The JAK2 46/1 Haplotype Analysis In Essential Thrombocythaemia and Polycythaemia Rubra Vera Reveals That CC Genotype Is Associated with a Higher JAK2V617F and c-MPL W515 Allele Burden

Genomic-wide association studies have identified the germline 46/1 haplotype as a predisposing allele associated with JAK2V617F positive myeloproliferative neoplasms (MPN). The present study analysed data on 856 JAK2V617F positive patients, 326 of which had complete clinical data.

To evaluate the JAK2 46/1 haplotype frequencies, JAK2V617F allele burden, c-MPL 515 mutation and risk of transformation.

Genomic DNA from whole peripheral blood or bone marrow patient samples was analysed as follows: JAK2V617F allele burden by Q-PCR, JAK2 exon 12 mutations by Q-PCR and PCR fragment analysis, MPL W515 L and K mutations by allele specific PCR. The 46/1 JAK2 mutation susceptibility haplotype (46/1) tagging SNP rs12343867 (susceptibility allele C) were analysed by pyrosequencing.

The allele frequency for the 46/1 tag SNP rs1234867 in the 856 patients was calculated for the total JAK2V617F cohort (0.48) and the clinical entities ET (0.34) and PRV (0.44) confirming that the 46/1 haplotype is greatly over represented in JAK2V617F MPD patients as compared to published the control population (Wellcome Trust Case Control Consortium (WTCCC) (0.24).

The Analysis of the 856 patients demonstrated that JAK2V617F and c-MPL W515L/K mutations co-existed in 16 patients(1.9%), the incidence of c-MPL W515L being twice as common as the c-MPL W515K mutations. There was no correlation between these mutations and age or 46/1 haplotype status. The JAK2V617F allele burden (AB) was lower in the c-MPL mutant patients, the average JAK2AB 31%. 3 out 4 c-MPL patients for which clinical information was available had a diagnosis of ET. No JAK2 exon 12 mutations were found in any of the 859 JAK2V617F positive samples suggesting that co-existing JAK2 exon 14 and exon 12 mutations are extremely rare.

The genotypic data in ET patients showed: C/C 12%, C/T 44%, T/T 44% and their respective JAK2V617 allele burden (AB) were 46%, 32%, 29%. The genotype data in PRV patients: C/C 18%, C/T 53%, T/T 28.6% and their respective AB were 47%, 31% and 39%. The median AB was 32% (n=121) for ET and 37% (n=103) for PRV.

Within a cohort of 255 patients (ET=138, PRV=117) 4% of ET and 6% of PRV patients transformed to acute myeloid leukaemia or myelofibrosis with no predominant haplotype association. In the ET patients, the median AB was 35%, there was no significant difference in the JAK2 V617F AB between those who transformed or not (p=0.45). Interestingly, on the whole ET group C/C genotype patients were more likely to have an allele burden >50% (p=0.058). In the PRV patients, the median AB was 48%. Again, the C/C genotype, PRV patients were more likely to have an AB>50% (p=0.06), although not reaching statistical significance.

The 46/1 haplotype in both clinical entities ET and PRV demonstrated a higher allele burden in the C/C genotype in comparison to the other genotypes. No predominant haplotype predicted the risk of transformation to a more aggressive disease such as MF or AML. The analysis also showed that c-MPL W515K/L mutations can co-exist with JAK2V617F. The c-MPL W515K/L mutations did not exhibit a positive correlation with a preferential 46/1, but was associated with a lower allele burden. No co-existing exon 12 and exon 14 mutations were found, suggesting the rarity of this occurrence.

No relevant conflicts of interest to declare.