Breast Cancer Research and Treatment

, Volume 107, Issue 1, pp 133–138

Circulating endothelial progenitor cells correlate to stage in patients with invasive breast cancer

Authors

  • Rakhi P. Naik
    • Division of Hematology/Oncology, Department of Medicine, New York Presbyterian HospitalWeill Medical College of Cornell University
  • David Jin
    • Department of Genetic Medicine, New York Presbyterian HospitalWeill Medical College of Cornell University
  • Ellen Chuang
    • Division of Hematology/Oncology, Department of Medicine, New York Presbyterian HospitalWeill Medical College of Cornell University
  • Ellen G. Gold
    • Division of Hematology/Oncology, Department of Medicine, New York Presbyterian HospitalWeill Medical College of Cornell University
  • Eleni A. Tousimis
    • Department of Surgery, New York Presbyterian HospitalWeill Medical College of Cornell University
  • Anne L. Moore
    • Division of Hematology/Oncology, Department of Medicine, New York Presbyterian HospitalWeill Medical College of Cornell University
  • Paul J. Christos
    • Department of Public Health, New York Presbyterian HospitalWeill Medical College of Cornell University
  • Tatiana de Dalmas
    • Division of Hematology/Oncology, Department of Medicine, New York Presbyterian HospitalWeill Medical College of Cornell University
  • Diana Donovan
    • Division of Hematology/Oncology, Department of Medicine, New York Presbyterian HospitalWeill Medical College of Cornell University
  • Shahin Rafii
    • Department of Genetic Medicine, New York Presbyterian HospitalWeill Medical College of Cornell University
    • Division of Hematology/Oncology, Department of Medicine, New York Presbyterian HospitalWeill Medical College of Cornell University
Preclinical Study/Clinical Trial/Epidemiology/Invited Commentary

DOI: 10.1007/s10549-007-9519-6

Cite this article as:
Naik, R.P., Jin, D., Chuang, E. et al. Breast Cancer Res Treat (2008) 107: 133. doi:10.1007/s10549-007-9519-6

Abstract

Tumor growth and metastasis is dependent on the formation and assembly of new blood vessels, a process known as neo-angiogenesis. Both pre-existing and circulating vascular cells have been shown to contribute to the assembly of tumor neo-vessels in specific tumors. Mobilization of endothelial progenitor cells (EPCs) from the bone marrow constitutes a crucial step in the formation of de novo blood vessels, and levels of peripheral blood EPCs have been shown to be increased in certain malignant states. However, the role of circulating EPCs in breast cancer is largely unknown. We recruited twenty-five patients with biopsy-proven invasive breast cancer at Weill Cornell Breast Center to participate in a pilot study investigating the correlation of circulating EPCs to extent of disease and initiation of chemotherapy. For each patient, a baseline sample was drawn before systemic treatment, and for seventeen of those patients, a second sample was taken after the first round of chemotherapy. Levels of peripheral blood EPCs, as defined by co-expression of CD133 and VEGFR2, were quantified by flow cytometry. Breast cancer patients with stage III & IV disease had statistically higher levels of circulating EPCs than did patients with stage I & II disease (median = 165,000 EPCs/5 × 106MNCs vs. median = 6,920 EPCs/5 × 106MNCs, respectively, P < 0.0001). In addition, in late-stage patients, levels of EPCs demonstrated a statistically significant drop after initiation of chemotherapy (median = 162,500 EPCs/5 × 106MNCs [pre] vs. median = 117,500 EPCs/5 × 106MNCs [post], P = 0.01). These results suggest that circulating EPCs may serve as a potential tumor biomarker in breast cancer and that EPCs may represent a plausible target for future therapeutic intervention.

