Trial Design and Oversight
We conducted a randomized, double-blind, placebo-controlled trial to evaluate postexposure prophylaxis with hydroxychloroquine after exposure to Covid-19.12 We randomly assigned participants in a 1:1 ratio to receive either hydroxychloroquine or placebo. Participants had known exposure (by participant report) to a person with laboratory-confirmed Covid-19, whether as a household contact, a health care worker, or a person with other occupational exposures.
Trial enrollment began on March 17, 2020, with an eligibility threshold to enroll within 3 days after exposure; the objective was to intervene before the median incubation period of 5 to 6 days. Because of limited access to prompt testing, health care workers could initially be enrolled on the basis of presumptive high-risk exposure to patients with pending tests; however, on March 23, eligibility was changed to exposure to a person with a positive polymerase-chain-reaction (PCR) assay for SARS-CoV-2, with the eligibility window extended to within 4 days after exposure.
This trial was approved by the institutional review board at the University of Minnesota and conducted under a Food and Drug Administration Investigational New Drug application. In Canada, the trial was approved by Health Canada; ethics approvals were obtained from the Research Institute of the McGill University Health Centre, the University of Manitoba, and the University of Alberta.
We included participants who had household or occupational exposure to a person with confirmed Covid-19 at a distance of less than 6 ft for more than 10 minutes while wearing neither a face mask nor an eye shield (high-risk exposure) or while wearing a face mask but no eye shield (moderate-risk exposure). Participants were excluded if they were younger than 18 years of age, were hospitalized, or met other exclusion criteria (see the Supplementary Appendix, available with the full text of this article at NEJM.org). Persons with symptoms of Covid-19 or with PCR-proven SARS-CoV-2 infection were excluded from this prevention trial but were separately enrolled in a companion clinical trial to treat early infection.
Recruitment was performed primarily with the use of social media outreach as well as traditional media platforms. Participants were enrolled nationwide in the United States and in the Canadian provinces of Quebec, Manitoba, and Alberta. Participants enrolled themselves through a secure Internet-based survey using the Research Electronic Data Capture (REDCap) system.13 After participants read the consent form, their comprehension of its contents was assessed; participants provided a digitally captured signature to indicate informed consent. We sent follow-up e-mail surveys on days 1, 5, 10, and 14. A survey at 4 to 6 weeks asked about any follow-up testing, illness, or hospitalizations. Participants who did not respond to follow-up surveys received text messages, e-mails, telephone calls, or a combination of these to ascertain their outcomes. When these methods were unsuccessful, the emergency contact provided by the enrollee was contacted to determine the participant’s illness and vital status. When all communication methods were exhausted, Internet searches for obituaries were performed to ascertain vital status.
Randomization occurred at research pharmacies in Minneapolis and Montreal. The trial statisticians generated a permuted-block randomization sequence using variably sized blocks of 2, 4, or 8, with stratification according to country. A research pharmacist sequentially assigned participants. The assignments were concealed from investigators and participants; only pharmacies had access to the randomization sequence.
Hydroxychloroquine sulfate or placebo was dispensed and shipped overnight to participants by commercial courier. The dosing regimen for hydroxychloroquine was 800 mg (4 tablets) once, then 600 mg (3 tablets) 6 to 8 hours later, then 600 mg (3 tablets) daily for 4 more days for a total course of 5 days (19 tablets total). If participants had gastrointestinal upset, they were advised to divide the daily dose into two or three doses. We chose this hydroxychloroquine dosing regimen on the basis of pharmacokinetic simulations to achieve plasma concentrations above the SARS-CoV-2 in vitro half maximal effective concentration for 14 days.14 Placebo folate tablets, which were similar in appearance to the hydroxychloroquine tablets, were prescribed as an identical regimen for the control group. Rising Pharmaceuticals provided a donation of hydroxychloroquine, and some hydroxychloroquine was purchased.
