Abstract P5-17-03: Quality of life (QoL) results from the TURANDOT trial comparing two bevacizumab (BEV)-containing regimens as first-line treatment for HER2-negative metastatic breast cancer (mBC)

Abstract

Abstract Background: BEV has been shown to significantly improve the efficacy of both paclitaxel (PAC) and capecitabine (CAP) as first-line therapy for HER2-negative mBC. TURANDOT, a randomized phase III trial comparing BEV-PAC vs BEV-CAP, includes extensive QoL evaluation. Methods: Patients (pts) with HER2-negative mBC who had received no prior chemotherapy for mBC were randomized to receive either BEV-PAC (BEV 10 mg/kg d1 & 15 + PAC 90 mg/m2 d1, 8, & 15 q4w) or BEV-CAP (BEV 15 mg/kg d1 + CAP 1000 mg/m2 bid d1-14 q3w) until disease progression, unacceptable toxicity, or withdrawal of consent. The primary endpoint is overall survival (OS); QoL, assessed using the EORTC QLQ-C30, is a secondary endpoint. Pts were asked to complete QLQ-C30 at baseline, q12w during study therapy, at the end of treatment, and 28 days after discontinuing from the study. Results: On-treatment QoL questionnaires were available from all 561 of the treated pts at baseline, 394 at week 12, 266 at week 24, 186 at week 36, 129 at week 48, and <100 at subsequent assessments. For time points up to week 48, there were more questionnaires available from the BEV-PAC than the BEV-CAP arm; the pattern reversed thereafter. Mean global health score was similar in the two treatment arms with little change from baseline over time. In the BEV-PAC arm, mean baseline score was 58.0, remaining above the baseline score up to week 60; mean changes from baseline ranged from +8.3 at week 96 (n = 4) to −9.2 at week 60; mean changes for time points with data from >50 pts ranged from +1.5 to +2.6. In the BEV-CAP arm, mean baseline score was 56.0; the mean score remained above the baseline mean score until week 96. Mean change from baseline ranged from −3.9 at week 84 to +4.9 at week 96. Mean scores for most individual symptom scales were low (typically <20) at baseline and remained low until week 96, suggesting that symptoms evaluated in QLQ-C30 did not represent a major burden to pts. The highest mean baseline symptom scores in both treatment arms were for fatigue (34.1 with BEV-PAC vs 38.8 with BEV-CAP), pain (30.4 vs 34.3, respectively), and insomnia (29.3 vs 32.4, respectively). Mean scores for these symptom scales showed no meaningful increase (representing a deterioration in QoL) over time. Mean scores for the appetite loss and dyspnea scales slightly favored the BEV-CAP arm from week 36 onwards. Scores for financial difficulties fluctuated over time. Physical, role, emotional, cognitive, and social functioning scales showed slight or no change over time and no difference between treatment regimens. Conclusion: Extensive QoL evaluation in the TURANDOT trial revealed no clear difference in global health score and little or no difference in functioning and symptom scales between the treatment regimens. Importantly, BEV-containing therapy was not associated with a deterioration in global health score in either arm. This has important implications when considering the potential benefit of a treatment, particularly in the first-line mBC setting, where clinically meaningful endpoints continue to be debated. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P5-17-03.