Remdesivir and molnupiravir were the only 2 repurposed antivirals that were approved for emergency use during the COVID-19 pandemic. Both drugs received their emergency use authorization on the basis of a single industry-funded phase 3 trial, which was launched after evidence of in vitro activity against SARS-CoV-2. In contrast, for tenofovir disoproxil fumarate (TDF), little in vitro evidence was generated, no randomized trials for early treatment were done, and the drug was not considered for authorization. Yet, by the summer of 2020, observational evidence suggested a substantially lower risk for severe COVID-19 in TDF users compared with nonusers. The decision-making process for the launching of randomized trials for these 3 drugs is reviewed. Observational data in favor of TDF was systematically dismissed, even though no viable alternative explanations were proposed for the lower risk for severe COVID-19 among TDF users. Lessons learned from the TDF example during the first 2 years of the COVID-19 pandemic are described, and the use of observational clinical data to guide decisions about the launch of randomized trials during the next public health emergency is proposed. The goal is that gatekeepers of randomized trials make better use of the available observational evidence for the repurposing of drugs without commercial value.