Keywords

AngiogenesisCirculating endothelial progenitor cellsBreast cancer

Introduction

The importance of angiogenesis in solid tumor progression is undisputed. Neoplastic survival depends on an extensive vascular network to provide a continuous oxygen supply and toxic waste removal [1]. As a result, microvessel recruitment is essential for tumor maintenance and growth beyond 2 mm3 in size [2, 3]. In light of the success of monoclonal antibody treatments such as human epidermal growth factor receptor-2 (HER-2) and anti-angiogenic therapy such as anti-VEGF-A in advanced breast malignancies, the identification of novel targets against neoangiogenesis in breast cancer has elicited particular interest. Several lines of evidence suggest that direct mobilization of endothelial progenitor cells (EPCs) from the bone marrow constitutes a vital step in the formation of de novo blood vessels in malignant states and that inhibition of EPC recruitment in neoplastic conditions has been found to effectively impair growth and progression of tumors [47]. In this regard, EPCs holds promising therapeutic potential in breast cancer patients and offers a potential predictive indicator of tumor growth and progression.

Several recent studies have found that EPCs can be detected and quantified in peripheral blood samples. Generally, these circulating EPCs are identified by phenotypic co-expression of CD133 (AC133), a cell surface marker unique to hematopoetic stem cells, and vascular endothelial growth factor receptor-2 (VEGFR2/KDR), an endothelial cell marker [811]. Using this definition, increased levels of circulating EPCs have been observed in select malignant diseases such as multiple myeloma and myelofibrosis with myeloid metaplasia [1216]. Given the potential predictive and therapeutic value of EPCs in breast cancer, we designed a pilot study to investigate baseline and post-treatment levels of EPCs in patients with breast malignancy. We hypothesized that the level of circulating EPCs in breast cancer patients would correlate to extent of disease and that EPC levels would decrease after initiation of systemic therapy.

Patients and methods

Twenty-five women with pathologically confirmed invasive breast cancer at the Weill Cornell Breast Center were recruited for participation in this study. All but two patients were recruited after undergoing definitive breast surgery for their malignancy. The remaining two patients were scheduled to undergo neo-adjuvant chemotherapy and, therefore, had baseline samples drawn after diagnostic biopsy alone. For each patient, a baseline sample was drawn prior to systemic therapeutic intervention, and for seventeen of those patients, a follow-up sample was obtained after initiation of systemic treatment. Chemotherapy was administered in the neo-adjuvant, adjuvant, or metastatic setting, and all blood samples were drawn in conjunction with routine labs. The study protocol was approved by the New York Presbyterian Hospital-Weill Medical College of Cornell University Institutional Review Board, and informed consent was obtained on all patients.

Peripheral venous blood (10 ml) was collected in EDTA-containing tubes. The peripheral blood mononuclear cell (PBMC) layer was isolated using Ficoll density-gradient centrifugation within 12 h of blood collection. Mononuclear cells were then labeled with FITC-conjugated anti-CD14 (BD Pharmingen, San Jose, CA), PE-conjugated anti-CD14 (BD Pharmingen, San Jose, CA), or PE-conjugated anti-CD133 (Miltenyi Biotec, Auburn, CA) + FITC-conjugated anti-VEGFR2 (BD Pharmingen, San Jose, CA). The frequency of circulating EPCs was determined by measurement of cells exhibiting CD133 and VEGFR2 co-expression after gating for CD14-positive cells. Flow cytometry was performed using a Beckman Coulter (Fullerton, CA) flow cytometer, and data was analyzed using associated software. Results were reported as number of EPCs per 5 million mononuclear cells (MNCs).

Statistical analysis was performed using SPSS Version 13.0 software. To account for small sample size, medians of baseline EPC levels between stage groups were compared using the Kruskal–Wallis and Wilcoxon rank-sum tests. Comparisons between pre- and post-chemotherapy EPC values were made with the Wilcoxon signed rank test. All P values are two-sided with statistical significance evaluated at the 0.05 α level.

Results

Demographics and tumor characteristics of the included patients are summarized in Table 1. The clinical classifications and staging of each patient were assessed using the Revision of the American Joint Committee on Cancer Staging System for Breast Cancer [17]. No statistical difference in patient age, hormone receptor status, or Her2neu status was found among stage groups.
Table 1

Patient and tumor characteristics by stage

Characteristics

Stage I (N = 7)

Stage II (N = 6)

Stage III (N = 5)

Stage IV (N = 7)

Median age (years)

37 (range 31–51)

41 (range 32–53)

63 (range 49–65)

55 (range 47–60)

Tumor size (T)

