President Donald Trump has enthusiastically pushed the use of two malaria drugs — one in combination with an antibiotic — to treat COVID-19, the disease caused by the novel coronavirus. But there is currently only limited evidence to suggest the drugs are effective against the new virus.
Over a series of press conferences and on Twitter, Trump has repeatedly expressed his optimism that two existing malaria drugs — chloroquine and its safer derivative hydroxychloroquine — will work in COVID-19 patients. Both drugs are also used to treat lupus and rheumatoid arthritis.
Calling himself a “big fan” of hydroxychloroquine, Trump said in a March 20 press conference that he had “seen things that are impressive,” even as Dr. Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases, cautioned that the evidence supporting the drug was anecdotal — and that it would take a large, well-done clinical trial to prove its effectiveness.
As we have written previously, Trump left the misleading impression that the FDA had approved chloroquine to treat COVID-19. In fact, there are no FDA-approved drugs to treat COVID-19, although existing drugs can be legally prescribed by doctors to treat other ailments for which they weren’t approved in what’s called off-label prescribing.
After the release of a small clinical trial in France, Trump also trumpeted combining hydroxychloroquine with the antibiotic azithromycin, which is sold under the brand name Zithromax.
“HYDROXYCHLOROQUINE & AZITHROMYCIN, taken together, have a real chance to be one of the biggest game changers in the history of medicine,” Trump wrote in a March 21 tweet. “Hopefully they will BOTH (H works better with A, International Journal of Antimicrobial Agents)….. ….be put in use IMMEDIATELY,” he added, referring to the journal in which the French study appeared.
But as we’ll explain, the trial has significant caveats and limitations — and far more data is needed to know whether the drugs are safe and effective to treat COVID-19.
In a press conference the following day, Trump exaggerated the existing evidence for the pharmaceuticals being sent to New York state for COVID-19 clinical trials.
Trump, March 22: There’s been some tremendous signs that this could work. Now, again, you know, some doctors think it should go for years in testing. But, you know, this has been something that’s been around for many years. It’s been phenomenal — a strong, powerful drug for malaria. But we think it might work on this, based on evidence — based on very strong evidence. We’re going to see. We’re going to know on — you know, sometime after Tuesday.
It’s not clear which drug Trump had in mind, but New York state said in a press release that it received shipments of hydroxychloroquine, azithromycin and chloroquine.
Trump is entitled to his opinion of how the trials will turn out, but he overstates the evidence supporting the drugs, which is slim.
Lab Studies in Cultured Cells
At least two studies have found that hydroxychloroquine has antiviral effects against the new coronavirus, or SARS-CoV-2, in cells grown in the lab. Publishing in Clinical Infectious Diseases on March 9, researchers in Beijing found that hydroxychloroquine was better than chloroquine in preventing the virus from replicating in cultured monkey cells.
Another Chinese group reported on March 18 that hydroxychloroquine blocked entry of the virus into the same kind of cultured cells, and also appeared to prevent the virus from being released from cells. The drug, however, was less effective against the novel coronavirus than chloroquine.
These are suggestive findings, but as Peter Hotez, a professor and dean of the National School of Tropical Medicine at Baylor College of Medicine, told us, the experiments by themselves don’t mean all that much.
“Hydroxychloroquine has some activity in a test tube against the virus,” he said. “But it’s hard to know how that corresponds to clinical disease.”
In the past, for example, chloroquine has been found to inhibit a variety of viruses in cells, including influenza and chikungunya. But when tested in animals or people, those results didn’t hold. In fact, in nonhuman primates, chloroquine treatment made their chikungunya infections worse, not better.
The authors of the second SARS-CoV-2 report also cautioned that while hydroxychloroquine is usually considered less toxic than chloroquine, overdoses are still possible — and in their experiment, the drug killed a fair number of cells at the levels needed to produce antiviral effects. That feature, they wrote, “requires careful designing and conducting of clinical trials to achieve efficient and safe control of the SARS-CoV-2 infection.”
The rest of the so-called evidence for hydroxychloroquine is largely anecdotal — essentially, people around the world trying it, and claiming it has worked.
Trump, for example, tweeted a link to New York Post story about a 52-year-old Florida man with COVID-19 who said hydroxychloroquine had saved his life. The president later appeared to reference the man in his press briefing, falsely suggesting that the drug had been “approved” to treat COVID-19 “in record-setting time.”
But as Fauci has explained in press briefings, without a controlled clinical trial, in which a treatment group can be compared with a control group, it’s impossible to say whether the drug actually did anything — or whether the person improved for other reasons.
One that has been published — a small randomized controlled trial of 30 COVID-19 patients with mild to moderate symptoms — found “no significant reduction in time to clinical improvement or viral clearance in the hydroxychloroquine arm compared to the conventional therapy control group,” according to a World Health Organization report.
The authors of the trial concluded that the drug needed to be tested in a larger group of patients.
The French Study
It turns out that the French study Trump cited was, in at least one respect, less rigorous than the Chinese clinical trial.
The study, which was done at a hospital in Marseille, France, calls itself a clinical trial, but it wasn’t a randomized controlled trial, and therefore, not the type of study that can actually determine whether a treatment is effective.
Rather than randomly selecting which patients would receive the treatment — thereby minimizing any differences between groups that could affect the outcome — the study used those who refused the hydroxychloroquine treatment as the control group. Controls also included patients at other hospitals in southern France.
The study was very small, ultimately including just 20 patients who received hydroxychloroquine — six of whom also received azithromycin — to compare against the 16 patients who did not.
