Genomic profiling and genotype-phenotype correlations in myeloproliferative neoplasms

Myeloproliferative neoplasms (MPNs) are disorders of the stem cell, due to acquired mutations causing a clonal proliferation of one or more hemopoietic progenitor in the bone marrow. According to the World Health Organization (WHO) 2016 classification of myeloid neoplasms, Ph-negative MPNs include polycythemia vera (PV), essential thrombocythemia (ET), primary myelofibrosis (PMF), and prefibrotic myelofibrosis (prePMF). A somatic mutation in JAK2, CALR or MPL genes is found in the great majority of patients with MPNs. These driver mutations constitutively activate the JAK/STAT pathway, resulting in increased phosphorylation of its substrates and leading to increased cytokine responsiveness of myeloid cells. MPNs may progress to more aggressive diseases. This evolution is typically associated with the acquisition of somatic variants in genes involved in different pathways, including DNA methylation and regulation of chromatin structure, transcriptional regulators, signaling pathway and splicing factors. The phenotype of MPNs seems to be related to the combination of driver and subclonal variants and their order of acquisition but the molecular and clinical correlations are not yet completely understood. This work is a genotype-phenotype study aimed to correlate biological and clinical features in a cohort of 509 MPN patients, diagnosed with ET, prePMF and PMF at the UOC Ematologia, Fondazione IRCCS Policlinico San Matteo, between 1985 and 2019. DNA sequence variants were studied through a NGS approach using the Illumina Nextera Rapid Capture Custom Enrichment Kit and HiSeq2500 platform. The panel targeted the coding sequence of 81 genes known to be involved in myeloid neoplasms. Overall, 589 additional somatic variants were detected. Compared to ET and prePMF, PMF showed a larger proportion of patients carrying at least one additional somatic variant, a higher average number of variants per patient and a greater involvement of high molecular risk genes. ET, prePMF and PMF showed different mutational landscapes: in ET the most commonly involved pathway was DNA methylation genes, while in prePMF RNA splicing genes were often affected, together with DNA methylation; in PMF chromatin structure, DNA methylation and RNA splicing were the most recurrent mutated pathways. This finding suggests that prePMF and PMF share molecular features with MDS, since RNA splicing is often involved in MDS development. No significative association between driver mutation and additional variants was found, except for mutations in the spliceosome genes, which did not occur in CALR- mutated patients. The correlations of the additional variants with the clinical picture highlighted a negative impact of high-risk genes on OS and on the progression of the disease. In conclusion, these data suggest that not only clinical and histopathological criteria, but also distinct mutation patterns might differentiate ET, prePMF and PMF. In the era of precision medicine, an accurate diagnosis and prognostic stratification will be useful in the choice of a proper treatment for each patient