Association Between EZH2 and Other Acquired Mutations In Myelofibrosis and Myelodysplastic/Myeloproliferative Neoplasms

Abstract 625

We recently identified EZH2 as the major target of chromosome 7q acquired uniparental disomy (aUPD) in myeloproliferative neoplasm (MPN) and myelodysplastic syndromes (MDS). To determine the prevalence of EZH2 mutations we screened all coding exons for mutations in total of 624 cases with myeloid disorders (MPN, n=157; MDS, n=154; MDS/MPN, n=219; AML, n=54, CML in transformation, n=40) and found 49 monoallelic or biallelic EZH2 mutations in 42 individuals, most commonly MDS/MPN (27/219; 12%), primary or secondary myelofibrosis (4/30; 13%) and MDS (9/154; 6%). To determine if EZH2 mutations might co-operate with other known abnormalities or whether they might be mutually exclusive, we tested the mutational status of TET2, ASXL1, CBL, RUNX1, CEBPA, FLT3, NPM1, and WT1 in 187 of the 219 MDS/MPN cases that were screened for EZH2. We also tested an additional cohort of 52 primary myelofibrosis cases for both EZH2 and JAK2 V617F mutations. Of the 187 MDS/MPN cases (CMML, n=97; atypical CML, n=68; MDS/MPN-U, n=22), mutations were seen most frequently in TET2 (67/187; 36%), followed by ASXL1 (38/187, 20%; not including cases with the controversial c.1934dupG variant), RUNX1 (27/187; 14%), EZH2 (25/187; 13%), CBL (22/175; 13%), FLT3 (8/187; 4%), CEBPA (7/187; 4%), NPM1 (6/187; 3%) and WT1 (2/187; 1%). Sixty six (35%) cases tested negative for mutations in all 9 genes. Of the 25 cases with EZH2 mutations, 22 (88%) had mutations in at least one other gene, most frequently TET2 (n=11) and ASXL1 (n=10). EZH2 mutations were also seen in combination with mutations in CBL (n=5), CEBPA (n=4), RUNX1 (n=3) and FLT3 (n=2), however there was no significant difference in the frequency of other mutations on comparison of EZH2 mutated and EZH2 unmutated cases. When the analysis was restricted to the 10 cases with homozygous EZH2 mutations, a similar heterogeneity was observed with mutations in CBL, RUNX1, CEPBA and TET2 only (n=1 for each gene), ASXL1 only (n=2), TET2+ASXL1 (n=1), TET2+ASXL1+RUNX1 (n=1) or no other mutation (n=2). Analysis of CFU-GM from one case that tested positive for both EZH2 and TET2 mutations revealed a complex pattern with an EZH2 mutation clearly preceding the sequential acquisition of two TET2 mutations. Of the 82 primary and secondary myelofibrosis cases, 9 (11%) tested positive for an EZH2 mutation. Of these, 5 were positive for JAK2 V617F and 4 were negative. In 2 cases both EZH2 and JAK2 V617F were homozygous indicating that the predominant clone must harbor both mutations. Overall, these data indicate a complex interaction between different abnormalities with little indication of co-operativity or functional redundancy. Whilst these observations will need to be refined by detailed analysis of single clones, they do suggest that the development of both myelofibrosis and MDS/MPN requires functional alterations in multiple pathways.