor–positive tumors, or HER2/neu-negative tumors, a CTC count of less than five CTCs at baseline was significantly associated with longer PFS. In patients who underwent chemotherapy (alone or with hormone therapy or trastuzumab therapy) or who had nonvisceral disease, hormone receptor–positive tumors, or HER2/neu-negative tumors, a CTC count of less than five CTCs at baseline was significantly associated with longer OS.
Prediction of PFS and OS at First Visit After Initiation of Therapy
Sixty patients had blood drawn at the first follow-up visit. Fifteen of these patients (25%) had ≥ five CTCs at this time. These patients had a significantly shorter median PFS compared with patients with less than five CTCs (2.1 months; 95% CI, 1.4 to 9.2 months, v 9.5 months; 95% CI, 6.7 to > 18 months, respectively; P = .0057) and a significantly shorter median OS (11.1 months; 95% CI, 6.2 to > 18 months, v > 18 months, respectively; P = 0.0012; Fig 1B and 1E). Further analyses revealed that a CTC count of less than five CTCs at the first follow-up visit was significantly associated with longer PFS in all subgroups except hormone receptor–positive disease and nonvisceral metastases (Table 4). Despite the large difference in the median PFS times, the lack of significance in the hormone receptor–positive disease group is largely a result of the small number of patients with ≥ five CTCs in this subgroup. Hormone receptor status remained the only factor not associated with improved prognosis in this analysis with regards to OS. Fifteen patients (25%) who demonstrated a reduction of CTCs to less than five showed improvement in their median PFS compared with patients with ≥ five CTCs at the first follow-up visit, although this improvement did not reach statistical significance (9.5 v 2.1 months, respectively; P = .1308; Fig 2).
Prediction of PFS and OS at First Reassessment of Disease Status by Imaging After Initiation of Therapy
Seventy-nine patients had their blood drawn at the restaging visit. Seventeen of these patients (22%) had ≥ five CTCs at this time. These patients had a significantly shorter median PFS from initiation of therapy compared with patients with less than five CTCs (1.8 months; 95% CI, 1.2 to 2.2 months, v 8.9 months; 95% CI, 6.1 to 11.9, respectively; P = .0001) and a significantly shorter median OS from initiation of therapy (11.1 months; 95% CI, 4.6 to > 18 months, v > 18 months, respectively; P = .0001; Figs 1C and 1F). Further analyses showed that a CTC count of less than five CTCs at the restaging visit was significantly associated with longer PFS in all patients except those who received hormone therapy (alone or with trastuzumab) or those who had nonvisceral disease (Table 4). Furthermore, except for patients who received hormone therapy (alone or with trastuzumab therapy), a CTC count of less than five CTCs at the restaging visit was significantly associated with longer OS in all patients. Twenty-three patients (29%) who demonstrated a reduction of CTCs to below five showed improvement in PFS compared with patients with ≥ five CTCs at the restaging visit (8.1 v 1.8 months, respectively; P = .0332; Fig 2).
Univariate and Multivariate Analysis of Predictors of PFS and OS
Univariate and multivariate Cox proportional hazards regression was performed to assess the association between factors of interest and PFS or OS. In univariate analysis, HER2/neu status, type of first-line therapy, and CTC levels at baseline, at first follow-up visit, and at restaging visit predicted PFS (Table 5). Hormone receptor status, type of first-line therapy, and CTC levels at baseline, at first follow-up visit, and at restaging visit also predicted OS.
We used stepwise Cox regression analysis at a stringency level of P < .05 to both include and exclude factors for the combination of the baseline CTC count, first follow-up CTC count, or restaging CTC count with the other factors to predict PFS and OS. Several of the clinical factors maintained their association in the multivariate analysis (Table 6). CTC level was the strongest predictor.
