Myelodysplastic syndromes are a heterogeneous group of diseases with clonal hematopoiesis. Although patients are stratified by International Prognostic Scoring Systems (IPSS, IPSS-R) according to their risk of transformation to acute myeloid leukemia, part of lower-risk patients (LR-MDS) progresses rapidly. Early identification of patients at risk of rapid progression is crucial to initiation of effective treatment. Currently, prognostic scoring systems incorporate clinical variables such as bone marrow blasts, cytogenetics, and the number of cytopenias in IPSS, or hemoglobin, neutrophil, and platelet levels in IPSS-R. The aim of our work was to identify molecular markers at diagnosis that indicate the risk of rapid progression in LR-MDS and to determine their role in the progression.
We sequenced 214 diagnostic samples from LR-MDS patients according to IPSS. Pathogenic mutations were identified in 64% of the cases. We used statistics and a machine learning approach to determine the genes with the main effect on overall survival (OS) and progression-free survival (PFS). The mutated RUNX1 gene was shown to affect the most rapid progression. We demonstrated the great effect of the implementation of the RUNX1 mutational status into IPSS-R (Figure 1A). RUNX1 was also the most frequently mutated gene in the group of patients who progressed within 5 years from diagnosis.
Furthermore, we aimed to uncover signaling pathways underlying malignant transformation. We sequenced the transcriptome of bone marrow (BM) hematopoietic stem cells (CD34+) from the time of diagnosis of 8 RUNX1-mutated LR-MDS (mutR-LR), 29 RUNX1-unmutated LR-MDS (wtR-LR), 20 higher-risk MDS patients (HR) and 13 healthy controls. Clustering analysis showed that gene expression profiles of mutR-LR CD34 + cells were different from those of wtR-LR; however, they resembled HR expression profiles (Figure 1B). Differential expression analysis and GSEA between mutR-LR and wtR-LR CD34 + cells showed deregulation of more than 4000 genes. In mutR-LR, downregulation of pathways associated with antitumor cellular response – the DNA damage response (DDR), cellular senescence, chromatin and gene silencing, apoptosis, cellular response to stress, telomere maintenance, and hypoxia – was observed (Figure 1C). We hypothesized that these mechanisms play a role in LR-MDS cells contributing to maintaining the lower-risk state. Our data suggest the role of RUNX1 as a tumor suppressor in LR-MDS and show the impact of RUNX1 mutations, direct or indirect, in disrupting a biological antitumor barrier.
In the next step, we measured senescence-associated β-galactosidase (SA-β-gal) activity in several BM sorted cell types and showed its significantly higher levels in wtR-LR, particularly in CD14+ monocytes (Figure 1D). We also showed the activation of DDR by detection of γH2AX in formalin-fixed paraffin-embedded BM section of wtR-LR (Figure 1E). The data suggest that while some wtR-LR BM progenitors activate the DDR, some wtR-LR BM cells suffer more DNA damage and undergo senescence.
Our work is an example of how relevant molecular data can improve risk stratification of MDS patients. Based on our data, we are able to identify patients at risk of rapid progression and choose proper follow-up and treatment strategy. LR-MDS patients with RUNX1 mutations at diagnosis should be intensively monitored despite the low-risk score. Our results also show that RUNX1 mutations disrupt antitumor cellular defense in hematopoietic stem cells and contribute to rapid progression in LR-MDS.