VEXAS Syndrome and Thrombosis: Findings of Inflammation, Hypercoagulability, and Endothelial Dysfunction

VEXAS (vacuoles, E1 enzyme, X-linked, autoinflammatory, somatic) syndrome is a novel autoinflammatory syndrome due to a ubiquitin like modifier activating enzyme 1 (UBA1) somatic mutation, recently discovered in 2020.[1] This mutation affects the major E1 enzyme that initiates ubiquitin conjugation of cellular proteins meant for degradation by proteasomes, where decreased ubiquitination causes accumulation of prominent intracellular vacuoles seen in myeloid and erythroid precursors in the bone marrow.[2] This autoinflammatory disease has a significant hematologic malignancy and thrombotic burden, with an increased incidence of venous thromboembolism (VTE) (36.4%), where deep vein thrombosis (DVT) is more common than pulmonary embolism (PE), with a lower incidence of arterial thrombosis (1.6%).[3] However, limited information is available on the potential mechanisms of thrombosis in patients with VEXAS syndrome. We describe two male patients with VEXAS syndrome diagnosed in Singapore and performed biomarkers evaluating the hemostatic, inflammatory, and endothelial function.

Patient 1 is a 69-year-old Chinese male with a past medical history of hyperlipidemia, hypothyroidism, fatty liver, and stage 3A chronic kidney disease who presented in July 2019[3] with prolonged fever, urticarial eruption, painful cervical lymphadenitis, migratory arthralgia, myalgia, macrocytic anemia, and raised C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR). He was treated with methotrexate 12.5 mg/wk and a tapering course of prednisolone from 0.5 mg/kg/d. He subsequently developed relapsing polychondritis and had recurrent flares of urticaria and painful lymphadenopathy, diagnosed on excision biopsy to be suggestive of Kikuchi's disease. In September 2020, he presented with acute right lower limb swelling with extensive DVT from the right common femoral vein to the popliteal vein seen on Doppler ultrasound. Computer tomography scans performed to exclude malignancy showed concurrent bilateral PE. His blood investigations showed worsening macrocytic anemia, mild thrombocytopenia, and persistently raised CRP and ESR. Antiphospholipid antibodies (anti-cardiolipin IgG and IgM, anti-β2-glycoprotein I, lupus anticoagulant) were negative. Antithrombin and homocysteine levels were normal. The acute thrombosis likely resulted in a mildly depressed protein C activity at 65% (reference range: 70–150%) and protein S activity at 53% (reference range: 65–130%).

Given the history of prolonged fevers, rash, relapsing polychondritis, polyarthritis, and cytopenias, Sanger sequencing of peripheral blood and buccal swab samples was performed to confirm the clinical suspicion of VEXAS syndrome. This revealed a somatic mutation of UBA1 variant p.Met41leu (C121A- > C), confirming the diagnosis of VEXAS syndrome. A retrospective review of his bone marrow examination performed 1 year ago also revealed characteristic vacuolation in the myeloid series. Hydroxychloroquine 200 mg/d was added to his treatment regime of methotrexate and prednisolone. Anticoagulation of his VTE initially was with enoxaparin 1 mg/kg twice daily subcutaneously and later switched to apixaban 5 mg twice daily. His right lower limb swelling resolved clinically without any evidence of postthrombotic syndrome. After 6 months of therapeutic anticoagulation, the apixaban dose was reduced to 2.5 mg twice daily without any further thrombotic sequelae.

