Assessment of Systematic JAK2V617F Screening in 996 Patients With Venous Thromboembolism

JAK2V617F mutation induces constitutive activation of the JAK–STAT signaling pathway. It is the most prevalent molecular driver of BCR::ABL1 negative myeloproliferative neoplasms (MPN). Thrombotic events constitute the main complications of MPN, contributing to both morbidity and mortality [1], with a risk of arterial and venous thromboembolism (VTE) increased by 3-fold and 10-fold, respectively [2]. The presence of JAK2V617F mutation is associated with a higher thrombotic risk in this population [3]. It is recommended to systematically perform a blood count in patients with a history of VTE requiring a thrombophilia assessment. In cases of splanchnic thrombosis (ST) with no obvious cause, MPN driver mutations should be screened. In such patients, an MPN is diagnosed in 30%–40% of cases [4]. Importantly, JAK2V617F mutation can also be identified in individuals without overt MPN. It represents one of the most common clonal hematopoiesis of indeterminate potential (CHIP) [5]. Previous studies suggested a higher risk of thrombosis in these patients [6]. In the present work, we evaluated the role of the JAK2V617F mutation in the management of patients with a history of VTE. We conducted 1/a retrospective study of patients tested for the JAK2V617F mutation over a 10-year period to describe their clinical profiles; 2/a prospective monocentric study in which we systematically tested patients with VTE for the JAK2V617F mutation. For the retrospective study, patients were tested for JAK2V617F mutation using quantitative polymerase chain reaction (qPCR) in the Molecular Biology Laboratory of Marseille between January 2010 and December 2020. All patients had at least one blood count with or without bone marrow evaluation. Patients were classified as having MPN based on the 2022 WHO diagnostic criteria [7] (Supplementary method). After exclusion of patients presenting other myeloid hematological malignancies, patients who did not meet the diagnostic criteria for MPN were classified as having a normal blood count (NBC) (Supplementary method). Regarding the prospective phase, we included all consecutive patients aged over 18 referred to our Reference Center for Thrombophilia of Marseille between January 2022 and December 2023, with a history of VTE. VTE events were documented through spiral computed tomographic scanning angiography, ventilation/perfusion lung scan, duplex ultrasound, and/or venography. We performed conventional thrombophilia screening and JAK2V617F mutation screening by qPCR in all included patients (Supplementary method). The study was approved by the local Ethics Committee at APHM Hospital, Marseille, France (PADS24-88_dgr). In total, 5357 patients have been tested for the JAK2V617F mutation, and 733 (13.7%) were positive. Among them, 674 (92.0%) were diagnosed with MPN, and 59 patients (8.0%) were NBC (Figure S1). Patients positive for the JAK2V617F mutation were older than JAK2 wild-type (WT) patients (65.4 vs. 55.9 years, p < 0.001). In JAK2V617F positive patients, as compared to MPN patients, NBC patients were younger (56.1 vs. 66.3 years, p < 0.001) and had lower variant allele frequency (VAF) (5.3 vs. 22.6%; p < 0.001) (Table S1). Thrombosis was, by far, the most frequent clinical indication for JAK2 testing in NBC patients as it was observed in 57.6% of cases (n = 34) (Table S2). Two thirds of thrombosis events were venous. ST was the most frequent venous location (n = 11; 45.8%). Interestingly, deep vein thrombosis (DVT) of the lower limb and/or pulmonary embolism (PE) was the second most frequent location (n = 10; 41.7%). In patients with arterial thrombosis (n = 10), stroke was, by far, the first location (n = 7; 70.0%). Median VAF was below 2% in all groups but ST (6.20%) (Table S3). During the prospective phase, we included 996 patients with VTE: 481 with isolated DVT, 434 with PE (± DVT), 41 with cerebral venous thrombosis (CVT), and 40 with ST (Table 1). JAK2V617F mutation was detected in 3.8% of patients (n = 38). Among these, 10 were diagnosed with MPN, and 28 remained classified as having a NBC (Table 1 and Figure S1). Median age was similar in NBC and MPN patients: 58 versus 57 years (p = 0.58). However, the VAF was lower in NBC patients: 0.13% versus 3.87% (p = 0.02). In NBC patients, 6 (21.4%) presented with VAF higher than 2%. The most common sites of thrombosis in NBC patients were PE (46.4%, n = 13) and DVT (35.7%, n = 10), while in MPN patients, DVT (50%, n = 5) and ST (30%, n = 3) predominated. Notably, the proportion of ST was higher in MPN patients compared to JAK2-WT and NBC patients (30.0% vs. 3.5% and 10.7%, respectively) (Table 1). JAK2V617F mutation was identified as the third most frequent thrombophilia defect (n = 38, 3.8%) following factor V Leiden (FVL) (n = 120, 12.0%) and G20210A prothrombin mutation (PTM) (n = 73, 7.3%). This held true across all thrombosis sites except ST, where it was the most frequent defect (n = 6, 15.0%) (Figure 1 and Table S4). Of note, in NBC patients, 24 (85.7%) had no other thrombophilia abnormalities, making JAK2V617F mutation an isolated finding (Table 1). To our