A deadly virus, a disowned scientific paper, and patients testifying to a miracle cure. There is no shortage of information about whether or not hydroxychloroquine can help patients beat COVID-19.
The question is whether any of this evidence gets us closer to a cure.
On one side are doctors and health campaigners who say the drug’s history as a malaria treatment means it is safe, even if its efficacy is promising but unproven. “What do you have to lose?” is how President Trump regularly poses the question.
“I may take it,” Trump said recently. “We’re just hearing really positive stories, and we’re continuing to collect the data.”
On the other are virologists, statisticians, and a slew of scientists who say studies so far show no benefit from taking the drug.
The controversy cuts to the heart of the scientific method, and the way complex medical information is disseminated to a public desperate for good news, according to Joshua Sharfstein, vice dean at Johns Hopkins University’s Bloomberg School of Public Health. “The idea that someone got medicine and then got better does not mean that the medicine made them better,” he said.
The potential role of an anti-malaria drug hit the headlines in mid-March, following a study in France. Researchers in Marseille reported that of 14 patients treated with hydroxychloroquine, more than half were virus free after six days. Six treated with hydroxychloroquine plus azithromycin were all “cured.”
By way of comparison, only two of 16 “control” patients, who did not receive the treatment, were found to be virus free.
The trial ended on March 16. A pre-print (a way of rushing out important research ahead of peer review) appeared almost immediately.
And on March 18, the results were broadcast on Fox News. New York lawyer Gregory Rigano, who has since dropped his claim that he was an adviser to Stanford Medical School, described the research to Tucker Carlson as a “well-controlled, peer-reviewed study that showed a 100% cure rate against coronavirus.”
Trump referred to it for the first time the next day, calling it a potential “game changer.”
However, the study was not peer reviewed or “well-controlled,” according to scientists such as Dr. Katherine Seley-Radtke, professor of medical chemistry at the University of Maryland, Baltimore County, and an expert on antivirals and coronaviruses.
“A lot of people questioned the paper because it didn’t have proper controls, it was done on a very small sample size, and there were just too many things that were wrong with it,” she said. “That’s why a number of people raised red flags about it.”
Studies dating back to 2006 suggested hydroxychloroquine had no impact on coronaviruses, although it might help reduce inflammation in patients, added Seley-Radtke.
The study was flawed because it did not use double-blind, randomized trials, in which patient and doctor were unaware who received the therapy and who was in a control group. In addition, the control patients were not drawn from the same population. Some were treated at a different location (where other aspects of their treatment might have differed) or had underlying conditions that made them unsuited to treatment.
The paper also omitted six of the treated patients from its conclusions, which might have significantly altered the results. One died, three were transferred to intensive care, and one ended treatment because of nausea.
Doctors in Paris last week reported that their effort to replicate the Marseille results had failed.
Finally, the International Society of Antimicrobial Chemotherapy, whose journal published the paper online, said it did not meet its “expected standard,” especially relating to patient inclusion. “Although ISAC recognizes it is important to help the scientific community by publishing new data fast, this cannot be at the cost of reducing scientific scrutiny and best practices.”
Other studies have also raised questions. A recent trial in Wuhan, China, included a control group and found improved symptoms among those who received hydroxychloroquine, but offered no data on whether viral load had decreased.
An earlier study of 30 patients in Shanghai found no impact on levels of the virus between those who received treatment and those that did not.
At the end of last month, the Food and Drug Administration cited “anecdotal evidence” in granting emergency approval for doctors to use two anti-malarial drugs on COVID-19 patients.
On Tuesday, Trump used one such anecdotal case, the story of a Michigan state representative who recovered after taking hydroxychloroquine, to urge patients to seek out the drug.
“Four hours later, she awoke, and she said, ‘I feel better,’ and then shortly thereafter she felt great. This is a woman who thought she was going to die.”
Sharfstein said using unproven treatments might make sense on a case-by-case basis in the absence of anything better.
“But that shouldn’t be a national policy: Doctors can try things,” he said. “The national policy should be: Let’s get evidence to figure out what works.”
Failing to do that, he added, risked overlooking other treatments with greater potential.
“We should let the evidence sort out where we are really going to put a big investment,” he said.
