To the editor, Major venous thrombosis in essential thrombocythemia (ET), with an estimated incidence of 0.6 per 100 patient‐years, accounts for one third of all thrombotic events.1.De Stefano V. Finazzi G. Barbui T. Antithrombotic therapy for venous thromboembolism in myeloproliferative neoplasms.Blood Cancer J. 2018; 8: 65Crossref PubMed Scopus (32) Google Scholar In addition, the incidence of major hemorrhage is estimated at 0.79 per 100 patient‐years.2.Finazzi G. Carobbio A. Thiele J. et al.Incidence and risk factors for bleeding in 1104 patients with essential thrombocythemia or prefibrotic myelofibrosis diagnosed according to the 2008 WHO criteria.Leukemia. 2012; 26: 716-719Crossref PubMed Scopus (127) Google Scholar A recent systematic review highlighted the variable risk of recurrent venous events in myeloproliferative neoplasms (MPN), which ranges from 0% to 33%.3.Hamulyák E.N. Daams J.G. Leebeek F.W.G. et al.A systematic review of antithrombotic treatment of venous thromboembolism in patients with myeloproliferative neoplasms.Blood Adv. 2021; 5: 113-121Crossref PubMed Scopus (23) Google Scholar Age > 60 years, thrombosis history, and JAK2 mutation are established risk factors for venous thrombosis in ET.4.Barbui T. Vannucchi A.M. Buxhofer‐Ausch V. et al.Practice‐relevant revision of IPSET‐thrombosis based on 1019 patients with WHO‐defined essential thrombocythemia.Blood Cancer J. 2015; 5Crossref Scopus (150) Google Scholar Thrombosis and survival assessments by the International Prognostic Score for Essential Thrombocythemia (IPSET) consider thrombosis history without accounting for site, along with age > 60 years and JAK2 mutation (revised IPSET‐thrombosis) or leukocytosis ≥11 × 109/L (IPSET‐survival).4.Barbui T. Vannucchi A.M. Buxhofer‐Ausch V. et al.Practice‐relevant revision of IPSET‐thrombosis based on 1019 patients with WHO‐defined essential thrombocythemia.Blood Cancer J. 2015; 5Crossref Scopus (150) Google Scholar, 5.Passamonti F. Thiele J. Girodon F. et al.A prognostic model to predict survival in 867 World Health Organization‐defined essential thrombocythemia at diagnosis: a study by the international working group on myelofibrosis research and treatment.Blood. 2012; 120: 1197-1201Crossref PubMed Scopus (190) Google Scholar Regardless of the association of splanchnic and cerebral venous thrombosis with recurrence, impact on overall survival is unknown.6.De Stefano V. Vannucchi A.M. Ruggeri M. et al.Splanchnic vein thrombosis in myeloproliferative neoplasms: risk factors for recurrences in a cohort of 181 patients.Blood Cancer J. 2016; 6Crossref PubMed Scopus (60) Google Scholar, 7.Gangat N. Guglielmelli P. Betti S. et al.Cerebral venous thrombosis and myeloproliferative neoplasms: a three‐center study of 74 consecutive cases.Am J Hematol. 2021; 96: 1580-1586Crossref Scopus (4) Google Scholar The current study focuses on venous thrombosis in ET with the following objectives: (1) describe the site‐specific prevalence, phenotypic, and genotypic correlations; (2) determine risks of recurrent venous thrombosis and major hemorrhage; and (3) evaluate the prognostic impact of venous thrombosis on overall survival, fibrotic and leukemic transformation, and arterial events. The current retrospective series was conducted following institutional review board approval and included 737 adult patients with ET evaluated at the Mayo Clinic from 1967–2021 and confirmed to fulfill the 2016 World Health Organization (WHO) criteria.8.Arber D.A. Orazi A. Hasserjian R. et al.The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia.Blood. 