    T1

7

3

1

N/A

    T2

N/A*

3

1

N/A

    T3

N/A

0

3

N/A

Lymph node status (N)

    N0

7

2

0

N/A

    N1

N/A

4

1

N/A

    N2

N/A

N/A

2

N/A

    Unknown

  

2**

 

Hormone receptor status

    Positive

4

3

3

4

    Negative

3

3

2

3

Her2/neu status

    Positive

0

2

3

1

    Negative

7

4

2

6

    ER/PR/Her2neu negative

3

2

0

1

Sites of metastasis

    1 site

N/A

N/A

N/A

1

    2 sites

N/A

N/A

N/A

3

    >3 sites

N/A

N/A

N/A

3

* Not applicable

** Patients received neo-adjuvant chemotherapy

For stage I–III patients who had undergone definitive breast surgery, median time from surgery to obtainment of baseline sample was 47 days (range 8–348 days). The median baseline circulating EPC level found in stage I to IV patients was 5,830, 8,510, 165,000, and 160,000, respectively (all numbers reported as EPCs/5 × 106MNCs). The median circulating EPC level in stage III and IV breast cancer patients was significantly higher than that of stage I and II patients (165,000 EPCs/5 × 106MNCs vs. 6,920 EPCs/5 × 106MNCs, respectively, P < 0.0001) (Fig. 1). No significant difference was observed between EPC values of stage III patients who had undergone definitive breast surgery verses those who undergone diagnostic biopsy alone. Although no statistical difference between circulating EPC levels for stage I vs. II and stage III vs. IV was detected, all other inter-stage comparisons were significant. No trends were observed between EPC values and age, hormone status, or Her2/neu status (data not shown).
https://static-content.springer.com/image/art%3A10.1007%2Fs10549-007-9519-6/MediaObjects/10549_2007_9519_Fig1_HTML.gif
Fig. 1

Baseline endothelial progenitor cells (EPCs) by stage: Pre-chemotherapy levels of EPC’s by stage in patients with invasive breast cancer. The median EPC level of Stage III (open circles) & stage IV (closed diamonds) patients was statistically higher than that of stage I (closed circles) & stage II (open diamonds) patients (*P < 0.001 by Wilcoxon rank-sum test). Bars represent median EPC levels

All but one of the stage I–III patients for whom a follow-up blood sample was taken were administered a doxorubicin and cyclophosphamide-based regimen. The remaining patient received cyclophosphamide, methotrexate, and 5-FU (CMF). All stage IV patients were administered an ixabepilone-based chemotherapy regimen on an unrelated study protocol. Patients with stage III and IV disease demonstrated a 28% drop in circulating median EPC level after administration of the first round of chemotherapy, which represented a statistically significant difference between pre- and post-chemotherapy values (162,500 EPCs/5 × 106MNCs vs. 117,500 EPCs/5 × 106MNCs, respectively, P = 0.01) (Fig. 2). On the other hand, patients with early-stage breast cancer experienced only a 2% drop in circulating EPC level after chemotherapy initiation, and as such, the difference did not reach statistical significance. Among the stage IV patients (n = 5), those who demonstrated a clear response to chemotherapy as defined by a 30% reduction in measurable disease by CT scan by Response Evaluation Criteria in Solid Tumors (RECIST) [18] showed a substantially greater decrease in EPC level than did those with stable disease (90% decrease vs. 21% decrease, respectively). Further study with a larger sample size will be required to validate this observation.
https://static-content.springer.com/image/art%3A10.1007%2Fs10549-007-9519-6/MediaObjects/10549_2007_9519_Fig2_HTML.gif
Fig. 2

Pre- vs post-chemotherapy by endothelial progenitor cells (EPCs) by stage: Pre- and post-chemotherapy EPC levels by stage in patients with invasive breast cancer. There was a statistically significant difference between median pre- (closed diamonds) and post-chemotherapy (open circles) values in stage III/IV patients (*P = 0.01) but not stage I/II patients (P = 0.23) [by Wilcoxon signed rank test]. Bars represent median EPC levels