DISCUSSION
The data reported in two subsets of patients included in this analysis also indicate the need for future investigations using CTCs. First, the detection of CTCs at baseline was of prognostic significance in patients with hormone receptor–positive disease and without evidence of visceral disease (worse PFS and OS, respectively), which suggests that detection of CTCs is more than simply a marker of tumor burden. More than two thirds of breast cancer tumors express the estrogen receptor, and the majority of these tumors was estrogen dependent for growth.20 Several endocrine modalities are available for the management of hormone receptor–positive disease, in particular for postmenopausal women.21,22 Despite the superior efficacy demonstrated with aromatase inhibitors in postmenopausal women, the management of hormone-sensitive MBC remains disappointing, with an overall response rate of approximately 20% to 40% and, more importantly, a time to progression of approximately 9 to 11 months.21 For these patients, who are candidates for various endocrine treatments based on the expression of their hormone receptor status in the primary or metastatic tumor, there is no known prognostic marker that could be used for risk stratification and treatment selection. Furthermore, patients with hormone receptor–positive disease but with visceral disease sometimes are selected for treatment with cytotoxic agents based on the assumption of possible aggressive malignancy. The detection of CTCs at time of diagnosis may help predict which patients would most benefit from earlier initiation of chemotherapy.
Second, in this study, patients with HER2/neu-positive tumors who were treated with trastuzumab-based regimens had the highest proportional reduction in CTC levels at first follow-up and at the time of restaging, confirming the recognized efficacy of this treatment modality. These data indicate that the use of this technology during follow-up of patients with MBC receiving systemic therapy can provide extremely useful information by identifying two groups of patients. The first group is constituted of women who have a high probability of obtaining prolonged benefit from their treatments (disappearance of CTCs), and the other group is represented by patients in whom the treatment may produce a symptomatic improvement and/or an objective remission (or even stability of disease by imaging) but will not be associated with a significant prognostic improvement (persistent detection of ≥ five CTCs). On the contrary, patients with stable disease and disappearance of CTCs (approximately 25% of these patients) may still benefit from continuation of their palliative treatment, despite lacking criteria for objective remission.
In summary, this analysis demonstrates that the detection of CTCs in patients with MBC about to start first-line treatment is associated with significant prognostic information. Furthermore, the persistence of CTCs at 3 to 4 weeks after the treatment is started and at the time of restaging continues to be significantly associated with prognosis, particularly in women with hormone receptor–negative disease and women who are receiving chemotherapy. These data suggest the value of this technology in the identification of chemotherapy-resistant patients who could benefit from early treatment change and/or more investigational approaches.
Authors’ Disclosures of Potential Conflicts of Interest
NOTES
Supported by Immunicon Corporation, Huntingdon Valley, PA.
Authors’ disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Ellis M, Hayes DF, Lippman ME: Treatment of metastatic disease, in Harris J, Lippman M, Morrow M, et al (eds). Diseases of the Breast (ed 2). Philadelphia, PA, Lippincott-Raven, 2000, pp 749-799
2. Greenberg PA, Hortobagyi GN, Smith TL, et al: Long-term follow-up of patients with complete remission following combination chemotherapy for metastatic breast cancer. J Clin Oncol 14:2197-2205, 1996
3. Nieto Y, Cagnoni PJ, Shpall EJ, et al: Phase II trial of high-dose chemotherapy with autologous stem cell transplant for stage IV breast cancer with minimal metastatic disease. Clin Cancer Res 5:1731-1737, 1999
4. Rivera E, Holmes FA, Buzdar AU, et al: Fluorouracil, doxorubicin, and cyclophosphamide followed by tamoxifen as adjuvant treatment for patients with stage IV breast cancer who have no evidence of disease. Breast J 8:2-9, 2002
5. Swenerton KD, Legha SS, Smith L, et al: Prognostic factors in metastatic breast cancer treated with combination chemotherapy. Cancer Res 39:1552-1562, 1979
6. Hortobagyi GN, Smith TL, Legha SS, et al: Multivariate analysis of prognostic factors in metastatic breast cancer. J Clin Oncol 1:776-786, 1983
