Terminal erythroid differentiation (TED) is a distinct developmental stage of human erythropoiesis during which proerythroblasts (pro) sequentially differentiate into early basophilic erythroblasts (EBs), late basophilic erythroblasts (LBs), polychromatic erythroblasts (polys) and orthochromatic erythroblasts (orthos). Orthos upon enucleation generate reticulocytes which mature to form red blood cells. Using a recently developed flow cytometry based method, we examined TED in more than 200 samples from 126 unique MDS patients and identified two distinct subsets: two-third of the patients showed sufficient number of cells undergoing TED (TED+) while the remaining one-third had too few cells undergoing TED (TED-) despite having adequate numbers of hematopoietic cells for flow analysis. Compared to TED+, the TED- patients were associated with higher myeloid:erythroid (M:E) ratio (mean 5.7:1), more profound anemia (P=0.0003), higher blasts (P=0.0030), but lower absolute neutrophil count (P=0.0245). We also found that the TED- cases were associated with significantly worse overall survival (56 versus 103 months, P <0.0001).

To identify differences at the molecular level between the TED+ (n=23) and TED- (n=19) patients, we sequenced the RNA from bone marrow mononuclear cells using NGS. To identify the biological processes and pathways that are deregulated in TED- patients, we performed Gene Set Enrichment Analysis (GSEA) and Database for Annotation, Visualization and Integrated Discovery (DAVID) analyses to identify significantly differentially expressed genes (>0.5 and <-0.5 log fold change) using DESeq2 and edgeR packages.

PCA analysis using Shifted logarithm transformation (ntd) of RNAseq data identified several gene clusters; in general TED+ samples were closer to each other than TED- samples. A total of 2711 genes showed differential expression, 1572 down regulated and 1139 up regulated (Fig panel A). Both GSEA and DAVID analyses of significantly down-regulated genes showed a marked enrichment of biological processes including heme biosynthesis, erythroid differentiation, and cell cycle, all of which are associated with TED. Significantly up-regulated genes were associated with apoptosis, interferon signaling, TNFα, IL6, and IL2 signaling. Although, it is not clear which cell types are producing these cytokines, one likely cell type could be T-cells. We saw a significantly higher percentage of CD4+ cells in the bone marrows of TED- group (P=0.0162), as assessed by immunophenotyping of various T, B, and plasma cell populations. Also, a CIBERSORT analysis of RNAseq data which estimates the abundance of immune cells in a mixture of cell population showed a significantly higher number of CD4+ cell proportion in TED- group (P=0.0109).

Using GSEA ranking we identified 100 genes (top 50 up-regulated and top 50 down-regulated) that cluster TED+ and TED- samples into two groups (Fig panel B). Also, using a 10-gene panel of highly expressed genes during TED, we saw a clustering of TED+ and TED- groups.

Molecular characterization, using RNAseq data, of TED+ and TED- groups identified several biological pathways deregulated in TED- cases. Given that the RNAseq was done using BMMNC, which is a heterogenous mixture of many different cell types including cells in various TED stages, this "down-regulation" may reflect the loss of cells undergoing TED as observed using flow cytometry. Upregulation of genes involved in cytokine signaling, specially of TNF alpha pathway genes may be a reflection of increased infiltration of CD4+ T cells. Enrichment of genes involved in apoptosis in TED- cases likely represents excessive erythroid cell death and it is likely that this increased apoptosis is due to increased TNF signaling.

Conclusion: Distinct RNA expression profiles were associated with presence or absence of cells undergoing TED in MDS patients. Pathways associated with apoptosis and TNF were upregulated while those related to heme synthesis and erythroid differentiation were downregulated in TED- cases. We conclude that presence or absence of terminal erythroid differentiation identifies two distinct clinical entities within MDS patients with unique molecular profiles that can be identified through RNA sequencing.

Disclosures

Ali: Onconova Therapeutics: Consultancy; Kura Oncology: Consultancy. Coutinho: Kura Oncology: Consultancy. Hoehn: Johnson & Johnson, LLC: Equity Ownership; Janssen: Employment. Jurcic: Seattle Genetics: Consultancy, Research Funding; Novartis: Membership on an entity's Board of Directors or advisory committees; Merck: Consultancy; Kura Oncology: Research Funding; Incyte: Consultancy; Genentech: Research Funding; Forma Therapeutics: Research Funding; Celgene: Research Funding; Alexion Pharmaceuticals: Consultancy; Actinium Pharmaceuticals, Inc.: Membership on an entity's Board of Directors or advisory committees, Research Funding; Daiichi-Sankyo: Research Funding; Astellas Pharma, Inc: Research Funding; Syros Pharmaceuticals: Research Funding; Amgen: Consultancy. Raza: Onconova Therapeutics: Research Funding, Speakers Bureau; Genoptix: Speakers Bureau; Kura Oncology: Research Funding; Janssen R&D: Research Funding; Novartis: Speakers Bureau; Syros Pharmaceuticals: Research Funding; Celgene Inc.: Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution