Health-related quality of life in the ENDEAVOR study: carfilzomib-dexamethasone vs bortezomib-dexamethasone in relapsed/refractory multiple myeloma

Abstract

We examined effects of carfilzomib-dexamethasone (Kd56) versus bortezomib-dexamethasone (Vd) on health-related quality of life (HR-QoL) in relapsed/refractory multiple myeloma (MM) patients from the ENDEAVOR study. HR-QoL was assessed by the European Organisation for Research and Treatment of Cancer QoL Questionnaire (QLQ-C30), MM-specific module (QLQ-MY20), and Functional Assessment of Cancer Therapy/Gynecologic Oncology Group-Neurotoxicity (FACT-GOG-Ntx) "Additional Concerns" neurotoxicity subscale. The QLQ-C30 Global Health Status (GHS)/QoL scale and seven prespecified subscales were compared between groups using mixed model for repeated measures. Of 929 randomized patients, 911 with ≥1 post-baseline assessment were included. Kd56 was associated with statistically significant improvements in GHS/QoL, fatigue, pain, side effects, and FACT/GOG-Ntx scores versus Vd, although mean differences did not meet thresholds for clinical significance. The Kd56 group had longer time to deterioration (TTD) in GHS/QoL (median 3.7 versus 2.8 months, p = 0.0046), physical function (5.6 versus 3.7 months, p = 0.0390), nausea/vomiting (17.6 versus 8.2 months, p = 0.0358), side effects (6.4 versus 3.7 months p < 0.0001), and FACT/GOG-Ntx (11.1 versus 5.5 months, p = 0.0004). Overall, Kd56 resulted in statistically but not clinically significant improvements in mean GHS/QoL scores versus Vd. Treatment with Kd56 versus Vd also significantly prolonged TTD in GHS/QoL, physical function, nausea/vomiting, side effects, and FACT/GOG-Ntx.

Conflict of interest statement

H.L. has received research funding from Amgen and Takeda; received consulting fees from Amgen, Takeda, BMS, Celgene, and Janssen and has served on the speakers’ bureau for Amgen, Takeda, BMS, Celgene, and Janssen. P.M. has received consulting fees from Amgen, Celgene, Takeda, Janssen, BMS, and Novartis. M.A.D. has received consulting fees from Amgen, Celgene, Janssen, Takeda, and Novartis. M.V.M has received consulting fees from Janssen, Celgene, Amgen, Takeda, BMS, and Glicomimetics. M.K. has received consulting fees from Celgene, Takeda, Amgen, BMS, Janssen, and Chugai, and has received research funding from Celgene. R.H. reports research grants from Celgene, Amgen, Novartis, and Takeda; and consulting fees from Janssen, Amgen, and BMS. S.F. and J.B. are employees and equity owners of Amgen. K.C. has received consulting fees from Amgen, Celgene, BMS, and Endomag. K.W. has served on advisory boards for Amgen, Celgene, and Janssen; received consulting fees from BMS, Celgene, Janssen, Novartis, Onyx, and Takeda; and has received research funding from Janssen and Celgene.

Figures

Fig. 1

Subject disposition (CONSORT diagram). PRO patient-reported outcome

Fig. 2. Adjusted least squares mean treatment difference in QLQ-C30 and QLQ-MY20 scores, and functional assessment of the FACT/GOG-Ntx subscale.

a GHS/QoL and functional domains from the QLQ-C30. *: b Symptom domains from the QLQ-C30 and QLQ-MY20. †: c FACT/GOG-Ntx. ‡: Horizontal bars indicate 95% confidence intervals. The analysis was performed based on a linear mixed-effects model. The model includes the fixed, categorical effects of treatment, the randomization stratification factors—prior proteasome inhibitor treatment, lines of prior treatment, International Staging System stage and choice of route of bortezomib administration, and random effects of subject intercept and coefficient on time. The least squares mean estimates are the overall estimates under the assumption that the treatment effect is the same across visits. Prespecified, between-group MIDs for QLQ-C30 subscales were 5 for GHS/QoL, 6 for physical functioning, 7 for role functioning, 6 for fatigue, 4 for nausea/vomiting, and 7 for pain. Between-group MIDs for QLQ-MY20 scales using the SEM as a proxy were 9 for disease symptoms and 7 for side effects of treatment. The MID for the FACT/GOG-Ntx score is estimated to be between 3.3 and 4.4 points. *: Subscales were scored as directed by the EORTC scoring manual where higher scores indicate better HR-QoL and better functioning. †: Subscales were scored as directed by the EORTC scoring manual where higher scores indicate more severe symptoms (QLQ-C30) or more symptoms (QLQ-MY20). ‡: The Ntx subscale is scored from zero to 44, with lower scores indicating more neurotoxic symptoms. EORTC: European Organisation of Research and Treatment of Cancer; FACT/GOG-Ntx: Functional Assessment of Cancer Therapy/Gynecologic Oncology Group-Neurotoxicity subscale; GHS/QoL: Global Health Status/Quality of Life domain; HR-QoL: health-related quality of life; Kd56: carfilzomib (56 mg/m2) and dexamethasone; MID: minimum important difference; QLQ-C30: Quality of Life Questionnaire-Core 30-item module; QLQ-MY20: Quality of Life Questionnaire-multiple myeloma specific 20-item module; SEM: standard error of the mean; Vd: bortezomib and dexamethasone