7. Beitsch PD, Clifford E: Detection of carcinoma cells in the blood of breast cancer patients. Am J Surg 180:446-449, 2000
8. Fehm T, Sagalowsky A, Clifford E, et al: Cytogenetic evidence that circulating epithelial cells in patients with carcinoma are malignant. Clin Cancer Res 8:2073-2084, 2002
9. Gaforio JJ, Serrano MJ, Sanchez-Rovira P, et al: Detection of breast cancer cells in the peripheral blood is positively correlated with estrogen-receptor status and predicts poor prognosis. Int J Cancer 107:984-990, 2003
10. Austrup F, Uciechowski P, Eder C, et al: Prognostic value of genomic alterations in minimal residual cancer cells purified from the blood of breast cancer patients. Br J Cancer 83:1664-1673, 2000
11. Terstappen LWMM, Rao C, Gross S, et al: Peripheral blood tumor cell load reflects the clinical activity of the disease in patients with carcinoma of the breast. Int J Oncol 17:573-578, 2000
12. Ashworth TR: A case of cancer in which cells similar to those in the tumors were seen in the blood after death. Aust Med J 14:146-147, 1869
13. Carey RW, Taft PD, Bennett JM, et al: Carcinocythemia (carcinoma cell leukemia): An acute leukemia-like picture due to metastatic carcinoma cells. Am J Med 60:273-278, 1976
14. Myerowitz RL, Edwards PA, Sartiano GP: Carcinocythemia (carcinoma cell leukemia) due to metastatic carcinoma of the breast: Report of a case. Cancer 40:3107-3111, 1977
15. Gallivan MV, Lokich JJ: Carcinocythemia (carcinoma cell leukemia): Report of two cases with English literature review. Cancer 53:1100-1102, 1984
16. Cristofanilli M, Budd GT, Ellis M, et al: Circulating tumor cells predict progression free survival and overall survival in metastatic breast cancer. N Engl J Med 351:781-791, 2004
17. Allard WJ, Matera J, Miller MC, et al: Tumor cells circulate in the peripheral blood of all major carcinomas but not in healthy subjects or patients with non-malignant diseases. Clin Cancer Res 15:6897-6904, 2004
18. Momburg F, Moldenhauer G, Hammerling GJ, et al: Immunohistochemical study of the expression of a Mr 34,000 human epithelium specific surface glycoprotein in normal and malignant tissues. Cancer Res 47:2883-2891, 1987
19. Kagan M, Howard D, Bendele T, et al: Circulating tumor cells as cancer markers, a sample preparation and analysis system, in Diamandis EP, Fritsche HA, Lilja H, et al (eds): Tumor Markers: Physiology, Pathobiology, Technology and Clinical Applications. Washington, DC, American Association for Clinical Chemistry Press, 2002, pp 495-498
20. Buzdar A, Vergote I, Sainsbury R: The impact of hormone receptor status on the clinical efficacy of the new-generation aromatase inhibitors: A review of data from first-line metastatic disease trials in postmenopausal women. Breast J 10:211-217, 2004
21. Bonneterre J, Thurlimann B, Robertson JF, et al: Anastrozole versus tamoxifen as first-line therapy for advanced breast cancer in 668 postmenopausal women: Results of the Tamoxifen or Arimidex Randomized Group Efficacy and Tolerability study. J Clin Oncol 18:3748-3757, 2000
22. Mouridsen H, Gershanovich M, Sun Y, et al: Phase III study of letrozole versus tamoxifen as first-line therapy of advanced breast cancer in postmenopausal women: Analysis of survival and update of efficacy from the International Letrozole Breast Cancer Group. J Clin Oncol 21:2101-2109, 2003(Massimo Cristofanilli, Da)
The University of Michigan Comprehensive Cancer Center, Ann Arbor, MI
The Cleveland Clinic, Cleveland, OH
Washington University, St Louis, MO
University of Arizona, Phoenix, AZ
Immunicon Corporation, Huntingdon Valley, PA
ABSTRACT
PATIENTS AND METHODS: One hundred seventy-seven patients with measurable MBC were enrolled onto a prospective study. Eighty-three of the 177 patients were entering first-line treatment, and these patients are the focus of this analysis. CTCs from 7.5 mL of whole blood drawn before treatment initiation (baseline) and monthly thereafter for up to 6 months were isolated and enumerated using immunomagnetics.
RESULTS: The mean (± standard deviation) follow-up time was 11.1 ± 4.4 months (median, 12.2 months). Forty-three patients (52%) had ≥ five CTCs at baseline. The median PFS was 7.2 months (95% CI, 4.9 to 9.4 months), and the median OS was more than 18 months. Patients with ≥ five CTCs at baseline and at first follow-up (4 weeks) had a worse prognosis than patients with less than five CTCs (baseline: median PFS, 4.9 v 9.5 months, respectively; log-rank, P = .0014; median OS, 14.2 v > 18 months, respectively; log-rank, P = .0048; first follow-up: median PFS, 2.1 v 8.9 months, respectively; log-rank, P = .0070; median OS, 11.1 v > 18 months, respectively; log-rank, P = .0029). CTCs before and after the initiation of therapy were strong, independent prognostic factors.
