A Major Study That Raised Concerns About Hydroxychloroquine Has Been Retracted
The study has been retracted by scientists who say they can no longer vouch for its data.
Researchers have retracted a massive study that said that hydroxychloroquine, a malaria drug hyped as a potential COVID-19 cure by President Donald Trump, could pose serious risks to coronavirus patients.
The study, published in May in the medical journal Lancet, concluded that the malaria drug was ineffective against the virus, a finding in line with several other studies. For the first time, it also linked hydroxychloroquine to a higher risk of death. Days after the publication of the study, which claimed to be based on 96,000 hospitalized COVID-19 patients across six continents, the World Health Organization paused human tests on the drug.
But the study quickly drew concerns from outside scientists, more than 180 of whom signed a letter outlining perceived inconsistencies and demanding that the authors share their full dataset, which had come from a little-known health data analytics company called Surgisphere Corporation. Last week, the authors corrected some of their data but claimed that their conclusions remained unchanged.
On Thursday, however, three of the four researchers reversed course and acknowledged they “can no longer vouch for the veracity of the primary data sources.”
In a notice in the Lancet, they wrote that they had sought to have outside experts independently audit the data, but that Surgisphere, which claimed to have an electronic database with patient outcomes from about 670 hospitals, would not hand over its full dataset due to “client agreements and confidentiality agreements.”
Sapan Desai, the fourth co-author of the study and the founder of Surgisphere, was not listed alongside the other authors who retracted the paper.
A second paper based on Surgisphere data was also retracted from the New England Journal of Medicine on Thursday. The study, published by several of the same researchers, was about cardiovascular disease in COVID-19 patients, not about hydroxychloroquine.
For instance, the company’s handful of employees appeared to include a science fiction writer and an adult model, according to the Guardian. The study also claimed to be based in part on data from Australia, but the Guardian reported that it could not confirm with several of that country’s health agencies that they had provided data to the study.
The Lancet study had an immediate impact. Two major hydroxychloroquine clinical trials — one by the WHO, another in the United Kingdom — were put on pause. And the governments of France, Belgium, and Italy banned the drug from being used as a coronavirus treatment.
But scientists questioned whether the study was robust enough to justify those decisions. “Many of us in the scientific community were just very angry at seeing a poorly written and executed study published in the Lancet, given loads of publicity, and then having a hugely negative impact on carefully planned clinical trials around the world,” James Watson, a Thailand-based statistician with the University of Oxford’s Centre for Tropical Medicine and Global Health, told BuzzFeed News last week.
The Lancet study was not alone in finding that hydroxychloroquine is an ineffective treatment for COVID-19, and in linking the drug to serious heart problems. Several other studies have previously reached similar conclusions, and the FDA has acknowledged the cardiac risks of this and a related drug, chloroquine, warning that the drugs not be used outside of a hospital setting.
The latest such study, published Wednesday, found no evidence that hydroxychloroquine was helpful in preventing COVID-19.
As Eric Topol, a cardiologist at the Scripps Research Translational Institute, told BuzzFeed News last week, “Everything points to a drug that has no efficacy. There’s no sign that it helps anyone. We know it has significant side effects that are worrisome.”
But the now-retracted Lancet study was also the first to link the drug to a higher rate of deaths, a link that is now unsubstantiated or at least highly in question.