Patient 2 is a 63-year-old Chinese male with a past medical history of provoked distal lower limb DVT with previous anticoagulation, gout, benign prostatic hyperplasia, and erosive gastritis. He presented in 2018 with fever, loss of weight, rashes with medium vessel vasculitis and panniculitis seen on skin biopsy, lymphadenopathy, macrocytic anemia, and pulmonary infiltrates with organizing pneumonia seen on lung biopsy. Inflammatory markers interleukin (IL)-1 receptor antagonist, interferon γ-induced protein 10 (IP-10), macrophage inflammatory protein-1α, IL-6, interferon-γ (IFN-γ), and monocyte chemoattractant protein-1 were elevated on a 27-plex Luminex assay after starting prednisolone and azathioprine as a steroid-sparer. IP10 was markedly elevated at 34,539 pg/mL (reference range: 0–954 pg/mL). Autoimmune antibody screen showed the presence of antinuclear antibody (1:320, mixed homogenous and nucleolar pattern) and a single positive lupus anticoagulant. Repeat lupus anticoagulant and antiphospholipid antibodies were negative. A somatic mutation of UBA1 variant p.Met41Thr was detected, confirming the diagnosis of VEXAS syndrome. The patient required high doses of prednisolone of at least 55 mg daily and failed the following steroid sparers: (1) anakinra (large local reaction), (2) azathioprine (due to worsening anemia), (3) methotrexate (stopped due to concerns of methotrexate pneumonitis), (4) tocilizumab (stopped due to paradoxical reaction with cervical lymphadenopathy); and (5) tofacitinib (ineffective). Ruxolitinib was used eventually as alternative steroid sparer with some efficacy in reducing the steroid requirement. The patient finally underwent haploidentical stem cell transplantation in view of the high-dose prednisolone requirement and costly ruxolitinib treatment.

Informed consent was obtained from patient 1 and patient 2. We performed venesection of 20 mL of whole blood per patient in the outpatient setting when they were well. Patient 1 was on prednisolone 15 mg daily, methotrexate 7.5 mg/wk, and hydroxychloroquine 200 mg daily, and patient 2 was on prednisolone 25 mg/d and anakinra 100 mg/d at the time of blood sampling. Coagulation tests were performed on platelet poor plasma using the STA R Max Series coagulation analyzer (Diagnostica Stago, France) and Sysmex CN-6000 automated coagulation analyzer (Sysmex Corporation, Kobe, Japan). Prothrombin time (PT) was measured with STA Neoplastine CI Plus 10, activated partial thromboplastin time with STA Cephascreen, fibrinogen (modified Clauss) with STA Liquid FIB, D-dimer with STA Liatest D-Dimer. Clotting factor levels (factor V and VIII) were measured with STA Deficient V and VIII, respectively. von Willebrand factor (VWF) antigen was assayed with an immunoturbidimetric method using STA Liatest VWF: Ag kit and VWF Glycoprotein 1B binding activity was performed using INNOVANCE VWF Ac assay (Siemens). Thrombin generation was performed using the ST-Genesia (Stago, Asnières-sur-Seine, France). Clot waveform analysis was performed with Sysmex CN-6000 with parameters obtained from PT using Innovin (Siemens Healthcare, Marburg, Germany) and for aPPT, Dade Actin FSL (Siemens Healthcare) as per International Society on Haemostasis and Thrombosis Scientific and Standardization Committee recommendation. Endothelial inflammatory markers including intercellular adhesion molecule 1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), and P-selectin were performed (R & D Systems, ThermoFisher Scientific). Tests for NETosis were also performed using Quant-iT PicoGreen dsDNA assay (ThermoFisher Scientific, P7589) for plasma cell-free DNA, western blotting for plasma citrullinated histone H4 (H4Cit, with primary antibody from Millipore, 07-596), and cell-based NETosis propensity assay using neutrophils isolated by density gradient centrifugation using Histopaque-1119 (Sigma, 11191) and Percoll (Cytiva, 17089101).

The blood samples showed significant hypercoagulability with elevated factor VIII levels, hyperfibrinogenemia, and raised endothelial biomarkers including ICAM-1, VCAM-1, VWF:Ag, and VWF glycoprotein 1B binding activity in our two patients, with normal IL-6 levels ([Table 1]). This correlated with the global hemostatic tests performed, with thrombin generation in patient 2 having an elevated peak and increased endogenous thrombin potential (in the presence of thrombomodulin), and clot wave analysis for both patients showing features of hypercoagulability with increased clot velocity (minimum 1), acceleration (minimum 2), deceleration (maximum 2), and delta change. Patient 1 was on long-term apixaban; hence, thrombin generation was not expected to be significantly elevated due to the anti Xa effect. Signs of NETosis were examined, where the PicoGreen assay showed no increase in plasma cell-free DNA levels, which was corroborated by a lack of H4Cit signal as examined by western blotting. Cell-based NETosis propensity assay using neutrophils freshly isolated from patient blood showed that there was no exacerbation of NET formation and peptidylarginine deiminase 4 activity as reflected by immunostaining of H4Cit. The absence of demonstrable NETosis was likely because both patients were on high doses of prednisolone, which would have inhibited neutrophil function at the time of blood sampling.

