[Prognostic value of early normalization of CA 125 during chemotherapy in stages III and IV ovarian tumors].

The value of early CA 125 assays and analysis of its diminution kinetics during chemotherapy have been the subject of numerous studies. In contrast, routine utilization of CA 125 assays in clinical practice remains controversial or at best difficult to apply because the definitions and prognostic values associated with CA 125 assays vary greatly from one study to the next. This study was designed to determine whether serial CA 125 assays during induction chemotherapy for ovarian carcinoma, using simple evaluation criteria directly applicable in routine clinical practice such as early normalization (level < 35 UI/ml) are predictive of response to treatment or improved survival. This retrospective longitudinal analysis concerned a historical population of 140 patients with ovarian carcinoma stages III and IV treated at the Antoine-Lacassagne Cancer Center between 1978 and 1993. All the patients were treated by chemotherapy based on platinum salts every 21 days. Serum CA 125 assays were performed both before and after surgery and during each chemotherapy cycle. Eighty-four patients had a pre-operative CA 125 assay. No difference is observed in survival as a function of their preoperative CA 125 concentration (p = 0.4). Sixty-seven patients had a CA 125 assay the 6th week after initiation of chemotherapy, 62 the 9th week and 47 the 18th week. Normalization of CA 125 the 6th week (p = 0.0001), the 9th week (p = 0.0008) and the 18th week (p = 0.03) after the initiation of the chemotherapy cycle are correlated with survival. The median survival in our study is 42 months if the CA 125 is below 35 UI/ml the 6th week versus 13 months if the level of CA 125 remains more than 35 UI/ml. In all, 66 of the 105 FIGO stage III patients underwent second-look surgery. Normalization of CA 125 levels is correlated with the absence of any gross residual tumor at the second-look procedure, the 6th week of chemotherapy (p = 0.0019), the 9th week of chemotherapy (p = 0.0003) and the 18th week of chemotherapy (p = 0.0015). This correlation is not confirmed when the presence of histologic residual tumor in biopsy specimens obtained during second-look surgery is taken into consideration. Overall, 88% of patients whose CA 125 levels failed to normalize during evaluation at the second cycle of chemotherapy have residual tumor at second-look surgery. Outside of clinical trials, repeated early CA 125 assays to determine the chemosensitivity and the prognosis of patients with ovarian carcinoma are of little interest compared to a single CA 125 assay at the 6th week after initiation of chemotherapy. This approach seems to be a good compromise between the information sought and its practical use. However the interest of early modification of chemotherapy regimen after 2 cycles, if the level of CA 125 remains more than 35 UI/ml, will have to be showed.