Luspatercept Improves Quality of Life and Reduces Red Blood Cell Transfusion Burden in Patients with Non-Transfusion-Dependent β-Thalassemia in the BEYOND Trial

Introduction: Patients (pts) with non-transfusion-dependent β-thalassemia (NTDT) suffer from chronic anemia due to ineffective erythropoiesis. Low hemoglobin (Hb) levels in pts with anemia are associated with development of clinical complications and poor quality of life (QoL). There are no approved treatments for NTDT-associated anemia. Luspatercept has been shown to increase Hb levels in pts with NTDT in the BEYOND study (Taher AT, et al. HemaSphere 2021;5[Suppl 2];Abstract S101). This Hb increase was correlated with improvement in the NTDT Patient-Reported Outcomes (NTDT-PRO) Tiredness/Weakness (T/W) domain. Here we report the effect of luspatercept on red blood cell transfusion (RBCT) burden and QoL outcomes in pts with NTDT in the BEYOND study.

Methods: Eligible pts with Hb ≤ 10 g/dL were randomized to luspatercept (1.0-1.25 mg/kg) or placebo (PBO) subcutaneously for ≥ 48 weeks (wk). The disease-specific instrument NTDT-PRO, measuring chronic anemia symptoms in pts with NTDT, was administered to pts as a daily diary starting on the 7 days prior to Dose 1, Day 1, until wk 24, and thereafter over the 7 days prior to dosing at every other dosing visit. The 36-item Short Form Health Survey Version 2 (SF-36 v2) and Functional Assessment of Chronic Illness Therapy-Fatigue, Fatigue subscale (FACIT-F FS) were completed by the pts at screening and at every other dosing visit thereafter starting with day 1, dose 1. Mean change from baseline in NTDT-PRO T/W and Shortness of Breath (SoB) domain scores and in FACIT-F FS score over a continuous 12-wk interval during wk 13 to 24 and wk 37 to 48 was assessed using analysis of covariance (ANCOVA) with treatment group and other relevant covariates in the model. The SF-36 v2 physical component summary (PCS) and mental component summary (MCS) mean change from baseline at wk 24 and wk 48 were analyzed using a mixed-effect repeated measures model with intercept and time as the random effect. The model included change from baseline scores up to wk 48 as the outcome, treatment group assignment, time, baseline scores, randomization stratification factor(s), time-by-baseline score, and time-by-treatment group assignment terms. Descriptive statistics were used to ascertain the number of RBC units, RBCT events, reason for RBCT, and time to first RBCT event.

Results: Of 145 pts, 96 were randomized to luspatercept and 49 to PBO. NTDT-PRO analyses were conducted in the intention-to-treat population. The QoL-evaluable populations for FACIT-F FS and SF-36 v2 analyses (luspatercept vs PBO) were: 94 vs 47 and 92 vs 46 pts, respectively. Baseline domain scores of the QoL measures were comparable between the 2 treatment arms. Mean change from baseline (least squares mean [LSM]) in NTDT-PRO T/W domain score (wk 13-24) was −0.68 vs −0.20 for luspatercept vs PBO, respectively (LSM difference [LSMD] −0.48; 95% confidence interval [CI] −1.03 to 0.08; P = 0.0924). Improvement was more pronounced in the luspatercept arm during wk 37-48 (Table).

Within the NTDT-PRO SoB domain, mean change from baseline (wk 13-24) was −0.46 vs 0.02 for luspatercept vs PBO, respectively (LSMD −0.49; 95% CI −1.02 to 0.04; P = 0.0721) and during wk 37-48 was −0.59 vs 0.47 for luspatercept vs PBO, respectively (LSMD −1.07; 95% CI −1.80 to −0.33; P = 0.0047) (Table).

Favorable QoL improvements were also observed in the luspatercept arm vs PBO and increased over time for both FACIT-F FS and SF-36 v2 assessments (Table).

During wk 1-48, 79/96 (82.3%) luspatercept vs 22/49 (44.9%) PBO pts were RBCT-free (P < 0.0001); 5/13 (38.5%) and 1/7 (14.3%) pts treated with luspatercept and PBO, respectively, who received RBCTs at baseline, became RBCT-free during wk 1-48. Among all pts who received RBCTs during wk 1-48, pts receiving luspatercept required fewer RBCTs compared with pts receiving PBO (median 1.0 RBCTs [range 1.0-6.0] vs 2.0 [1.0-7.0], respectively), and fewer RBC units (median 2.0 units [range 1.0-8.0] vs 2.5 [1.0-13.0], respectively). Time to first RBCT (median [range] days) was longer for pts receiving luspatercept vs PBO (336 [2.0-336.0] vs 260.0 [9.0-336.0], respectively) and, in both arms, the main reasons for receiving RBCTs were anemia and surgery.

Conclusions: In pts with NTDT, luspatercept improved RBCT burden and QoL compared with PBO. The improvement was more pronounced over time and consistent across 3 QoL-measuring instruments. Most pts receiving luspatercept remained RBCT-free during wk 1-48 of treatment.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal