Pretreatment CA-125 and risk of relapse in advanced ovarian cancer.

Abstract

A previous report suggested the nadir serum CA-125 level within the group of patients with ovarian cancer who achieved normalization of CA-125 accurately defined the risk of relapse. Using similar CA-125 subgroups, we sought to determine if the baseline CA-125 level before initiation of maintenance chemotherapy in women achieving a clinically-defined complete response to primary chemotherapy would be of prognostic value. Patients included in this retrospective analysis had been treated on one of two previously reported trials of maintenance chemotherapy (three v 12-monthly cycles of paclitaxel; oral altretamine), with a baseline CA-125 level of < or = 35 u/mL. Progression-free survival (PFS) from study entry was analyzed by the Cox regression model. The distribution of premaintenance baseline CA-125 levels for the 384 patients was 58%, 34%, and 8% for values of (A) < or = 10 u/mL, (B) 11 to 20 u/mL, and (C) 21 to 35 u/mL, respectively. The baseline CA-125 was highly statistically significant, either as a categoric variable (P < .001) or as a continuous variable (P < .0001). Median PFS was 24 months, 17 months, and 7 months for groups (A), (B), and (C), respectively. There was no evidence the CA-125 effect differed by trial or treatment in an interaction analysis (P = .70). The baseline CA-125 level before initiation of maintenance chemotherapy strongly predicts the risk of subsequent relapse. Patients with premaintenance baseline CA-125 values < or = 10 u/mL have a superior PFS compared with higher levels in the normal CA-125 range.