Open Collections

Abstract

Background: A single measurement or a summary of a limited number of measurements of CA125 was considered in the prediction of clinical outcomes for patients with ovarian cancer. We aimed to identify the classes of patients with advanced ovarian cancer based on their CA125 trajectory and to investigate the heterogeneity of clinical outcomes among the patients in the different classes. Methods: CA125 trajectory classes were identified by latent-class mixed models based on values collected within 3 and 6 months post-treatment for 819 women with advanced ovarian cancer enrolled in a randomized trial. Results: Based on their CA125 values during the first 6 months of treatment, the patients with low CA125 levels at baseline that remained low during treatment had the best clinical outcome (a median survival of 83 months and a progression-free survival of 34 months). In contrast, the patients with high CA125 values at baseline with a modest decrease during treatment had the highest risk of death and progression (hazard ratio [95% confidence interval]: 4.83 [3.56, 6.54] for overall survival and 5.15 [3.87, 6.87] for progression-free survival). Conclusions: Longitudinal trajectories of CA125 may provide more direct information for the prognoses of patients with advanced ovarian cancer undergoing chemotherapy treatment.