CONCLUSION: Detection of CTCs before initiation of first-line therapy in patients with MBC is highly predictive of PFS and OS. This technology can aid in appropriate patient stratification and design of tailored treatments.
INTRODUCTION
Despite years of clinical research, the odds of achieving complete response for patients with metastatic breast cancer (MBC) are extremely low.1,2 Only a few patients who achieve a complete response after chemotherapy remain in this state for prolonged periods of time, with some patients remaining in remission beyond 20 years.2 These long-term survivors are usually young, have an excellent performance status, and, more importantly, have limited metastatic disease.3,4 The majority of patients with metastatic disease respond transiently to conventional treatments and develop evidence of progressive disease within 12 to 24 months of initiating treatment.1 For these patients, systemic treatment has not translated into a significant improvement in survival but has substantially improved their quality of life. Several clinical factors have been proposed but never prospectively validated that would help in the prediction of long-term outcome and efficacy of treatments.5,6
Circulating tumor cells (CTCs) can be detected in blood from patients with metastatic and primary carcinomas.7-11 Over the past several years, the development of immunomagnetic platforms has permitted accurate enumeration of CTCs at extremely low frequencies.11 In several case reports, the presence of CTCs has been associated with shortened survival times.12-15 We conducted a prospective, multicenter, double-blind study to determine the clinical significance of CTCs in patients with measurable MBC starting a new course of systemic therapy. The overall results of this study have been published elsewhere.16 This article represents the detailed analysis of only the cohort of patients with newly diagnosed MBC about to start first-line therapy. We believe this is an important analysis because patients who are newly diagnosed and about to start first-line therapy will derive the most benefit from appropriate risk stratification and treatment planning. Furthermore, an early prediction of treatment efficacy could have an impact in their quality of life.
PATIENTS AND METHODS
Before the initiation of therapy, patients had imaging evaluation (including computed tomography scans of chest and abdomen) of their metastatic sites and a baseline blood draw for enumeration of CTCs. Serial blood specimens were collected at roughly monthly intervals for a period of up to 6 months. Reassessments of disease status by the same modalities used at baseline were conducted every 9 to 12 weeks, depending on treatment type and schedule. Disease status was assessed using WHO criteria without knowledge of the patients’ CTC results.
Isolation and Enumeration of CTCs
Blood samples were drawn into 10-mL EDTA Vacutainer tubes (Becton Dickinson, Franklin Lakes, NJ), to which a cell preservative was subsequently added. Samples were maintained at room temperature and processed within 72 hours after collection. All CTC evaluations were performed without knowledge of the patients’ clinical status in one of two central laboratories (Immunicon, Huntingdon Valley, PA, or IMPATH Predictive Oncology, Los Angeles, CA) or at selected participating centers. CTCs were immunomagnetically enriched from 7.5 mL of blood using ferrofluids coated with antibodies targeting the epithelial cell adhesion molecule.17,18 Isolated cells were fluorescently labeled with the nucleic acid dye 4',6-diamidino-2-phenylindole (DAPI) and monoclonal antibodies specific for leukocytes (CD45-allophycocyanin) and epithelial cells (cytokeratin 8,18,19–phycoerythrin) and analyzed by the CellSpotter Analyzer17,19 (Veridex LLC, Warren, NJ).
CTC Definition
The CellSpotter Analyzer presents a gallery of images of candidate CTCs present in a sample. To qualify as a CTC, an object must be round or oval, have a nucleus (as determined by positive 4',6-diamidino-2-phenylindole staining) contained within the cytoplasm (as determined by positive cytokeratin 8,18,19–phycoerythrin staining), and lack expression of CD45 (as determined by negative CD45-allophycocyanin staining). Results are always expressed as the number of cells per 7.5 mL of whole blood.17,19 Cell size ranges from 4 μm to more than 30 μm, and large heterogeneity in morphology is observed.17
Statistical Analysis
Fisher’s exact test was used to test for statistically significant differences in the proportions of patients with less than five or ≥ five CTCs at the blood draw time points of baseline, first follow-up visit (3 to 4 weeks), and first restaging visit after baseline (9 to 12 weeks). Time to disease progression or patient death was defined as the time elapsed between the date of the baseline blood draw and the date of clinical progression or death or, if neither progression nor death was observed during the follow-up period, the date of the last follow-up visit. Kaplan-Meier survival plots were generated based on CTC levels at baseline and follow-up blood collections, and the curves were compared using log-rank testing. Cox proportional hazards regression was used to determine univariate and multivariate hazard ratios for selected potential predictors of progression-free survival (PFS) and overall survival (OS). The cutoff of five CTCs to distinguish patients with favorable and unfavorable prognoses has been described previously.16 P < .05 was considered significant.
