THE PROGNOSTIC ROLE OF PREOPERATIVE CHROMOGRANIN A IN CLINICALLY LOCALISED PROSTATE CANCER TREATED WITH RADICAL PROSTATECTOMY

Introduction & Objectives: The aim of the present study was to analyze the prognostic values of preoperative CgA as a marker of poor prognosis and recurrence alter radical prostatectomy. Moreover we attempted to find a correlation with well known prognostic variables. Material & Methods: This study comprises 306 patients with clinically localized prostate cancer prospectively recruited who underwent radical prostatectomy from January 2000 to May 2005. A blood sample for the determination of serum preoperative Chromogranine A value (radioimmuno assay CGA-RIACT) was obtained in all cases. All radical prostatectomy specimens were formalin fixed, coated with India ink, weighed and serially perpendicular sectioned, staged according to the 2002 American Joint Committee on Cancer (AJCC) staging system. Spearman correlation test was used to compare CgA values and other continuous variables. Kruskal-wallis test was used to analyze CgA levels differences among 3 or more groups of patients (PSA, GS, Stage), while the Mann Whitney test was used for 2 grouping variables (status). The probability of survival was estimated by the Kaplan-Meier method, with the log-rank test used to estimate differences among levels of the analyzed variables. Results: Median (mean, 25%,75% percentile) CgA level for the 306 patients included was 68 ng/ml (100.1, 47.9-98.7). Correlation between age and CgA levels was positive and statistically significant (p<0.001). Moreover, patients were divided in two groups based on the median age (<68 and ≥68 years). The difference was statistically significant (p=0.002). The comparison of preoperative CgA values among patients grouped according to pathological stage (pT2, pT3a, pT3b, pT4, N+), Gleason score (GS 2-6, 7, 8-10) and preoperative PSA (< 1Ong/ml, 10-20 ng/ml, > 20 ng/ml) did not achieve statistically significant differences. The mean (median, range) follow up was 21.18 months (18, 1-55 months). Of the 281 patients included in the survival analysis, 208 (74%) were free from biochemical recurrence (NED) and 73 (26%) presented a biochemical recurrence (PROG). The difference between CgA levels among NED and PROG patients was not statistically significant. Patient were stratified on the bases of the normal value of CgA (123ng/ml) and then according to the cut-off value of 68 ng/ml, but these were both unable to achieve significant risk stratification. Conclusions: Studies on a possible prognostic role in localized hormone naive prostate cancer have provided conflicting results. In our study population, larger than those published to date, we found a significant positive correlation of serum CgA with age, but no significant statistical correlation with other available variables and progression free survival. A possible explanation of our results could be that NE cells are not enough to raise circulating levels of CgA when dosed in patients with localized prostate cancers.