TET2 mutations predict thrombotic risk in patients with polycythemia vera

Cardiovascular events, including thrombosis are a common complication in patients with myeloproliferative neoplasms (MPN) like polycythemia vera (PV) and essential thrombocythemia (ET).¹ Approximately, 34-39% of patients with PV will develop an adverse cardiovascular event during the course of the disease, worsening their prognosis.¹ Currently, the risk of thrombosis is stratified according to patient age (higher risk in > 60 years) and history of thrombotic events in patients with PV.² For ET the driver Janus kinase 2 (JAK2) V617F mutation is also included in thrombotic risk stratification due to its association with an increased risk of thrombosis.²

Non-driver mutations in the DNA-methyltransferase 3A (DNMT3A), tet methylcytosine dioxygenase 2 (TET2), and the additional sex combs like 1 (ASXL1) genes, collectively known as DTA genes, have been associated with clonal hematopoiesis (CHIP) which increases the risk of developing coronary disease and myocardial infarction.¹ Nevertheless, there is limited data on the role of DTA mutations in the prediction of thrombotic risk in patients with MPN. For this, Adrián Segura-Díaz and colleagues from the Hospital Universitario de Gran Canaria, Gran Canaria, ES, used next generation sequencing (NGS) to investigate whether pathogenic non-driver mutations can be used as predictors of thrombosis in MPN patients.¹ The results of this study were published in Cancers and are summarized below.¹

Study design

Initial MPN cohort

• NGS of samples from N = 68 patients with the following diagnoses:
  • PV: n = 16
  • ET: n = 25
  • Primary myelofibrosis (MF): n = 16
  • Secondary MF: n = 11
• Median patient age was 68 years (range, 43-90) and 38.2% were males

Case-control PV study

• N = 55 age-matched patients with PV, comprising 25 patients with, and 30 patients without thrombotic events. Twelve patients were also included in the first analysis described above

Results

Initial MPN cohort

• In a median follow-up of 50 months (range, 1-200), thrombotic events after diagnosis occurred in 32.4% of the patients. These corresponded to:
  • Arterial thrombosis: 48% (n = 12 events)
  • Venous thrombosis: 52% (n = 13 events)
• Thrombotic events based on diagnosis occurred in:
  • 43.8% of PV patients
  • 40.0% of ET patients
  • 18.5% of MF patients (primary and secondary)
• NGS revealed that pathogenic non-driver mutations in genes other than JAK2, calreticulin (CALR) or MPL were observed in 48.5% of patients with a mean number of 0.63 mutations/patients and with 11% of patients harbouring more than one
  • The most common non-driver gene mutations were in DTA genes:
    • DNMT3A: 14.0%
    • TET2: 32.6%
    • ASXL1: 14.0%
• Mutations in DTA genes were observed in:
  • 62.5% of patients with PV
  • 12% of patients with ET
  • 56.3% of patients with primary MF
  • 45.5% of patients with secondary MF
• When looking at all MPN patients as a whole, no association between DTA mutations and thrombosis was observed
• When examining each disease separately, the presence of DTA mutations was significantly associated with thrombosis in patients with PV (p = 0.031; Pearson’s chi-square test) with 85.7% of patients with PV and thrombotic events harbouring at least one DTA mutation
• Secondary MF patients with DTA mutations also had an increased risk of thrombosis (p = 0.028; Fischer’s exact test) with 87.5% of them having a thrombotic event and a concomitant DTA mutation

Case-control PV study

• The DTA mutation-thrombosis association in PV was further confirmed in a case-control study with 55 age-matched patients
• The number of patients with thrombotic events and DTA mutations are shown in Table 1
  • In summary, NGS verified the significant association of DTA mutations and thrombosis in these PV patients (p= 0.0024; 95% CI, 2.02-22.32; Fischer’s exact test)
• TET2 mutations alone were significantly associated with thrombotic events (p= 0.031; OR= 3.56; 95% CI, 1.15-11.83) with 65% of patients with a thrombotic event harbouring a TET2 mutation
  • When TET2 mutations were removed from the logistical regression analysis, there was no significant association between ASXL1 and DNMT3A mutations and thrombosis in patients with PV

Table 1. Frequency of thrombotic events in patients with DTA and non-DTA mutated polycythemia vera¹

Conclusions

This NGS study revealed that pathogenic mutations in DTA genes (DNMT3A, TET2, ASXL1) are significantly associated with the occurrence of thrombotic events in patients with PV. More specifically, the case control study suggested that TET2 mutations specifically are responsible for this association in patients with PV. The authors suggest that mutational analysis in DTA genes should be considered in current risk algorithms at diagnosis to identify patients with PV at higher risk of developing thrombosis.