As the coronavirus outbreak continues to spread worldwide and more people become critically ill, scientists are racing to find a treatment that will help turn the tide. Dozens of medicines are in clinical trials in China—and now in the U.S.—to treat the disease, officially named COVID-19. Some are antiviral drugs that are already used to narrowly target other viruses. Experts say these medications are unlikely to do much against the novel coronavirus. Other drugs being tested—such as the broad-spectrum antiviral remdesivir, developed by Gilead Sciences—could prove quite effective, some evidence suggests. But only the rigorous, controlled clinical studies now underway will be able to confirm this possibility.

At the time of this writing, the COVID-19 outbreak has sickened more than 82,000 people globally and killed more than 2,800 of them. No vaccine or direct treatment currently exists. The more than 80 clinical trials being conducted in China involve drugs that were developed to treat illnesses such as HIV/AIDS, malaria and Ebola. These candidates include HIV antivirals called protease inhibitors, which work by blocking enzymes the virus needs to replicate, and a malaria drug called chloroquine, which is not an antiviral but has shown some efficacy against COVID-19 in a lab dish. Yet experts say drugs that specifically target other pathogens are unlikely to work well enough.

“The mistake generally made these days is to think that [just] any antiviral would be effective against [the coronavirus]. This is, of course, not true,” says Erik De Clercq, an emeritus professor of medicine at KU Leuven in Belgium, who helped discover the HIV antiviral tenofovir. De Clercq believes scientists should focus on developing compounds tailored to the new virus. “Instead of being in a hurry [to test] all known compounds—what they now call ‘repurposing a compound,’—we really need new compounds that are specific for [the coronavirus] and would be the subject of clinical trials,” he says. But until such compounds can be developed and tested, De Clercq says he is hopeful that remdesivir—an experimental drug that was originally developed to treat Ebola and has also proved effective against the SARS and MERS viruses in vitro—could be effective. (Gilead, which manufactures remdesivir, developed tenofovir and other antiviral drugs based on compounds De Clercq co-discovered.)

Remdesivir works by inhibiting an enzyme known as an RNA-dependent RNA polymerase, which many RNA viruses—including coronaviruses—use to replicate themselves. In contrast, retroviruses, such as HIV, are RNA viruses that use an enzyme called reverse transcriptase, which creates DNA from an RNA blueprint. But our own cells also rely on enzymes that transcribe DNA, so it is much harder to inhibit such enzymes without harming our own cells. Because coronaviruses use RNA-dependent enzymes, an antiviral such as remdesivir has a good chance of working against them, De Clercq notes.

Timothy Patrick Sheahan, an assistant professor of epidemiology at the University of North Carolina Gillings School of Global Public Health, is among those in the U.S. working on antiviral drugs for COVID-19. Like De Clercq, he is skeptical that many of the antivirals already on the market would work. “I’m doubtful that existing approved medications for other infectious diseases will have some magical property against this new coronavirus,” he says. “Most antiviral drugs are developed to be exquisitely sensitive and potent against one specific thing.” And part of that development process involves getting rid of “off-target” effects—even though they might inhibit other viruses. Sheahan also notes that the coronavirus research community “has suffered from a lack of randomized controlled trials.” Existing antiviral drugs were also tried against SARS (severe acute respiratory syndrome), which was first identified in 2003, and MERS (Middle East respiratory syndrome), first reported in 2012. But Sheahan says those studies were not well-controlled. In contrast, the current outbreak will give scientists a chance to test these drugs in a much more rigorous way by using randomized controlled trials, he says.

Sheahan and his colleagues have published several papers showing that remdesivir is effective against SARS, MERS and related bat coronaviruses, as well as some of the common cold coronaviruses. They are currently testing it on the new virus. Sheahan’s lab is also working with a group at Emory University to develop another broad-spectrum antiviral that works similarly to remdesivir: it mimics a nucleic acid used by the RNA polymerase enzyme and tricks the virus into incorporating the drug into its genome instead. His team is planning to submit some of its work for publication soon.

On a compassionate-use basis, remdesivir was given to the first known U.S. coronavirus patient: a man in Washington State who had recently returned from the outbreak’s epicenter in Wuhan, China. And he has made a good recovery. But that patient is, of course, only a single person, and a larger sample size will be needed to determine the drug’s efficacy. Two trials of remdesivir are currently underway in China: one for severe cases of COVID-19 and the other for mild or moderate cases. Results for both trials are expected in April. Another clinical trial is planned in the U.S., and it will be run by the University of Nebraska Medical Center and the National Institute of Allergy and Infectious Diseases. That trial will be conducted at up to 50 sites around the world and will test remdesivir against a placebo.

Lisa Gralinski, an assistant professor of epidemiology and colleague of Sheahan’s at the Gillings School, is also optimistic that remdesivir is a promising candidate for treating the new coronavirus. “I think it will probably be really effective” if you can get it to the patient within the first or second week, she says. But “you’re not going to be able to come in and give this drug to someone who’s approaching end-stage lung disease and improve their outcome.” At that point, the lung damage is no longer being caused by viral replication but is happening because of the body’s own immune response—so an antiviral would likely not be effective. Yet if enough of the drug is available, Gralinski says, she would give it at the time of diagnosis.

As for developing new antivirals, she thinks there probably will not be a big enough market to make them commercially feasible. “This is the largest human coronavirus severe-disease outbreak we’ve ever seen,” Gralinski says. But the numbers are low enough that it is still “not a viable thing to treat for a pharmaceutical company.” As with previous outbreaks, such as Zika, the virus could burn itself out before the new drug is developed—and there would no longer be a need for it. But, she adds, “if we already have something that’s mostly through development, like has luckily been the case with remdesivir, you can get it to people very rapidly.” Even if the drug proves to be effective, however, producing enough of it and distributing it to everyone in need is not guaranteed.

Read more about the coronavirus outbreak here.