Impact of concomitant hydroxychloroquine use on safety and efficacy of remdesivir in moderate COVID-19 patients

ABSTRACT

Background: Remdesivir (RDV) has been shown to shorten recovery time and was well tolerated in patients with severe COVID-19. Hydroxychloroquine (HQN) is an experimental treatment for COVID-19. Effects of coadministration of HQN with RDV have not been studied and are relevant given the long half-life (∼22 days) of HQN. We report the impact of concomitant HQN and RDV use on clinical outcomes and safety in patients with moderate COVID-19. Methods: We enrolled hospitalized patients with confirmed SARS-CoV-2 infection, oxygen saturation >94% on room air, and radiological evidence of pneumonia. Patients were randomized 111 to receive 5d or 10d of intravenous RDV once daily plus standard of care (SoC), or SoC only. We compared patients on concomitant HQN (HQNpos) vs not (HQNneg). Clinical recovery was evaluated using Cox proportional hazards. Covariate adjustment included age, sex, race, region, symptom duration, oxygen support status and obesity. Recovery and adverse events (AEs) were assessed through death, discharge, or d14. Results: Of 584 patients, 199 (34%) received HQN (5d RDV n=57 [30%];10d RDV, n=49 [25%];SoC n=93 [47%]). Through median follow-up of 13d (range 1-41d), HQNpos patients on 5d or 10d RDV had a lower recovery rate (adjusted HR [95% CI] 0.78 [0.59, 1.03], p=0.09) with longer median time to recovery (8 vs 6 days) compared to HQNneg. HQNpos compared to HQNneg patients in 5d RDV showed a trend of reduced recovery rate (HR 0.69 [0.45,1.04], p=0.080);such an effect was not observed in 10d RDV or SoC (Table 1). More HQNpos than HQNneg patients had AEs in RDV (5/10d) or SoC arms evaluated separately, and all arms combined. This difference was significant for AEs and SAEs for all arms combined after covariate adjustment (Table 2). Conclusion: In moderate COVID-19 patients, concomitant HQN may delay recovery on RDV and showed no impact on recovery with SoC alone. The AE profile of HQNpos patients was worse than that observed for HQNneg patients, regardless of RDV treatment.