Increased FXII activation in endovascular treated stroke patients on periprocedural antithrombotic treatment

Abstract Endovascular treatment (EVT) for large vessel occlusion in acute ischemic stroke patients (AIS) improves reperfusion rates, yet a considerable number of patients do not recover despite successful recanalization. Microvascular occlusions may underlie these reperfusion failures after EVT and periprocedural antithrombotic agents - unfractionated heparin (UFH) and/or acetylsalicylic acid (ASA)- may be administered to mitigate these complications. However, the MR CLEAN MED Trial, investigating the safety and efficacy of periprocedural UFH, ASA or UFH+ASA in AIS patients undergoing EVT, found an association between administration of these antithrombic agents and a higher rate of symptomatic intracranial hemorrhages. Our recent work showed that EVT elicited activation of the contact system and microthrombi formation in a large animal stroke model. Neutrophil activation and NETosis may also impact secondary thromboinflammation, potentiating thrombogenesis even in the presence of UFH and ASA, while exacerbating bleeding risk. Thus, safer and more effective antithrombotic drugs are needed. Prior studies implicate neutrophil extracellular traps (NETs) in AIS through direct cellular activation, coagulation stimulation, and inflammation. Heparin can dismantle NETs, potentially improving outcomes; however, links between NETs and coagulation activity in AIS patients remain unclear. To better characterize the thrombo-inflammatory mechanisms in humans we carried out a prespecified substudy of the MR CLEAN MED Trial, addressing plasma biomarkers, exploring the full coagulation cascade alongside markers of neutrophil activation/NETosis. The clinical outcomes included function (in)dependence (mRS), mortality, stroke severity (NIHSS) and follow-up infarct volume (FIV). Plasma samples collected at three time points (T0 = baseline; T1 = 1 hour post-EVT; T2 = 24 hours post-EVT) of 216 patients from the MR CLEAN MED Trial, were included. Coagulation activity was assessed by measuring activated coagulation factor (F) XII (FXIIa), FXIa, FIXa, FXa, FVIIa and thrombin (T) in complex with their natural inhibitors (antithrombin (AT), C1-esterase inhibitor (C1Inh) or alpha-1-antitrypsin (a1AT)) using ELISA. NETs markers (Histone-DNA, MPO-DNA complex, citrullinated Histone H3) were measured with ELISA. Correlations between coagulation activity and NETs were analyzed via Pearson correlation; related-Samples Friedman's two-way ANOVA by ranks and multivariable regression was used for analyzing differences in coagulation and associations with outcomes, including interaction terms for coagulation activity and NETs. Among patients receiving EVT only (n=53), FXIIa:AT and FXIIa:C1Inh showed no significant changes over time. In the UFH (n=60) and UFH+ASA (n=55) groups, FXIIa:AT significantly increased at T1 compared to baseline (UFH: 35.8 pM [17.2–111.2] vs. 116.5 pM [52.9–284.5], p=0.002; UFH+ASA: 31.2 pM [20.8–76.1] vs. 94.8 pM [45.2–182.0], n.s.), returning to baseline by T2 (31.2 pM [16.2–51.5], p=0.000; 28.5 pM [12.6–57.3], p=0.016, respectively). ASA-treated patients (n=48) showed a significant increase in FXIIa:C1Inh at T1 (1210.2 pM [990.1–1368.0] vs. 1478.4 pM [1224.3–1590.6], p=0.031). FXIa:C1Inh showed a similar pattern in the ASA group, but not in UFH-treated patients. In the UFH+ASA group, FXIa:a1AT decreased at T1 versus baseline (29.5 pM [13.7–65.1] vs. 18.1 pM [3.6–37.4], p=0.071), returning near baseline at T2 (33.5 pM [5.4–47.0], n.s.). Standard care showed no change in FXIIa or FXIa, but FVIIa:AT increased at T1 (153.3 pM [80.2–386.7] vs. 302.3 pM [190.0–761.3], n.s.). Coagulation activity showed low to moderate correlation with NETs markers in all groups. In the UFH+ASA group, there was a moderate to strong correlation between FXIIa:AT and FXIa:a1AT, FXIa:α1AT and FXa:AT, and FXIa:a1AT and histone-DNA. No significant associations were found between coagulation activity and outcomes. In summary, periprocedural antithrombotic treatment in EVT-treated stroke patients increased FXII activation compared to EVT only. However, coagulation activity did not associate with outcomes, nor were interactions found between NETs presence and coagulation activity on outcome measures. Based on these and previous animal data, FXIIa or downstream FXIa ± NETosis inhibition may hold promise to improve periprocedural antithrombotic management, balancing efficacy and bleeding risk.