Effect of JAK2 V617F on thrombotic risk in patients with essential thrombocythemia: measuring the uncertain

Current data about thrombotic risk in ET patients harboring the JAK2 V617F mutation remain partially inconclusive.1, 2 A systematic literature review of MEDLINE up to February 2008 to identify studies of ET in the JAK2 era was conducted using the following search algorithm: JAK2 AND (essential OR thrombocytosis OR thrombocythemia OR thrombosis). All searches were limited to studies of humans published in English. A manual search of abstracts was initially conducted and relevant studies were retrieved in full text. In addition, a manual review of references was carried out to identify any additional relevant articles. To be included in the analysis, studies had to report the prevalence of thrombosis in JAK2 V617F patients and in wild-type carriers with ET.

Weighted averages were reported as Odds Ratios (ORs) along with their 95% Confidence Intervals (95%CIs) to quantify the effect of JAK2 positivity on the thrombotic risk in each study. Major thrombotic events were extracted, including strokes and transient ischemic attacks, myocardial infarctions and angina pectoris, peripheral artery occlusion, deep vein thrombosis and pulmonary embolism.

Pooled ORs were calculated according to the Mantel-Haenszel method for fixed effects (FE) and DerSimonian-Laird for random effects (RE). Statistical heterogeneity was measured using the χ Q test (p < 0.10 is considered representative of significant statistical heterogeneity) and the I2 statistic, as previously described.3 To establish the effect of clinical heterogeneity between studies, subgroup analysis was performed. Although the selection of a random- vs. fixed-effects model remains controversial, a fixed-effects model appears more appropriate whenever heterogeneity is limited.

A total of 492 relevant studies were initially retrieved. Among them, 17 studies (see Online Supplementary Appendix) met the inclusion criteria. Incidence figures for thrombosis vary from 17% to 43%, and JAK2 V617F positivity varies from 37% to 71%. A significant association of JAK2 mutation with thrombosis was evident in half of these studies whereas no such correlation was documented in the remaining studies (Table 1A).

Meta-analysis of 2,905 patients with ET and 778 patients with thrombosis (Table 1B), showed JAK2 V617F patients have a two-fold risk of developing thrombosis (OR_RE 1.84, 95%CI 1.40–2.43) with significant heterogeneity between studies (I2 = 42.5%).

The statistical heterogeneity reported should be considered a reflection of clinical heterogeneity between different study populations, type (prospective vs. retrospective) and variability in follow-up. Moreover, JAK2 V617F patients are older at diagnosis, have higher hemoglobin levels, higher leukocyte counts and lower platelet counts.2, 4–6 Leukocytosis is regarded as an additional factor for thrombosis whereas thrombocytosis is not,5, 6 and additional evidence is provided to support the role of granulocytes in MPD-associated thrombosis.7 The above observations are consistent with the superior effectiveness of hydroxyurea (a non-specific myelosuppressive agent) compared to anagrelide (a platelet-specific cytoreductive agent) in high risk ET8,9 and supported by the lack of an increased risk of thrombosis associated with extreme thrombocytosis in otherwise low-risk ET.10 Age is a well-established confounder in thrombotic risk, and cardiovascular risk factors may vary between study groups. However, 2 recent studies have demonstrated that the presence of cardiovascular risk factors does not modify thrombotic risk in patients with ET who experience first-time thrombosis.11, 12 In fact, recurrent thrombosis is again predicted by age (>60 years) and thrombosis history12 factors with well-established prothrombotic effect in ET.6, 11

Finally, the allele burden of the mutated JAK2 gene, the effect of which cannot be estimated, may account for the diversity between studies. Results so far remain contradictory.2, 13

Table 1A.Characteristics of studies included in the analysis.

Given the exaggerating effect of smaller studies, larger series (>100 patients) were analyzed separately (8 studies, 2,394 patients, 627 with thrombosis) and the effect remained significant (OR_FE 1.77, 95%CI 1.46–2.15) with no evidence of variability between studies (I2 = 0), suggesting a lack of true clinical heterogeneity.

Table 1B.Pooled effect of JAK2 V617F positivity on thrombotic risk in 2905 patients with ET. The squares and lines show the estimated Odds Ratios and their 95% CIs. The size of each square is proportional to the amount of information (weight) available in the subgroup. Overall estimates are shown by a diamond, with the width representing the 95% CI.

JAK2 V617F was also associated with an increased risk of venous (OR_FE 2.09, 95%CI 1.44–3.05), arterial thrombosis (OR_FE 1.96, 95%CI 1.43–2.67), and for thrombosis at presentation (OR_FE 1.88, 95%CI 1.38–2.56), effects without significant heterogeneity (Supplementary Online Table S2 A,B,C). A history of thrombosis is a well-defined risk factor for recurrent thrombosis in ET, and on clinical grounds this association may contribute to the increased risk of thrombosis thereafter.

This analysis represents the cumulative evidence on JAK2 association with thrombosis in ET, with all its inherent biases and weaknesses. Therefore, this study is hypothesis-generating but cannot prove direct causality. Results should be interpreted with caution, as it is still unclear if this particular association of JAK2 V617F with thrombosis is independent of other confounding factors with known significant effect. The effect on JAK2 mutation is probably mediated through a distinct prothrombotic phenotype, with a predilection for both venous and arterial events, that includes leukocytosis, older age and thrombosis at presentation, features that are well-established risk factors of thrombosis even in the pre-JAK2 era.