2026 年 33 巻 3 号 p. 313-322
Aim: This study attempted to clarify the prevalence and clinical characteristics of Janus kinase 2 V617F (JAK2) gene mutations in patients with cerebrovascular diseases.
Methods: We prospectively enrolled patients who were admitted to or referred to our department with cerebrovascular disease due to suspected major cerebral artery disease or small-vessel occlusion within 30 days of onset between January 1, 2021, and April 30, 2024, and who consented to undergo a JAK2 mutation analysis. We investigated the prevalence of JAK2 mutations based on the clinical subtype of stroke and the presence or absence of major cerebral artery disease. We also examined the clinical characteristics of patients with positive JAK2 mutation.
Results: Among 316 consecutive inpatients (216 males; median age, 74 years old), JAK2 mutations were detected in 4 of 102 (3.9%) patients with large artery atherosclerosis, 2 of 101 patients (2.0%) with small-vessel occlusion, and 2 of 113 (1.8%) with other stroke subtypes. A multiple logistic regression analysis showed that patients with the positive JAK2 mutation had significantly higher hematocrit values (odds ratio [OR], 1.32; 95% confidence interval [CI], 1.07–1.62; p = 0.010), platelet counts (OR, 1.19; 95% CI, 1.07–1.31; p = 0.001), and thrombomodulin levels (OR, 1.08; 95% CI 1.01–1.15; p = 0.025) at admission than patients with the negative JAK2 mutation.
Conclusions: The frequency of JAK2 mutations is very low among patients with major cerebral artery diseases and small-vessel occlusion. The mechanisms underlying stroke onset in patients with the positive JAK2 mutation may involve factors beyond hematopoietic cells, such as endothelial dysfunction.
Janus kinase 2 (JAK2) mutations are most frequently observed in myeloproliferative neoplasms (MPNs) and promote the proliferation of hematopoietic cells through constitutive activation of signal transduction pathways, such as those mediated by the erythropoietin receptor. This activation can potentially lead to thrombosis, including cerebrovascular events. JAK2 mutations have been detected in approximately 95% of polycythemia vera cases, 60% of essential thrombocythemia cases, and 50% of primary myelofibrosis cases. Although JAK2 mutations are considered an independent risk factor for thrombosis, recent studies using murine disease models have shown that JAK2 mutations may also promote atherosclerosis1-4). These findings suggest that JAK2 mutations may contribute to the development of cardiovascular and cerebrovascular diseases not only through hematopoietic cell proliferation but also via mechanisms such as atherosclerosis.
Previously, we reported that JAK2 mutations are associated with cerebral arterial stenosis in patients with MPNs5). However, there are few reports on the prevalence of JAK2 mutations in patients with cerebrovascular disease6, 7). Therefore, the present study investigated the prevalence of JAK2 mutations in patients with ischemic stroke, particularly non-cardioembolic stroke, and clarified the clinical characteristics associated with the presence or absence of these JAK2 mutations.
We enrolled 377 consecutive patients admitted to our department between January 1, 2021, and May 31, 2024, with cerebrovascular disease caused by suspected major cerebral artery disease (MCAD) or small-vessel occlusion (SVO) within 30 days of onset. Among these, 52 patients were unable to provide their informed consent, and 9 refused to participate. As a result, 316 patients were included in this study for the analysis.
This study was reviewed and approved by the Ethics Committee of Kawasaki Medical School Hospital (approval number: 5071) and was conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all participants.
Clinical Subtype ClassificationTo evaluate ischemic stroke subtypes, all patients underwent head magnetic resonance imaging (MRI), head and neck magnetic resonance angiography (MRA), a 12-lead electrocardiogram (ECG), 24-hour Holter ECG or long-term (≥ 3-day) portable ECG monitoring, transthoracic or transesophageal echocardiography, carotid ultrasonography, and blood testing. Blood tests included a complete blood count, alanine aminotransferase, creatinine, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglycerides, blood glucose, hemoglobin A1c, B-Type Natriuretic Peptide, von Willebrand factor antigen, von Willebrand factor activity, D-dimer, total homocysteine, thrombomodulin, protein C, protein S, antithrombin III levels, plasminogen, anticardiolipin antibodies, and a JAK2 V617F gene mutation analysis.
