Clinical phenotypes associated with circulating tumor cell enumeration in metastatic castration-resistant prostate cancer.

BACKGROUND: The presence of ≥5 circulating tumor cells (CTCs) is prognostic for shorter survival in men with metastatic castration-resistant prostate cancer (mCRPC). However, some men have low CTCs despite widespread disease, suggesting heterogeneity in CTC phenotype or detection. The aim of this study was to evaluate the association of CTC enumeration with clinical disease characteristics and overall survival in men with mCRPC at our institution. DESIGN: CTCs were enumerated using the CellSearch method in a prospective correlative study in men with mCRPC starting a new systemic therapy. The primary objective was to determine the clinical phenotype of the subset of men with mCRPC who have a poor prognosis and low CTCs. Secondary end points included associations of CTCs with survival and known prognostic biomarkers, before therapy and at progression. RESULTS: At baseline, median CTC count was 16 cells and prostate-specific antigen (PSA) level was 178 ng/ml. At progression, median CTC count was 42, PSA level was 245 ng/ml, levels of lactate dehydrogenase and alkaline phosphatase rose, and level of hemoglobin dropped. The median overall survival for this heavily pretreated population was 11.2 months, and the multivariable hazard ratio for death of men with CTCs<5 vs.≥5 was 0.43 (95% CI: 0.24-0.77). Median progression-free survival was 4.4 months. CTC enumeration modestly correlated with lactate dehydrogenase and alkaline phosphatase levels but only weakly correlated with PSA and hemoglobin levels. We were unable to identify a consistent subgroup of poor prognosis men with a low number of CTCs. CONCLUSION: CTC enumeration appears to be prognostic in men with mCRPC and describes a phenotype of hematogenous dissemination that cannot be predicted based on standard clinical and laboratory assessments.

Duke Scholars

Author Rhonda L Bitting Medicine, Medical Oncology

Author Susan Halabi Biostatistics & Bioinformatics, Division of Biostatistics

Author Daniel James George Medicine, Medical Oncology

Author Andrew John Armstrong Medicine, Medical Oncology