Health Related Quality of Life Results from the Open-Label, Randomized, Phase III Endeavor Trial Evaluating Carfilzomib and Dexamethasone Versus Bortezomib and Dexamethasone in Patients with Relapsed or Refractory Multiple Myeloma

Introduction: The randomized, phase 3 study ENDEAVOR (NCT01568866; N=929) demonstrated a clinically meaningful and statistically significant improvement in progression-free survival for patients with relapsed or refractory multiple myeloma who were treated with carfilzomib and dexamethasone (Kd) versus bortezomib and dexamethasone (Vd; median, 18.7 vs 9.4 months; hazard ratio: 0.53; 95% confidence interval [CI]: 0.44-0.65; P<0.0001) (Dimopoulos et al. Lancet Oncol. 2016;17:27−38). Patient-reported outcomes (PROs) were included as exploratory endpoints in the ENDEAVOR study. Here, we present results of a prespecified analysis of health-related quality of life (HR-QoL) in the ENDEAVOR trial.

Methods: HR-QoL was assessed by 3 validated instruments: the European Organization of Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire-Core 30-item questionnaire (QLQ-C30), the EORTC Quality of Life Questionnaire-multiple myeloma specific 20-item module (QLQ-MY20), and the Functional Assessment of Cancer Therapy/Gynecologic Oncology Group-Neurotoxicity (FACT/GOG-Ntx) "Additional Concerns" neurotoxicity subscale. These instruments were assessed prior to treatment administration on day 1 of cycle 1, and then every 28 days until disease progression, withdrawal of consent, or commencement of other nonstudy anticancer treatment. Due to differing treatment cycle lengths, the PRO assessments coincided across groups every 12 weeks. The primary PRO hypothesis was superiority of Kd over Vd for the Global Health Status/Quality of Life (GHS/QoL) scale of the QLQ-C30. Seven further subscales were prespecified from the QLQ-C30 (fatigue, nausea/vomiting, pain, physical functioning, role functioning) and the QLQ-MY20 (disease symptoms, side effects of treatment).

PRO subscales were compared between groups using a restricted maximum likelihood-based mixed model for repeated measures (MMRM). Two sensitivity analyses were performed for the GHS/QoL scale to evaluate the robustness of the MMRM to missing data. Clinical interpretation for the EORTC QLQ-C30 subscales was guided by pre-specifying minimum important differences (MIDs) based on evidence-based guidelines (5 points for the GHS/QoL scale). For the QLQ-MY20 subscales, the standard error of measurement was used as a proxy for the MID. The proportion of patients who had improved (≥5 points) from baseline on the GHS scale was summarized at each coinciding time point.

Results: Baseline completion of the QLQ-C30 questionnaire was similar between groups (Kd, 87.7%; Vd, 84.3%). Compliance was high when calculated for all patients expected to provide a questionnaire at each time point (ie, alive and on-study), ranging from 73.1% to 93.9%. Median duration on study treatment was 40 weeks and 27 weeks for Kd and Vd patients, respectively. Using the MMRM model, Kd was associated with statistically significantly higher GHS/QoL scores compared with Vd (p<0.0001). However, the overall treatment difference point estimate of 3.5 (95% CI, 2.0-5.1) did not reach the pre-defined MID. When including the treatment by time interaction (p=0.28) to estimate the treatment difference at timepoints where HR-QoL assessments coincided with day 1 of a cycle, the point estimates increased over time, with the differences at week 60 and 72 reaching clinical significance (5.4 and 5.8, respectively) (Figure). Results from the two sensitivity analyses confirmed findings from the MMRM analysis. Statistically significant benefits were observed in favor of the Kd group for fatigue (P=0.04), pain (P=0.02), side effects (P<0.0001) and NTx subscales (p=0.0002), although these differences did not meet MID thresholds. The proportion of patients reaching a ≥5 point improvement in the GHS scale was numerically higher in the Kd group up to week 48, although the difference between the groups did not reach statistical significance.

Conclusions: This analysis of PROs in the ENDEAVOR study demonstrated that Kd was statistically superior to Vd on the QLQ-C30 GHS/QoL scale, with clinically meaningful differences observed at later timepoints but not on average overall. For patients remaining on longer term treatment, the clinical benefits of Kd compared with Vd were associated with better GHS/QOL. Although not meeting MID thresholds, statistically significant benefits were also observed in favor of the Kd group for other aspects of HR-QoL.

View largeDownload slide

View largeDownload slide

Close modal