Table 1 Summary of hemostatic, endothelial, and inflammatory parameters of two patients with VEXAS syndrome

		Reference range	Patient 1	Patient 2
Hemostatic tests	PT (s)	11.7–14.0	14.7	29.6
	aPTT (s)	27.0–37.0	27.3	29.6
	Thrombin clotting time (s)	15.0–18.0	15.6	15.8
	Fibrinogen (g/L)	1.8–4.5	4.62	6.68
	D-dimer (µg/mL)	<0.50	0.44	0.61
	Factor VIII (%)	60–150	268	308
	Factor V (%)	70–120	107	111
Endothelial markers	ICAM-1 (ng/mL)	<95	115.6	186.5
	VCAM-1 (ng/mL)	<187	222.8	322.4
	P-Selectin (ng/mL)	<103	59.3	57.0
	VWF:Ag (%)	56–160	201	278
	Glycoprotein-1B Ac	56–162	233	296
Inflammatory markers	Interleukin-6 (pg/mL)	<2	1.45	0.215
Thromboscreen without thrombomodulin (absolute)	Lag time (min)	1.42–2.97	4.2	3.81
	Peak height (nM)	97–290	157.5	315.2
	Endogenous thrombin potential (nM*min)	741–1654	994	1514
Thromboscreen with thrombomodulin	Lag time (min)	1.45–3.38	6.68	4.36
	Peak height (nM)	11–193	118.7	269.4
	Endogenous thrombin potential (nM*min)	50–806	497	1136
Clot waveform analysis	Min1 (PT)	1.95–5.67	6.127	8.514
	Min2 (PT)	0.97–2.93	3.135	4.4
	Max2 (PT)	0.75–2.35	2.47	3.578
	Delta change (PT) (%)	6.52–17.28	19.1	25.1
	Min1 (aPTT)	2.86–6.78	7.392	9.281
	Min2 (aPTT)	0.46–1.10	1.251	1.522
	Max2 (aPTT)	0.37–0.93	1.065	1.329
	Delta change (aPTT) (%)	25.21–63.09	63.2	80.3

Abbreviations: aPTT, activated partial thromboplastin time; ICAM, intercellular adhesion molecule; Max, maximum; Min, minimum; PT, prothrombin time; VCAM, vascular cell adhesion molecule; VWF, von Willebrand factor.

Note: Bold values are abnormal.

We had hypothesized in our previous review[4] ([Fig. 1]) that the development of VTE originates from somatic mutations occurring at methionine-41 (p.Met41) within the UBA1 gene. These mutations impact the major E1 enzyme responsible for initiating the ubiquitin conjugation of cellular proteins designated for degradation through proteasomes.[1] A reduction in ubiquitination processes leads to the accumulation of intracellular vacuoles, resulting in the presence of abnormal hematopoietic myeloid precursors, aberrant neutrophils, and monocytes. These, in turn, activate innate immune pathways through cytokine release, thus inciting severe inflammation.

Fig. 1 Virchow's triad showing key molecular and cellular mechanisms of thrombosis in VEXAS syndrome (Author's original figure from Reference [3]).

Evidence for increased inflammation is seen in the immunoprofiling of T cells showing an inversion of the CD4:CD8 ratio, consistent with acquired T cell activation associated with inflammation,[1] [4] as well as an increased proportion of spliced X-box binding protein 1 in monocytes and the activation of the unfolded protein response (UPR), resulting in atypical differentiation of monocytes and the loss of non-classical (CD14 dimCD16 + ) and intermediate monocyte populations.[1] Sustained overactivation of the UPR eventually induces cellular stress, intensifies the inflammatory response, and culminates in cellular apoptosis.[5] In addition to IFN-γ, type I IFN can induce monocyte production of IP10 and the UBA1 mutation has been thought to induce inflammation through the activation of UPR, leading to a type I IFN response.[6] [7] The markedly elevated IP10 in patient 2 may be in keeping with a type I IFN/IP10 axis, but IFNα was not measured in our Luminex assay.

