Neuroendocrine immunophenotype as predictor of clinical recurrence in 110 patients with prostate cancer.

We evaluated the relationship between NE expression and well-known prognostic factors and assessed whether tumor relapse after radical surgery correlates with the extent of NE differentiation. Radical prostatectomy specimens from 110 patients with clinically localized prostate cancer were assessed. Patients were followed up every three months for the first two years after surgery and six monthly for 5 additional years until failure, or for a mean of 48 months from the time of surgery for those who did not experience failure. The percentage of cells showing CgA immunoreactivity was evaluated using a visual quantitative method. Tumor staining was categorized as positive if greater than 10 percent and negative if less than 10 percent of tumor cells were stained, to ensure that only cases with significant positivity were included in the positive group. The median follow-up was 5.4 years (range 1.8 to 7.2). The median time to clinical recurrence was 7.5 years and the median time to biochemical recurrence was 2.8 years. Of 31 patients (28 percent) who experienced a PSA recurrence, 15 developed a clinical recurrence. The mean preoperative PSA level was 9 ng/ml (range 2.7 to 25). Most cases were well differentiated (Gleason score less than 7), intraprostatic (less than pT2) tumors. Immunoreactivity in >or= 10 percent of the cells was seen in 17.2 percent (n=19) of the tumor specimens. The preoperative PSA level, Gleason score, use of neoadjuvant or adjuvant therapy, lymphnode positivity were not statistically associated with NE expression. Only the primary pathologic stage appeared to be associated with CgA staining in the primary tumor (p=0.001). On the univariate analysis NE expression did not predict biochemical recurrence free survival, whereas it was associated with clinical recurrence. NE differentiation in clinically localized prostate cancer can be associated with failure after definitive surgical treatment, even if no conclusions can be drawn regarding its value as an independent prognostic factor.