Randomized phase II trial of weekly ixabepilone with or without biweekly bevacizumab for platinum-resistant or refractory ovarian/fallopian tube/primary peritoneal cancer

Abstract

<h3>Objectives:</h3> Ixabepilone is a microtubule-stabilizing agent that may retain activity in paclitaxel-treated patients. The goal of this multicenter randomized phase II study was to assess the activity and safety of ixabepilone with bevacizumab compared to ixabepilone alone in patients with platinum-resistant/refractory ovarian, fallopian tube, or primary peritoneal cancer. An exploratory objective was to examine the role of prior treatment with bevacizumab and tumor expression of class III ß-tubulin (TUBB3) by immunohistochemistry as a predictive biomarker. <h3>Methods:</h3> Participants were randomly assigned to receive ixabepilone 20 mg/m² days 1, 8, 15 with (IXA+BEV) or without (IXA) bevacizumab 10 mg/kg days 1, 15 every 28 days. Patients were stratified by receipt of prior BEV. The primary endpoint was progression-free survival (PFS). Overall survival (OS), safety, and response rates served as secondary endpoints. <h3>Results:</h3> A total of 78 patients were randomized from March 2017-July 2020. Among 76 evaluable patients who received IXA+BEV (n=39) compared to IXA (n=37), the objective response rate was 33% (n=13) versus 8% (n=3) (<i>P</i>=0.004), with clinical benefit durable at 6 months in 37% (n=14) and 3% (n=1) (<i>P</i><0.001). The addition of BEV significantly improved both PFS (median 5.5 vs 2.2 months, HR=0.33, 95% CI 0.19-0.55, <i>P</i><0.001) [Fig. 1a] and OS (median 10.0 vs 6.0 months, HR=0.52, 95% CI 0.31-0.87, <i>P</i>=0.006) [Fig. 1b]. Both regimens were well-tolerated. TUBB3 expression did not predict response in either arm. Subgroup analyses revealed minimal effect of prior BEV on response to IXA+BEV [Fig 1c/d]. <h3>Conclusions:</h3> IXA+BEV is a well-tolerated, effective combination for treatment of platinum/taxane-resistant/refractory ovarian cancer that extends both PFS/OS relative to IXA monotherapy. Prior receipt of BEV should not preclude use of IXA+BEV. TUBB3 is not a predictive biomarker for response to IXA+BEV.