Predictive value of serum CA-125 levels in patients with persistent or recurrent epithelial ovarian cancer or peritoneal cancer treated with bevacizumab on a Gynecologic Oncology Group phase II trial

Objective: To compare two methods of determining therapeutic response and disease progression - modified Gynecologic Cancer Intergroup (GCIG) criteria based on CA-125 and Radiographic Evaluation Criteria in Solid Tumors (RECIST), in a phase II trial of bevacizumab for patients with recurrent or persistent epithelial ovarian and peritoneal carcinoma. Methods: Patients were treated with bevacizumab 15 mg/kg every 21 days. Modified GCIG definitions of progression and response were retrospectively applied and compared to RECIST-defined progression and response. The prognostic significance of CA-125- and RECIST-defined responses and progressions were explored. Results: Sixty-two patients were evaluable by RECIST, 59 for progression by CA-125, and 45 for response by CA-125. Median progression-free survival (PFS) by RECIST and progression-free interval (PFI) by CA-125 were 4.7 and 5.2 months respectively. However, 12.9% of those with CA-125 defined progression remained progression-free according to RECIST for at least 8 months. Thirteen of 62 patients (21%) had response by RECIST and 14/45 (31%) by CA-125. Time dependent analyses indicated that progression by CA-125 was associated with a 5.2 fold increased risk of progression by RECIST, and response by CA-125 had a 5 fold decrease in risk of progression by RECIST. Landmark and time dependent analyses showed prognostic value of responses by CA-125 and RECIST. Conclusions: In this study, disease assessment by RECIST and CA-125 appears to correlate in general. However, approximately 10% of patients might demonstrate progression earlier by CA-125.