Study Identifies Germline Variants That Predispose Patients to Developing Myeloproliferative Neoplasms

Though the JAK2 V617F mutation is present in the majority of patients with myeloproliferative neoplasms (MPNs) like polycythemia vera (PV) and primary myelofibrosis (PMF), there are additional molecular abnormalities that contribute to an individual’s susceptibility for developing MPN, according to a genome-wide association study (GWAS) published in Blood.

Authors, led by David A. Hinds, PhD, from the DNA genetic testing and analysis company 23andMe, Inc., identified four genes – JAK2, SH2B3, TET2, and CHEK2 – wherein both inherited variation and somatic mutation contributed to V617F clonal hematopoiesis and/or MPNs.

“We have identified inherited loci … that predispose to both age-related JAK2 V617F clonal hematopoiesis in the general population, as well as MPN independent of V617F status,” Dr. Hinds and co-authors wrote. “These genes impact diverse biologic pathways such as cellular aging (TERT), JAK-STAT signaling (JAK2, SH2B3), epigenetic regulation (TET2), and DNA damage repair and/or tumor suppressor function (CHEK2).”

Patients with MPNs – including PV, PMF, essential thrombocythemia (ET), post-PV/ET myelofibrosis, systemic mastocytosis, chronic myelogenous leukemia, chronic eosinophilic leukemia, and hypereosinophilic syndromes – were invited to participate in a free online MPN research initiative between August 2001 and December 2013. Participants (n=1,451) were given free 23andMe Personal Genome Service (PGS), in which customers submit a saliva DNA sample, which is genotyped using a genome-wide single nucleotide polymorphism (SNP) array that conducts ancestry analyses and reports on health-related genetic information.

An additional cohort of 23andMe PGS customers who agreed to participate between January 2011 and November 2013 served as population controls (n=252,637). A third replication cohort consisted of 23andMe customers enrolled between January 2014 and February 2015 (n=169,467).

DNA extraction and genotyping of participants’ saliva samples were performed by the National Genetics Institute, using a SNP array platform with custom probes for the V617F somatic mutation.

Participants self-reported phenotype data through web-based background surveys that included information about the specific diagnosis, treatment, and disease status. The researchers assessed the presence of V617F by comparing mutation detection from the SNP arrays with participants’ self-reported V617F status.

A total of 726 unrelated individuals (all of whom had >97% European ancestry, according to ancestry analysis) were selected from the original MPN participant group, with the following self-reported diagnoses:

  • ET (n=246)
  • PV (n=258)
  • PMF (n=37)
  • Related/overlapping MPN diagnoses (n=135)

Of the population controls, 497 individuals had evidence of V617F mutation. The prevalence of V617F varied from 0.03 percent among those <30 years to 0.46 percent among those >60 years. These participants tended to be older than those in the original MPN cohort (70% vs. 31% were >60 years).

“The 0.2 percent prevalence rate of JAK2 V617F in the control cohort is roughly double the estimated age-adjusted prevalence rate of combined JAK2 V617F-positive MPN in the United States based on data from two large health-care plans,” Dr. Hinds and colleagues explained, noting that the “excess prevalence of JAK2 V617F clonal hematopoiesis among unselected controls likely represents a combination of persons with an undiagnosed MPN, individuals who may develop a future MPN, and those who will never develop a hematologic disorder.”

An additional 342 other self-reported phenotypes were assessed for V617F carrier status in the population cohort, adjusting for age, gender, and genetic principal components. “A few of these control individuals, while not enrolled in the MPN research initiative, independently reported having an MPN,” the authors reported. “This was strongly associated with V617F carrier status (odds ratio [OR] = 2.38, P=5.6×10−27).” V617F carriers also reported more blood clots (OR=1.8; p=0.014), less anemia (OR=0.45; p=0.019), and higher prevalence of stroke (OR=1.9; p=0.03), though these associations were not significant after adjustment.

The researchers compared the 726 MPN cases with the total control group, and the 497 V617F carriers in the control population with the remaining 252,150 V617F non-carrier controls. In both analyses, there was a strong association in the 9p24.1 region around JAK2 and 5p22 region in TERT. In addition, SH2B3 was suggestive (p<10-6) in the MPN analysis and showed genome-wide significance in the analysis of V617F carriers among the control group.

“Motivated by the consistency of results from these two analyses,” Dr. Hinds and colleagues then performed a combined GWAS of the 726 MPN cases plus the V617F carriers in the control population (n=1,223), compared with the remaining V617F non-carrier controls (n=252,150).

The strongest signal was in the germline JAK2 46/1 haplotype (rs59384377; OR=2.46; p=6.6×10–89), which has previously been associated with V617F-positive MPN. This variant had the strongest association and the highest prevalence among patients with PV, but the smallest association and the lowest prevalence in ET.

Genome-wide significant associations were also identified in:

  • TERT variant rs7705526 (OR=1.8; p=1.1×10–32)
  • SH2B3 variant rs7310615 (OR=1.4; p=3.1×10−14)
  • upstream TET2 variant rs1548483 (OR=2.0; p=2.0×10−9)

These associations were confirmed in an analysis of the 446 JAK2 V617F carriers with the 169,021 V617F non-carriers in the replication cohort.

Then, in a joint analysis of the combined GWAS and replication results, Dr. Hinds and colleagues identified additional genome-wide significant predisposition alleles associated with CHEK2 (OR=4.4; p=7.5×10-8) and ATM (OR=2.2; p=6.5×10-7), which was consistent in both MPN cases and V617F carriers. Furthermore, all SNP odds ratios were similar for MPN patients and for controls who were V617F carriers.

“These data indicate that the same germline variants not only endow individuals with a predisposition to MPN, but also to JAK2 V617F clonal hematopoiesis, a more common phenomenon that may foreshadow the development of an overt neoplasm,” the authors concluded.

The use of saliva samples rather than blood samples is a limitation of this study, because the variable DNA composition between saliva and blood samples could have affected the sensitivity of the assay. “Although we do not expect this test to match the sensitivity of targeted blood-based assays, it may remain attractive as a screen followed by secondary testing,” they wrote.

The study was sponsored by 23andMe, Inc.


Hinds DA, Barnholt KE, Mesa RA, et al. Germline variants predispose to both JAK2 V617F clonal hematopoiesis and myeloproliferative neoplasms. Blood. 2016 June 30. [Epub ahead of print]