Objectives: To study neuroendocrine (NE) tumor cell differentiation in prostate cancer in relation to failure after radical prostatectomy.
Methods: Radical prostatectomy specimens from 103 of 111 patients randomized to 3-month neoadjuvant luteinizing hormone-releasing hormone-analogue treatment (neoadjuvant group) or to surgery alone (control group) were available for analysis. Immunohistochemistry using antibodies to chromogranin A (CGA) enabled detection of tumor cells with NE differentiation. NE differentiation was scored as NE-negative (0 to 1+) or NE-positive (2 to 3+). The number of CGA-positive cells/cm(2) tumor area on the slides was assessed in a separate analysis. The patients were followed up for 39 months after surgery, and a prostate-specific antigen value of 0.5 ng/mL or greater in two consecutive blood samples was considered biochemical failure.
Results: Kaplan-Meier analysis stratified for neoadjuvant hormonal treatment showed the failure rate to be significantly greater among those with NE-positive tumors than among those with NE-negative tumors. However, the number of CGA-positive cells/cm(2) was not a variable of prognostic significance. Instead, both NE differentiation and the CGA-positive cell count correlated with the tumor area on the slides (P = 0.0001). Multivariate analysis revealed the tumor area on the slide (P <0.0001) and positive surgical margins (P = 0.03) to be the only significant predictors of biochemical failure.
Conclusions: The extension of NE differentiation in prostate cancer correlates with tumor volume and is not an independent prognostic factor of failure after radical prostatectomy.