RESULTS
Prevalence of CTCs and Response to Therapy As Assessed by Imaging
We have previously reported that detection of circulating epithelial cells using CellSearch (Veridex) is rare in healthy women (mean ± SD, 0.1 ± 0.2 per 7.5 mL of blood) and patients with benign disease (mean ± SD, 0.1 ± 0.9 per 7.5 mL of blood).16 Patients who were started on a chemotherapy regimen (alone or in combination, n = 59) as their first-line treatment had a higher percentage of CTCs detected at baseline compared with patients receiving hormone therapy (alone or in combination, n = 24; P = .015; Table 2). Patients who were started on chemotherapy alone (n = 37) as their first-line treatment had a higher percentage of CTCs detected at baseline compared with patients receiving hormone therapy alone (n = 23), although this difference was not statistically significant (P = .064). There was no significant difference in CTC levels at any time point between patients with or without visceral metastasis (P ≥ .371) or between patients receiving hormone therapy alone (n = 23) or combined trastuzumab therapy (n = 17; P ≥ .108). However, patients who were started on chemotherapy alone (n = 37) had a higher percentage of CTCs detected at the first follow-up and first imaging visit blood draws compared with patients receiving combined trastuzumab therapy (n = 17; P ≤ .064).
Patients treated with chemotherapy (alone or in combination, n = 59) showed a significant decrease in CTCs from baseline to the first follow-up (61% to 27%, P = .003) and first imaging visit blood draws (61% to 24%, P = .004). Patients treated with hormone therapy (alone or in combination with trastuzumab therapy, n = 24) also showed a significant decrease in CTCs from baseline to the first follow-up (29% to 19%, P = .018) and first imaging visit blood draws (29% to 17%, P = .003). Patients treated with chemotherapy alone (n = 37) showed a significant decrease in CTCs from baseline to the first follow-up (57% to 37%, P = .001) and first imaging visit blood draws (57% to 31%, P = .004). The largest decrease in CTCs from baseline to the first follow-up (59% to 7%, P = .600) and first imaging visit blood draws (59% to 0%, P = not applicable) was observed in patients receiving trastuzumab-combined regimens.
CTCs and Imaging to Assess Response to Therapy
Four of the 83 patients died before the first follow-up imaging visit; all of these patients had detectable CTCs, and their counts were 11, 24, 456, and 4,648. Objective evaluation of response to therapy was obtained by a centralized review of the imaging studies for 70 of the 79 patients with a follow-up disease status assessment (Table 3). The response to therapy for the other nine patients was determined by the clinical site. At the first follow-up imaging visit, 20 (25%) of the 79 patients were classified as having a partial response, with 11 of these patients (55%) having ≥ five CTCs before the initiation of therapy but only one patient (5%) having ≥ five CTCs at the first follow-up imaging visit. In contrast, of the 20 patients (25%) classified as having progressive disease, 14 patients (70%) had ≥ five CTCs before the initiation of therapy, and similar results were observed at the first follow-up blood draw and imaging visit. The remaining 39 patients (49%) were classified as having stable disease; 14 of them (36%) had ≥ five CTCs at baseline, and only three patients (8%) had persistent CTCs at restaging, possibly suggesting a therapeutic benefit despite the lack of criteria for objective remission.
At the first follow-up blood draw, the median PFS for patients with less than five CTCs was 9.5 months v 2.1 months for patients with ≥ five CTCs (P = .0057), and the median OS for patients with less than five CTCs was more than 18 months v 11.1 months for patients with ≥ five CTCs (P = .0012; Fig 1). At the first follow-up imaging visit, the median PFS for patients with less than five CTCs was 8.9 months v 1.8 months for patients with ≥ five CTCs (P = .0001), and the median OS for patients with less than five CTCs was 18 months v 11.1 months for patients with ≥ five CTCs (P = .0001; Fig 1). In summary, detection of CTCs 3 to 4 weeks after the initiation of therapy predicts treatment efficacy as determined by traditional imaging modalities 5 to 6 weeks later. The strong correlation between the presence of CTCs and poor prospects for survival suggests that the prognostic value of CTCs may be more relevant than the information obtained by imaging.
