In this photo illustration, tablets on a blister pack of Hydroxychloroquine are displayed on April 10, 2020 in Rio de Janeiro, Brazil. (BUDA MENDES/GETTY IMAGES VIA BLOOMBERG)

Scientists are raising questions about a study that linked antimalarial drugs to increased heart risks and death among COVID-19 patients.

Backed by US President Donald Trump, who has said he took a course of hydroxychloroquine to ward off the virus, the drugs have become a political and medical bone of contention

The study, published last week in The Lancet medical journal, found high rates of dangerous side effects in patients treated with hydroxychloroquine and chloroquine, and some work on testing the drugs in patients has been suspended based on the concerns it raised. A group of some 120 researchers signed a letter pointing out inconsistencies, calling on the authors and The Lancet to reveal more details about their analysis.

Backed by US President Donald Trump, who has said he took a course of hydroxychloroquine to ward off the virus, the drugs have become a political and medical bone of contention. A separate scientific article that had lent support to using them against the coronavirus was withdrawn by its authors last week because of the controversy.

Not long after The Lancet study was published, the World Health Organization suspended a portion of a global study of potential COVID-19 treatments that included the antimalaria drugs because of safety concerns. On Friday, the WHO released a summary of its results to date, saying it found little or no benefit from use of the drugs compared with standard care, whether or not they were paired with an antibiotic. The agency cautioned that the evidence was of “very low certainty.”

French drugmaker Sanofi said Friday that it was halting patient recruitment in its trials of hydroxychloroquine for treating COVID. The company also said it will stop releasing the drug for off-label use in COVID-19 “until current safety concerns are cleared by WHO.”

Researchers have said that hydroxychloroquine and chloroquine need to be tested against the coronavirus in randomized clinical trials before they can be recommended. Only Gilead Sciences Inc.’s remdesivir has been shown to benefit patients by this measure.

ALSO READ: Hydroxychloroquine: WHO sees safety findings by mid-June

The Lancet study, led by Mandeep Mehra of Brigham and Women’s Hospital in Boston, looked at the medical records of 96,000 coronavirus patients around the world, about 15,000 of whom were treated with antimalarials. Some of the patients also received antibiotics that are thought to enhance the benefit of the antimalarial drugs

High Doses

The letter pointed out 14 major shortcomings in the paper, including that computer code used to analyze the data wasn’t made public, and that no information was included on the medical centers that contributed data. It also said that rates of deaths reported from Africa seemed “unlikely,” that the daily doses purportedly received by some US patients appeared higher than recommended, and that the level of chloroquine use in some continents was “implausible.”

The letter called on the US company holding the data for the study, Chicago-based Surgisphere, to provide greater detail, and for an independent evaluation of the analysis. When scientists asked to see more of the data, according to the letter, Mehra replied that the authors weren’t able to share it under agreements with countries, governments and hospitals.

Surgisphere addressed the building controversy in a statement on its website before the scientists sent their letter Thursday. The company said it works with top-tier institutions “that have a tertiary care level of practice and provide quality healthcare that is relatively homogenous around the world. As with most corporations, the access to individual hospital data is strictly governed.”

READ MORE: Trial of hydroxychloroquine in virus patients halted by WHO

The Lancet, in a post on Twitter, said it had made corrections to the study but that its results and conclusions remained unchanged. The corrections included how patients were counted from different regions, and the journal said that an incorrect appendix table had been included.

“There have been no changes to the finding of the paper,” according to the post.