Discussion

Our pilot study demonstrates that the absolute levels of circulating EPCs in breast cancer patients appear to correlate with stage of disease and furthermore, that EPC levels decrease with initiation of chemotherapy in late-stage patients. A pilot study from another group suggests the same phenomenon in a locally advanced breast cancer setting receiving cytotoxic chemotherapy [11]. This pilot data set also suggests that levels of circulating EPCs correlate to therapeutic response in breast cancer patients with measurable disease, although a larger study is needed to validate this observation. While the numbers of patients enrolled in this pilot study are small, the magnitude of difference between EPC levels for early and late-stage patients is exceedingly large. Specifically, our study suggests that baseline EPC levels in stage III patients, both before or after mastectomy, are significantly higher than levels in patients with stage II disease. In addition, because the peripheral blood EPC values in stage III patients are comparable to those in stage IV patients, these data could potentially suggest a true biologic difference between early and late stage groups.

Research in mouse models suggests that EPC levels correlate to tumor growth and that chemotherapeutic agents can affect bone marrow mobilization of EPCs [19, 20]. In patients with multiple myeloma, a decrease in peripheral blood EPCs, as defined by colony-forming units (CFUs) was shown to correspond to length of thalidomide treatment [12]. Our observations that levels of EPCs correlate to extent of disease in breast cancer and that circulating levels decreased with initiation of therapy suggest that peripheral blood EPCs may serve as a predictor of overall treatment response and tumor progression.

Circulating EPCs have also been implicated in repair of vascular injury, and recent evidence has demonstrated a correlation between increased EPC levels and adverse cardiovascular outcomes [2123]. Similarly, in the context of multiple myeloma, increased peripheral blood EPCs have been shown to correlate to levels of M protein and B 2-microglobin, two validated markers of disease activity [12]. If EPC levels correspond to the virulence of malignant disease, then perhaps early-stage breast cancer patients with high EPCs levels may warrant aggressive adjuvant intervention to prevent recurrence. In fact, this may represent a subgroup of patients who might benefit from anti-angiogenic therapy concomitant with adjuvant chemotherapy. Recently, genetic signatures of breast cancer tumors have been used to quantify the likelihood of disease recurrence in patients with node-negative, estrogen-positive tumors [24]. Since circulating EPCs may reflect the ability of a tumor to recruit the vascular infrastructure required to grow and metastasize, they may serve as a surrogate marker for disease recurrence and prognosis in a way similar to the evolution of use of tumor signatures.

With the success of the anti-vascular endothelial growth factor (VEGF-A) monoclonal antibody bevacizumab in early trials of colorectal, lung, renal cell and breast carcinoma, angiogenesis as potential therapeutic target in malignancy has proven fruitful [2529]. As a critical step in vasculogenesis by tumors, therefore, VEGFR2+ cells such as endothelial progenitors and mature endothelial cells offer a promising target for therapy. ZD6474, a VEGFR2 tyrosine kinase inhibitor, has undergone phase I trials and is now in the phase II testing [30]. Other promising VEGFR tyrosine kinase inhibitors are also currently in development and have shown marked anti-tumor activity against a variety of solid tumor types in preclinical models both as a single agent and in combination with chemotherapy [3133]. In the therapeutic context, patients with increased levels of circulating EPCs may reflect a subpopulation with high potential to derive benefit from VEGFR2 inhibitors.

Most studies have relied on cell surface markers to define the number of circulating progenitors. One shortcoming of our current and published studies is the lack of functional studies to quantify the number of circulating EPCs with repopulation potential. More sophisticated assays, such as late outgrowth colony formation (CFU-EC), may be necessary to determine the functional contribution of circulating EPCs to the assembly of new vessels.

The results from this small preliminary pilot study suggest that circulating EPCs may serve as a potential tumor biomarker in breast cancer patients to assess tumor burden and response to therapy. Adequately powered studies are needed to properly train, test, and validate this hypothesis.

Acknowledgements

Supported by the Mentored Medical Student in Clinical Research Program (General Clinical Research Center/National Institutes of Health Grant M01RR00047), Madeline & Stephen Anbinder Clinical Scholar Award, and Anne Moore Breast Cancer Research Fund.

We thank David Nanus, MD, for comments on the manuscript.

Copyright information

© Springer Science+Business Media, LLC 2007