BACKGROUND: Response prediction in patients undergoing chemotherapy for ovarian cancer is an important issue, since the cure rate is only about 15-20%. We attempted to develop a semi-empirical model to predict response in individual cases after the first cycle of chemotherapy. MATERIAL/METHODS: This prospective study included 51 cases of advanced ovarian cancer. A method was standardized to estimate ovarian tumor volume accurately from CT scan films. This permits the inclusion of patients who have undergone CT scan elsewhere. Patients underwent 4-6 cycles of chemotherapy and tumor volume was estimated after each cycle. This yielded a tumor regression curve for each patient. RESULTS: Percent reduction in tumor volume after the first chemo-cycle was a significant prognostic factor in multivariate analysis. Depending upon the rate of regression patients could be classified into Fast Regressing FR (n=29) and Moderately Regressing MR (n=16), whereas 6 patients showed Progressive Disease (PD) despite ongoing chemotherapy. The median survivals for the FR, MR and PD groups were 29.3, 18.9 and 8.5 months respectively. We found that 'percent reduction in volume after first chemo-cycle' could categorize a patient as FR/MR/PD correctly in 94.1% of cases. This parameter could also detect 5 out of 6 inherently resistant PD cases, who would otherwise undergo further chemotherapy, since early detection of resistance by clinical monitoring is quite difficult. CONCLUSIONS: An individual patient at risk for shorter survival and with inherent drug resistance can be identified after the first cycle of chemotherapy.

Keywords: Ovarian Neoplasms - drug therapy