S270: LONGER-TERM ANALYSIS OF EFFICACY OF LUSPATERCEPT VERSUS PLACEBO IN PATIENTS WITH TRANSFUSION-DEPENDENT BETA-THALASSEMIA ENROLLED IN THE BELIEVE STUDY

Résumé fourni par la source

Background: Ineffective erythropoiesis and anemia are hallmarks of β-thalassemia. Luspatercept, a first-in-class erythroid maturation agent, has shown efficacy in the phase 3 BELIEVE (NCT02604433) trial in patients (pts) with β-thalassemia requiring regular red blood cell transfusions (RBCT). However, longer-term analyses of efficacy are yet to be reported. Aims: To observe and report the longer-term effect of luspatercept treatment on transfusion burden and liver iron concentration (LIC). Methods: Pts ≥18 years of age with β-thalassemia or hemoglobin E/β-thalassemia (compound β-thalassemia mutation and/or multiplication of α-globin genes allowed) requiring regular RBCT (6–20 RBC units/24 wk before randomization with no transfusion-free period >35 d) were randomized 2:1 to luspatercept 1.0 mg/kg (titration up to 1.25 mg/kg allowed) or placebo subcutaneously every 3 wk. Response was defined as RBCT burden reduction of ≥33% (≥33% response) or ≥50% (≥50% response) from baseline (BL; ≥2 units) during a rolling 12- or 24-wk interval. LIC was determined by MRI. Results: As of Jan 5, 2021, 125/224 (55.8%) pts randomized to luspatercept completed 144 wk of treatment and 6 (2.7%) pts completed 192 wk. The main reasons for treatment discontinuation were pt withdrawal (23.7% luspatercept vs 11.6% placebo), adverse events (10.3% vs 1.8%), and lack of efficacy (1.8% vs 7.1%). More pts receiving luspatercept vs placebo had ≥33% or ≥50% response during any rolling 12- or 24-wk interval up to the cutoff date (all P<0.0001); the number of responders receiving luspatercept was higher compared with previous data cuts (Table). The median (95% CI) longest duration of response for ≥33% and ≥50% responders (rolling 12-wk interval) in the luspatercept group increased from the primary data cut to 114.0 (107.0–137.0) and 99.0 (95.0–104.0) d, respectively (Table); median (95% CI) longest duration of response for placebo pts was 90.0 (86.0–94.0) and 86.0 (84.0–103.0) d, respectively. The median (range) total duration of response for ≥33% and ≥50% responders (rolling 12-wk interval) in the luspatercept group was 586.0 (84–1300) and 357.5 (84–1267) d, respectively, and 171.0 (84–627) and 169.0 (84–485) d in the placebo group. The 6 pts who completed 192 wk of luspatercept required on average 6.41 (SD 4.32) fewer RBC units during wk 145–192 compared with BL. Overall mean change from BL in transfusion window for ≥50% responders (any 24-wk interval) was increased by 9.88 d (SD 22.04, n=51). More pts receiving luspatercept (27/224 [12.1%]) than placebo (2/112 [1.8%]) achieved RBCT independence (RBC-TI) ≥8 wk (P=0.0015), an increase from previous data cuts (Table). Median (95% CI) longest durations of RBC-TI were 72.0 (62.0–103.0) d for luspatercept and 71.5 (62.0–NA) for placebo, though the placebo result was driven by 2 pts. Longer-term treatment with luspatercept resulted in a decreasing trend in LIC compared with BL for ≥33% responders (rolling 12-wk interval) with a mean (SD) change from BL in LIC at wk 144 (n=25) of −3.73 (9.08) mg/g dry weight. Image:Summary/Conclusion: Continued treatment with luspatercept resulted in more pts experiencing reduced RBCT burden and longer duration of response compared with previous data cuts. RBC units transfused decreased over the longer-term treatment period and the time between transfusions increased compared with BL for ≥50% responders. Longer-term luspatercept treatment also resulted in more pts experiencing RBC-TI ≥8 wk and a potential decreasing trend in LIC; LIC analysis is ongoing. These data indicate that pts continue to benefit from longer-term luspatercept treatment.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.