2016; 127: 2391-2405Crossref PubMed Scopus (5723) Google Scholar Follow‐up information including vascular events and disease progression was updated in May 2021. Patients with major venous thrombosis confirmed on imaging studies were stratified into three cohorts based on first event: cerebral venous thrombosis (CVT); splanchnic venous thrombosis (SVT) inclusive of hepatic, portal, splenic and mesenteric venous thrombosis; and other venous thrombosis (other VT), the latter included deep venous thrombosis, pulmonary embolism, and central retinal vein thrombosis. Comparison between categorical variables was performed by chi square test and continuous variables by Wilcoxon/Kruskal–Wallis tests. A Cox proportional hazards model was used to compute multivariable analyses. p‐value ≤.05 was considered significant. JMP Pro 16.0.0 software package (SAS Institute) was utilized for all analyses. Of 737 patients with ET (median age 58 years, range 18–90; 63% females), first major venous thrombosis was documented in 131 (18%; 2 per 100 patient‐years) with 35 (27%), 48 (37%), and 48 (37%) events prior to, at, or after ET diagnosis, respectively, and were provoked in 42 (32%) cases in the context of surgery (n = 24), immobilization/trauma (n = 5), pregnancy (n = 5), solid tumor (n = 4), hospitalization (n = 3), or oral contraceptive use (n = 1). Site‐specific frequencies were 2% (n = 16), 5% (n = 38), and 11% (n = 77) for CVT, SVT, and other VT, respectively. Table 1 outlines presenting clinical and laboratory features including phenotypic and genotypic comparison of ET patients with CVT, SVT, versus other VT. Patients with CVT in comparison to those with other VT were more likely to be younger (median age 39 vs. 67 years; p = .02), display lower platelet count (median 624 vs. 789 × 109/L; p = .0003), and less likely to have hypertension (6% vs. 58%, p = .01) or smoking history (0% vs. 30%; p = .02; Table 1). JAK2V617F mutational frequency, albeit higher with CVT and SVT, did not significantly differ from those with other VT (83% and 82% vs. 69%; p = .6 and p = .1, respectively). On the other hand, SVT patients compared to those with other VT, were more likely to be younger (median age; 40 vs. 67 years, p = .0001), present with major hemorrhage prior to/at the time of ET diagnosis (24% vs. 8%, p = .03) and less likely to be CALR mutated (7% vs. 23%, p = .03). In addition, the incidence of CALR mutation was lower in patients with VT versus those without VT (17% vs. 28%, p = .05). Next generation sequencing investigated the presence of high‐risk molecular variants: SRSF2, U2AF1, ASXL1, EZH2, and IDH2. SRSF2, ASXL1, and IDH2 mutations were present in one (3%), three (8%), and one (3%) patients, respectively.TABLE 1Presenting clinical and laboratory characteristics of 737 patients with essential thrombocythemia (ET) stratified by first major venous thrombotic eventVariablesAll patients n = 737No venous thrombosis (VT) (anytime) n (%) 606 (82)Venous thrombosis (VT) (anytime) n (%) 131 (18)Cerebral venous thrombosis (CVT) (anytime) n (%) 16 (2)Splanchnic venous thrombosis (SVT) (Anytime) n (%) 38 (5)Other venous thrombosis (VT)aOther venous thrombosis includes deep venous thrombosis (n = 43), pulmonary embolism (n = 20), deep venous thrombosis and pulmonary embolism (n = 12), and retinal vein thrombosis (n = 2). (Anytime) n (%) 77 (11)p‐value univariate/multivariate VT vs. No VTp‐value univariate/multivariate CVT vs. Other