Abstract
Background: Azvudine, as an antiviral drug, has been approved for the treatment of COVID-19, and multiple randomized controlled trials (RCTs) and retrospective cohort studies have been conducted. This study aimed to systematically evaluate the efficacy and safety of Azvudine in treating COVID-19 patients. Methods: As of December 1, 2023, we searched databases including PubMed, Web of Science, Ovid, ICTRP, Cochrane Library, Clinical Trials, MedRxiv, and Springer Link for relevant RCTs and retrospective cohort studies. EndNote X9 was used for literature screening and management, and R software was employed for meta-analysis. Results: A total of 1142 COVID-19 patients from five RCTs were included, with 575 patients receiving Azvudine treatment. Azvudine significantly reduced the hospitalization time and the time to nucleic acid conversion to negative in patients with mild to moderate COVID-19. However, compared to the control group, Azvudine did not significantly reduce the incidence of adverse events (AEs) (risk ratio: 0.89, 95% confidence interval [CI]: 0.80, 1.00). Additionally, eight ongoing clinical trials were included to evaluate the efficacy and safety of Azvudine. In fourteen retrospective cohort studies, a total of 6602 COVID-19 patients were analyzed, with 3118 patients receiving Azvudine treatment. Azvudine significantly reduced all-cause mortality (odds ratio [OR]: 0.49, 95% CI: 0.38, 0.63). The incidence of AEs in the Azvudine group and the Nirmatrelvir/Ritonavir group was 4.13% (60/1453) and 5.08% (67/1319), respectively, indicating that Azvudine significantly reduced the incidence of AEs compared to Nirmatrelvir/Ritonavir (OR: 0.68, 95% CI: 0.47, 0.98). Conclusions: Azvudine significantly reduced the hospitalization time and the time to nucleic acid conversion to negative in COVID-19 patients and significantly lowered all-cause mortality (Grading of Recommendations Assessment, Development, and Evaluation [GRADE]: high-certainty evidence). In terms of safety, Azvudine demonstrated a favorable safety profile (GRADE: moderate-certainty evidence because of suspected publication bias and residual confounding). Further large-scale studies are needed to validate its efficacy and safety.