NEJM Data Show Luspatercept Alleviates Anemia in Patients With MDS

Habib Dable

Treatment with luspatercept-aamt (Reblozyl) reduced the severity of anemia in patients with very low- to intermediate-risk myelodysplastic syndromes (MDS) with ring sideroblasts who require red blood cell (RBC) transfusions, according to findings from the phase III MEDALIST trial that have now been published in the New England Journal of Medicine (NEJM).¹

Additionally, the benefit with luspatercept extended to patients with refractory disease or those who were unlikely to respond to or were intolerant of erythropoiesis-stimulating agents (ESAs).

The published findings showed that 38% of patients in the luspatercept group had RBC-transfusion independence (RBC-TI) for ≥8 weeks compared with 13% of those in the placebo group (P <.001), meeting the primary endpoint of the study.

The key secondary endpoint of RBC-TI for ≥12 weeks was met by a higher percentage of patients in the luspatercept group than in the placebo group at 28% versus 8% for weeks 1 through 24, and 33% versus 12% for weeks 1 through 48 (P <.001 for both). In an analysis that applied the International Working Group (IWG) 2018 criteria, 19% of luspatercept-treated patients compared with 4% of patients on placebo had RBC-TI for ≥16 weeks during weeks 1 through 24; these rates were 28% and 7%, respectively, during weeks 1 through 48.

The study met an additional secondary endpoint, which was hematologic improvement-erythroid (HI-E) for ≥8 weeks, as assessed by IWG 2006 criteria. In weeks 1 through 24, HI-E responses occurred in 53% of patients on luspatercept compared with 12% of those in the placebo group. Fifty-nine percent of patients treated with luspatercept had an HI-E response in weeks 1 through 48 versus 17% of patients on placebo. During the double-blind period, no significant changes in neutrophil or platelet counts were observed.

“It is truly an honor to see work that began in Acceleron laboratories more than a decade ago and advanced successfully through a collaboration with Bristol-Myers Squibb now highlighted in the New England Journal of Medicine,” Habib Dable, president and chief executive officer of Acceleron, the company co-developing luspatercept with Bristol-Myers Squibb, stated in a press release.² “Publication of the MEDALIST trial results in a journal of such eminence will certainly help to raise awareness among physicians of a potential new therapeutic option for treating anemia in certain patients with MDS.”

In December 2019, the FDA’s Oncologic Drugs Advisory Committee (ODAC) scheduled a hearing to review a supplemental biologics license application (sBLA) for luspatercept as a treatment for this patient population. However, later that month, the FDA waived the ODAC review of the application.³ The agency’s action date for a final decision on the sBLA is April 4, 2020.

The double-blind, placebo-controlled, phase III MEDALIST trial included 229 patients who were categorized with International Prognostic Scoring System (IPSS)-Revised very low-, low-, or intermediate-risk MDS with ring sideroblasts and required RBC transfusions. Patients were randomized to receive luspatercept at a dose of 1.0 to 1.75 mg/kg (n = 153) or placebo (76) subcutaneously every 3 weeks.

The median age was 71 years (range, 26-95) and 63% of patients were male. SF3B1 mutations were identified in 93% of patients in the luspatercept group and 86% of those on the placebo arm. Ninety-five percent of patients previously received ESAs and 48% had received prior iron chelation treatment. The data cutoff date was May 8, 2018.

Ninety percent (n = 52) of patients in the luspatercept group who had a response had their first response at the starting dose of 1.00 mg/kg, and 7% of patients responded after dose increases at 1.33 mg/kg (n = 2) and 1.75 mg/kg (n = 2).

RBC-TI for ≥8 weeks occurred in 80% of patients on luspatercept who received <4 units per 8 weeks, in 37% of those who received 4 to 6 units per 8 weeks, and in 9% of patients who received ≥6 units per 8 weeks. In the placebo group, these RBC-TI rates were 40%, 4%, and 3%, respectively.