VTp‐value univariate/multivariate SVT vs. Other VTp‐value univariate/multivariate CVT vs. SVTAt or prior to diagnosis, n = 83At or prior to diagnosis, n = 15At or prior to diagnosis, n = 2At or prior to diagnosis, n = 40Median age (range), years58 (18–90)58 (18–90)56 (18–89)39 (18–85)40 (18–88)67 (28–89).6<.0001/.02<.0001/.0001.8Age > 60 years, n (%)340 (46)278 (46)62 (47)2 (13)13 (34)47 (61).8.0002/.05.006/.03.09Female, n (%)464 (63)384 (63)80 (61)9 (56)22 (58)49 (64).6.6.5.9Hemoglobin, median (range), g/dl13.8 (9–16.3)13.9 (11–16.3)13.6 (9–16.1)13.5 (10.2–15.7)13.1 (9–16.1)13.8 (11.1–16).02/.07.8.01/.17.5Leukocytes, median (range), ×109/L9 (4–32)9 (4–28)9 (4–32)6 (5–11)10 (4–32)9 (4–27).09.001/.44.2.0008/.14Leukocytes ≥ 11 × 109/L, n (%)157/724 (22)130/599 (22)27/125 (22)011/35 (31)16/77 (21).9.02/.44.2.004/.02Platelets, median (range), ×109/L800 (230–3460)809 (434–3460)730 (230–3330)624 (427–1072)724 (230–3330)789 (457–2869).6.002/.0003.8.03/.01Cardiovascular risk factors, n (%)364/681 (53)291/551 (53)73/130 (56)1/16 (6)18/38 (47)54/76 (71).5<.0001/.02.01/.34.001/.09Diabetes mellitus61/679 (9)49/549 (9)12/130 (9)03/38 (8)9/76 (12).9.05/.14.5.13Hypertension279/680 (41)226/550 (41)53/130 (41)1/16) (6)8/38 (21)44/76 (58).9<.0001/.01.0001/.17.15Smoking160/676 (24)124/546 (23)36/130 (28)013/38 (34)23/76 (30).2.001/.02.7.001/.01Driver mutation status, n (%)n = 733n = 606n = 127n = 12n = 38n = 77JAK2V617F445 (61)351 (58)94 (74)10 (83)31 (82)53 (69).003/.01.6.1.1CALR194 (26)172 (28)22 (17)1 (8)3 (7)18 (23).005/.05.09.03/.03.8MPL20 (3)19 (3)1 (1)001 (1).4.7.3–Triple negative (TN)74 (10)64 (11)10 (8)1 (8)4 (11)5 (6).3.3.3.6Next generation sequencing, n (%)n = 245n = 206n = 39n = 2n = 7n = 30SF3B1651001.9.8.5–SRSF2761001.9.8.5–U2AF1220000.4–––ASXL11293012.4.7.5.6EZH2330000.3–––IDH2321010.4–.06.6TP53431001.6.8.5–SH2B3101001.05/.03.8.5–DNMT3A16142011.7.8.3.6TET221147016.03/.03.5.7.6CSF3R220000.4–––RUNX1220000.4––‐ThrombophiliacThrombophilia testing included Factor V Leiden, prothrombin gene mutation, Protein C, S, Antithrombin deficiency, anti‐phospholipid antibody, paroxysmal nocturnal hemoglobinuria (PNH)., n (%)5/111 (5)2/15 (14)1/35 (3)2/61 (3)–.3.8.3Antiphospholipid antibody2/111 (2)1/15 (7)1/35 (3)0––––Prothrombin G20210A (HET)1/111 (0.5)1/15 (7)00––––Factor V Leiden (HET)2/111 (2)002/61 (3)––––Revised IPSET‐thrombosisdInternational Prognostic Score for thrombosis in Essential Thrombocythemia (IPSET‐thrombosis)., n (%)n = 737n = 606n = 131n = 16n = 38n = 77Very low167 (23)155 (26)12 (9)02 (5)10 (13)<.0001.1.5.4Low167 (23)155 (26)12 (9)1 (6)5 (13)6 (8)Intermediate85 (11)81 (13)4 (3)01 (3)3 (4)High318 (43)215 (35)103 (79)15 (94)30 (79)58 (75)Major arterial thrombosis at or prior to ET diagnosisbMajor arterial thrombosis includes myocardial infarction, angina, cerebrovascular accidents, transient ischemic attack, peripheral arterial thrombosis, aortic thrombosis, mesenteric artery thrombosis, central retinal thrombosis., n (%)96 (13)72 (12)24 (18)2 (13)3 (9)19 (25).05/.01.3.02/.41.6Major arterial thrombosis after ET diagnosisbMajor arterial thrombosis includes myocardial infarction, angina, cerebrovascular accidents, transient ischemic attack, peripheral arterial thrombosis, aortic thrombosis, mesenteric artery thrombosis, central retinal thrombosis., n (%)109 (15)82 (13)27 (21)1 (6)5 (13)21 (27).04/.63.04/.27.07.4Major hemorrhage at or prior to ET diagnosiseMajor hemorrhage includes bleeding events that require red cell transfusion support, resulted in ≥2 g/dl decline in hemoglobin