The JAK2 46/1 Haplotype Predisposes to Myeloproliferative Neoplasms Characterized by Diverse Mutations.

Abstract 433

A common constitutional JAK2 haplotype termed 46/1 (also known as GGCC) predisposes to V617F JAK2-positive myeloproliferative neoplasms (MPN) but the underlying mechanism is obscure. Two hypotheses have been postulated: (i) ‘hypermutability' of JAK2 on 46/1 compared to other haplotypes and (ii) a functional difference of JAK2 on 46/1 that positively interacts with V617F and thus provides ‘fertile ground' for development of an MPN. To investigate these possibilities we analyzed patients with essential thrombocythemia entered into the PT-1 studies. As expected, 46/1 was highly overrepresented in V617F positive cases (n=404) compared to population controls (n=1492; P=3.9×10^-11) and in informative individuals V617F preferentially arose on the 46/1 chromosome (P<0.0001). This haplotype was also overrepresented in cases without V617F (n=347, P=0.009), with an excess seen for individuals with MPL exon 10 mutations as well as those who were V617F and MPL exon 10 negative. Analysis of further MPL-positive, V617F negative cases confirmed an excess of 46/1 alleles (n=176, P=0.004) however we found no association between MPL mutations and MPL haplotype. An excess of 46/1 was also seen in cases with JAK2 exon 12 mutations (n=69, P=0.002) and in informative individuals these mutations preferentially arose on the 46/1 chromosome (P=0.029). No association between 46/1 and clinical or laboratory features was seen in the PT-1 cohort either with or without V617F. An excess of 46/1 in JAK2 exon 12 mutated cases is compatible with both the hypermutability and fertile ground hypotheses, but the excess in MPL mutated cases argues against the former. However, on analysis of peripheral blood leukocytes we detected no difference in sequence, splicing or expression of JAK2 on 46/1 compared to JAK2 on other haplotypes suggesting that any functional difference of JAK2 on 46/1, if it exists, must be relatively subtle. Finally, since 46/1 is seen at comparable frequencies in different MPN subtypes it does not explain the phenotypic diversity associated with V617F JAK2. In an analysis of 138 MPL mutated cases (ET, n=99; MF, n=36; other MPN, n=3) we found that the median W515K/L mutation levels as determined by pyrosequencing in ET (21%) were significantly lower than those seen in MF (46%; P<0.0001). However we were unable to confirm previously reported differences in SNP frequencies (rs1524395 at 7p11, rs2279784 at 3q21 and rs12500918 at 4q31) between cases with ET (n=763) and PV (n=163). In summary, our data provide further evidence for the fertile ground hypothesis whereby a proposed subtle functional difference of JAK2 on 46/1 makes the development of an MPN more likely when a JAK2 mutation arises on this haplotype, or when JAK2 46/1 is present with a MPL exon 10 mutation. JAK2 and MPL mutation levels, but not 46/1 status, are related to disease phenotype.