Anticoagulant vs. antiviral therapy, recovery and neurodamage in Long COVID: a real-world prospective cohort study

ABSTRACTRecent estimates consider that Long COVID affects 400 million people worldwide, of which <10% recover after two years. Predictive biomarkers and effective therapies are urgently needed for the clinical management of Long COVID. Neurological manifestations (poor memory, brain fog) are frequent, while neurodamage and neuroinflammatory markers UCHL1, GFAP and NFL were found to be increased in some but not all Long COVID cohorts. However, a direct link between two major Long COVID disease mechanisms, viral persistence and coagulation defects, and CNS damage is unclear. Therefore, we investigated the effect of combined anticoagulant (Asaflow + Clopidogrel) vs. antiviral (Paxlovid) treatment upon clinical outcome and neurodamage markers in Long COVID patients. Using multivariate logistic regression, we found that second-line antiviral treatment is the single best predictor of clinical recovery in a real-world Long COVID cohort (n=106) with long-term (220 person-years) follow-up. Our results support a putative role for viral persistence in Long COVID pathogenesis but also indicate a large therapeutic window for Paxlovid and other antivirals. A rapid decline in neurodamage markers (GFAP/NFL) upon antiviral treatment suggests CNS damage in Long COVID might be therapeutically targeted and should be considered in ongoing and future clinical trials.