Lack of reliability of CA125 response criteria with anti-VEGF molecularly targeted therapy

BACKGROUND. CA125 is an accepted indicator of epithelial ovarian cancer (EOC) response and is used to monitor patients treated with cytotoxic chemotherapy. It is uncertain how CA125 is affected by molecularly targeted drugs. In this pilot study, the authors analyzed the utility of CA125 to predict disease behavior in patients who were receiving sorafenib, a Raf-kinase/VEGFR2 inhibitor, and bevacizumab, an anti-VEGF monoclonal antibody. METHODS. Fifteen of 42 patients had recurrent EOC. Patients received sorafenib 200 mg orally twice daily or D1-5 of 7 and bevacizumab 5 mg/kg to 10 mg/kg intravenously every 2 weeks for 28-day cycles. Computed tomography (CT) scans were performed every 2 cycles for restaging, and CA125 was measured monthly. CA125 concentrations were retrospectively analyzed as a function of clinical behavior. RESULTS. Fourteen of 15 patients had abnormal CA125 concentrations at study entry (median 1056 U/mL; range, 67 U/mL to 9813 U/mL). Seven (47%) patients had partial response by imaging criteria. Five of these 7 patients had partial response by CA125 criteria (71% sensitivity). Eight (53%) patients would have had partial responses if CA125 criteria were used; only 5 were confirmed by CT (63 % specificity). Imaging and CA125 criteria combined yielded a higher total response rate of 10 of 15 (67%). Three patients with objective partial response by imaging lasting >20, >22, and >24 cycles would have terminated treatment prematurely if CA125 had been used. CONCLUSIONS. CA125 changes may not correspond to imaging response criteria for ECC patients who are receiving sorafenib and bevacizumab. Caution is recommended when using CA125 as a response criterion of molecularly targeted agents until prospective studies validate CA125 changes with objective imaging response results.