Blood samples for the JAK2 gene mutation analysis were collected as 5 mL of whole blood in tubes containing ethylenediaminetetraacetic acid. The samples were sent to a commercial laboratory (SRL Inc., Tokyo, Japan), where DNA was extracted using the QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany), and the JAK2 mutation was quantified using allele-specific quantitative polymerase chain reaction (PCR; ipsogen JAK2 DX reagent; Sysmex Corporation, Hyogo, Japan). The lower limit of detection was 0.042%8). The clinical subtypes of ischemic stroke were classified according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria as large artery atherosclerosis (LAA), small-vessel occlusion (SVO), cardioembolism (CE), stroke of other determined etiology (O), and stroke of undetermined etiology (U). In this study, branch atheromatous disease (BAD) was classified as an SVO9).
MRI ProtocolMRI examinations—including diffusion-weighted imaging (DWI), fluid-attenuated inversion recovery (FLAIR), and time-of-flight (TOF) MRA—were performed using a 3-T scanner (Ingenia 3.0T CX; Philips Medical Systems, Best, Netherlands). DWI scan parameters were as follows: repetition time (TR), 3500 ms; echo time (TE), 75 ms; b-values, 0 and 1000 s/mm2; field of view (FOV), 240×225 mm2; acquired voxel size, 1.88×1.58×5.00 mm3; reconstructed voxel size, 0.94×0.94×5.00 mm3; flip angle, 90°. FLAIR scan parameters were as follows: TR, 10,000 ms; TE, 120 ms; inversion time, 2700 ms; FOV, 240×192 mm2; acquired voxel size, 0.79×1.11×5.00 mm3; reconstructed voxel size, 0.43×0.43×5.00 mm3; flip angle, 90°. Brain TOF-MRA scan parameters: TR, 25 ms; TE, 3.45 ms; FOV, 200×190 mm2; acquired voxel size, 0.52×1.02×1.05 mm3; reconstructed voxel size, 0.28×0.28×0.50 mm3; flip angle, 18°. Neck TOF-MRA scan parameters: TR, 20 ms; TE, 3.45 ms; FOV, 200×200 mm2; acquired voxel size, 0.66×1.03×1.60 mm3; reconstructed voxel size, 0.39×0.39×0.80 mm3; and flip angle, 18°. Brain and neck TOF-MRA were performed to evaluate the location and degree of major arterial stenosis or occlusion. MCAD was defined as stenosis (≥ 50%) or occlusion of the extracranial or intracranial internal carotid artery, vertebral artery, basilar artery, or proximal segments of the anterior, middle, or posterior cerebral arteries. MCAD evaluations were independently performed by two stroke specialists (N.O. and Y.Y.) who were blinded to the clinical information. Inter-rater agreement for MCAD evaluation was previously confirmed to be high (Cohen’s kappa = 0.85)5). In cases of disagreement between the evaluators, the final MCAD diagnosis was made based on consensus.
CovariatesThe following patient information was collected: stroke risk factors (age, sex, hypertension, diabetes mellitus, dyslipidemia, atrial fibrillation, current smoking and drinking status); history of stroke and other thrombotic events (hemorrhagic and ischemic stroke, transient ischemic attack [TIA], ischemic heart disease, peripheral artery disease, pulmonary embolism, and deep vein thrombosis), and other vascular history (cerebral and aortic aneurysm/dissection); pre-onset treatment (antithrombotic drugs, cytoreductive drugs, statins); ischemic stroke clinical subtype; admission blood test results, head MRI and head/neck MRA findings; National Institutes of Health Stroke Scale (NIHSS) scores at admission and discharge; modified Rankin Scale (mRS) scores before onset, at discharge, and three months post-onset; and antithrombotic therapy at discharge.
Statistical AnalysesTo compare clinical characteristics according to JAK2 mutation status in patients with ischemic stroke, Fisher’s exact test was used for categorical variables, and the Mann–Whitney U test was applied for univariate analyses of continuous variables. Categorical variables are expressed as numbers and percentages and continuous variables as medians with interquartile ranges (IQRs). Furthermore, to examine factors independently associated with JAK2 mutation–positive ischemic stroke, a multivariate logistic regression analysis was performed using variables that showed associations with p＜0.1 in a univariate analysis, and odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. A threshold of p＜0.1 was adopted for variable selection in the multivariate analysis, given the exploratory nature of this study on JAK2 mutations in ischemic stroke. This approach ensures inclusion of variables with potential trends that may reveal significant associations when adjusted for other factors. In addition, only variables available at admission were considered for the multivariate analysis, as the aim was to identify factors present at admission that could guide future decisions on JAK2 mutation testing in non-cardioembolic stroke patients.
