Human leukocyte antigen (HLA) alleles as predictive factors for benefit from lenalidomide in acute myeloid leukemia (AML).

Objectives: Lenalidomide is an active agent in acute myeloid leukemia (AML); response rates are about 15-30%. There are no well-defined predictive factors for benefit from lenalidomide in AML. One of the mechanisms of lenalidomide is natural killer (NK) cell activation; hence human leukocyte antigen (HLA) class I alleles (serving as killer immunoglobulin-like receptor ligands) could play a predictive role. We here evaluate the same when lenalidomide was used as a bridge to transplant.

Methods: Consecutive AML patients started on lenalidomide as bridge to transplant between Aug-2013 to Aug-2018 were included in this single centre retrospective analysis. The starting dose and schedule of lenalidomide were at the discretion of the treating clinician. Lenalidomide was scheduled to be stopped about 2-4 weeks prior to planned transplant admission (or was stopped earlier if there was intolerance). For this study, event was defined as progression/relapse while on lenalidomide or within 4 weeks of stopping the drug. The primary endpoint was event free survival (EFS). Those who underwent transplant without an event were censored on the day of transplant. Toxicities and post-transplant outcomes were secondary endpoints.

Results: Twelve patients (8 males, median age 29 years) were included. At start of lenalidomide, 7 had complete remission (CR)-1 (measurable residual disease or MRD by flow cytometry was positive in 3, negative in 3, and 1 unknown), 4 CR-2 (all MRD negative) and 1 active disease. In the whole cohort, median EFS was not reached with projected 3 year EFS being 80%. There was a significantly reduced risk of event with HLA A*24 (0% vs 75%, P=0.018) or with HLA B*40 (0% vs 60%, P=0.045). Only 1 patient needed discontinuation due to toxicities (cytopenias). Among patients who underwent transplant, grade II-IV acute graft versus host disease (GVHD) was seen in 83%.

Conclusions: This is first study to show that HLA alleles may have a bearing on the effect of lenalidomide in AML and could serve as predictive biomarkers. These findings need to be confirmed in larger prospective studies.