or involved critical organs., n (%)41/720 (6)24/592 (4)17/128 (13)2/16 (13)9/37 (24)6/75 (8).0002/.0001.6.02/.03.3Major hemorrhage after ET diagnosiseMajor hemorrhage includes bleeding events that require red cell transfusion support, resulted in ≥2 g/dl decline in hemoglobin or involved critical organs., n (%)85/720 (12)56/592 (9)29/128 (23)1/16 (6)10/37 (27)18/75 (24)<.0001/.0003.08.7.06Note: p Values < .05 are significant and in bold.a Other venous thrombosis includes deep venous thrombosis (n = 43), pulmonary embolism (n = 20), deep venous thrombosis and pulmonary embolism (n = 12), and retinal vein thrombosis (n = 2).b Major arterial thrombosis includes myocardial infarction, angina, cerebrovascular accidents, transient ischemic attack, peripheral arterial thrombosis, aortic thrombosis, mesenteric artery thrombosis, central retinal thrombosis.c Thrombophilia testing included Factor V Leiden, prothrombin gene mutation, Protein C, S, Antithrombin deficiency, anti‐phospholipid antibody, paroxysmal nocturnal hemoglobinuria (PNH).d International Prognostic Score for thrombosis in Essential Thrombocythemia (IPSET‐thrombosis).e Major hemorrhage includes bleeding events that require red cell transfusion support, resulted in ≥2 g/dl decline in hemoglobin or involved critical organs. Open table in a new tab Note: p Values < .05 are significant and in bold. Treatments for CVT, SVT, and other VT included systemic anticoagulation alone (0%/28%/33%), systemic anticoagulation plus cytoreductive therapy (50%/28%/31%), systemic anticoagulation plus aspirin (30%/6%/9%), systemic anticoagulation plus aspirin and cytoreductive therapy (10%/9%/17%), aspirin and/or cytoreductive therapy (10%/19%/5%). Warfarin was utilized in 49 (51%) of patients, direct oral anticoagulants in 20 (21%), enoxaparin in 9 (9%), heparin in 4 (4%), and fondaparinux in 1 (1%) of cases. Cytoreductive therapies included hydroxyurea in 47 (49%) and anagrelide in 7 (7%) patients. At a median follow‐up of 8.7 years (range: 1–49.4 years), subsequent venous thrombosis was documented in 64 (9%; 1 per 100 patient‐years) with recurrent events in 16 of 83 (19%; 2.2 per 100 patient‐years) with prior venous thrombosis. Recurrence rates were highest with SVT (32%, n = 9) followed by other VT (15%, n = 6), and CVT (7%, n = 1) (p < .0001; Figure 1A). Provoked (n = 22) versus unprovoked (n = 61) venous event did not impact recurrence (23% and 18%, p = .27); results were similar when events were site‐stratified. On univariate analysis of predictors of subsequent venous thrombosis, SVT (p < .0001, hazard ratio [HR] 4.3), age > 60 years (p = .01, HR 1.9), and JAK2 mutation (p = .01, HR 1.9) emerged significant, while gender differences were not significant (p = .10); SVT and age > 60 years retained significance on multivariable analysis. In patients with SVT, recurrent events included SVT (n = 6, 66%), pulmonary embolism (n = 1, 10%), deep vein thrombosis (n = 1, 10%), and CVT (n = 1, 10%); four patients were already on systemic anticoagulation at the time of recurrence (n = 2 anticoagulation alone, n = 1 anticoagulation plus aspirin, n = 1 anticoagulation plus cytoreductive therapy), and three on cytoreductive therapy. Overall, recurrence occurred in 10/47 (21%) patients receiving systemic anticoagulation, 0/4 (0%) patients receiving cytoreduction alone, 3/16 (19%) patients receiving systemic anticoagulation plus cytoreduction, and 3/16 (19%) patients with neither anticoagulation nor cytoreduction (p = .60). Blood counts at the time