All statistical analyses were performed using the JMP software program, version 14.0.0 (SAS Institute Inc., Cary, NC, USA). Statistical significance was set at p＜0.05.
Among the 316 patients with cerebrovascular disease due to suspected MCAD or SVO within 30 days of onset, 297 had ischemic stroke, and 19 had TIA. MCAD was present in 188 patients (59.5%), and JAK2 mutations were found in 8 patients (2.5%). Of these 8 patients with JAK2 mutations, 2 had a history of JAK2 mutation-positive MPNs, while 6 (1.9%) were identified as having de novo JAK2 mutations. The JAK2 mutation rates according to ischemic stroke clinical subtype were as follows: 4/102 (3.9%) in LAA, 2/101 (2.0%) in SVO, 1/16 (6.3%) in O, and 1/61 (1.6%) in U (Fig.1). No cases positive for JAK2 mutations were found among the 17 patients with CE (Fig.1). Among the 188 patients with MCAD, 7 (3.7%) had a JAK2 mutation, compared to only 1 of 128 patients (0.8%) without MCAD (Fig.1). However, this difference was not statistically significant.
LAA, large artery atherosclerosis; SVO, small vessel occlusion; CE, cardioembolism; O, stroke of other determined etiology; U, stroke of undetermined etiology; TIA, transient ischemic attack; MCAD, major cerebral artery disease.
The complete results of the univariate analyses for all variables are detailed in Table 1. As shown in Table 1, a univariate analysis showed no significant differences in stroke risk factors, stroke history, or pre-admission treatment between the groups with and without JAK2 mutations. However, blood tests at admission showed that the JAK2 mutation-positive group had significantly higher white blood cell counts, hematocrit values, and platelet counts than the JAK2 mutation-negative group, as well as significantly elevated thrombomodulin levels, a marker of endothelial dysfunction (Table 1). While no significant differences were observed in infarct size, number, or distribution between the groups, the JAK2 mutation-positive group had a significantly higher proportion of posterior cerebral artery lesions among patients with MCADs (Table 1). In a multivariate analysis, the JAK2 mutation-positive group showed ORs of 1.32 (95% CI 1.07–1.62, p = 0.010) for hematocrit, 1.19 (1.07–1.31, p = 0.001) for platelet count, and 1.08 (1.01–1.15, p = 0.025) for thrombomodulin (Table 2).
| Variables | Total patients (N = 316) | JAK2 V617F mutation | p value | |
|---|---|---|---|---|
| Yes (N = 8) | No (N = 308) | |||
| Age, years, median (IQR) | 74 (66–81) | 79 (63–87) | 74 (66–81) | 0.409 |
| Stroke risk factors, n (%) | ||||
| Male | 216 (68.4) | 6 (75.0) | 210 (68.2) | 1.000 |
| Hypertension | 267 (84.5) | 7 (87.5) | 260 (84.4) | 1.000 |
| Diabetes mellitus | 100 (31.7) | 3 (37.5) | 97 (31.5) | 0.711 |
| Dyslipidemia | 245 (77.5) | 7 (87.5) | 238 (77.3) | 0.689 |
| High-risk sources of CE | 45 (14.2) | 1 (12.5) | 44 (14.3) | 1.000 |
| Current Smoking | 82 (26.0) | 2 (25.0) | 80 (26.0) | 1.000 |
| Alcohol | 83 (26.3) | 2 (25.0) | 81 (26.3) | 1.000 |
| Stroke history, n (%) | ||||
| Ischemic stroke or TIA | 96 (30.4) | 3 (37.5) | 93 (30.2) | 0.703 |
| Hemorrhagic stroke | 10 (3.2) | 0 (0) | 10 (3.3) | 1.000 |
| Thrombotic events＊, n (%) | 109 (34.5) | 3 (37.5) | 106 (34.4) | 1.000 |
| Other vascular history, n (%) | ||||
| Cerebral aneurysm/dissection | 5 (1.6) | 0 (0) | 5 (1.6) | 1.000 |
| Aortic aneurysm/dissection | 14 (4.4) | 0 (0) | 14 (4.5) | 1.000 |
| Treatments before admission, n (%) | ||||
| Antithrombotic therapy | 103 (32.5) | 3 (37.5) | 100 (32.5) | 0.719 |
| Statin | 99 (31.3) | 1 (12.5) | 98 (31.8) | 0.443 |
