Prophylaxis Against COVID-19 With Hydroxychloroquine and Chloroquine: Comment on the Article by Putman et al.

To the Editor:

We read with great interest the article by Dr. Putman and colleagues, which reviews data from 45 studies evaluating hydroxychloroquine (HCQ), chloroquine (CQ), anakinra, and interleukin‐6 (IL‐6) inhibitor therapies in the treatment of COVID‐19 (1). Except anakinra, none of the other therapies decreased the risk of death in hospitalized COVID‐19 patients. We would like to discuss the evidence evaluating the role of HCQ as prophylaxis against SARS–CoV‐2 infections. The in vitro antiviral effect of antimalarials suggested a role in preventing disease progression (2). A meta‐analysis of 5,577 participants from 5 randomized controlled trials suggested that outpatient treatment with HCQ (as opposed to treatment in hospitalized patients) reduced the incidence of the composite outcome of SARS–CoV‐2 infection, hospitalization, and death; serious adverse events were not reported and cardiac arrhythmias were rare (3). Several studies suggested that HCQ had no benefit in the prophylaxis against SARS–CoV‐2 (4, 5, 6, 7, 8) (Table 1). Only 1 open‐label, controlled trial showed lower incidence of COVID‐19 in the HCQ group; however, imperfect methodology raises concerns over its validity (9). The bulk of the evidence suggests that HCQ has limited or no utility in the prophylaxis for SARS–CoV‐2 infections.

Table 1

Studies of CQ and HCQ as prophylaxis against SARS–CoV‐2*

Author (ref.), date of publication, country	Methods	Population	Group	Dosage	Outcome measures	Results
Boulware et al (6), June 3, 2020, US and Canada	Double‐blind, placebo‐controlled, randomized	821 exposed within 4 days	HCQ (n = 414); placebo (n = 407)	800 mg/day; 600 mg/day for 4 days	Incidence of laboratory‐confirmed or symptomatic illness	Symptomatic COVID‐19 in 49 (11.8%) of 414 receiving HCQ vs. 58 (14.3%) of 407 receiving placebo (P = 0.35)
Salvarani et al (8), August 6, 2020, Italy	Population‐based study	2,251,903 residents of Bologna, Modena, and Reggio Emilia	CQ/HCQ (n = 4,408); general population (n = 2,112,319)	≥1 prescription of CQ or HCQ for approved indications	Incidence of testing; incidence of positive result; probability of being found positive once tested	OR for testing 1.09 (95% CI 0.94–1.28); OR for Dx 0.94 (95% CI 0.66–1.34); OR for positive result 0.83 (95% CI 0.56–1.23)
Rajasingham et al (5), September 18, 2020, US and Canada	Double‐blind, placebo‐controlled, randomized	1,483 HCWs	Once weekly HCQ (n = 494); twice weekly HCQ (n = 495); placebo (n = 494)	400 mg in 2 doses; 400 mg once weekly; 400 mg twice weekly for 12 weeks	Incidence of COVID‐19–free survival	COVID‐19–free survival in 39 (7.9%) of 494 receiving placebo vs. 29 (5.9%) of 494 receiving HCQ once weekly (HR 0.72, 95% CI 0.44–1.16) vs. 29 (5.8%) of 495 receiving HCQ twice weekly (HR 0.74, 95% CI 0.46–1.19)
Abella et al (4), September 30, 2020, US	Double‐blind, placebo‐controlled, randomized	132 HCWs	HCQ (n = 66); placebo (n = 66)	600 mg/day for 8 weeks	SARS‐CoV‐2 positive by nasopharyngeal swab within 8 weeks	Terminated early for futility; COVID‐19 in 4 (6.3%) of 64 receiving HCQ vs. 4 (6.9%) of 61 receiving placebo
Dhibar et al (9), November 6, 2020, India	Open‐label, controlled	317 contacts of confirmed cases	HCQ (n = 132); control (n = 185)	800 mg for 1 day; 400 mg once/week for 3 weeks	Incidence of laboratory‐confirmed or symptomatic illness	COVID‐19 in 14 (10.6%) of 132 receiving HCQ vs. 36 (19.5%) of 185 controls (RR 0.50, 95% CI 0.25–0.99)
Mitjà et al (7), November 24, 2020, Spain	Open‐label, cluster‐randomized	2,314 contacts of index case patients	HCQ (n = 1,116); control (n = 1,198)	800 mg for 1 day; 400 mg/day for 6 days	PCR‐confirmed symptomatic COVID‐19 within 14 days	COVID‐19 in 64 (5.7%) of 1,116 receiving HCQ vs. 74 (6.2%) of 1,198 controls (RR 0.86, 95% CI 0.52–1.42)

CQ = chloroquine; HCQ = hydroxychloroquine; OR = odds ratio; 95% CI = 95% confidence interval; Dx = diagnosis; HCWs = health care workers; HR = hazard ratio; RR = relative risk; PCR = polymerase chain reaction.

Sixteen clinical studies investigating the preventive role of CQ or HCQ in the setting of COVID‐19 have been registered around the world; 5 studies were completed and others were terminated, suspended, or withdrawn (10). Of note, our own double‐blind, placebo‐controlled, randomized trial to determine the efficacy of CQ in preventing symptomatic COVID‐19 among New York–Presbyterian Hospital health care workers was terminated early because of lack of enrollment. CQ‐naive health care workers with moderate or high risk of exposure to COVID‐19 were randomized to receive placebo or CQ (500 mg daily for 1 week followed by 500 mg weekly). Study participants were followed up for 3 months to record SARS–CoV‐2 infections and safety events. The primary end point was the percentage of patients with COVID‐19 infections. The study started in April 2020 but progressed at a glacial pace due to safety concerns and loss of interest in prophylactic use of CQ and HCQ as the number of COVID‐19 infections decreased in New York City over the summer. We enrolled 9 participants and terminated the study early. An additional double‐blind, placebo‐controlled, randomized trial of HCQ (800 mg on day 1 followed by 400 mg for 4 additional days) for post–COVID‐19 exposure prophylaxis in asymptomatic household contacts was halted in early May 2020. This study was designed to have no contact between COVID‐19–exposed participants and study staff. Based on our Institutional Review Board requirement that prolongation of the QTc interval be ruled out prior to enrollment, the trial was effectively terminated before even 20 of the planned 1,600 participants were enrolled.

As vaccination of the health care force is almost complete in the US, the prophylactic roles of CQ and HCQ have clearly become less relevant. Most treatment data on HCQ in COVID‐19 to date have not shown clinical benefit, and final prevention data from the HERO‐HCQ trial (11), which remains active, will hopefully elucidate the role of HCQ in the prophylaxis against the development of SARS–CoV‐2 infections and conclude this chapter.

The authors acknowledge the support of Rising Pharma for the donation of the chloroquine used in one of the studies discussed in this publication.