of recurrent VT were available in nine patients, of which platelet count was below 450 × 109/L, 450–600 × 109/L and above 1000 × 109/L in three, four, and two patients, respectively. Incidence rates for major hemorrhage were higher among patients with SVT (27%) and other VT (24%) compared to those with CVT (6%; p = .06 and p = .08, respectively). Moreover, major hemorrhage was more likely to occur with anticoagulation plus aspirin (33%) versus anticoagulation alone (21%; p = .06). Neither history of venous thrombosis nor site of event impacted overall survival, fibrotic and leukemic transformation, or arterial thrombosis; the superior overall survival and lower incidence of arterial thrombosis observed in patients with SVT was fully accounted for by differences in age (Figure 1B–E). The management of MPN‐associated venous thrombosis continues to be an unmet clinical need. Prior studies are limited by the retrospective design, combining arterial and venous events and inclusion of all MPN subtypes.9.Valerio De S. Tommaso Z. Elena R. et al.Recurrent thrombosis in patients with polycythemia vera and essential thrombocythemia: incidence, risk factors, and effect of treatments.Haematologica. 2008; 93: 372-380Crossref Scopus (260) Google Scholar However, given the differential risk of venous thrombosis among patients with ET, polycythemia vera (PV), and primary myelofibrosis (PMF), interpretation and extrapolation of results from the former studies poses a major challenge. The current study is the foremost to report on site‐specific major venous thrombosis in ET patients fulfilling the WHO 2016 diagnostic criteria, thus excluding cases with pre‐fibrotic myelofibrosis. Our study provides overall and site‐specific incidence and recurrence rates for major venous thrombosis, and illustrates phenotypic and genotypic distinctions, confirming the association of CVT and SVT with younger age and JAK2V617F mutation.6.De Stefano V. Vannucchi A.M. Ruggeri M. et al.Splanchnic vein thrombosis in myeloproliferative neoplasms: risk factors for recurrences in a cohort of 181 patients.Blood Cancer J. 2016; 6Crossref PubMed Scopus (60) Google Scholar, 7.Gangat N. Guglielmelli P. Betti S. et al.Cerebral venous thrombosis and myeloproliferative neoplasms: a three‐center study of 74 consecutive cases.Am J Hematol. 2021; 96: 1580-1586Crossref Scopus (4) Google Scholar Moreover, patients with VT, compared to those without VT, were more likely to harbor TET2 and SH2B3 mutations. Noteworthy observations include a high rate of recurrent venous thrombosis; risk was heightened five‐ and two‐fold with SVT, in relation to those without venous thrombosis or other VT, respectively. Importantly, despite high recurrence rates, major venous thrombosis did not adversely impact overall survival, in contrast to PV.10.Tefferi A. Rumi E. Finazzi G. et al.Survival and prognosis among 1545 patients with contemporary polycythemia vera: an international study.Leukemia. 2013; 27: 1874-1881Crossref PubMed Scopus (431) Google Scholar Taken together, our findings reinforce the need to consider site of venous thrombosis in prognostic risk assessment in ET. NG and AT designed the study, collected data, performed analysis, and co‐wrote the paper. AS and NS collected data and performed analysis. CAH performed review of bone marrow biopsies. KB, ME, APW, and AP contributed patients. VDS, AMV, and TB provided critical input. All authors reviewed and approved the final draft of the paper. The authors report no conflicts of interest.