President Trump has often mentioned hydroxychloroquine as a potential treatment for Covid-19. Last month he announced that he was taking the antimalaria drug, also known as HCL, as a prophylaxis after two White House staffers tested positive for the virus. Mr. Trump’s critics in the press seem to be on a mission to discredit the therapy. Now a prestigious medical journal may be joining in.
The Lancet published a study on May 22 that purported to find a 30% increased risk of death for hospitalized Covid-19 patients treated with HCL or chloroquine, a related treatment. “Study says drug hailed by Trump is harmful,” the Washington Post reported. The World Health Organization suspended its HCL trial. France, Belgium and Italy announced they would prohibit the drug for the treatment of Covid patients.
In an open letter to the Lancet’s editors and the study’s authors, some 120 doctors, statisticians and epidemiologists write that the headlines about the study “have caused considerable concern to participants and patients enrolled in randomized controlled trials” evaluating the drugs. Thus many researchers have scrutinized the data, and the “scrutiny has raised both methodological and data integrity concerns.”
Antimalarials have been long been known to increase risk for cardiac arrhythmias, particularly in patients with underlying heart conditions. Infection by the novel coronavirus can also cause inflammation of the heart muscle in severely ill patients, which can result in arrhythmias. Yet the drugs have been used safely for 60 years to treat and prevent malaria as well as for rheumatoid arthritis and lupus.
The Lancet study aggregated raw data collected by Surgisphere, a public-health analytics company, from 96,032 hospitalized Covid-19 patients treated at 671 hospitals on six continents. Yet its raw data and code haven’t been shared with other researchers for review, despite Lancet’s pledge in March to do so for all Covid-19 studies it publishes.
“There was no mention of the countries or hospitals that contributed to the data source and no acknowledgments of their contributions. A request to the authors for information on the contributing centres was denied,” the study’s critics noted. Additionally, “there was no ethics review.”
The study’s authors say Surgisphere’s contracts with hospitals don’t allow them to share individual patient data. But the authors won’t even share data that are aggregated by hospital. This information is important to determining the influence of confounding variables such as disease severity, drug dosage and underlying conditions.
Other problems include that more deaths were reported from five Australian hospitals collectively than had occurred in the entire country during the study period. The study showed little variation in patient outcomes or underlying risk factors by continent, which would be odd given health disparities among different populations.
Also curious: The average reported dose of HCL was 100 milligrams higher than FDA guidelines—which is 800 milligrams of hydroxychloroquine sulfate on the first day of treatment and then 400 milligrams daily for four to seven days—even though two-thirds of the data ostensibly came from hospitals in North America. That leaves two possibilities: The data are inaccurate, or doctors were treating patients with dangerously high doses.
The authors on Friday corrected some glaring errors. They noted, for instance, that a hospital in Asia was lumped in with the Australian death totals. But the authors also stand by the results. Researchers worry that the study’s publicity will make it harder to recruit patients for randomized clinical trials that are necessary to determine whether the drug may benefit some patients.
HCL has been found to interfere with the virus’s replication in vitro, though observational studies have produced mixed results. Two studies from France found most patients recovered rapidly and suffered few serious side effects. Another, also from France, found no benefit, though its subjects were older and received higher dosages.
It is possible that HCL may prevent infections or help younger patients with mild cases who aren’t experiencing immune overreactions known as cytokine storms. HCL is among the few drugs being studied as a prophylactic. “There’s so much swirling around it, people won’t want to enter those trials,” David Smith, an infectious-disease specialist at the University of California in San Diego, told Nature magazine. “In which case, it will be an open question that won’t get answered.”
Lancet’s decision to publish the study with little apparent scrutiny also suggests politics may be influencing its scientific judgment. An unsigned editorial in May criticized President Trump’s “inconsistent and incoherent national response” to the pandemic and asserted that “Americans must put a president in the White House come January, 2021, who will understand that public health should not be guided by partisan politics.” The Lancet editors should focus on healing themselves.
Ms. Finley is a member of the Journal’s editorial board.