| Laboratory findings at admission, median (IQR) | ||||
| WBC, ×102/mm3 | 69.5 (57.6–86.1) | 92.7 (72.7–135.3) | 69.1 (57.3–85.4) | 0.010＊＊ |
| Hematocrit, % | 42.0 (38.3–44.7) | 44.6 (42.5–47.3) | 41.9 (38.2–44.6) | 0.038＊＊ |
| Platelets, ×104/mm3 | 22.6 (17.6–27.6) | 42.0 (33.2–54.6) | 22.2 (17.5–27.1) | ＜0.001＊＊ |
| MPV, fL | 9.8 (9.2–10.4) | 9.8 (9.2–10.4) | 9.8 (9.2–10.0) | 0.471 |
| PDW, fL | 10.5 (9.3–11.9) | 10.5 (9.5–11.3) | 10.5 (9.3–12.0) | 0.763 |
| vWF antigen, % | 163 (126–213) | 153 (94–207) | 163 (127–213) | 0.538 |
| vWF activity, % | 158 (121–217) | 127 (83–153) | 162 (122–219) | 0.059 |
| D-dimer, μg/mL | 0.7 (0.5–1.7) | 0.6 (0.5–0.8) | 0.7 (0.5–1.7) | 0.316 |
| Total homocysteine, nmol/mL | 12.6 (9.9–16.0) | 12.7 (10.7–17.8) | 12.6 (9.8–16.0) | 0.695 |
| Thrombomodulin, U/mL | 13.3 (9.3–17.1) | 19.7 (16.6–21.6) | 13.1 (9.3–16.9) | 0.003＊＊ |
| ALT, IU/L | 17 (13–25) | 25 (14–30) | 17 (13–24) | 0.172 |
| Creatinine, mg/dL | 0.84 (0.69–1.04) | 1.00 (0.89–1.04) | 0.83 (0.69–1.05) | 0.125 |
| C-reactive protein, mg/dL | 0.13 (0.07–0.31) | 0.07 (0.05–0.33) | 0.13 (0.07–0.31) | 0.265 |
| LDL-C, mg/dL | 117 (92–143) | 106 (90–140) | 117 (92–144) | 0.640 |
| HDL-C, mg/dL | 48 (41–58) | 47 (43–60) | 48 (41–58) | 0.951 |
| TG, mg/dL | 119 (84–171) | 150 (102–220) | 118 (84–170) | 0.244 |
| Glucose, mg/dL | 119 (104–153) | 121 (105–153) | 119 (104–153) | 0.883 |
| Hemoglobin A1c, % | 6.0 (5.7–6.6) | 5.9 (5.6–6.8) | 6.0 (5.7–6.6) | 0.672 |
| BNP, pg/mL | 33.9 (15.4–82.4) | 21.2 (11.3–39.6) | 34.0 (15.6–83.2) | 0.282 |
| Infarct size, number, and distribution on MRI, n＊＊＊ (%) | ||||
| Max diameter, 15 or more mm | 171 (57.6) | 4 (50.0) | 167 (57.8) | 0.726 |
| Number, 2 or more | 139 (46.8) | 5 (62.5) | 134 (46.4) | 0.481 |
| Multiple vascular territories | 19 (6.4) | 0 (0) | 19 (6.6) | 1.000 |
| Stenosis or occlusion of the major cerebral arteries on MRA, n (%) | ||||
| Any major artery | 188 (59.5) | 7 (87.5) | 181 (58.8) | 0.149 |
| Internal carotid artery | 113 (35.8) | 4 (50.0) | 109 (35.4) | 0.463 |
| Anterior cerebral artery | 16 (5.1) | 1 (12.5) | 15 (4.9) | 0.343 |
| Middle cerebral artery | 77 (24.4) | 3 (37.5) | 74 (24.0) | 0.409 |
| Posterior cerebral artery | 34 (10.8) | 3 (37.5) | 31 (10.1) | 0.044＊＊ |
| Basilar artery | 13 (4.1) | 0 (0) | 13 (4.2) | 1.000 |
| Vertebral artery | 34 (10.8) | 1 (12.5) | 33 (10.7) | 0.602 |
| Neurological severity, median (IQR) | ||||
| NIHSS score at admission | 2 (1–5) | 3 (1–9) | 3 (1–5) | 0.628 |
| NIHSS score at discharge | 1 (0–3) | 3 (0–6) | 1 (0–3) | 0.192 |
| Functional outcomes, median (IQR) | ||||
| premorbid mRS | 0 (0–2) | 0 (0–2) | 0 (0–2) | 0.918 |
| mRS at discharge | 2 (1–3) | 3 (1–4) | 2 (1–3) | 0.436 |
| mRS 3 months after stroke | 1 (0–3) | 3 (1–4) | 1 (0–3) | 0.065 |
JAK indicates Janus kinase; IQR, interquartile range; CE, cardioembolism; TIA, transient ischemic attack; WBC, white blood cells; MPV, mean platelet volume; PDW, platelet distribution width; vWF, von Willebrand factor; ALT, alanine aminotransferase; LDL, low-density lipoprotein; HDL, high-density lipoprotein; TG, triglycerides, BNP, B-type natriuretic peptide; MRI, magnetic resonance image; MRA, magnetic resonance angiography; IQR, interquartile range; NIHSS, National Institutes of Health Stroke Scale; mRS, modified Rankin Scale
＊Thrombotic events include ischemic stroke, TIA, ischemic heart disease, peripheral artery disease, pulmonary embolism, and deep vein thrombosis. ＊＊Significant at p＜0.05. ＊＊＊Number of patients excluding 19 cases of TIA.
