Inherited Predisposition to Somatic JAK2V617F Mutation — Probably Not the Cause of PV

Kilpivaara O, Mukherjee S, Schram AM, et al. A germline JAK2 SNP is associated with predisposition to the development of JAK2(V617F)-positive myeloproliferative neoplasms. Nat Genet. 2009;41:455-9.   Chronic myelogeneous leukemia (CML), polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF) are all clonal myeloproliferative disorders (MPDs) with many similarities but distinct clinical phenotypes. Studies of the Philadelphia chromosome, an acquired chromosomal lesion associated with CML, led to the discovery of the BCR-ABL gene and its characterization as a gain-of-function mutated tyrosine kinase. BCR-ABL is highly specific for CML, facilitating accurate diagnosis and development of spectacularly successful targeted therapy. In 2002, our group discovered an acquired chromosomal lesion in PV, uniparenteral disomy of chromosome 9p.¹  This work preceded the identification of JAK2V617F, another gain-of-function tyrosine kinase mutation, which is present in almost all PV patients.

The precise pathogenic role of JAK2V617F in PV is not clear since many patients with ET and PMF also have the mutation, as do a small portion of patients with other hematologic disorders, such as chronic neutrophilic leukemia and refractory anemia with ringed sideroblasts and thrombocytosis. Some compelling evidence suggests that JAK2V617F may not be the disease-initiating mutation in PV, including its absence in some patients with sporadic PV. In familial PV there is an absence of linkage to 9p, suggesting an independent germline predisposition,²  yet in familial PV, affected members can be either JAK2 V617F-negative or positive. Interestingly, in these families, individual affected members can have different MPDs (PV and ET, PV and CML, etc.).³  Furthermore, in sporadic PV, not all of the clonal PV cells are JAK2 V617F-positive.  Erythropoietin-independent erythroid colonies (BFU-Es) — a hallmark of PV — are mostly homozygous for JAK2 V617F, but some are heterozygous and others have no mutation.⁴  Importantly, in acute leukemic transformation of JAK2 V617F MPDs the blast cells are frequently negative for the mutation.^5,6  These diverse observations strongly suggest that somatic mutation of the JAK2 gene is neither the initiating nor sole pathogenic process in PV.

Three papers published simultaneously in the March issue of Nature Genetics provide an important new contribution to our understanding of PV. Remarkably, by different routes, groups led by Nicholas Cross in the UK, Robert Kralovics in Vienna, and Ross Levine in New York described the same principal finding that specific germline haplotypes of the JAK2 locus in chromosome 9p were associated with the presence of the JAK2 V617F somatic mutation in MPD cells, but not necessarily with the origin of the first clonal event of these MPDs. The haplotypes were not associated with increased JAK2 transcript levels or with augmented erythroid proliferation in vitro, suggesting that by themselves they do not contribute to the genesis of PV. Interestingly, the Cross group was able to study a family with two affected members, both having the JAK2V617F mutation, and found that one was on a specific haplotype that favors JAK2 V617F somatic mutation while the other was not. This is consistent with a previous report showing absence of linkage of familial PV to 9p.²  Kralovics and his colleagues also studied the genomic composition of individual erythroid colonies from MPD subjects and found that occurrence of the JAK2V617F mutation was independent of haplotype. Unlike the other two teams, Levine’s group reported that a specific JAK2 haplotype contributes significantly to the excess risk of JAK2V617F-positive familial MPDs, a statement needing independent confirmation.