Abstract
Lenalidomide is an effective new agent for the treatment of Myelodysplastic Syndrome (MDS). In phase II clinical trials, patients treated with Lenalidomide achieved unprecedented rates of hematologic and cytogenetic responses, particularly in patients with interstitial deletions of chromosome 5q. More than 80% of patients with MDS do not have 5q deletions, though approximately 25% of these patients respond to Lenalidomide.3–5 Neutropenia and thrombocytopenia are the primary toxicities of Lenalidomide, and are severe in some cases. MDS is a highly heterogeneous disorder, and no diagnostic assays predict which patients without 5q deletions will respond to Lenalidomide. Using gene expression profiling, we identified a molecular signature that predicts Lenalidomide response. Biologic characterization of the predictive genes demonstrated that the response signature is comprised of a cohesive set of erythroid-specific genes: patients who respond have a lower expression of the erythroid signature in bone marrow mononuclear cells. To facilitate clinical implementation of a gene expression-based assay, we evaluated the signature using two methods, quantitative reverse transcriptase-polymerase chain reaction (RT-PCR) and a bead-based detection technology. Using both methods, the erythroid gene expression signature was expressed at lower levels in patients who responded to Lenalidomide. The bead-based methodology employs ligation mediated amplification (LMA), a methodology that is capable of highly multiplexed reactions with preservation of the relative abundance of transcripts, with amplicon detection on Luminex microspheres, already in use for clinical diagnostic assays. This assay is capable of detecting a large gene expression signature (currently up to 100 genes) in a single, well-controlled assay. With further validation, LMA with fluorescent microsphere detection may become the superior technology for clinical diagnostic assays of multi-gene signatures. We tested the erythroid signature in an independent set of 12 samples using LMA with fluorescent microsphere detection. Again, all responders had a lower expression of the erythroid genes than responders. Validation in a larger cohort of patients is underway. These studies suggest that Lenalidomide-responsive patients without 5q deletions have a defect in erythroid differentiation analogous to the ineffective erythropoiesis in patients with 5q deletions, and that an erythroid gene expression signature predicts Lenalidomide activity in MDS.
Disclosures: Lenalidomide is approved for the treatment of patients with myelodysplastic syndromes and deletions of 5q. Our study identifies a molecular signature of patients without 5q deletion who respond to Lenalidomide.
Author notes
Corresponding author