Inflammation in VEXAS syndrome and clonal hematopoiesis can lead to a hypercoagulable state and is associated with an imbalance in procoagulant and anticoagulant factors. We observed hypercoagulability in the blood specimens of our two patients, noting elevated factor VIII and fibrinogen levels. Obiorah et al[4] in their case series of 16 patients reported 9 patients with VTE, with 3 out of 5 patients having high factor VIII levels and 1 patient having a high factor IX level, with no significantly abnormal VWF levels. Elevated factor VIII levels are associated with increased risk of venous thrombosis via enhanced thrombin formation with increased risk of vascular thrombosis, where patients with factor VIII activity levels ≥ 150 IU/dL are associated with a five- to six-fold increased risk of venous thrombosis compared with levels < 100 IU/dL.[8]

Both patients had raised endothelial markers ICAM-1, VCAM-1, VWF:Ag, and glycoprotein 1B binding activity, with the presence of vasculitis in VEXAS syndrome in patient 2, suggesting evidence for ongoing endothelial dysfunction. In a recent literature review vascular inflammation with involvement of small, medium, and large vessels was identified[9] and in a report by Karadeniz et al of two patients with VEXAS syndrome there was severe thickening of the walls of large veins detected on Doppler ultrasound, suggestive of venulitis.[10]

In summary, given the findings of inflammation, hypercoagulability and endothelial dysfunction, long-term anticoagulation treatment in VEXAS patients with known thrombosis, together with anti-inflammatory and immunosuppressive therapies, may provide an appropriate therapeutic approach to VEXAS syndrome. Further well-designed studies evaluating larger cohorts of patients with VEXAS syndrome are required to assess the baseline prothrombotic and inflammatory states preceding immunosuppressive therapy and in VEXAS patients with recurrent thrombosis.

Authors' Contributions

B.E.F. and S.H.T. had full access to all data and take responsibility for the integrity of the data. Drafting of the manuscript: B.E.F. Performing of laboratory tests: C.L.L.S., B.P.L.L., L.D.W., S.L.W., L.L.G. Acquisition, analysis, or interpretation of data and critical revision of manuscript for important intellectual content: All authors.

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Patient Consent

Informed consent was obtained by the study team.

Publication History

Article published online:
05 January 2024

• References

• 1 Beck DB, Ferrada MA, Sikora KA. et al. Somatic mutations in UBA1 and severe adult-onset autoinflammatory disease. N Engl J Med 2020; 383 (27) 2628-2638
• 2 Fan BE, Cao L, Gallardo CA. et al. Myeloid and lymphoid vacuolation in VEXAS syndrome. Am J Hematol 2021; 96 (08) 1056-1057
• 3 Oo TM, Koay JTJ, Lee SF, Lee SMS, Lim XR, Fan BE. Thrombosis in VEXAS syndrome. J Thromb Thrombolysis 2021; 53 (04) 965-970
• 4 Obiorah IE, Patel BA, Groarke EM. et al. Benign and malignant hematologic manifestations in patients with VEXAS syndrome due to somatic mutations in UBA1. Blood Adv 2021; 5 (16) 3203-3215
• 5 Ivanova EA, Orekhov AN. The role of endoplasmic reticulum stress and unfolded protein response in atherosclerosis. Int J Mol Sci 2016; 17 (02) 193
• 6 Kokuzawa A, Nakamura J, Kamata Y, Sato K. Potential role of type I interferon/IP-10 axis in the pathogenesis of anti-MDA5 antibody-positive dermatomyositis. Clin Exp Rheumatol 2023; 41 (02) 275-284
• 7 Beck DB, Werner A, Kastner DL, Aksentijevich I. Disorders of ubiquitylation: unchained inflammation. Nat Rev Rheumatol 2022; 18 (08) 435-447
• 8 Jenkins PV, Rawley O, Smith OP, O'Donnell JS. Elevated factor VIII levels and risk of venous thrombosis. Br J Haematol 2012; 157 (06) 653-663
• 9 Watanabe R, Kiji M, Hashimoto M. Vasculitis associated with VEXAS syndrome: a literature review. Front Med (Lausanne) 2022; 9: 983939
• 10 Karadeniz H, Cerit M, Avanoğlu Güler A, Tufan A, Kanthi Y. Venous inflammation might be one of the features of VEXAS syndrome and associated thrombosis. Rheumatology (Oxford) 2023; 62 (09) e269-e270