Background. The importance of immuno-histological detection of neuroendocrine differentiation in prostatic adenocarcinoma with respect to disease at presentation and Gleason grade is gaining acceptance. There is limited literature on the relative significance of three commonly used markers of NE differentiation i.e. Chromogranin A (CgA), Neuron specific enolase (NSE) and Synaptophysin (Syn). In the current work we have assessed the correlation of immuno-histological detection of neuroendocrine differentiation in prostatic adenocarcinoma with respect to disease at presentation and Gleason grade and to determine the relative value of various markers. Materials and methods. Consecutive samples of malignant prostatic specimens (Transurethral resection of prostate or radical retropubic prostatectomy) from 84 patients between January 1991 and December 1998 were evaluated by immunohistochemical staining (PAP technique) using selected neuroendocrine tumor markers i.e. Chromogranin A (CgA), Neuron specific enolase (NSE), and Synaptophysin (Syn). According to the stage at diagnosis, patients were divided into three groups. Group (i) included patients who had organ confined disease, group (ii) included patients with locally invasive disease, and group (iii) with distant metastasis. NE expression was correlated with Gleason sum and clinical stage at presentation and analyzed using Chi-Square test and one way ANNOVA. Results. The mean age of the patients was 70 ± 9.2 years. Group I had 14 patients, group II had 31 patients and group III had 39 patients. CgA was detected in 33 cases, Syn in 8 cases, and NSE in 44 cases. Expression of CgA was seen in 7% of group I, 37% in group II and 35% of group III patients (p 0.059). CgA (p 0.024) and NSE (p 0.006) had a significantly higher expression with worsening Gleason grade. Conclusion. CgA has a better correlation with disease at presentation than other markers used. Both NSE and CgA had increasing expression with worsening histological grade this correlation has a potential for use as a prognostic indicator. Limitations in the current work included small number and retrospective nature of work. The findings of this work needs validation in a larger cohort.