IPSS and JAK2 Status Could Predict Thrombosis Risk in Primary Myelofibrosis

Analysis shows that International Prognostic Scoring System score and the presence of a JAK2 mutation was linked to risk of thrombosis in patients with primary myelofibrosis. Investigators observed a benefit for ruxolitinib in patients.

Risk of major venous and arterial thrombosis may be identified based on International Prognostic Scoring System (IPSS) score and JAK2 mutation status of patients with primary myelofibrosis (MF), which can inform anti-thrombotic treatment, according to findings from the risk-stratification ERNEST study (NCT04153305) displayed in a poster presentation during the 2021 ASH Annual Meeting and Exposition.

Patients who had the highest risk for thrombosis also had a low/intermediate-1 IPSS score and a JAK2 mutation.1 Moreover, investigators observed a trend that indicated a benefit for the use of ruxolitinib (Jakafi) compared with hydroxyurea to reduce vascular risk, which they recommended to be explored further in larger studies.

Patients enrolled in the trial with primary and secondary MF and who received treatment for MF, including ruxolitinib, were evaluated. In the cohort (n = 1010), the mean age was 63.7 years with 60% males. A breakdown of MF type revealed 58% of patients with primary MF, 22% with post-polycythemia vera (post-PV) MF, and 20% with post-essential thrombocytopenia (post-ET) MF.

Patients were stratified using IPSS and MYSEC, a prognostic model used to predict survival in secondary MF. In the low category, 16.3% were classified according to IPSS, and 37.9% were classified by MYSEC. In the intermediate-1 category, 33.5% were classified according to IPSS, and 44.8% were classified according to MYSEC. In the high category, 23.3% were classified according to IPSS, and 2.6% were classified by MYSEC. In the intermediate-2 category, 26.9% were classified according to IPSS, and 14.7% were classified according to MYSEC.

When the types of treatments were reviewed, 40.8% of patients received no treatments, and 48.2% had received only hydroxyurea. Among patients who were treated with ruxolitinib, 6.8% were previously treated with hydroxyurea, and 3.9% received ruxolitinib alone. Only 0.3% of patients had received interferon.

A total of 108 cases (10.7%) of thrombosis were reported, according to lead author Tiziano Barbui, MD, of the Foundation for Research Ospedale Papa Giovanni XXIII in Bergamo, Italy. Overall, there were 50 cases (46.3%) of arterial thrombosis and 58 cases (53.7%) of venous thrombosis.

“The incidence of stroke [24.0%] and transient ischemia attacks [18.0%] were notable in patients with arterial thrombosis,” Barbui said. “Among patients who experienced venous thrombosis, 19.0% experienced splanchnic thrombosis,” he continued. In addition, 34.5% reported deep venous thrombosis.

Investigators reported 59 (10.1%) cases of thrombosis among patients with post MF, 22 (10.6%) cases of thrombosis among patients with post ET MF, and 27 (12.3%) cases among patients with post PV MF. The incidence rate (95% CI) among patients with post MF, post ET MF, and post PV MF was 1.91% (1.4%-2.49%), 1.60% (0.98%-2.60%), and 2.79% (1.90%-4.10%), respectively.

Univariate analysis revealed that age was a factor in primary thrombosis risk (P = .026), as was low IPSS score (P = .020) and JAK2-positivity (P = .002), said Barbui. “Interestingly, among patients with primary MF, those patients in the low and intermediate-1 categories had a 63% risk of thrombosis, but those in the high or intermediate-2 risk only had a 37% risk. We failed to find significant thrombosis predictors in post ET and post PV MF and that is why we focused on primary MF,” Barbui said.

When reviewing the cumulative incidence function for IPSS and JAK2-positivity, investigators reported a 13% incidence for the low/intermediate-1 category and 7.5% incidence in patients in the high/intermediate-2 category. Similarly, the incidence among patients with JAK2-positive disease was 14% compared with 4.5% among patients who were JAK2-negative.

Having a JAK2 mutation was confirmed as an independent risk factor for thrombosis according to multivariate analysis, said Barbui (subdistribution hazard ratio [sHR], 3.12%; 95% CI, 1.40%-6.94%; P = .005). He noted that the low/intermediate-1 IPSS categories increased the risk of thrombosis but was not statistically significant.

Regarding treatment effects, multivariate analysis indicated that the use of ruxolitinib showed a non-significant trend towards a protection of approximately 67% compared with hydroxyurea (sHR, 0.33%; 95% CI, 0.08%-1.32%, P = .117).

Low IPSS score, intermediate-1 risk, and JAK2 mutation suggest that the clonal cellular phase of myelofibrosis is a risk factor for thrombosis. Further evaluation of the role of rituximab is also warranted.


1. Barbui T, Ghirardi A, Carobbio A, et al. The interaction between IPSS score and JAK2 mutation identifies patients at different vascular risk in primary myelofibrosis. Presented at: 2021 ASH Annual Meeting and Exposition; December 11-14, 2021; Atlanta, GA. Abstract 236