Global Health Status Inadequately Captures Longitudinal Changes in Quality of Life in Patients Receiving Carfilzomib for Multiple Myeloma: Electronic Patient-Reported Outcomes (ePROs) from a Prospective Observational Study

Background

Carfilzomib (CFZ) is a second-generation proteasome inhibitor with activity in multiple myeloma (MM) that is associated with cardiotoxicity and dyspnea. The pathophysiology and risk factors for developing these toxicities, and their effect on health-related quality of life (HRQoL), are unknown. We are conducting a prospective observational study (NCT04827563) of longitudinal changes in cardiovascular and endothelial physiology among patients with MM being treated with CFZ. One of our aims is to utilize longitudinal electronic patient-reported outcomes (PROs) to assess how global HRQoL and specific quality of life domains change during CFZ treatment, as well as to determine if PRO scores correlate with development of cardiovascular toxicities.

Methods

Patients with MM receiving CFZ for the first time are prospectively enrolled and undergo the following study procedures prior to CFZ initiation and then every 2 weeks for the first 6 weeks of CFZ therapy (C1D1, C1D15, C2D1, C2D15): 24-hour ambulatory blood pressure monitoring, echocardiography, pulse wave velocity, and EndoPAT endothelial function assessment. Validated PRO instruments of cancer- and myeloma-specific HRQoL (EORTC QLQ-C30 and QLQ-MY20) and dyspnea (FACIT Dyspnea-10) are prospectively administered by email at all aforementioned timepoints and 3 and 6 months after enrollment (C4D1, C7D1). Cardiovascular adverse events (CVAEs), including dyspnea, are prospectively assessed at all timepoints by CTCAE v5.0 and adjudicated by a cardio-oncologist. PRO scores were analyzed by one-way repeated measures ANOVA. Correlations were calculated by Pearson or tetrachoric coefficients.

Results

Ten patients have completed 6 months of follow-up. The median age was 57 years (range: 43-66), including 9 newly-diagnosed MM, 6 men, and 9 with pre-existing cardiovascular risk factors. All patients received CFZ-containing quadruplet regimens in 28-day cycles, with 8 receiving twice-weekly CFZ. ePRO survey completion rate was 95%.

Six patients (60%) developed any-grade CVAEs, with 4 (40%) developing grade 3-4 CVAEs, most commonly grade 3 HTN and grade 3 HFpEF. Four patients (40%) developed treatment-emergent dyspnea by CTCAE, 2 with grade 1 dyspnea and 2 with grade 3 dyspnea at maximum. Most patients developed dyspnea by C1D15, and 2 patients had ongoing grade 1 dyspnea at last follow-up.

There was no significant change in Global Health Status over the first 6 months of CFZ treatment on EORTC (p = .35, Figure 1). However, there was a significant improvement in disease-related myeloma symptoms (p = .02, Figure 2) and significant worsening of treatment-related side effects (p = .03, Figure 2) over that same time period, with improvement in side effects back to baseline by C7D1. In addition, there were significant improvements in pain (p = .03) and future perspective (p = .03). Changes in disease symptoms, pain, and future perspective, but not side effects, were also clinically significant based on minimally-important differences of the PRO instruments. There were no significant longitudinal changes in other EORTC functioning or symptom domains. In addition, there were no significant changes in dyspnea severity (p = .52) or functional limitations from dyspnea (p = .24) on FACIT Dyspnea.

There were 7 patients (70%) with treatment-emergent dyspnea as assessed by increasing FACIT scores from baseline. There was a moderate positive correlation between PRO-assessed dyspnea and incidence of any-grade CVAEs (tetrachoric 0.53, p<.05) and a moderate negative correlation between PRO-assessed dyspnea and Global Health Status (Pearson -0.50, p< .0001).

Conclusions

Although global HRQoL as assessed by EORTC Global Health Status did not change over the first 6 months of CFZ treatment in this prospective cohort, there were statistically and clinically meaningful changes in several quality of life domains. This suggests that global HRQoL in myeloma trials may inadequately describe dynamic changes in symptom burden during treatment; capture of multiple domains of HRQoL is warranted. The FACIT Dyspnea instrument captured more patients with treatment-emergent dyspnea than CTCAE and worsening dyspnea scores were associated with increased incidence of CVAEs and inferior HRQoL, implying a possible role of this PRO instrument in prediction of CFZ-associated cardiotoxicity and identification of patients with low-grade but bothersome dyspnea.

Cooperrider:Abbvie: Other: Current employment by spouse. Wolfe:BMS: Speakers Bureau. Derman:COTA Healthcare: Membership on an entity's Board of Directors or advisory committees; Janssen: Consultancy; Sanofi: Consultancy. Jakubowiak:Janssen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees. DeCara:UptoDate: Patents & Royalties.