Treatment outcomes of patients (pts) with small cell or neuroendocrine carcinoma (SC/NE) of the prostate: A single-institution experience.

Abstract

111 Background: Neuroendocrine carcinoma of the prostate, including the small cell subset, is aggressive with poor prognosis. Most reported series are from large databases (e.g. Surveillance, Epidemiology, and End Results) which lack granular details on clinical history. We examine our institution’s 58 pt series with SC/NE of the prostate. Methods: We identified pts with SC/NE of the prostate treated with platinum-based chemotherapy. Pts must have had biopsy-proven SC/NE histology. Pt history, including whether it was de novo or treatment-emergent SC/NE (t-SC/NE) disease, stage at SC/NE diagnosis, and treatment history were recorded. For localized SC/NE treated with definitive intent, disease-free survival (DFS) was measured from SC/NE diagnosis to relapse or death. For metastatic SC/NE, systemic therapies were noted. Overall survival (OS) was defined from SC/NE diagnosis to death, estimated by Kaplan-Meier method and compared by log-rank test. Results: Fifty-eight pts with SC/NE were identified (median age 66). Nineteen (32.8%) had de novo SC/NE, while the rest had t-SC/NE. Forty-four (75.9%) pts had metastatic disease at SC/NE diagnosis. Most had pure small cell (n=15) or neuroendocrine (n=21) histology, with the remaining (n=22) having a mixed component (e.g. adenocarcinoma). Of 14 pts without distant metastases at SC/NE diagnosis, 10 received definitive local therapy: most commonly chemotherapy followed by radiation (n=3), chemotherapy followed by surgery (n=3), or surgery alone (n=2). Median DFS for pts treated with definitive local therapy was 11.7 months (mos). For all pts, OS was longer for de novo vs treatment-emergent SC/NE (26.5 vs 10.9 mos, p=0.0001) and mixed vs pure SC/NE histology (21.0 vs 11.0 mos, p=0.0002). Among 45 pts treated with first line (1L) carboplatin and etoposide (Table 1) for metastatic SC/NE, OS did not differ between those receiving maintenance immunotherapy (IO) vs not (12.5 vs 11.0 mos, p=0.9). Conclusions: In one of the largest non-database series of SC/NE prostate cancer pts, OS was longer for de novo SC/NE and mixed histology. Maintenance IO after carboplatin + etoposide was not associated with improved OS. Treated with 1L carboplatin + etoposide for metastatic SC/NE (n=45) Number of patients Maintenance IO 18 No maintenance IO 27 Best response to carboplatin + etoposide (with or without maintenance IO)  Complete response 3  Partial response 20  Stable disease 8  Progressive disease 5  Not evaluable 9