Efficacy and Safety of Luspatercept Treatment in Patients with Myelodysplastic Syndrome/Myeloproliferative Neoplasm with Ring Sideroblasts and Thrombocytosis (MDS/MPN-RS-T): A Retrospective Analysis from the MEDALIST Study

Introduction: Patients (pts) with MDS/MPN-RS-T have limited treatment options for anemia due to ineffective erythropoiesis. Luspatercept, the first-in-class erythroid maturation agent that enhances late-stage erythropoiesis, is approved by the FDA for treatment of anemia in adult pts with lower-risk (LR) MDS with ring sideroblasts (RS) or MDS/MPN-RS-T after erythroid-stimulating agent (ESA) failure. In the randomized, double-blind, phase 3 MEDALIST study, luspatercept significantly reduced transfusion burden vs placebo in pts with LR-MDS (NCT02631070; Fenaux P, et al. N Engl J Med 2020;382:140-51).

Here, we assess the efficacy and safety of luspatercept in pts with MDS/MPN-RS-T enrolled in the MEDALIST study.

Methods: Eligible pts were ≥ 18 years; had IPSS-R-defined Very low-, Low-, or Intermediate-risk MDS with RS; were refractory, intolerant, or unlikely to respond to ESAs (serum erythropoietin > 200 U/L); and required regular RBC transfusions. Pts were randomized 2:1 to luspatercept (1.0 mg/kg, titration to 1.75 mg/kg) or placebo administered subcutaneously every 3 wks. The primary endpoint was achievement of RBC transfusion independence (RBC-TI) ≥ 8 wks during Wks 1–24.

Results: A retrospective analysis identified 23/229 (10.0%) pts enrolled in the MEDALIST trial who had MDS/MPN-RS-T, per WHO 2016 criteria (Arber DA, et al. Blood 2016;127:2391-405); 14 received luspatercept and 9 received placebo. Pts in this subgroup received a median of 4.0 RBC units/8 wks (range 2.0–11.5) during the 16 wks prior to treatment. At baseline, pts had a median hemoglobin (Hb) level of 7.7 g/dL (range 7.0–9.0), a median leukocyte count of 5.1 × 10⁹/L, a median platelet count of 447.0 × 10⁹/L, and 18 (78.3%) pts had serum erythropoietin levels < 200 U/L (Table).

In the luspatercept arm, 9/14 (64.3%) pts with MDS/MPN-RS-T achieved the primary endpoint of RBC-TI for ≥ 8 wks during Wks 1–24, compared with 2/9 pts (22.2%) receiving placebo (odds ratio 11.3; 95% confidence interval [CI] 1.19, 106.12; P = 0.028).

Pts receiving luspatercept were significantly more likely to achieve clinical benefit (achieving RBC-TI ≥ 8 wks and/or modified hematologic improvement-erythroid [mHI-E] per IWG 2006 criteria [≥ 4 units/8 wks reduction in RBC transfusion in pts with ≥ 4 units/8 wks baseline RBC transfusion burden; Hb increase by ≥ 1.5 g/dL] during Wks 1–24 in pts with < 4 units/8 wks baseline RBC transfusion burden), compared with pts receiving placebo (78.6% vs 33.3%; P = 0.034). Median time from the start of clinical benefit response to end of treatment was 94.6 wks (range 8.0–150.0) in the luspatercept arm and 23.9 wks (range 23.7–57.9) in the placebo arm. During Wks 1–24, mHI-E was achieved by 10 luspatercept pts (6 were high transfusion burden [HTB; defined as baseline transfusion burden ≥ 4 units/8 wks] and 4 were low transfusion burden [LTB; defined as baseline transfusion burden < 4 units/8 wks]) and 1 placebo pt (1/5 HTB). RBC-TI ≥ 8 wks was achieved by 4/8 HTB pts receiving luspatercept (vs 0/5 placebo) and 5/6 LTB pts (vs 2/4 placebo).

After 24 wks, pts in the luspatercept arm had a mean Hb increase of +1.7 g/dL compared with an increase of +0.9 g/dL in pts in the placebo arm (least squares [LS] mean difference +0.85 g/dL; 95% CI −1.13, +2.82). Greater reductions from baseline in mean serum ferritin levels were seen with luspatercept (−121.8 μg/L) compared with placebo (−91.9 μg/L) over Wks 9–24 (LS mean difference −90.1; 95% CI −758.4, 578.2). Pts in the luspatercept arm had median platelet counts of 467.5 × 10⁹/L and median leukocyte counts of 6.5 × 10⁹/L post 24 wks of treatment, compared with pts in the placebo arm with 514.0 × 10⁹/L and 6.2 × 10⁹/L, respectively.

The incidence of specific TEAEs (occurring in ≥ 1 patient) are as follows: fatigue (1/14 [7.1%] luspatercept vs 1/9 [11.1%] placebo), dizziness (7/14 [50.0%] vs 0/9), dyspnea (3/14 [21.4%] vs 0/9), nausea (6/14 [42.9%] vs 2/9 [22.2%]), arthralgia (1/14 [7.1%] vs 0/9), diarrhea (6/14 [42.9%] vs 1/9 [11.1%]), and hypertension (3/14 [21.4%] vs 0/9). In the luspatercept arm, 1/14 (7.1%) pts experienced ≥ 1 thromboembolic event (transient ischemic attack) and 1/9 (11.1%) pts in the placebo arm progressed to AML (as of July 1, 2019).

Conclusions: Luspatercept demonstrated clinical efficacy in pts with MDS/MPN-RS-T with a generally well-tolerated safety profile. These data support the clinical benefits of luspatercept in this patient population with otherwise limited treatment options.