Prediction of PFS and OS Before Initiation of Therapy
In all 83 patients, the median PFS time was approximately 7.2 months (95% CI, 4.9 to 9.4 months), and the median OS time was more than 18 months. Figure 1 shows the Kaplan-Meier plots for prediction of PFS and OS using the baseline CTC counts (Figs 1A and 1D, n = 83 patients), the first follow-up CTC counts (Figs 1B and 1E, n = 60 patients), and the CTC counts at restaging (Figs 1C and 1F, n = 79 patients). Forty-three of the patients (52%) had ≥ five CTCs per 7.5 mL of blood at baseline. These patients had a significantly shorter median PFS (~ 4.9 months; 95% CI, 2.0 to 8.1 months) and median OS (~ 14.2 months; 95% CI, 11.1 to > 18 months) compared with patients with less than five CTCs per 7.5 mL of blood (median PFS, ~ 9.5 months; 95% CI, 6.1 to > 18 months; P = .0014; median OS, > 18 months; P < .0048). We further analyzed PFS and OS by patient characteristics (Table 4). Baseline CTC count was almost significantly associated with PFS in patients whose first-line therapy was chemotherapy (alone or with hormone therapy or trastuzumab therapy; P = .0464; Table 4). In patients with visceral disease, hormone receptor–positive tumors, or HER2/neu-negative tumors, a CTC count of less than five CTCs at baseline was significantly associated with longer PFS. In patients who underwent chemotherapy (alone or with hormone therapy or trastuzumab therapy) or who had nonvisceral disease, hormone receptor–positive tumors, or HER2/neu-negative tumors, a CTC count of less than five CTCs at baseline was significantly associated with longer OS.
Prediction of PFS and OS at First Visit After Initiation of Therapy
Sixty patients had blood drawn at the first follow-up visit. Fifteen of these patients (25%) had ≥ five CTCs at this time. These patients had a significantly shorter median PFS compared with patients with less than five CTCs (2.1 months; 95% CI, 1.4 to 9.2 months, v 9.5 months; 95% CI, 6.7 to > 18 months, respectively; P = .0057) and a significantly shorter median OS (11.1 months; 95% CI, 6.2 to > 18 months, v > 18 months, respectively; P = 0.0012; Fig 1B and 1E). Further analyses revealed that a CTC count of less than five CTCs at the first follow-up visit was significantly associated with longer PFS in all subgroups except hormone receptor–positive disease and nonvisceral metastases (Table 4). Despite the large difference in the median PFS times, the lack of significance in the hormone receptor–positive disease group is largely a result of the small number of patients with ≥ five CTCs in this subgroup. Hormone receptor status remained the only factor not associated with improved prognosis in this analysis with regards to OS. Fifteen patients (25%) who demonstrated a reduction of CTCs to less than five showed improvement in their median PFS compared with patients with ≥ five CTCs at the first follow-up visit, although this improvement did not reach statistical significance (9.5 v 2.1 months, respectively; P = .1308; Fig 2).
Prediction of PFS and OS at First Reassessment of Disease Status by Imaging After Initiation of Therapy
Seventy-nine patients had their blood drawn at the restaging visit. Seventeen of these patients (22%) had ≥ five CTCs at this time. These patients had a significantly shorter median PFS from initiation of therapy compared with patients with less than five CTCs (1.8 months; 95% CI, 1.2 to 2.2 months, v 8.9 months; 95% CI, 6.1 to 11.9, respectively; P = .0001) and a significantly shorter median OS from initiation of therapy (11.1 months; 95% CI, 4.6 to > 18 months, v > 18 months, respectively; P = .0001; Figs 1C and 1F). Further analyses showed that a CTC count of less than five CTCs at the restaging visit was significantly associated with longer PFS in all patients except those who received hormone therapy (alone or with trastuzumab) or those who had nonvisceral disease (Table 4). Furthermore, except for patients who received hormone therapy (alone or with trastuzumab therapy), a CTC count of less than five CTCs at the restaging visit was significantly associated with longer OS in all patients. Twenty-three patients (29%) who demonstrated a reduction of CTCs to below five showed improvement in PFS compared with patients with ≥ five CTCs at the restaging visit (8.1 v 1.8 months, respectively; P = .0332; Fig 2).