Additionally, the authors noted that the RBC-TI rates with luspatercept at ≥8 weeks during weeks 1 to 24 were similar regardless of number of baseline mutations: 1 (36.4%), 2 (34.9%), 3 (42.4%), and 4 or 5 (33.3%).⁴

Regarding safety, the most common adverse events (AEs) found in ≥10% of the luspatercept and placebo arms were fatigue (27% vs 13%), diarrhea (22% vs 9%), asthenia (20% vs 12%), nausea (20% vs 8%), dizziness (20% vs 5%), and back pain (19% vs 7%). Grade ≥3 AEs occurred in 42% and 45% of patients on luspatercept and placebo, respectively; the most common were fatigue (5% with luspatercept and 3% with placebo) and injury (5% and 3%, respectively). Thirty-one percent of patients on luspatercept and 30% of those on placebo experienced ≥1 serious AE.

Longer-term analyses from MEDALIST were presented at the 2019 ASH Annual Meeting.⁵ At a data cutoff date of July 1, 2019, results showed that the RBC-TI at ≥8 weeks was 47.7% with luspatercept and 15.8% with placebo (odds ratio [OR], 5.978; 95% CI, 2.840-12.581; P <.0001). At ≥6 units per 8 weeks, the RBC-TI was 21.2% with luspatercept and 6.1% with placebo (OR, 4.17; 95% CI, 0.89-19.60; P = .0547). In those who received 4 to 6 units per 8 weeks, the RBC-TI was 48.8% and 8.7% with luspatercept and placebo, respectively (OR, 10.00; 95% CI, 2.07-48.28; P = .0013). In patients who received <4 units per 8 weeks, the RBC-TI was 84.8% with luspatercept and 40.0% with placebo (OR, 8.36; 95% CI, 2.51-27.83; P = .0002).

Additionally, the longer follow-up showed that of the 73 patients treated with luspatercept achieving RBC-TI ≥8 weeks during the entire treatment period, 69.9% had ≥2 separate response periods, 38.4% of patients had ≥3 separate response periods, and 20.5% had ≥4 separate response periods.

“Patients with lower-risk myelodysplastic syndromes with ring sideroblasts for whom erythropoiesis-stimulating agents are not effective or are not an option have limited treatment options available beyond continued transfusions,” the authors concluded in the NEJM article. “Luspatercept significantly reduced the transfusion burden in a substantial proportion of these patients and was associated with mainly low-grade toxic effects.”

Luspatercept is also being evaluated in the ongoing COMMANDS trial (NCT03682536), which is evaluating the efficacy and safety of luspatercept compared with epoetin alfa for the treatment of anemia due to IPSS-R very low-, low-, or intermediate-risk MDS in ESA-naïve patients who require RBC transfusions.

Luspatercept is currently approved for the treatment of anemia in adult patients with beta thalassemia who require regular RBC transfusions.

References

1. Fenaux P, Platzbecker U, Mufti GJ, et al. Luspatercept in patients with lower-risk myelodysplastic syndromes. N Eng J Med. 2020;382(2):140-151. doi: 10.1056/NEJMoa1908892.
2. New England Journal of Medicine publishes results of pivotal phase 3 Reblozyl (luspatercept-aamt) MEDALIST Trial [news release]: Princeton, NJ and Cambridge, MA. Bristol-Myers Squibb. Published January 9, 2020. https://bit.ly/2sgGqrA. Accessed January 9, 2020.
3. Bristol-Myers Squibb and Acceleron Pharma provide update on FDA advisory committee for reblozyl (luspatercept-aamt) [news release]: Princeton, NJ and Cambridge, MA. Bristol-Myers Squibb. Published December 6, 2019. https://bit.ly/34SCgEh. Accessed December 6, 2019.
4. Fenaux P, Platzbecker U, Mufti GJ, et al. Luspatercept in patients with lower-risk myelodysplastic syndromes [supplementary appendix]. N Engl J Med. 2020;382(2):140-151. doi: 10.1056/NEJMoa1908892.
5. Fenaux P, Mufti GJ, Buckstein R, et al. Assessment of longer-term efficacy and safety in the phase 3, randomized, double-blind, placebo-controlled medalist trial of luspatercept to treat anemia in patients (pts) with revised International Prognostic Scoring System (IPSS-R) very low-, low-, or intermediate-risk myelodysplastic syndromes (MDS) with ring sideroblasts (RS) who require red blood cell (RBC) transfusions. Blood. 2019;134(suppl_1):841. doi: 10.1182/blood-2019-123064.