P693: LUSPATERCEPT RESTORES EFFECTIVE ERYTHROPOIESIS AND PROVIDES SUPERIOR AND SUSTAINED CLINICAL BENEFIT VS EPOETIN ALFA: BIOMARKER ANALYSIS FROM THE PHASE 3 COMMANDS STUDY

Abstract

Background: Anemia is commonly observed in myelodysplastic syndromes (MDS) due to ineffective erythropoiesis (IE). Increased dysplastic erythroid precursors (EPs) and abortive differentiation in bone marrow (BM) are hallmarks of IE in MDS. Luspatercept is approved to treat anemia in patients (pts) with erythropoiesis-stimulating agent (ESA)-refractory, transfusion-dependent lower-risk MDS (LR-MDS). ESAs promote proliferation of EPs, whereas luspatercept enhances EP maturation. COMMANDS (NCT03682536) is a phase 3, open-label, randomized trial comparing the efficacy and safety of luspatercept vs epoetin alfa in ESA-naïve pts with LR-MDS. The COMMANDS study met its primary endpoint of red blood cell transfusion independence ≥12 wk and mean hemoglobin increase ≥1.5 g/dL during wk 1–24, showing the superiority of luspatercept vs epoetin alfa, herein referred to as “ESA” (58.5% vs 31.2%; P<0.0001). Aims: To differentiate the mechanism of action of luspatercept from ESAs and determine its correlation with clinical benefit. Methods: Cytomorphology assessments were performed on BM aspirates at baseline (BL), wk 24, and wk 48. Complete blood count and reticulocytes (retics) were measured in peripheral blood. BM mononuclear cells (BMMC) were subjected to bulk RNA sequencing (RNA-Seq). Erythroferrone (ERFE) and growth differentiation factor 15 (GDF15) serum levels were measured by ELISA. Distribution of continuous variables between the 2 groups were compared using the Wilcoxon rank sum test. Results: Cytomorphology analysis showed a significant increase of EPs at wk 24 compared with BL in both luspatercept and ESA arms (P<0.05; Fig. A). Compared with BL, EP increase in the luspatercept arm was gradual and sustained at wk 24 (P=0.031) and wk 48 (P=1.7e-05), and retics significantly increased at wk 24 and 48 (BL=40.5 vs 61.5 and 63 × 109/L, respectively; P<0.0001). Unlike luspatercept, ESAs increased EPs, but retics (BL=46 × 109/L) were unchanged at wk 24 (42 × 109/L; NS), and at wk 48, indicating luspatercept overcomes IE by removing the block in differentiation. There was an association between increased EPs and retics and durable clinical benefit at wk 48 for luspatercept vs ESAs in both ring sideroblast (RS)+ and RS− pts, albeit delayed in RS− pts. In the luspatercept vs ESA arms, serum ERFE increased (median 7.2, 10.1 ng/mL; P<0.05; vs median 9.1, 10.4 ng/mL; NS) as did GDF15 levels (median 5.0, 7.0 ng/mL vs median 4.5, 5.4 ng/mL; P<0.05), further suggesting expansion of EPs. Gene set enrichment analysis on BMMC RNA-Seq (Fig. B) revealed that upregulation of genes expressed in both early and late EPs at BL was favorable for response to luspatercept (normalized enrichment score [NES] >2, padj ≤0.01) and enrichment of late EPs in the ESA arm was unfavorable (NES<−2, padj=0.006). Luspatercept treatment led to downregulation of TGFβ signaling (NES=−5.55, padj=0.01 vs 0.56 in ESA; Fig. C), HALLMARK apoptosis (NES=−1.56, padj=0.01 vs 0.64 in ESA), and KEGG spliceosome (NES=−2.4 vs 1.5 in ESA; padj=0.01 vs 0.07 in ESA), which were indicative of a pharmacodynamic effect. Summary/Conclusion: Compared with ESAs, response to luspatercept led to sustained increase of EPs and retics over 48 wk, suggesting a differential effect of inhibiting the TGFβ signaling pathway to unblock the erythroid maturation defect in MDS. Luspatercept acts on different erythroid stages, including expansion and maturation of EPs, for sustained clinical benefit. These findings underscore novel insights into mechanistic differentiation of luspatercept vs ESAs correlating with superior clinical benefit of luspatercept in pts with LR-MDS. Keywords: Anemia, Myelodysplastic syndrome, Erythropoieisis, Erythroid differentiation