| Variables | Odds Ratio (95%CI) | p value |
|---|---|---|
| WBC, ×102/mm3 | 1.02 (0.99 ― 1.06) | 0.230 |
| Hematocrit, % | 1.32 (1.07 ― 1.62) | 0.010＊ |
| Platelets, ×104/mm3 | 1.19 (1.07 ― 1.31) | 0.001＊ |
| vWF activity, % | 0.99 (0.97 ― 1.01) | 0.330 |
| Thrombomodulin, U/mL | 1.08 (1.01 ― 1.15) | 0.025＊ |
| Posterior cerebral artery | 2.09 (0.17 ― 26.13) | 0.567 |
CI indicates confidence interval; WBC, white blood cells; vWF, von Willebrand factor
＊Significant at p＜0.05.
Detailed Clinical Features of the Eight Patients with the JAK2 Mutation
Table 3 summarizes the clinical characteristics of the eight patients with JAK2 mutations. The median age was 79 (range 50–90) years old, and 6 patients were male. Seven patients had multiple risk factors for stroke, 1 had atrial fibrillation, and 3 had a history of ischemic stroke. The median JAK2 mutation allele burden was 19.8%, with 2 patients having very low values (＜0.1%). The clinical stroke subtypes were as follows: 4 (50%) LAA, 2 (25%) SVO, and 2 (25%) other subtypes, showing the diversity of stroke subtypes among the patients, despite LAA being the most common. However, 7 (88%) had MCAD, and 4 (50%) had stenosis or occlusion in non-culprit vessels. In Case 1, the culprit vessel occlusion improved on follow-up MRA, although moderate stenosis remained. Five patients (63%) did not receive antithrombotic therapy before stroke onset, and all received antiplatelet therapy for secondary prevention. The median admission NIHSS score was 3 (IQR: 1–9), and the 3-month mRS score was 3 (1–4).
| Case no. | Age, years | Sex | TOAST＊ |
Stroke risk ＊ ＊factors and MPNs |
History of thrombosis |
WBC, 102/ mm3 |
Hct, % |
Platelets, 104/ mm3 |
TM, U/mL |
JAK2 V617F allele burden |
Topographic features on MR image |
Major cerebral arteries with 50-99% stenosis or occlusion |
Pre-onset treatment | Antithrombotic treatment after stroke |
NIHSS score on ad |
mRS at 3M |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 74 | M | LAA | HT | Ischemic stroke | 64.4 | 45.5 | 40.5 | 14.5 | 16.387 | Multiple, small infarctions in Lt hemisphere | Lt MCA＊＊＊ Rt MCA | Clopidogrel | Aspirin Warfarin | 4 | 3 |
| 2 | 85 | M | LAA |
HT, DM, DL, ET |
Ischemic stroke | 74.9 | 50.8 | 46.6 | 16.1 | 33.892 | Multiple, small infarctions in Rt hemisphere |
Rt extracranial ICA Lt VA |
Cilostazol Hydroxyurea Pitavastatin | Aspirin Clopidogrel | 1 | 2 |
| 3 | 88 | F | LAA |
HT, DM, DL, ET |
None | 135.7 | 40 | 57.3 | 18.1 | 23.21 | Large infarction in Lt striate capsular area | Lt MCA | None | Aspirin | 11 | 4 |
| 4 | 90 | M | U | HT, DL, AF | Ischemic stroke | 92.6 | 47.5 | 66.7 | 20.5 | 24.038 | Multiple, small infarctions in Rt PCA territory |
Rt extracranial ICA Rt PCA |
Apixaban | Aspirin Apixaban | 3 | 5 |
| 5 | 71 | F | SVO | HT, DL | None | 71.9 | 42.6 | 36.4 | 19.2 | 16.079 | Small infarction in Lt thalamus | NA | None | Prasugrel | 1 | 1 |
| 6 | 60 | M | SVO | HT, DL | None | 92.7 | 46.8 | 23.6 | 20.2 | 0.379 | Small infarction in Lt side of pons | Rt intracranial ICA | None | Aspirin | 3 | 2 |
| 7 | 50 | M | Other | DL, SM | None | 143.4 | 43.6 | 43.5 | 22 | 38.272 | Multiple, small infarctions in Lt hemisphere | Lt ex＊t＊r＊a＊cranial ICA | None | Aspirin | 1 | 0 |
| 8 | 83 | M | LAA |
HT, DM, DL, SM |
None | 133.9 | 42.5 | 32.1 | 22.7 | 0.668 | Multiple infarctions in Rt ACA territory | Rt ACA | None | Aspirin | 15 | 3 |
TOAST indicates Trial of Org 10172 in acute stroke treatment; MPNs, myeloproliferative neoplasms (2 ETs, essential thrombocythemia); WBC, white blood cells; Hct, hematocrit; TM, thrombomodulin; JAK, Janus kinase; MR, magnetic resonance; NIHSS, National Institutes of Health Stroke Scale; ad, admission; mRS, modified Rankin scale; Lt, left; Rt, right; MCA, middle cerebral artery; ICA, internal carotid artery; VA, vertebral artery; PCA, posterior cerebral artery; ACA, anterior cerebral artery; and NA, not applicable
