Molecular pathogenesis of MDS: genomic and epigenetic instability

Abstract

New high-throughput technologies have facilitated tremendous progress in the understanding of molecular pathogenesis of myelodysplastic syndrome (MDS) and related myeloid malignancies, including MDS/myeloproliferative overlap syndromes (MDS/MPN) and secondary myeloid leukemia (sAML). In addition to better mapping chromosomal abnormalities, including microdeletions and gains, the pathogenetic and clinical role of somatic uniparental disomy has been recognised. Targeted and unbiased genomic screens have led to the identification of new somatic mutations and realisation that multiple mutational events are involved in or facilitate clonal evolution. This molecular heterogeneity explains the pathomorphological diversity of MDS and the variability of clinical courses seen in this disease. Genetic/genomic lesions tend to involve specific pathways but may converge or link individual aspects of the pathogenesis of MDS. IDH1/2, DNMT3A and TET2 mutations affect methylation and hydroxymethylation of DNA, while EZH2, EED, SUZ12, ASXL1 and UTX mutations may lead to dysregulation of histone methylation. Such mutations may constitute the previously hypothetical link between genomic and epigenetic instability. CBL and mutation in phosphothyrosine kinase receptors may affect proliferation and mutations in TP53 and RAS pathways impair apoptotic responses. The recently discovered mutations in genes of spliceosomal machinery further add to the complexity of molecular mechanisms of clonal dominance and leukaemic transformation. Spliceosomal mutations, through patterns of misplacing, unsplicing and aberrant alternative spicing, can lead to haploinsufficiency of a variety of tumour suppressor genes (TSG) and missplicing patterns generated by individual mutations likely determine clinical phenotypes. The key pathogenetic mechanism in MDS, haploinsuffcient expression of TSG, can be due not only to inactivating/hypo-morphic mutations, promoter silencing and hemizygous deletions but also to mis/unsplicing with subsequent degradation of aberrant RNA. The challenges of the next years will include identification of lesions with clinical relevance and incorporation of such defects into the diagnostic and prognostic schemes. Furthermore, clinically applicable targeted or high-throughput methods have to be designed and introduced into clinical practice.