EXPRESSION OF INDUCILE NITRIC OXIDE SYNTHASE AND CYCLOOXYGENASE-2 IN OVARIAN CANCER: CORRRELATION WITH CLINICAL OUTCOME

Objectives. Cyclooxygenase-2 (COX-2) and inducible nitric oxide synthase (iNOS) play a critical role in cancer development. We investigated NOS and COX-2 expression in relation to clinical outcome in 78 International Federation of Gynecology and Obstetrics (FIGO) stage III ovarian serous carcinoma with a low grade of differentiation (W). Methods. Disease-free interval and cause-specific survival rates (Kaplan-Meier method) were compared using the log rank test. A multivariate analysis (Cox-proportional hazards models) was used to determine the independent effect of each variable on prognosis. Fisher's exact test was used to analyze the distribution of NOS and COX-2 expression according to clinical complete response to chemotherapy and to the presence of a brief disease-free interval (less than or equal to 12 months). Results. Overall 60 and 125 months cause-specific survival rates were 32% and 11%, respectively. In univariate analysis, NOS (P= 0.005 and P = 0.003, respectively), COX-2 (P = 0.002 and P = 0.007, respectively), residual disease after surgery (P = 0.017 and P = 0.032, respectively), and FIGO stage (P = 0.008 and P = 0.025, respectively) were associated with survival and a disease-free interval. In multivariate analysis (Cox proportional hazards models), the factors that were found to be significantly independent predictors of disease relapse as well as survival were NOS (P = 0.014 and P = 0.001, respectively), COX-2 expression (P = 0.007 and P = 0.029, respectively), and FIGO stage (P = 0.026 and P = 0.05, respectively). NOS and COX-2 expressions were correlated with a brief disease-free interval (P 0.001) and clinical complete response to first-line chemotherapy (P = 0.038 and P = 0.033, respectively). Conclusions. The evaluation of NOS and COX-2 expression may give additional prognostic information concerning the clinical outcome of patients with ovarian carcinoma and may encourage them to select more tailored therapies. (C) 2004 Elsevier Inc. All rights reserved.