Much of the news coverage of the COVID-19 epidemic has pointed to the promise of the anti-malarial drug hydroxychloroquine to combat the novel coronavirus that causes COVID-19. Hydroxychloroquine is one of three drugs that researchers have flagged as possible treatments for COVID-19.
Faculty in the Department of Epidemiology and Biostatistics are leading clinical trials of two of those drugs, hydroxychloroquine and the antibiotic azithromycin. Other UCSF faculty are leading trials of these drugs for hospital patients; UCSF is also leading a trial of the third drug of interest, an experimental anti-viral called remdesivir. Remdesivir is given by infusion, so is only available to hospitalized patients.
Why these drugs? When facing a new disease, doctors first look to existing drugs, knowing it will take months or years to develop a new drug. Hydroxychloroquine sparked interest because it had been shown to be active in a petri dish experiment against the SARS virus, which is related to the coronavirus that causes COVID-19.
It was logical, then, for doctors in Marseille, France, to try hydroxychloroquine when they faced very sick patients with no proven therapy to offer. For some of the patients, the French doctors also prescribed the antibiotic azithromycin, which is commonly given to patients with bacterial pneumonia. Azithromycin has anti-inflammatory properties. It may also help patients recover by clearing any secondary bacterial infections.
The French doctors’ results, published in the International Journal of Microbial Agents, suggested that hydroxychloroquine reduced patients’ viral load, and the two drugs in combination reduced viral load substantially. Their findings were far from conclusive, given the small number of patients, lack of randomization, short follow-up time and no clear relationship between viral load and clinical outcomes. The journal has since disavowed the paper, and a second research group gave the same drugs and found no benefit.
But with COVID-19 running roughshod over the globe, these limited results sparked a lot of interest in the two drugs. A controlled randomized trial is a logical next step.
Many of the UCSF trials are structured to be able to adapt as more evidence emerges. If a drug shows clear benefits, for instance, the trial would likely shut down its placebo arm. The Proctor Foundation trial will provide a single 1-gram dose of azithromycin – as compared to more common three-pill, 1.5-gram regiments – making it useful in finding the right dose of the antibiotic to treat the viral COVID-19.
Kelly, an infectious disease doctor and Ebola researcher, does his clinical work at the Veterans Administration (VA), which has approved and funded a three-arm trial of azithromycin, hydroxychloroquine and a placebo. Some of the 300 patients will also receive COVID-19 test kits by mail for patients to test themselves.
For safety reasons, neither study will give any one patient both drugs. Hydroxychloroquine can have heart-related side effects. And while combining it with azithromycin had a stronger anti-viral effect in the French study, it also magnifies the risk of cardiac side effects. Because COVID-19 can also cause damage to the heart, cardiac side effects are particularly worrisome. The FDA requires EKG monitoring for patients taking both drugs, which isn’t feasible in the current circumstances.
With multiple trials of these drugs at UCSF alone, one could be forgiven for thinking that perhaps it is just regulatory bureaucracy that stands between these drugs and clinical use against COVID-19. Not so. There is incredible pressure to do something, anything, for patients, partly because Trump’s cheerleading is prompting coronavirus patients and their families to ask for the drugs.
“After the French trial everybody who gets sick, they get hydroxychloroquine. Many providers will just give it to them,” Kelly said. “We need trials. We need things fast, sure, but we also need good data.”
Kelly and his colleague Salomeh Keyhani, MD, at the VA saw a clinical trial was a way to provide these best-guess drugs for keen patients while also contributing to the larger need for evidence on what works.
Others are likely making similar calculations, leading to multiple competing trials of the same two drugs. That also poses challenges.
“You could take a lot of smaller trials and do a metanalysis, but that really only works if you have protocol standardization, and I haven’t been seeing that,” said Oldenburg.
There hasn’t been time.
“The pace of getting a trial started under these conditions is just next-level, and that affects the ability to reach out to other groups,” Oldenburg added.
Many of the UCSF trials are structured to be able to adapt as more evidence emerges. If a drug shows clear benefits, for instance, the trial would likely shut down its placebo arm. The Proctor Institute trial will provide a single 1-gram dose of azithromycin – as compared to more common three-pill, 1.5-gram regiments – making it useful in finding the right dose of the antibiotic to treat the viral COVID-19.
“Say hydroxychloroquine becomes standard of care, then we could keep testing azithromycin as a non-inferiority trial. There are ways you can keep the science going and also keep the care going,” Kelly said. “All this off-label use stuff is not necessarily good for patients.”