Missed Opportunities on Emergency Remdesivir Use

On May 1, the US Food and Drug Administration (FDA) issued an Emergency Use Authorization (EUA) for remdesivir for the treatment of hospitalized adults and children with severe coronavirus disease 2019 (COVID-19). An EUA permits the use of an unapproved drug, biologic, or medical device and may be issued by the FDA when the secretary of Health and Human Services has declared a public health emergency “that has a significant potential to affect national security or the health and security of United States citizens living abroad.”¹ As seen with the FDA’s recent revocation of the EUA for hydroxychloroquine, the agency has wide discretion to revoke or revise EUAs when appropriate for public health (the FDA does not have comparable authority to act so quickly with respect to drugs that have been approved).

In the case of remdesivir, the decision to grant an EUA followed the release of a summary interim finding of shorter time to recovery among patients treated with the drug in a National Institute of Allergy and Infectious Diseases (NIAID)–funded phase 3 placebo-controlled clinical trial. In the immediate wake of the EUA, attention focused on the equitable distribution of the limited supply of remdesivir.² However, 2 deeper concerns exist.

First, the EUA does not ensure continued robust evidence generation despite outstanding questions over the safety and efficacy of the drug. Subsequently published results of the NIAID trial based on 1059 patients showed a median 4-day shorter recovery time associated with remdesivir, but did not show a significant increase in survival among patients receiving it.³ A separate phase 3 placebo-controlled trial in China involving 237 patients did not find significant clinical benefits from use of remdesivir in a similar patient population (but was reportedly stopped early because of inadequate enrollment),⁴ and currently unpublished results of another phase 3 trial reportedly did not identify significant clinical improvement among 393 hospitalized patients with moderate COVID-19 who received 10 days of treatment vs standard of care.⁵

Nevertheless, in light of the interim analysis of the NIAID-funded trial, remdesivir can now be offered to patients who had been receiving placebo in the trial, increasing the importance of assessments of the drug in clinical settings to answer unresolved therapeutic questions. Even though the FDA has broad power to impose “appropriate conditions with respect to the collection and analysis of information concerning the safety and effectiveness” of remdesivir,¹ the EUA does not contain terms adequate for monitoring the drug’s use or outcomes. Instead, under the EUA, physicians seeking to administer remdesivir must be given a 36-page fact sheet highlighting instructions for use, possible risks, and required reporting of suspected medication-related serious adverse events through the FDA’s passive surveillance system, which has historically been limited by underreporting and incomplete reporting.

Second, the pricing of remdesivir has not yet been set. Gilead, the manufacturer of remdesivir, has offered to donate 940 000 doses of the drug to the federal government, enough to treat approximately 85 000 to 157 000 patients. However, the manufacturer has stated that it will soon start charging for the drug, which analysts estimate will be used to treat thousands more individuals in the US and may be heavily stockpiled by the federal government. In May, the Institute for Clinical and Economic Review (ICER), an independent nonprofit group that performs cost-effective analyses for drugs, estimated that based on preliminary data and an assumption of a mortality benefit yet to be demonstrated, a cost-effective price for a 10-day course of remdesivir would be approximately $4500.⁶ ICER also cited data suggesting that a minimal price to recover the costs of drug production could be as low as $10, not including future spending on research and development specific to COVID-19, on the grounds that earlier research and development costs had already been recouped from Gilead's successful marketing of drugs for hepatitis C virus, for which remdesivir was originally developed. Gilead has yet to comment on these estimates for what the bounds of a reasonable price could be.

It is not too late to address either shortcoming. First, the FDA should revise its current EUA for remdesivir to require the creation of a patient registry that includes information on patient demographics, treatment dose and duration, and safety outcomes. Physicians who prescribe the drug would be responsible for reporting this information to the registry. Such registries are routinely required as part of risk evaluation and mitigation strategy (REMS) programs that the FDA imposes on drugs with special risks and can help in the identification of important safety signals. To avoid past missteps that occurred with REMS programs,⁷ the registry should be run by the FDA or an independent third party (as opposed to the manufacturer) with deidentified data made publicly available. This recommendation is consistent with the suggestion by White and Angus⁸ that much more could be learned from the “natural experiment” of the distribution of remdesivir via a state-level “lottery” and what will likely be other drugs in limited supply.

Such a registry would have the added benefit of enabling monitoring of disparities in access. COVID-19 has disproportionally affected racial and ethnic minorities in the US. In an analysis of a representative sample of 580 hospitalized patients with COVID-19 with available race or ethnicity data across 99 counties in 14 states, the Centers for Disease Control and Prevention reported that black and Hispanic individuals made up 41% of patients but only 32% of the catchment area populations.⁹ Historically, racial and ethnic minority populations have been slower to receive innovative treatments, an outcome that could be prevented or, at minimum, identified and remedied through transparent reporting.

The federal government should also commence negotiations with Gilead over an appropriate price for the drug. Remdesivir will not solve the pandemic, but it does appear to have some efficacy in reducing recovery time. This should be reflected in its price, which will serve as a signal to other innovators that their efforts will be fairly rewarded.

However, in the case of remdesivir, the price should also reflect the outsized role the US government had in its development. The phase 3 NIAID-designed and government-funded trial is only the most recent example of federal investment in the drug.¹⁰ In 2014, Gilead entered into a research contract with the US Army Medical Research Institute for Infectious Disease to scan Gilead’s library of molecules for potential treatments for Ebola virus disease. Among the candidates identified was remdesivir. Gilead conducted a phase 1 trial to evaluate the safety of the drug in humans, but more advanced studies, including investigations of the effectiveness of remdesivir against Ebola, were designed and funded by NIAID. Through these contributions, taxpayers assumed greater risk as clinical development progressed. This arrangement departs from the typical model of risk sharing, in which pharmaceutical manufacturers assume the primary costs of late-stage drug development, with the lure of market exclusivity and, thus, the ability to price what the market will bear. The trade-off for this off-loading of risk should be affordable access to the product. Such accounting would importantly still permit Gilead to profit from creating value but would be commensurate with its risk-taking.

Ultimately, the EUA for remdesivir presents a critical opportunity for the federal government to set a precedent for future emergency use authorization of investigational drugs in the COVID-19 pandemic and beyond. A framework to ensure adequate safety monitoring and reasonable pricing based on risk-taking, efficacy, and safety would be a positive legacy.

Corresponding Author: Ameet Sarpatwari, PhD, JD, Program On Regulation, Therapeutics, And Law (PORTAL), Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, 1620 Tremont St, Ste 3030, Boston, MA 02120 (asarpatwari@bwh.harvard.edu).

Published Online: June 24, 2020. doi:10.1001/jama.2020.11932

Conflict of Interest Disclosures: Ms Kaltenboeck reported receiving personal fees from AstraZeneca and SVB Leerink, a grant from Kaiser Permanente outside the submitted work, and serving on the Advisory Committee for the ICER report on drug price increases without supporting clinical evidence. No other disclosures were reported.

Funding/Support: Dr Sarpatwari and Dr Kesselheim’s work was supported by Arnold Ventures and the Harvard-MIT Center for Regulatory Science. Ms Kaltenboeck’s work was supported by Arnold Ventures.

Role of the Funder/Sponsor: Arnold Ventures and the Harvard-MIT Center for Regulatory Science had no role in the preparation, review, or approval of the manuscript or decision to submit the manuscript for publication.

Additional Contributions: We thank Jerry Avorn, MD (PORTAL), and Patricia J. Zettler, JD (The Ohio State University Moritz College of Law), for their comments on an earlier draft.