Inducible nitric oxide synthase (iNOS) expression and its prognostic value in prostate cancer.

Objectives: The expression of inducible nitric oxide synthase (iNOS) was evaluated in prostate cancer and the results were compared with other prognostic factors and patients outcome. Materials and Methods: Clinical and histopathological data and follow-up information of 198 prostate cancer (PC) patients treated between the years 1973 and 1992 at Kuopio University Hospital, Finland were collected from patient files. Archival tumor specimens were used for immunohistochemical analysis of iNOS. The expression of iNOS was analysed by light microscopy and the expression was scored into 3 grades (negative weak or strong). Results: iNOS was expressed in tumor cells and d in inflammatory cells inside and around the tumor. Normal and d hyperplastic prostate tissues adjacent to tumors were negative or weakly positive for iNOS. The strong iNOS expression in tumor cells was related to high T-classification (p=0.001), metastasis (p=0.06), high Gleason score (p=0.0004), DNA aneuploidy (p=0.0001) and perineural infiltration (p=0.0001). iNOS expression was not linked with the density of tumor infiltrating lymphocytes or the expression of p53. The mean values of Ki-67, mitotic index and S-phase fraction were higher in tumors strongly expressing iNOS. In univariate survival analysis the strong expression of iNOS was a significant predictor of poor survival in the entire cohort (p=0.0002) and in the M0 patients (p=0.008), but was not an independent predictor of survival in Cox's multivariate analysis. Conclusion: iNOS has been related to stimulative and suppressive effects on cancer cell growth but the prognostic value of iNOS has riot been previously studied in PC. Here we could demonstrate an association between strong iNOS expression and rapid cancer cell proliferation rate, dedifferentiation and advanced stage cancer. The strong iNOS expression was a predictor of poor survival in univariate analysis, but was inferior to established prognostic factors in multivariate analysis.