* TOAST classification includes large artery atherosclerosis (LAA), cardioembolism, small vessel occlusion (SVO), stroke of other determined cause (Other), and stroke of undetermined cause (U).
＊＊ Stroke risk factors include hypertension (HT), diabetes mellitus (DM), dyslipidemia (DL), atrial fibrillation (AF), current smoking (SM), and previous stroke (PS).
＊＊＊ Lt occlusive MCA, which was the culprit artery, regressed to moderate stenosis one week after admission.
＊＊＊＊ The etiology of the stroke was dissection of left extracranial ICA, which ultimately resulted in moderate stenosis.
This study investigated the prevalence of patients with positive JAK2 mutation with cerebrovascular disease caused by suspected MCAD or SVO within 30 days of onset. Few prospective studies have reported the prevalence of JAK2 mutations in patients with cerebrovascular disease. Notably, this is the first study to identify factors associated with JAK2 mutations in this patient population.
In the present study, the overall prevalence of JAK2 mutations was low across all stroke subtypes. Although healthy individuals were not included in this study, previous research reported that, in 54 healthy Japanese subjects, the JAK2 mutation measured using the same testing method as in the current study was below the lower limit of detection (0.042%)8). Other studies in the general population have reported JAK2 mutation-positive rates ranging from 0.2%–2%, although the detection methods vary10-12). Xu et al. detected JAK2 mutations in 37 samples (0.94%) when analyzing blood from 3,935 randomly selected visitors at a large Chinese hospital using PCR13). In venous thrombosis studies, Xavier et al. reported JAK2 mutations in 15 of 99 adult patients (15%) with splenic vein thrombosis14), whereas Lauwe et al. reported JAK2 mutations in 4 of 178 patients (2.3%) with deep vein thrombosis15). Regarding arterial thrombosis, Muendlein et al. reported JAK2 mutations in 6 of 997 individuals (0.6%) from the German population with no cardiovascular disease history, versus 21 of 1,589 (1.3%) of coronary artery disease patients12). In Japan, Yokokawa et al. analyzed JAK2 mutations in 832 consecutive cardiology inpatients, finding mutations in 15 patients (1.8%), all with low allele burdens (0.06%–1.73%), with significantly higher mutation rates in cardiovascular disease (13/462, 2.8%) than in non-cardiovascular disease (2/370, 0.5%)16). A retrospective study of 3,318 patients with ischemic stroke identified 17 with MPN-related stroke, of whom 16 (0.5%) were JAK2 mutation-positive17). Meanwhile, in a prospective study, Kristiansen et al. used high-sensitivity droplet digital PCR (mean detection limit: 0.0036%) to assess JAK2 mutations in 538 patients with acute ischemic stroke admitted to a single Danish neurology department. They found 61 patients with positive JAK2 (11.3%), significantly higher than the age- and sex-adjusted general population of 71/1,613 (4.4%)7). Compared to our finding of a 2.6% (8/316) JAK2 mutation prevalence in patients with stroke, the Danish report showed several-fold higher rates, though limiting the allele burden to ≥ 0.1% yielded a 4.8% (26/538) prevalence, and limiting to a ≥ 1% burden yielded a 2.8% (15/538) prevalence, similar to our findings. Combined with our report, the JAK2 mutation prevalence appears to be higher in patients with thrombosis, including those with ischemic stroke, than in the general population or hospital visitors.
