Neuroendocrine stains and proliferative indices of prostatic adenocarcinomas in transurethral resection samples.

Abstract

To determine whether the quantification of certain neuroendocrine and proliferative markers would help in the prognostic evaluation of prostatic adenocarcinomas obtained during transurethral resection of the prostate (TURP). Samples from two groups of patients with prostate cancer were examined. One group comprised 23 patients, of whom 12 were stage IV and 11 stage III, all with Gleason scores of > or = 7; this group was designated as high-grade, high-stage (HGHS). The second group comprised 10 consecutive patients with stage T1a adenocarcinoma of the prostate with Gleason scores of < or = 6, designated as low-grade, low-stage (LGLS) tumours. Tumour tissue from each TURP was stained with MIB-1 (an indicator of cell proliferation), neuron-specific enolase (NSE), chromogranin A (ChA) and synaptophysin (Syn), and 1000 cells counted to determine the percentages of positive cells in both benign and malignant tissue. The percentage of MIB-1-positive cells was designated as the proliferative index (PI). Patients were clinically followed to evaluate tumour progression, documented by rising prostate-specific antigen (PSA) levels, X-ray evidence of recurrent or metastatic carcinoma, or tissue biopsy showing malignancy. The mean number of neuroendocrine cells (NEC) for each marker and the mean PI were greater in the HGHS tumours than in the LGLS tumours or surrounding benign tissue of either group (P < 0.01). The LGLS tumours were remarkable for a having mean PI of about twice that of the benign tissue (2.9 and 1.3, respectively, P < 0.01); the NEC in these cases were more frequent in the benign than in the malignant tissue. There was no significant difference between the mean PIs and the mean percentages of NEC in the 14 HGHS tumours that progressed and the nine HGHS tumours that did not (P values 0.37-0.96). Although the PI assessed by MIB-1 and the number of NEC-positive cells were much higher in HGHS than LGLS tumours, this finding did not appear to have independent prognostic significance. The significance of the higher PI in LGLS tumours than in corresponding benign tissue is uncertain; LGLS tumours had fewer NEC than the surrounding benign tissue. The quantification of any of these four markers (MIB-1, NSE, ChA, Syn) was not prognostically helpful in these groups of cancers present in TURP specimens.