Univariate and Multivariate Analysis of Predictors of PFS and OS
Univariate and multivariate Cox proportional hazards regression was performed to assess the association between factors of interest and PFS or OS. In univariate analysis, HER2/neu status, type of first-line therapy, and CTC levels at baseline, at first follow-up visit, and at restaging visit predicted PFS (Table 5). Hormone receptor status, type of first-line therapy, and CTC levels at baseline, at first follow-up visit, and at restaging visit also predicted OS.
We used stepwise Cox regression analysis at a stringency level of P < .05 to both include and exclude factors for the combination of the baseline CTC count, first follow-up CTC count, or restaging CTC count with the other factors to predict PFS and OS. Several of the clinical factors maintained their association in the multivariate analysis (Table 6). CTC level was the strongest predictor.
DISCUSSION
The data reported in two subsets of patients included in this analysis also indicate the need for future investigations using CTCs. First, the detection of CTCs at baseline was of prognostic significance in patients with hormone receptor–positive disease and without evidence of visceral disease (worse PFS and OS, respectively), which suggests that detection of CTCs is more than simply a marker of tumor burden. More than two thirds of breast cancer tumors express the estrogen receptor, and the majority of these tumors was estrogen dependent for growth.20 Several endocrine modalities are available for the management of hormone receptor–positive disease, in particular for postmenopausal women.21,22 Despite the superior efficacy demonstrated with aromatase inhibitors in postmenopausal women, the management of hormone-sensitive MBC remains disappointing, with an overall response rate of approximately 20% to 40% and, more importantly, a time to progression of approximately 9 to 11 months.21 For these patients, who are candidates for various endocrine treatments based on the expression of their hormone receptor status in the primary or metastatic tumor, there is no known prognostic marker that could be used for risk stratification and treatment selection. Furthermore, patients with hormone receptor–positive disease but with visceral disease sometimes are selected for treatment with cytotoxic agents based on the assumption of possible aggressive malignancy. The detection of CTCs at time of diagnosis may help predict which patients would most benefit from earlier initiation of chemotherapy.
Second, in this study, patients with HER2/neu-positive tumors who were treated with trastuzumab-based regimens had the highest proportional reduction in CTC levels at first follow-up and at the time of restaging, confirming the recognized efficacy of this treatment modality. These data indicate that the use of this technology during follow-up of patients with MBC receiving systemic therapy can provide extremely useful information by identifying two groups of patients. The first group is constituted of women who have a high probability of obtaining prolonged benefit from their treatments (disappearance of CTCs), and the other group is represented by patients in whom the treatment may produce a symptomatic improvement and/or an objective remission (or even stability of disease by imaging) but will not be associated with a significant prognostic improvement (persistent detection of ≥ five CTCs). On the contrary, patients with stable disease and disappearance of CTCs (approximately 25% of these patients) may still benefit from continuation of their palliative treatment, despite lacking criteria for objective remission.
In summary, this analysis demonstrates that the detection of CTCs in patients with MBC about to start first-line treatment is associated with significant prognostic information. Furthermore, the persistence of CTCs at 3 to 4 weeks after the treatment is started and at the time of restaging continues to be significantly associated with prognosis, particularly in women with hormone receptor–negative disease and women who are receiving chemotherapy. These data suggest the value of this technology in the identification of chemotherapy-resistant patients who could benefit from early treatment change and/or more investigational approaches.
Authors’ Disclosures of Potential Conflicts of Interest
NOTES
Supported by Immunicon Corporation, Huntingdon Valley, PA.
