Determinants of lenalidomide response with or without erythropoiesis-stimulating agents in myelodysplastic syndromes: the HOVON89 trial

A randomized phase-II study was performed in low/int-1 risk MDS (IPSS) to study efficacy and safety of lenalidomide without (arm A) or with (arm B) ESA/G-CSF. In arm B, patients without erythroid response (HI-E) after 4 cycles received ESA; G-CSF was added if no HI-E was obtained by cycle 9. HI-E served as primary endpoint. Flow cytometry and next-generation sequencing were performed to identify predictors of response. The final evaluation comprised 184 patients; 84% non-del(5q), 16% isolated del(5q); median follow-up: 70.7 months. In arm A and B, 39 and 41% of patients achieved HI-E; median time-to-HI-E: 3.2 months for both arms, median duration of-HI-E: 9.8 months. HI-E was significantly lower in non-del(5q) vs. del(5q): 32% vs. 80%. The same accounted for transfusion independency-at-week 24 (16% vs. 67%), but similar in both arms. Apart from presence of del(5q), high percentages of bone marrow lymphocytes and progenitor B-cells, a low number of mutations, absence of ring sideroblasts, and SF3B1 mutations predicted HI-E. In conclusion, lenalidomide induced HI-E in patients with non-del(5q) and del(5q) MDS without additional effect of ESA/G-CSF. The identified predictors of response may guide application of lenalidomide in lower-risk MDS in the era of precision medicine. (EudraCT 2008-002195-10).