Building on the observed prevalence of JAK2 mutations, it is noteworthy that, among the 6 patients with newly identified JAK2 mutations, most did not meet the diagnostic criteria for MPN, except for Case 4 (Table 3). This aligns with our previous study5), which found that 33% of MPN patients with a history of stroke had blood counts within the normal range, although with high-normal platelet levels. These findings suggest that JAK2 mutations may contribute to stroke risk even before an MPN diagnosis. Importantly, our multivariate analysis indicates that, in patients with high-normal to mildly elevated blood cell counts not meeting MPN thresholds, elevated thrombomodulin levels at admission could be a key indicator for JAK2 mutation testing, offering clinical value in screening non-cardioembolic stroke patients. Furthermore, all individuals with JAK2 mutations in this study had conventional vascular risk factors, suggesting that these mutations are unlikely to cause stroke alone but may act as an adjunctive risk factor, potentially worsening thrombus formation at sites of endothelial dysfunction, such as atherosclerotic plaque. This synergistic effect warrants further investigation to clarify the mechanisms underlying stroke in the presence of JAK2 mutation.
In our study, among the 8 patients with JAK2-positive mutations, 2 had an allele burden of ≤ 2%. To date, the clinical significance of a low allele burden remains unclear. As previously mentioned, Yokokawa et al. reported that all patients with JAK2-positive mutations who had cardiovascular disease exhibited a low allele burden (≤ 2%)16). Similarly, Muendlein et al. found that, among 21 patients with JAK2-positive mutations who had coronary artery disease, the median allele burden was relatively low at 2.3% (mean 9.2%, standard deviation 20.5%), with 8 patients (38%) exhibiting a low allele burden of ≤ 2%12). In the study by Kristiansen et al., which included patients with acute ischemic stroke and the general population, the prevalence of a low allele burden (0.009% to 0.9%) was 7.1% (38/538) among stroke patients and 3.6% (58/1613) among age- and sex-adjusted general population controls, indicating an increased prevalence in stroke patients7). Surprisingly, a recent study indicated that driver mutations associated with hematologic malignancies, such as JAK2, may arise during embryonic development or early childhood and could remain latent for decades before manifesting clinically as MPNs18). Furthermore, JAK2 mutations are thought to contribute to a chronic inflammatory state, in addition to hematopoietic cell proliferation, thereby increasing the risk of vascular diseases2, 11). These prior findings suggest that even a low allele burden of JAK2 mutations may expose cerebral arteries to a prolonged, mild chronic inflammatory state, potentially contributing to the development of cerebrovascular events. Further studies with larger cohorts and longitudinal designs are warranted to elucidate the clinical impact of a low allele burden JAK2 mutations on cerebrovascular events.
While our previous study in patients with MPN showed that JAK2 mutations were associated with MCAD, the present study found higher JAK2 mutation rates in patients with MCAD than in those without MCAD, although the difference was not statistically significant. Ong et al. examined the stroke subtypes and JAK2 mutation rates in MPNs19). Among the 22 patients with ischemic stroke, 81% were JAK2 mutation-positive, with the following stroke subtypes: 16 (73%) LAA, 4 (18%) SVO, and 2 (9%) other subtypes. Conversely, Nagai et al. found 5 (22%) LAA cases among 23 patients with MPN with ischemic stroke, with only 2 being JAK2 mutation-positive20). These differences may have resulted from our evaluation of all cerebral arteries, not just the culprit vessels, and differences in stenosis evaluation methods, sample sizes, and study populations. Recently, Fukunaga et al. studied 21 patients with acute ischemic stroke with essential thrombocythemia and found cerebral artery stenosis (N = 14) or occlusion (N = 1) in 15 patients (71%)21). Interestingly, among the 14 patients with stenosis, 11 (79%) showed resolution of the stenosis, suggesting temporary thrombus formation as the underlying cause. In addition, carotid ultrasound revealed atherosclerotic lesions at the carotid bifurcations in 8 of 21 patients (38%), with a floating thrombus confirmed in 1 patient, supporting the hypothesis of thrombus formation on atherosclerotic lesions. While we did not perform follow-up MRI in all patients, in Case 1, the left middle cerebral artery occlusion at admission resolved to moderate stenosis after 1 week. Case 7 developed an ischemic stroke due to extracranial carotid dissection, which ultimately resulted in moderate carotid stenosis.
