Hydroxychloroquine and coronavirus: a guide to the scientific studies so far

With endorsements from a controversial French physician, Fox News and Donald Trump, hydroxychloroquine – an old anti-malarial drug that is today more commonly used to treat lupus – has received a disproportionate amount of attention as a potential treatment for Covid-19.

It has also become another partisan political wedge issue in the US: conservative politicians and media figures have hyped studies that support the theory that the drug is a potential treatment. And on Wednesday, Rick Bright, the head of a US government agency charged with investing in treatments and responses for pandemics, said he was forced out of his job over his resistance to the administration’s “misguided directives” promoting “broad use” of the drug, which he said “clearly lack scientific merit”.

Related: Trump stops hyping hydroxychloroquine after study shows no benefit

The hype around hydroxychloroquine is irresponsible, but the hope is understandable. There is as yet no drug that has been shown to be effective against the coronavirus, which has killed more than 180,000 people worldwide.

The limited evidence around hydroxychloroquine so far has come in a steady stream of scientific studies, often as soon as they are posted online as “preprints” – ie before they have gone through the rigorous vetting process known as peer review. None of the studies that have been released meet the gold standard for demonstrating a drug’s effectiveness – a large-scale, double-blinded randomized controlled trial (RCT), though multiple trials of that kind are under way.

While the world awaits those results, here’s a guide to some of the studies released thus far:

In vitro v in vivo

In early February, the journal Cell Research published a letter to the editor by Chinese scientists reporting the results of their experiments looking at whether existing drugs might be effective against the coronavirus. The scientists tested five drugs in vitro – meaning on cells infected with the virus in a laboratory rather than in humans – and found promising results for two: remdesivir and chloroquine. (Chloroquine is a close relative to hydroxychloroquine, which is considered safer.)

Caveats: There is a big difference between a drug showing promise in a laboratory and working in patients. Scientists have gotten similarly promising results with hydroxychloroquine against various viruses in past in vitro studies, including against the first Sars, but have yet to show its effectiveness against any virus in RCTs.

The controversial French study

Much of the media hype around hydroxychloroquine stems from a French study of the drug which purported to show significant reduction in viral load for patients treated with a combination of HCQ and azithromycin, a common antibiotic. The study was a clinical trial, meaning it involved actual patients, and underwent peer review before publication in the International Journal of Antimicrobial Agents (IJAA).

Caveats: There are numerous problems with this study’s design and the way its results were reported. The researchers downplayed the clinical outcomes – ie whether or not the patient improved, got worse or died – and instead based their analysis on measuring how long a patient was shedding the virus – ie whether the researchers were able to detect its presence from a nasal swab. All of the four patients with clear bad outcomes (three went to the ICU and one died) received the HCQ but were excluded from the viral shedding analysis. The researchers said that the remaining patients who received the drug shed virus for a shorter duration. This opened the door for people promoting the drug to mischaracterize their results and say they showed a “100% cure rate”.

The professional association affiliated with the IJAA has said the article “does not meet the society’s expected standard”, and the publisher has initiated additional independent peer review.

Meanwhile, the group of French scientists has continued to publish preprints reporting results on their continued use of hydroxychloroquine and azithromycin on patients in Marseille, without any control group.

A small Chinese clinical trial

In early April, Chinese scientists published a study looking at 62 patients with mild cases of Covid-19. Unlike the French trial, the study was randomized and the treatment and control groups were comparable. The study found a statistically significant difference in the time it took the two groups to recover, with patients who received HCQ showing shorter periods of fever and cough.

Related: Hydroxychloroquine: how an unproven drug became Trump’s coronavirus 'miracle cure'

Caveats: The number of patients involved was very small, and the researchers did not include any patients with severe or critical illness. This paper has not yet been peer-reviewed. The study authors concluded that their results “partially confirmed” the potential of HCQ as a treatment, but they said large-scale trials were still needed, as well as basic research to understand the mechanism by which the drug is affecting people.

A retrospective analysis of patient data from France

While awaiting the results of large-scale clinical trials, some scientists have performed retrospective analyses of real-life patients to emulate a clinical trial. This works by selecting a cohort of patients who received the drug and an equivalent cohort who did not and then comparing their outcomes. In a French retrospective analysis posted as a preprint in mid-April, researchers found “no evidence” that HCQ was an effective treatment when comparing 84 patients who received the drug and 97 who did not.

Caveats: Retrospective analyses are not a substitute for RCTs. One clear confounding factor is that researchers do not know why physicians decided to treat some patients with HCQ and others not. The paper has also not yet been peer-reviewed.

Red flags in an aborted Brazilian trial

While chloroquine and hydroxychloroquine are considered safe for their approved uses, they do have side-effects, including increasing the risk of heart arrhythmias in some people. In mid-April, a group of researchers in Brazil posted a preprint revealing that they had called a halt to a clinical trial of chloroquine at high doses after they observed high rates of arrhythmias and a large number of deaths.

The paper was published on 24 April by Jama Network Open following peer review, with the authors recommendation that the high dose of chloroquine “should not be recommended” for patients with severe Covid-19 due to the safety concerns. An accompanying editorial by three American physicians advised that the results “should prompt some degree of skepticism toward the enthusiastic claims about chloroquine and perhaps serve to curb the exuberant use”.

Caveats: The Brazilian trial was designed to compare the efficacy of two different doses of chloroquine, and it did not include a control group that only received a placebo. Researchers found the higher dose, which they chose because it was recommended by Chinese authorities, to be unsafe, but they are continuing to test the efficacy of a lower dose. They were not able to conclude definitively that the high dose of chloroquine was causing cardiac side effects, but described the “red flags” as sufficient to warrant their recommendation.

A retrospective analysis of US veterans

Similarly to the French retrospective analysis, researchers in the US looked at the outcomes for 368 patients who were treated for Covid-19 in the country’s Veterans Health Administration hospitals. The patients were broken into three groups: those who received HCQ, those who received HCQ and azithromycin, and those who did not receive HCQ. They found that HCQ, with or without azithromycin, did not reduce a patient’s need for a ventilator or reduce the risk of death. They also found that those who received HCQ alone had a higher risk of death than those who did not.

Caveats: The analysis was limited to male patients. As with the French retrospective analysis, this study is not a substitute for RCTs. One major confounding factor is that the patients who received HCQ were more severely ill. The researchers say that they accounted for this in their statistical analysis and still found a higher mortality rate for those receiving HCQ. The study has not yet been peer-reviewed.

US government recommendations

A panel of experts put together by the National Institute of Allergy and Infectious Diseases, the agency led by Dr Anthony Fauci, released treatment guidelines for Covid-19 on 21 April. The panel said that there was insufficient data to “recommend either for or against” hydroxychloroquine or chloroquine. However, the panel recommended against the use of hydroxychloroquine with azithromycin due to the risk of cardiac arrhythmia.

On 24 April, the Food and Drug Administration (FDA) issued a “drug safety communication” warning against the use of either hydroxychloroquine or chloroquine outside a hospital setting or clinical trial due to reports of “serious heart rhythm problems”.

“Hydroxychloroquine and chloroquine have not been shown to be safe and effective for treating or preventing Covid-19,” the communication stated.

• This article has been updated to correct the version of the drug that was used in the Brazilian trial. It was chloroquine, not hydroxychloroquine.

• Due to the unprecedented and ongoing nature of the coronavirus outbreak, this article is being regularly updated to ensure that it reflects the current situation at the date of publication. Any significant corrections made to this or previous versions of the article will continue to be footnoted in line with Guardian editorial policy.