Introduction: Myelofibrosis (MF) is a rare myeloproliferative neoplasm (MPN) characterized by bone marrow fibrosis, progressive bone marrow failure, and increased risk of acute myeloid leukemia. While MF arises from somatic driver mutations in JAK2, MPL, and CALR, some MPN patients may have a heritable component. To comprehensively examine the genetic etiology of MF, we performed the first integrative analysis of SNP array genotyping (using Infinium Global Screening Array), targeted long-read sequencing (using PacBio SMRT sequencing) and telomere length (TL, using qPCR assay).

Methods: Our study included 937 MF patients who received an allogeneic hematopoietic cell transplant (HCT) between 2000 and 2016 and had an available pre-HCT blood sample at the Center for International Blood and Marrow Transplant Research Repository. Somatic mosaic chromosomal alterations (mCAs, including deletions, duplications, or copy-neutral losses-of-heterozygosity (CNLOH)) were called with the Mosaic Chromosomal Alteration (MoChA) algorithm using raw genotyping intensity data. A genome-wide association study (GWAS) was restricted to include 827 MF patients of European ancestry and utilized 4,135 genetically-matched healthy controls.

Results: GWAS identified six independent MF susceptibility loci at genome-wide significance (P< 5×10 -8); four of which replicate prior MPN susceptibility loci [9p24.1(JAK2), 5p15.33(TERT), 3q25.33(IFT80), and 4q24(TET2)] and two novel MF loci [6p21.35(HLA-DQB1-AS1) and 17p13.1(TP53)] (Figure 1). A transcriptome-wide association analysis using whole blood GTEx data highlighted the 9p24.1 locus with increased JAK2 expression associated with elevated risk of MF (P= 2.18×10 -19). A strong colocalization statistic further indicated shared genetic component between eQTL and this JAK2 locus (HyPrColoc Posterior Probability= 0.6) (Figure 2).

Based on the strong signal identified at TERT (Figure 1), we investigated the relationship between MF risk and genetically-inferred telomere length using a panel of 19 germline variants previously found to be associated with telomere length. Of the 19 telomere-length associated variants investigated, 7 were found to be associated with MF risk (binomial P= 2.31×10 -5, linear trend P= 5.48×10 -4) (Figure 3). Both Mendelian randomization and genome-wide genetic correlation analyses further indicated that increased risk of MF was associated with longer inherited telomere length.

Utilizing available clinical mutation data on a subset of 185 patients, MF cases carrying the germline risk haplotype of the 9p24.1(JAK2) susceptibility locus were observed to more frequently have the JAK2 V617F mutation (71% vs 59%; P= 0.02). Targeted PacBio long-read sequencing around JAK2 provided further evidence of linkage between the germline risk allele and the JAK2 V617F mutation. Detectable autosomal mCAs were also abundant in MF cases with 67.4% having at least one mCA (compared to ~3% in the general population) and 27.6% having an mCA spanning JAK2 (mostly CNLOH) (Figure 4). In addition, using a binomial test for biased allelic imbalance, a cis relationship was identified at 9p24.1 in which the MF risk haplotype was predominantly duplicated by CNLOH (binomial P=1.36×10 -9). Regional sequencing of JAK2 further confirmed duplication of JAK2 V617F by CNLOH. Finally, we observed an inverse association between autosomal mCAs and qPCR measured telomere length (OR= 0.22, 95% CI= 0.07-0.65, P= 6.40×10 -3). These results were consistent by mCA chromosomal region and copy number state.

Conclusion: Our results suggest a molecular framework for the genetic etiology of MF in which both genetically-inferred telomere length and germline variation at JAK2 are associated with increased MF risk. The 9p24.1 risk haplotype predisposes to the acquisition of a somatic JAK2 V617F mutation in cis and subsequent duplication of JAK2 V617F by mCAs (usually CNLOH). This process leads to aberrant JAK2 activity and increased clonal proliferation, accelerating telomere length shortening and increasing genomic instability in patients with MF.

Disclosures

Gupta:AbbVie: Consultancy, Honoraria; Constellation Pharma: Consultancy, Honoraria; Roche: Consultancy; Pfizer: Consultancy; BMS-Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Sierra Oncology: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Incyte: Honoraria, Research Funding. Lee:Janssen: Other; Incyte: Research Funding; AstraZeneca: Research Funding; Kadmon: Research Funding; National Marrow Donor Program: Membership on an entity's Board of Directors or advisory committees; Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding; Pfizer: Research Funding; Syndax: Research Funding; Takeda: Research Funding; Amgen: Research Funding. Saber:Govt. COI: Other.

Author notes

 This icon denotes a clinically relevant abstract

Sign in via your Institution