If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.


Support nonprofit science journalism
C&EN has made this story and all of its coverage of the coronavirus epidemic freely available during the outbreak to keep the public informed. To support us:
Donate Join Subscribe


Drug Discovery


Why are antivirals so hard to develop?

With few good targets and limited financial incentives, the field of antiviral drug discovery has lagged. The COVID-19 pandemic might be awakening the world to the problem

by Laura Howes
May 20, 2021 | A version of this story appeared in Volume 99, Issue 19


Artist impression of influenza breaking out of a human cell.
Credit: Juan Gaertner/Science Source
An influenza virus breaking out of a human cell

Viruses are tricky. They’re small, mutate quickly, and make thousands and thousands of copies of themselves every day. Or rather, infected cells produce those new copies of the virus. Viruses can’t reproduce on their own—they sit inert until they can infect a cell.

And when it comes to finding drugs that can kill a virus, that’s part of the problem: viruses don’t have a lot of their own proteins and enzymes to target.

The handful of proteins and enzymes they do have might perform the same basic functions—allowing the virus to enter cells, replicate, and escape to do it all over again—but their sequences and structures differ even among viruses in the same class, let alone among families of viruses.

So even if you develop a drug against one virus, it is unlikely that you can use it to treat another. Add a lack of reliable animal models and a lack of investment, and it becomes clear that antiviral drug development is a complex problem.

From bug to drug

For decades, scientists didn’t know whether viruses had any of their own enzymes. Researchers assumed that hijacked cells just built new copies of a virus using the cells’ own enzymes and proteins. Then in 1967, scientists discovered the first viral enzyme, a poxvirus DNA-dependent RNA polymerase (Proc. Natl. Acad. Sci. U.S.A., DOI: 10.1073/pnas.58.1.134). That enzyme takes the virus’s DNA-encoded genome and transcribes it into RNA to start the production of new viral proteins. Once researchers identified those first viral enzymes, they realized they could start designing drugs to target viruses themselves. But progress was slow.

“Antiviral development has always lagged behind antibiotic drug development,” says Saye Khoo, an expert in antiviral pharmacology at the University of Liverpool. “And other than the topical treatments for warts and things like that, it’s really been pretty poor until acyclovir came along.”

There aren’t a lot of antivirals that have a broad-spectrum effect.
Ashley Brown, director, Antiviral Pharmacodynamics Laboratory, University of Florida

Acyclovir is used to treat herpes simplex infections, chicken pox, and shingles. Patented in 1974 and first approved for use in the 1980s, the drug is converted inside cells to a form that looks like a component of DNA. This tricks the virus’s DNA polymerase into incorporating a version of the drug into growing DNA chains, stopping replication.

And unlike earlier antivirals, acyclovir completely inhibits the viral DNA polymerase without stopping the cell’s own enzyme, meaning it causes very few major side effects. Researchers have been trying to take a similar approach ever since by targeting viral proteins that have no human equivalents. Marti Head, director of the Joint Institute for Biological Sciences at Oak Ridge National Laboratory (ORNL), likens understanding how viruses keep going to diagnosing engine trouble in your car. “If I’m running low on oil, I know from experience I can keep driving. And yet if my timing belt goes out, I’m dead in the water,” she says. Antiviral drug designers must determine which piece of the virus is most equivalent to the timing belt and whether it varies across viruses.

Every stage of infection and viral replication offers a chance to stick a wrench in the works. But because viruses code for only a few proteins of their own, there might be only one or two proteins that a drug can target. And those viral enzymes may have functions that host cells also perform. That overlap creates the potential for an antiviral to inadvertently harm healthy human cells.

Still, over the years, drug developers have found ways to safely target several key viral proteins—ones they turn to first when a new threat emerges. Much time has been devoted to enzymes responsible for RNA or DNA being copied—like the target of acyclovir. That’s because those enzymes often look similar in several viruses, meaning an inhibitor could be useful in multiple infections. Reverse transcriptases, which some viruses use to transform RNA into DNA, have been a major target in HIV. And proteases, which cut up viral proteins in the cell, have been a target in both chronic infections, like HIV infection and hepatitis C, and acute ones, such as SARS-CoV-2 infection. “There aren’t a lot of antivirals that have a broad-spectrum effect,” says Ashley Brown, an expert in antivirals at the University of Florida. “You have to make your drug target a specific protein in a specific virus.” In other words, one drug for every bug. But the good news is that targeting a specific viral enzyme decreases the risk of also affecting a host enzyme, making a drug safer. “So when you do manage to get an inhibitor, the chance of it moving forward right the way through to the clinic and beyond is quite high,” says Eddy Littler, chief operating officer of the UK biotech ReViral, which is developing a drug for respiratory syncytial virus, or RSV.

Some drug developers are targeting host cells, a strategy that in theory could generate drugs that will work against whole viral families, says John Bamforth, the interim executive director of the Rapidly Emerging Antiviral Drug Development Initiative (READDI).

READDI’s targets are families of viruses that have pandemic potential, and the initiative aims to get five broad-spectrum antivirals through early safety studies in humans within 5 years. One way to do that, Bamforth says, is to target host biology. “If we can affect the host cell, then, to a degree, it doesn’t matter what the virus does,” he says. That approach makes drugs more flexible and viruses less likely to develop resistance. But focusing on the host cell comes with a trade-off: a higher risk of significant side effects.