The primary outcome was prespecified as symptomatic illness confirmed by a positive molecular assay or, if testing was unavailable, Covid-19–related symptoms. We assumed that health care workers would have access to Covid-19 testing if symptomatic; however, access to testing was limited throughout the trial period. Covid-19–related symptoms were based on U.S. Council for State and Territorial Epidemiologists criteria for confirmed cases (positivity for SARS-Cov-2 on PCR assay), probable cases (the presence of cough, shortness of breath, or difficulty breathing, or the presence of two or more symptoms of fever, chills, rigors, myalgia, headache, sore throat, and new olfactory and taste disorders), and possible cases (the presence of one or more compatible symptoms, which could include diarrhea).15 All the participants had epidemiologic linkage,15 per trial eligibility criteria. Four infectious disease physicians who were unaware of the trial-group assignments reviewed symptomatic participants to generate a consensus with respect to whether their condition met the case definition.15
Secondary outcomes included the incidence of hospitalization for Covid-19 or death, the incidence of PCR-confirmed SARS-CoV-2 infection, the incidence of Covid-19 symptoms, the incidence of discontinuation of the trial intervention owing to any cause, and the severity of symptoms (if any) at days 5 and 14 according to a visual analogue scale (scores ranged from 0 [no symptoms] to 10 [severe symptoms]). Data on adverse events were also collected with directed questioning for common side effects along with open-ended free text. Outcome data were measured within 14 days after trial enrollment. Outcome data including PCR testing results, possible Covid-19–related symptoms, adherence to the trial intervention, side effects, and hospitalizations were all collected through participant report. Details of trial conduct are provided in the protocol and statistical analysis plan, available at NEJM.org.
We anticipated that illness compatible with Covid-19 would develop in 10% of close contacts exposed to Covid-19.9 Using Fisher’s exact method with a 50% relative effect size to reduce new symptomatic infections, a two-sided alpha of 0.05, and 90% power, we estimated that 621 persons would need to be enrolled in each group. With a pragmatic, Internet-based, self-referral recruitment strategy, we planned for a 20% incidence of attrition by increasing the sample size to 750 participants per group. We specified a priori that participants who were already symptomatic on day 1 before receiving hydroxychloroquine or placebo would be excluded from the prophylaxis trial and would instead be separately enrolled in the companion symptomatic treatment trial.
Because the estimates for both incident symptomatic Covid-19 after an exposure and loss to follow-up were relatively unknown in early March 2020,9 the protocol prespecified a sample-size reestimation at the second interim analysis. This reestimation, which used the incidence of new infections in the placebo group and the observed percentage of participants lost to follow-up, was aimed at maintaining the ability to detect an effect size of a 50% relative reduction in new symptomatic infections.
An independent data and safety monitoring board externally reviewed the data after 25% and 50% of the participants had completed 14 days of follow-up. Stopping guidelines were provided to the data and safety monitoring board with the use of a Lan–DeMets spending function analogue of the O’Brien–Fleming boundaries for the primary outcome. A conditional power analysis was performed at the second and third interim analysis with the option of early stopping for futility. At the second interim analysis on April 22, 2020, the sample size was reduced to 956 participants who could be evaluated with 90% power on the basis of the higher-than-expected event rate of infections in the control group. At the third interim analysis on May 6, the trial was halted on the basis of a conditional power of less than 1%, since it was deemed futile to continue.
We assessed the incidence of Covid-19 disease by day 14 with Fisher’s exact test. Secondary outcomes with respect to percentage of patients were also compared with Fisher’s exact test. Among participants in whom incident illness compatible with Covid-19 developed, we summarized the symptom severity score at day 14 with the median and interquartile range and assessed the distributions with a Kruskal–Wallis test. We conducted all analyses with SAS software, version 9.4 (SAS Institute), according to the intention-to-treat principle, with two-sided type I error with an alpha of 0.05. For participants with missing outcome data, we conducted a sensitivity analysis with their outcomes excluded or included as an event. Subgroups that were specified a priori included type of contact (household vs. health care), days from exposure to enrollment, age, and sex.