knowledge, this is the first study to evaluate the systematic screening for JAK2V617F mutation in patients presenting with VTE. We report a prevalence of 3.8% positioning JAK2V617F mutation as the third most common thrombophilia defect after FVL and the PTM. In 85.7% of NBC patients, JAK2V617F mutation was the only thrombophilia defect, suggesting that it may represent an independent risk factor for VTE. This hypothesis is supported by a large study involving almost 50,000 individuals from the general population that demonstrated a higher rate of VTE (odds ratios (OR) = 3.1) in individuals with JAK2V617F-associated clonal hematopoiesis [8]. Moreover, a recent study of a large UK cohort reported a strong association between JAK2V617F mutation and VTE, with an OR of 6.58 [9]. Most NBC patients had VAF below the CHIP threshold of 2%: 55.9% and 78.6% in the retrospective and prospective studies, respectively (Tables 1 and S1). This result raises the question of the clinical relevance of a low VAF and its potential contribution to the thrombotic phenotype. A previous study has reported an increased risk of VTE (OR = 2.8) in patients without MPN when using a 1% threshold [10], suggesting a potential implication of JAK2V617F even for low VAF. To address this issue, we tested a population of 50 control patients, with no history of thrombosis, matched in age and sex with the NBC patients with thrombosis. No JAK2V617F mutation was found in the control group at the 0.05% threshold (data not shown). Moreover, we performed extended molecular profiling by sequencing 24 genes (Table S5) in 56 NBC patients from both retrospective and prospective studies. It is well known that the presence of CHIP in individuals without MPN, as well as the acquisition of additional mutations in patients with overt MPN, may modulate thrombotic risk. Some studies have reported an increased prevalence of CHIP in DNMT3A, TET2, or ASXL1 mutations in patients with unprovoked VTE [9]. Most NBC patients (78.6%) harbored no additional mutations (Table S6). This finding further supports the role of JAK2V617F mutation in the thrombotic phenotype of NBC patients. We predominantly found additional mutations in two genes, TET2 and DNMT3A, which together accounted for 66.6% (10 out of 15) of the identified mutations. In the retrospective study, we detected JAK2V617F mutation in 11 NBC patients with ST (Table S3). Despite normal blood counts, one might question whether these patients were truly NBC, as isotopic blood mass was not routinely performed at the time of the retrospective study. In cases of ST, splenic sequestration could obscure the diagnosis of true MPN, particularly masked Polycythemia Vera. This could explain the higher VAF observed in ST as compared to other thrombotic sites. The high prevalence of JAK2V617F mutation among VTE patients, along with prior evidence of its strong association with VTE, raises the question of whether this mutation should be systematically included in thrombophilia screening—regardless of blood counts or thrombotic site. Further studies are needed to evaluate both the cost-effectiveness of this strategy and the risk of recurrent VTE in NBC patients. In the absence of active cancer, discontinuing anticoagulant therapy after 3 to 6 months is a possible strategy. However, the risk of VTE recurrence in the context of CHIP remains unknown. Follow-up studies in NBC patients would be valuable to estimate this risk. In addition, the rate of progression to MPN in patients with low VAF should be addressed in follow-up studies. Whether the early identification of JAK2V617F mutation could improve patient management by facilitating the early diagnosis of MPN remains to be determined. In cases of high progression rates among patients with a low VAF, mutation screening may therefore require the use of a highly sensitive detection method (i.e., qPCR over NGS). In conclusion, the notable prevalence of the JAK2V617F mutation in patients with VTE raises the possibility that screening for this mutation might be a valuable addition to thrombophilia assessment, regardless of blood count and thrombosis location. Further follow-up of these patients could provide useful insights into the potential risk of VTE recurrence and progression to MPN. N.A. performed the experiments. N.A., J.Q., and A.T. analyzed the data. N.A., J.Q., G.V., P.S., and P.-E.M. wrote the paper. G.V., P.S., R.C., J.G., and P.-E.M. designed the concept of the work and supervised analyses and writing. R.A. helped recruit control patients. All authors read, corrected, and approved the final manuscript. The authors would like to thank all the staff of the clinical and biological hematology departments and the molecular biology department of Marseille University Hospitals. Informed written consent was obtained in accordance with the Declaration of Helsinki. The study was approved by the local Ethics Committee at APHM Hospital, Marseille, France. The authors declare no conflicts of interest. Data can be available on demand. Contact: [email protected]. Data S1: Assessment of systematic JAK2V617F screening in 996 patients with venous thromboembolism. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.