Studies of anemia in the myelodysplastic syndromes

Abstract

Background: The myelodysplastic syndromes (MDS) constitute a heterogeneous group of malignant bone marrow disorders, characterized by chronic anemia and increased risk of transformation to acute myeloid leukemia (AML). The first line therapy of anemia in MDS is erythropoietin (EPO) with or without granulocyte colony-stimulating factor (G-CSF). Recently, reports about adverse effects of EPO on survival in patients with solid tumors have resulted in questions about its role in MDS. Lenalidomide has a potent effect in 5q- syndrome, however, its mechanisms of action and long-term safety have not yet been studied sufficiently.

Aims: To assess the long-term efficacy and effects on outcome of treatment for anemia in MDS with EPO and G-CSF. To study the in vitro effects of lenalidomide on bone marrow progenitor cells from patients with low-risk MDS and del(5q). To investigate the presence of pre-treatment molecular lesions in patients with del(5q) low-risk MDS treated with lenalidomide, who subsequently underwent disease transformation.

Methods and results: We conducted a long-term follow-up of three studies of EPO and G-CSF treatment of anemia in MDS. The overall erythroid response rate was 39%, and the median response duration 23 months (range 3 to 116+ months). Patients with low-risk disease as well as complete erythroid responders had longer response duration. Most relapses were due to unknown factors; only 18% were attributable to a significant blast progression. Next, we evaluated the effect of treatment on survival and risk of leukemic evolution by comparing the treated cohort with untreated patients from two large datasets. In a multivariate analysis, we demonstrated that treatment with EPO and G-CSF was associated with improved survival in patients requiring <2 units of packed red blood cells (RBC) per month (hazard ratio [HR] 0.44; 95% confidence interval [CI], 0.29 to 0.66; P < 0.001). There was no association with the risk of AML evolution (HR, 0.89; 95%CI, 0.52 to 1.52; P = 0.66). We also studied the effects of lenalidomide on immature hematopoietic progenitor cells from patients with MDS and del(5q) in an erythroblast culture model. Lenalidomide inhibited the growth of malignant cells, while not affecting normal cells. Furthermore, lenalidomide affected gene expression of del(5q) progenitors, and up-regulated the tumor suppressor gene SPARC, located within the commonly deleted segment at 5q31. Finally, we describe two patients with 5q- syndrome, who initially responded well to lenalidomide but after two years unexpectedly developed progressive disease. Before treatment, we were able to demonstrate subclones of bone marrow cells with abnormal cytoplasmic nucleophosmin (NPMc+) and overexpression of p53, generally associated with high-risk MDS and AML. Both the NPMc+ and the p53 expressing subclones expanded at disease progression, and sequencing of TP53 confirmed a pre-treatment heterozygous mutation and a homozygous mutation at disease transformation.

Conclusions: Treatment with EPO and G-CSF in MDS (a) leads to long-term responses, (b) is associated with improved survival in patients requiring <2 units of RBC per month, and (c) does not alter the risk of AML evolution. Lenalidomide specifically inhibits the malignant bone marrow progenitors from patients with MDS and del(5q), and up-regulates the tumor suppressor gene SPARC which may be an important aspect of lenalidomide s mechanisms of action of as well as of disease pathogenesis. Patients with 5q- syndrome responding to lenalidomide and subsequently undergoing disease progression may already before treatment have molecular lesions affecting the genomic stability. The presence of such abnormalities could play a role in a future pre-treatment risk-stratification.