The study also did not test for clinical outcomes. Instead, researchers measured the amount of virus in patients over time using nasal swab samples, and found that after six days, those treated with hydroxychloroquine had less virus and were more likely to test negative for COVID-19.
According to the paper, on the sixth day, only two of the 16 untreated patients tested negative, compared with eight of the 14 hydroxychloroquine-only patients and all six of the dual-treated patients.
While the results may sound impressive, they don’t say anything about whether patients’ conditions or symptoms improved, and the lack of randomization means it’s not possible to attribute the differences in the virus amount, or what’s called viral load, to the treatments.
Many scientists identified other odd details and possible technical errors when they commented on the paper on the preprint site medRxiv. The paper, curiously, was posted on the preprint site on March 20, the same day it was published in the International Journal of Antimicrobial Agents, and had been accepted just a day after submission — raising questions about whether there was a peer-review process and how thorough it was.
The Immunology Institute at the Mount Sinai School of Medicine’s review noted that of the 26 original patients receiving hydroxychloroquine, six were lost to follow up, including three who transferred to the ICU, one who died, and one who decided to stop treatment because of side effects.
“Total length of clinical follow up was 14 days, but the data beyond day 6 post-inclusion are not shown,” the review reads, adding that the way the authors reported the viral loads was not standardized, making it “impossible to compare what is defined as a positive value between the patients in the control and treatment arms.” Many of the control patients also didn’t have nasal swabs taken every day.
Update, April 7: On April 3, the International Society of Antimicrobial Chemotherapy, which publishes the journal in which the French study appeared, issued a statement of concern about the paper, writing that the group “believes the article does not meet the Society’s expected standard, especially relating to the lack of better explanations of the inclusion criteria and the triage of patients to ensure patient safety.”
On Twitter, Australian National University geneticist Gaetan Burgio lambasted the study, saying, “In short, all this hype on the clinical trial is based on a open label, non randomized and underpowered clinical trial on [hydroxychloroquine] treatment against #COVID19 with viral load as an outcome that was not properly measured in 2/3 of the control cohort !!!”
To at least some scientists, then, the outsized attention this particular study has received is unwarranted. But it’s important to recognize that even if the French study had been done perfectly and had been randomized, it still wouldn’t be particularly strong evidence that the drugs work, since the study is so small.
Baylor’s Hotez said that when small studies are repeated with more patients, the results often don’t pan out. This is a normal part of science, but it’s why scientists like to see larger clinical trials before concluding a drug is safe and effective.
“The likelihood that this is actually going to be a useful therapy is not high, but it’s not zero, either,” Hotez said of the drug combo. “So we’ll just have to see how it works in larger clinical studies.”
According to the Centers for Disease Control and Prevention, several trials are planned in the U.S. to test hydroxychloroquine’s ability to treat or prevent SARS-CoV-2 infection. Currently, this includes a 1,600-person prevention study for people exposed to COVID-19 patients in New York and a similar study by the University of Minnesota involving 1,500 people across the U.S.
The World Health Organization is also coordinating a large multinational trial called SOLIDARITY, which among other potential COVID-19 treatments, will test chloroquine or hydroxychloroquine.
When it comes down to it, Hotez is not especially optimistic about the drugs Trump has promoted, and said he’d prefer a different experimental therapy for COVID-19.
“If I was a patient or a family member, hydroxychloroquine and azithromax would not be my first choice,” he said. “My first choice would be the plasma convalescent antibody.”
The idea behind convalescent plasma — which is hardly new, having been used in the 1918 influenza pandemic — is to harvest antibodies from the blood of people who have been infected with the virus and have recovered, and infuse them into sick patients to help them fight off the virus.
While the treatment doesn’t work for every disease and carries some risks, Hotez said it did work during the 2003 SARS epidemic, which was caused by a highly similar coronavirus. “And so this has a much higher probability of success,” he said.
It’s worth noting that although the drugs Trump has touted are generally well tolerated, they do have side effects, and it’s not clear if there might be more extreme or other negative effects specifically in COVID-19 patients.
As a March 24 viewpoint article in the Journal of the American Medical Association explains, it’s often assumed the benefits of “last resort” medications will offset the harms, but that’s not always the case.
“A common interpretation of off-label use and compassionate use of drugs is that if the patient died, they died from the disease, but if the patient survived, they survived because of the given drug,” the article reads. “This is not true.”
According to the article, because chloroquine and hydroxychloroquine can trigger abnormal heart rhythms and many of the sickest COVID-19 patients are elderly and may have heart problems, it’s possible the drugs could “increase the risk of cardiac death.”
Still, given the circumstances, some U.S. hospitals have included these drugs in their COVID-19 guidelines, as they view the potential benefits to outweigh the risks. The U.S. Society of Critical Care Medicine, however, has neither endorsed nor advised against the use of chloroquine and hydroxychloroquine, due to “insufficient evidence.”
More broadly, the promotion of unproven drugs for COVID-19 is having other unintended effects. In Nigeria, two people were hospitalized for chloroquine overdoses, and in Arizona, a man died and his wife was hospitalized after the couple ingested fish tank cleaner containing chloroquine. Many patients with autoimmune diseases, such as lupus, who rely on hydroxychloroquine to keep their symptoms at bay have also reported shortages of the drugs as orders related to COVID-19 have spiked.
Without naming any names, in a March 23 editorial H. Holden Thorp, the editor-in-chief of the prestigious journal Science, cautioned against excessively hyping therapies that remain speculative.
“Political overhyping of such approaches is extremely dangerous — it risks creating false expectations and depleting drugs needed to treat diseases for which they are approved,” he said. “And it sets science up to overpromise and underdeliver.”