Development of V617F JAK2 Associated Myeloproliferative Neoplasms Is a Non-Random Event That Is Strongly Dependent on JAK2 Haplotype

Epidemiological data and family studies have indicated that inherited factors may predispose to the development of myeloproliferative neoplasms (MPN). It has also been suggested that single nucleotide polymorphisms (SNPs) within JAK2 are associated with specific MPN subtypes. To explore the role of inherited factors in more detail, we initially performed quantitative analysis of a series of JAK2 SNPs in homozygous PV cases (%V617F >50%; n=73). Most mutant haplotypes could be read directly from the distorted allele ratios brought about by expansion of the homozygous clone. In many cases with 50–90% V617F, the residual wild type haplotype could also be read. Strikingly, of the 144 V617F alleles that could be determined, 111 (77%) had an identical core haplotype (subsequently designated 46/1) whereas only 9/76 (12%) residual wild type alleles were 46/1 (P = 1.9e-21, Fisher’s exact test). To explore this observation in more detail we first determined the haplotype structure of JAK2 using 14 SNPs genotyped by the Wellcome Trust Case Control Consortium (WTCCC) in 1500 UK blood donors. Nine haplotypes were inferred using the program PHASE that accounted for 94% of alleles, with a frequency of 0.24 for haplotype 46/1. Haplotype inference and tag SNP analysis revealed that 46/1 was also more frequent in heterozygous V617F positive MPD cases (135/354 alleles) compared to 188 locally sourced healthy controls (92/376 alleles; P = 0.0001) as well as the WTCCC cohort (P = 3.3e-8). Haplotype 46/1 was more frequent in all V617F positive disease entities compared to controls: PV (n=203; P=1.2e-16), ET (n=81; P=1.2e-9) and MF (n=41; P=8.0e-5) however there was no difference in the frequency of 46/1 between controls and V617F negative MPD / idiopathic erythrocytosis (n=123). To determine if heterozygous V617F also preferentially arose on a 46/1 allele as seen for homozygous cases, we developed an allele specific PCR between V617F and a SNP that tags this haplotype. In an analysis of 67 informative heterozygous V617F cases, 50 V617F alleles were 46/1 compared to only 17 residual wild type alleles (P=9.4e-9). We conclude that the 46/1 JAK2 haplotype is a strong predisposition factor for development of V617F associated MPDs (RR=2.6; 95% CI 2.3–2.9). The reason for this predisposition is currently unknown but it is likely that 46/1 is in linkage disequilibrium with an unknown constitutional functional variant that interacts with V617F JAK2.