Authors’ disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Ellis M, Hayes DF, Lippman ME: Treatment of metastatic disease, in Harris J, Lippman M, Morrow M, et al (eds). Diseases of the Breast (ed 2). Philadelphia, PA, Lippincott-Raven, 2000, pp 749-799
2. Greenberg PA, Hortobagyi GN, Smith TL, et al: Long-term follow-up of patients with complete remission following combination chemotherapy for metastatic breast cancer. J Clin Oncol 14:2197-2205, 1996
3. Nieto Y, Cagnoni PJ, Shpall EJ, et al: Phase II trial of high-dose chemotherapy with autologous stem cell transplant for stage IV breast cancer with minimal metastatic disease. Clin Cancer Res 5:1731-1737, 1999
4. Rivera E, Holmes FA, Buzdar AU, et al: Fluorouracil, doxorubicin, and cyclophosphamide followed by tamoxifen as adjuvant treatment for patients with stage IV breast cancer who have no evidence of disease. Breast J 8:2-9, 2002
5. Swenerton KD, Legha SS, Smith L, et al: Prognostic factors in metastatic breast cancer treated with combination chemotherapy. Cancer Res 39:1552-1562, 1979
6. Hortobagyi GN, Smith TL, Legha SS, et al: Multivariate analysis of prognostic factors in metastatic breast cancer. J Clin Oncol 1:776-786, 1983
7. Beitsch PD, Clifford E: Detection of carcinoma cells in the blood of breast cancer patients. Am J Surg 180:446-449, 2000
8. Fehm T, Sagalowsky A, Clifford E, et al: Cytogenetic evidence that circulating epithelial cells in patients with carcinoma are malignant. Clin Cancer Res 8:2073-2084, 2002
9. Gaforio JJ, Serrano MJ, Sanchez-Rovira P, et al: Detection of breast cancer cells in the peripheral blood is positively correlated with estrogen-receptor status and predicts poor prognosis. Int J Cancer 107:984-990, 2003
10. Austrup F, Uciechowski P, Eder C, et al: Prognostic value of genomic alterations in minimal residual cancer cells purified from the blood of breast cancer patients. Br J Cancer 83:1664-1673, 2000
11. Terstappen LWMM, Rao C, Gross S, et al: Peripheral blood tumor cell load reflects the clinical activity of the disease in patients with carcinoma of the breast. Int J Oncol 17:573-578, 2000
12. Ashworth TR: A case of cancer in which cells similar to those in the tumors were seen in the blood after death. Aust Med J 14:146-147, 1869
13. Carey RW, Taft PD, Bennett JM, et al: Carcinocythemia (carcinoma cell leukemia): An acute leukemia-like picture due to metastatic carcinoma cells. Am J Med 60:273-278, 1976
14. Myerowitz RL, Edwards PA, Sartiano GP: Carcinocythemia (carcinoma cell leukemia) due to metastatic carcinoma of the breast: Report of a case. Cancer 40:3107-3111, 1977
15. Gallivan MV, Lokich JJ: Carcinocythemia (carcinoma cell leukemia): Report of two cases with English literature review. Cancer 53:1100-1102, 1984
16. Cristofanilli M, Budd GT, Ellis M, et al: Circulating tumor cells predict progression free survival and overall survival in metastatic breast cancer. N Engl J Med 351:781-791, 2004
17. Allard WJ, Matera J, Miller MC, et al: Tumor cells circulate in the peripheral blood of all major carcinomas but not in healthy subjects or patients with non-malignant diseases. Clin Cancer Res 15:6897-6904, 2004
18. Momburg F, Moldenhauer G, Hammerling GJ, et al: Immunohistochemical study of the expression of a Mr 34,000 human epithelium specific surface glycoprotein in normal and malignant tissues. Cancer Res 47:2883-2891, 1987
19. Kagan M, Howard D, Bendele T, et al: Circulating tumor cells as cancer markers, a sample preparation and analysis system, in Diamandis EP, Fritsche HA, Lilja H, et al (eds): Tumor Markers: Physiology, Pathobiology, Technology and Clinical Applications. Washington, DC, American Association for Clinical Chemistry Press, 2002, pp 495-498
20. Buzdar A, Vergote I, Sainsbury R: The impact of hormone receptor status on the clinical efficacy of the new-generation aromatase inhibitors: A review of data from first-line metastatic disease trials in postmenopausal women. Breast J 10:211-217, 2004
21. Bonneterre J, Thurlimann B, Robertson JF, et al: Anastrozole versus tamoxifen as first-line therapy for advanced breast cancer in 668 postmenopausal women: Results of the Tamoxifen or Arimidex Randomized Group Efficacy and Tolerability study. J Clin Oncol 18:3748-3757, 2000
22. Mouridsen H, Gershanovich M, Sun Y, et al: Phase III study of letrozole versus tamoxifen as first-line therapy of advanced breast cancer in postmenopausal women: Analysis of survival and update of efficacy from the International Letrozole Breast Cancer Group. J Clin Oncol 21:2101-2109, 2003(Massimo Cristofanilli, Da)