To date, two cases of vertebral artery and internal carotid artery dissection have been reported in patients with JAK2 mutation-positive essential thrombocythemia22, 23). Experimental studies showed that JAK2 V617F knock-in mouse models demonstrated intrinsic alterations in megakaryocyte and platelet biology beyond mere cell count increases1). Furthermore, an association between JAK2 mutations in myeloid cells and the development of aortic aneurysms or dissections, mediated by increased matrix metalloproteinases and enhanced inflammation, has been suggested24, 25). Although the present study found no marked differences in the history of cerebral aneurysm/dissection or aortic aneurysm/dissection between patients with and without JAK2 mutations, it is plausible that the JAK2 mutation may have contributed to the onset of carotid dissection in Case 7. These findings suggest that, in patients with JAK2 mutations, the vascular wall may be more susceptible to inflammation and endothelial damage than in those without such mutations, potentially facilitating the formation of platelet thrombi at sites of vascular injury.
However, whether or not JAK2 mutations induce atherosclerosis remains unclear. Experimental mouse models have demonstrated that the JAK2 V617F mutation affects not only platelets but also neutrophils, macrophages, erythrocytes, endothelial cells, and other cell types, thereby causing atherosclerosis2-4, 26-28). While our study found no statistically significant difference in MCAD frequency according to the JAK2 mutation status, patients positive for JAK2 mutations independently showed higher thrombomodulin levels than those negative for JAK2 mutations. Falanga et al. previously reported that patients positive for JAK2 mutations with essential thrombocythemia had significantly higher thrombomodulin levels than mutation-negative patients29). Thrombomodulin, a membrane-bound glycoprotein expressed in endothelial cells, plays a crucial role in maintaining vascular homeostasis through anticoagulation, fibrinolysis regulation, inflammation and angiogenesis suppression, leukocyte adhesion regulation, endothelial cell migration/proliferation regulation, and endothelial permeability suppression30). When endothelial cells are damaged, soluble thrombomodulin is cleaved and released into the circulation, making it a reliable marker of endothelial dysfunction and atherosclerotic disease31-35). Therefore, elevated thrombomodulin levels in JAK2 mutation-positive patients may reflect underlying endothelial dysfunction and indicate an environment conducive to atherosclerosis. Ultimately, the JAK2 mutation may increase the risk of ischemic stroke through both 1) in situ thrombosis from blood cell proliferation and 2) atherosclerosis.
Among the limitations of this study, one patient (Case 2 in Table 3) with a JAK2 mutation was receiving cytoreductive therapy, which may have influenced the blood counts. Nevertheless, the significantly higher hematocrit and platelet counts in this patient suggest that this influence was unlikely to have substantially altered the overall findings. Since this was not a complete survey of all patients with ischemic stroke, the JAK2 mutation-positive rate among all patients with ischemic stroke remains unknown. In addition, caution is needed when comparing JAK2 mutation-positive rates with those in previous reports because of the different detection methods and sensitivities. While JAK2 mutation rates were low regardless of the stroke subtype, with no significant differences noted, patients with MCAD showed higher rates than those without MCAD. Given the low JAK2 mutation-positive rate in patients with ischemic stroke, multicenter studies with larger sample sizes are required.
The JAK2 mutation prevalence was very low in both LAA and SVO patients. Stroke mechanisms in JAK2 mutation-positive patients may involve factors beyond hematopoietic cells, suggesting special attention is needed for ischemic stroke risk assessment and preventive strategies in patients with JAK2 mutation. Importantly, our findings suggest that JAK2 mutations may contribute to stroke risk even before an MPN diagnosis is established.
We express our gratitude to Mari Okamoto for data collection and secretarial assistance. We also thank Editage (www.editage.com) for English language editing.
The authors report no conflicts of interest.
N.O. received funding from the KAWASAKI Foundation for Medical Science and Medical Welfare. This study was also supported by a research grant from the Japan Society for the Promotion of Science KAKENHI (grant number: 21K07429). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
