수술 전 전신치료 대상 유방암 환자에서 연속적 순환종양세포 수 분석

Abstract

PURPOSE: Circulating tumor cells (CTCs) are known to be associated with prognosis and response to therapy. Although most studies have focused on CTCs among metastatic cancers with greater tumor burden, there is insufficient information regarding operable breast cancers. We aimed to evaluate the clinical implication of CTC counts in patients with locally advanced breast cancer undergoing preoperative systemic therapy.

METHODS: In total, 207 patients with locally advanced breast cancer with no evidence of distant metastasis were prospectively enrolled from a single institute (Asan Medical Center, Seoul, Korea) between February 2014 and May 2017. All patients were undergoing neoadjuvant systemic therapy (NST), and CTC counts were calculated from 10 mL of blood drawn at three time points: before, during, and after systemic therapy. CTC isolation was performed using a Smart Biopsy™ System isolation kit; cells were stained for immunofluorescence microscopy with antibodies against EpCAM, cytokeratin, and CD45. Images of stained cells were captured using a fluorescence microscope with a 400× objective. Recurrence-free survival (RFS) and overall survival (OS) were analyzed in relation to CTC counts. Dynamic changes in CTC numbers associated with responses to therapy were also analyzed.

RESULTS: Mean follow-up period was 22.46 months, and mean age was 46.48 years. One or more CTCs were identified in 65.0% (132/203) of patients before NST, in 72.0% (135/186) during NST, and in 60.2% (103/171) after NST. In HER2-overexpressing breast cancers, one or more CTCs were detected in 89.3% (25/28) of patients before NST, 92.0% (23/25) during NST, and 95.7% (22/23) after NST, whereas 87.0% (20/23) of HER2-positive tumors were CTC-positive both before and after NST (p < 0.05). Initial tumor burden at diagnosis (clinical stage) did not correlate with CTC positivity, and overall, CTC count did not correlate with response to therapy. Using Response Evaluation Criteria in Solid Tumors (RECIST), 86.5% (179/204) of patients were responders (complete or partial response) and 12.1% (25/204) were non-responders (stable or progressive disease). A pathologic complete response was observed in 14.5% (30/207) of patients; however, no association was found between CTC counts/changes and radiological/pathological response to therapy. CTC count was also not correlated with overall prognosis. However, in hormone receptor-positive tumors, CTC detection before NST was associated with treatment response according to RECIST (responder vs. non-responder) using different cutoff CTC values of ≥1, ≥2, and ≥5 (p = 0.003, 0.017, and 0.023, respectively). Among patients lacking CTCs before therapy (35.2%, 50/144), appearance of CTCs after one or two cycles of systemic therapy (15.5%, 7/45; five missing values during NST) was associated with worse outcomes (hazard ratio, 16.46; 95% CI, 1.68–161.22; p = 0.016). Similarly, univariate analysis showed that the group displaying two consecutive ≥5 CTCs throughout the period (before, during, and after therapy) exhibited significantly worse RFS (hazard ratio, 22.83; 95% CI, 1.74–298.86; p = 0.017) and OS (98.4% vs 85.7%, p < 0.05; 98.4% vs.75.0%, p < 0.05) than the group with <5 CTCs at all three time points. This association was only significant among luminal-type breast cancers, whereas no association between CTC count and outcome was found among HER2 and triple-negative subtypes.

CONCLUSIONS: Although CTC count and its dynamic changes are known to reflect tumor burden, prognosis, and response to therapy in metastatic breast cancer, our findings support the limited value of CTC counts for patients with locally advanced breast cancers undergoing NST. Additional studies with a greater number of patients are necessary to conclude the prognostic implication of CTC because the literature reports conflicting results between different studies within this population.