Favipiravir versus Arbidol for COVID-19: A Randomized Clinical Trial

Chang Chen; Jianying Huang; Zhenshun Cheng; Jianyuan Wu; Song Chen; Yongxi Zhang; Bo Chen; Mengxin Lu; Yongwen Luo; Jingyi Zhang; Ping Yin; Xinghuan Wang

Our trial has several limitations. Firstly, for COVID-19, there is no clinically proven effective antiviral drug to serve as the control arm. Although Chinese guideline had recommended several options including Arbidol, 11 no RCT results on these drugs were reported. Arbidol was widely used by Chinese doctors in the beginning stage of this epidemic of COVID-19 (Jan. 1-30, 2020) based on in vitro evidence. 12 For ethical reasons, we chose Arbidol for the control arm. Secondly, observation time frame was limited due to the urgency of this epidemic. For the same reason, no relapse (including nucleic acid conversion, pyrexia, cough, or pneumonia progress by radiology) tracking were performed for the discharged patients. Thirdly, in the inclusion criteria, we did not force positive nucleic acid test as a necessity. The accuracy of nucleic acid assay was limited, which might due to multiple reasons including previous treatment, latency of onset, sampling method, biological specimen characteristics. This particular accuracy problem was a known issue among clinical practitioners across the world. It was estimated that the assay might have at most 30%-50% of sensitivity for patients in early stage of the disease, whilst contact history, clinical manifestations, radiology evidences, and lab results including leukopenia and lymphopenia could be confirmatory for these nucleic-acid-negative pneumonia patients. In the Chinese guideline, 11 patients meeting these criteria were considered as All rights reserved. No reuse allowed without permission. author/funder, who has granted medRxiv a license to display the preprint in perpetuity.