Fig. 3

Least square mean estimates and standard errors by treatment group (for the subset of visits with assessments at day 1 of a cycle) in QLQ-C30 Global Health Status/QoL. Note: analysis was performed based on a linear mixed-effects model. The model included the fixed, categorical effects of treatment (all baseline responses are modeled with a dummy treatment), the randomization stratification factors—prior proteasome inhibitor treatment (prior carfilzomib or bortezomib versus no prior carfilzomib or bortezomib treatment), lines of prior treatment (1 versus 2 or 3), International Staging System (1 versus 2 or 3) and choice of route of bortezomib administration (intravenous or subcutaneous), and random effects of patient intercept and coefficient on time and treatment-by-time interaction. The prespecified between group MID for GHS/QoL was 5. C1D1 cycle 1 day 1, HR-QoL health-related quality of life, Kd56 carfilzomib (56 mg/m2) and dexamethasone, QLQ-C30 Quality of Life Questionnaire-Core 30-item module, MID minimum important difference, QoL quality of life, Vd bortezomib and dexamethasone, W week

Fig. 4. Percentage of PRO responders for QLQ-C30 GHS/QoL.

a ≥5-point improvement in QLQ-C30 GHS/QoL scale from baseline. b ≥15-point improvement in QLQ-C30 GHS/QoL scale from baseline. Odds ratios are estimated using the Cochran-Mantel-Haenszel method stratified by the following randomization stratification factors: prior proteasome inhibitor treatment (yes versus no), lines of prior treatment (1 versus 2 or 3 lines), International Staging System stage (1 versus 2 or 3), choice of route of bortezomib administration (intravenous versus subcutaneous). GHS/QoL Global Health Status/Quality of Life, Kd56 carfilzomib (56 mg/m2) and dexamethasone, QLQ-C30 Quality of Life Questionnaire-Core 30-item module, Vd bortezomib and dexamethasone

Fig. 5. Estimated least squares mean change from baseline for the QLQ-C30 and QLQ-MY20 scores.

a GHS/QoL and functional domains from the QLQ-C30. †: b Symptom domains from the QLQ-C30 and QLQ-MY20 ‡. Vertical bars indicate 95% confidence intervals of the mean. *p < 0.05 change from baseline. †: Subscales were scored as directed by the EORTC scoring manual where higher scores indicate better HR-QoL and better functioning. ‡: Subscales were scored as directed by the EORTC scoring manual where higher scores indicate more severe symptoms (QLQ-C30) or more symptoms (QLQ-MY20). Clinically meaningful changes for improvements and worsening are indicated by dotted horizontal lines for the QLQ-C30 scales. SEM (±1 SEM) has been used in absence of anchor-based minimally important change for the QLQ-MY20. EORTC: European Organisation of Research and Treatment of Cancer; HR-QoL health-related quality of life, Kd56 carfilzomib (56 mg/m2) and dexamethasone, QLQ-C30 Quality of Life Questionnaire-Core 30-item module, QLQ-MY20 Quality of Life Questionnaire-multiple myeloma specific 20-item module, Vd bortezomib and dexamethasone

Fig. 6. Change from baseline GHS/QoL domain for patients with a partial response or better.

Positive change indicates improved GHS/QoL. The clinically meaningful threshold for improvements and worsening in the GHS/QoL domain was ±6. *p < 0.05 between-group differences. GHS/QoL Global Health Status/Quality of Life, Kd56 carfilzomib, lenalidomide, and dexamethasone, Vd bortezomib and dexamethasone