The challenges for antiviral drug discovery aren’t limited to designing a good inhibitor. The next step is to show that the drug works—first in cells, then in animals, and finally in humans. But the cell and animal models can present new obstacles for researchers working on viruses.

For example, researchers struggled for years to get hepatitis C to replicate in cells in a lab. Scientists solved that problem after some clever synthetic biology created self-replicating viral RNA from the hepatitis C virus, which allowed them to study how disrupting individual proteins affected the viral life cycle. And once the drug stops the virus in cells in a petri dish, then scientists need to test it in reliable animal models that can give physiologically relevant results.

Every virus uses cellular proteins and affects cellular processes downstream, complexity that can’t always be mimicked in an animal with a different biology than humans’. Gilead Sciences’ remdesivir, for example, worked well against the Ebola virus in small animal models and in nonhuman primates (Nature 2016, DOI: 10.1038/nature17180). It didn’t work so well in humans.

Several experts also cite the lack of good models as a stumbling block in anti­viral development. For example, Littler says mouse and rat models of RSV are not very predictive. ReViral tested its compound in a human challenge model, in which healthy people are purposely infected with RSV. It’s an artificial situation, he says, but it provides a good idea of how the drug will work in the clinic.

Brown, who uses in vitro and mathematical models to optimize therapeutic regimens for antiviral compounds, says that at every stage, drug designers need to think about what models can and can’t tell them. “These models are all necessary to move to the next stage, but you always have to be careful in your considerations on what’s going to translate” to the real world.

The viral life cycle
Drugs can be designed to interrupt a virus’s life cycle at many points between its entry in and exit from a cell.
Viruses enter cells, replicate, and then exit.
Credit: Yang H. Ku/C&EN
Source: Adapted from Nat. Rev. Microbiol. 2020, DOI: 10.1038/s41579-020-00468-6

From biology to business

Ultimately, biological problems make antiviral drug development tricky, but research over the past 50 years has shown that the right incentives can drive innovation.

Decades of investment in understanding and developing drugs for HIV and AIDS have been a major boost to the field of antivirals. Similarly, years of work transformed the prospects for people with hepatitis C, treatment for which used to entail nearly a year of drugs with harsh side effects. Now, a relatively short course of pills can cure it. “It’s a success story,” says ORNL’s Head, but “boy, was it a hard one to get to that success story.”

Both AIDS and hepatitis C are chronic and affect a broad number of people globally. Those factors created a large market to target for the pharmaceutical firms that could develop new and better treatments.

For acute viral infections, the time to take an antiviral is as soon as you’re infected. That small therapeutic window can shrink the market for a drug, as can the fact that a virus might affect areas of the world that might not have deep pockets to pay for expensive drugs.

While public funding can help fill the gaps, many investors also have a short attention span. “The funding goes with the interest in public health, you know, so you’d see it move around from Ebola to Zika, and then back to Ebola,” Brown says. “There’s no perfect solution.”

The current pandemic has the world considering how to approach antiviral drug development differently.

You would have thought, a year into the pandemic, a lot of small molecules would be creeping into the market.
Saye Khoo, professor of pharmacology, University of Liverpool

Littler’s biotech, ReViral, recently secured C-stage funding. He says that venture capital firms and investors are becoming more interested in investing in areas related to infectious disease. “There’s a lot of money around to invest in good programs, whether or not that’s private investment or public investment,” he says. Littler also says he’s noticed that pharma companies with virology expertise are putting more resources into antiviral development.

That awakening to the need for new antivirals is, unsurprisingly, due to the lack of treatment options for COVID-19. Only a handful of antivirals have shown promise against SARS-CoV-2 in clinical trials. Since March 2020, Khoo has been part of the UK’s Agile Coronavirus Drug Testing Initiative, which was created to accelerate the testing of drugs. When C&EN spoke to him in April 2021, Agile had just injected the first patient with VIR-7832, a monoclonal antibody developed by Vir Biotechnology and GlaxoSmithKline for treating COVID-19. It’s one of a handful of antibody drugs discovered amid the pandemic. But Khoo says he has been surprised at just how few small-molecule treatments against COVID-19 have appeared.

“You would have thought, a year into the pandemic, a lot of small molecules would be creeping into the market,” Khoo says. “Some of them are, but it’s largely not as much as we’d hoped for.”

Experts say the world should be considering what research can be done to prepare us for future pandemics and developing funding models to support that work.

Bamforth says that by the time READDI launched in April 2020, as the pandemic was spreading, interest in what they were doing had grown. READDI board members stressed that the initiative needed to double down on its efforts.

Bamforth expects that if READDI gets the funding and support it’s hoping for, some of the therapeutics the team develops will target existing infections, while other assets will be held ready and waiting for viruses that we have yet to encounter. Those prospective treatments can be put through safety studies in humans and then stockpiled for when a virus appears.

That’s an approach other people agree could work, but it needs an investment model to support it. “Once we get back to some sort of semblance of normality, we think there’ll be about a window of 6 months where people are still interested,” Bamforth says. The question is whether enough work can be done before that window closes.


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.