Neuroendocrine cells in prostate cancer correlate with poor outcomes: a systematic review and meta-analysis

Objectives To perform a systematic review and meta-analysis of the literature to understand the variation in the reporting of neuroendocrine staining and determine the influence of reporting neuroendocrine staining at diagnosis on patient outcomes. Methods Medical databases were searched to identify studies in which adenocarcinoma specimens were stained with any of the following four neuroendocrine markers: chromogranin A (CgA), neuron-specific enolase (NSE), synaptophysin and CD56. The prevalence of neuroendocrine staining and correlation of the prevalence of neuroendocrine staining to patient outcomes were analysed using a random-effects model. All statistical tests were two-sided. Results Sixty-two studies spanning 7616 patients were analysed. The pooled prevalence for the most common marker, CgA (41%), was similar to that of NSE (39%) and higher than that of synaptophysin (31%). The prevalence of CgA staining was significantly influenced by reporting criteria, where objective thresholds reduced the variation in prevalence to 26%. No correlation was found between CgA prevalence and tumour grade. Patients positive for CgA staining using objective criteria had more rapid biochemical progression (hazard ratio [HR] 1.98, 95% confidence interval [CI] 1.49 to 2.65) and poorer prostate cancer-specific survival (HR 7.03, 95% CI 2.55 to 19.39) compared to negative patients, even among those with low-risk cancers. Conclusion Discrepancies in the reported prevalence of neuroendocrine cells in adenocarcinoma are driven by the inconsistent scoring criteria. This study unequivocally demonstrates that when neuroendocrine cell staining is assessed with objective criteria it identifies patients with poor clinical outcomes. Future studies are needed to determine the exact quantifiable thresholds for use in reporting neuroendocrine cell staining to identify patients at higher risk of progression.