Early this year, when COVID-19 was still a localized outbreak in China, Eleanor Fish, an immunologist at the University of Toronto, reached out to colleagues in Wuhan to explore the possibility of evaluating interferon therapy in patients infected with the coronavirus. Fish has been studying interferons (IFNs)—proteins produced by the body in response to viral infections—for close to 35 years, and her previous favorable results with a synthetic version during the 2003 SARS outbreak in Canada prompted the idea.

Fish’s colleagues in China connected her with Qiong Zhou, a physician who was at the time treating COVID-19 patients at Wuhan Union Hospital. Zhou was very receptive to the proposal, Fish recalls. As this was an outbreak, there wasn’t any time to optimize an IFN for use against SARS-CoV-2, and Fish and Zhou had to make do with what was available and what had previously been tried successfully in other coronavirus infections. For a clinical study, Zhou’s team used IFN-alpha-2b, first approved by the US Food and Drug Administration (FDA) for the treatment of cancer in 1986, thanks to its immunomodulatory, antiproliferative, and antiangiogenic effects. The researchers tested the IFN along with arbidol, a widely used broad-spectrum antiviral drug, in 77 patients who were admitted to Union Hospital in January and February with a confirmed SARS-CoV-2 infection. Each of them had moderate symptoms and none required intensive care.

The results, published May 15 in Frontiers in Immunology, found that patients treated with IFN-alpha-2b alone or in combination with arbidol cleared the virus from their upper airways an average of seven days faster than the group given arbidol alone. Hand-in-hand with that, blood levels of inflammatory markers such as interleukin-6 and C-reactive protein were dramatically reduced in patients receiving IFN-alpha-2b. Fish adds that as-yet unpublished data show that IFN-alpha-2b also limited lung abnormalities as evidenced by CT scans.

“I think it is a decent clinical report, and they look out for inflammatory markers, which is important,” says Andreas Wack, an immunologist at the Francis Crick Institute in London who works on IFNs but was not involved in this study. That’s because IFN-alpha is known to induce a proinflammatory milieu. Fish says that in her study, IFN-alpha-2b–treated patients did not show evidence of a cytokine storm, one of the dangerous immune responses observed in some COVID-19 patients.

Jordan Feld, a hepatologist at Toronto General Hospital, says that though the study was not a randomized controlled trial, it nevertheless gives some promise to the idea of using IFNs in COVID-19. On July 20, a UK company called Synairgen reported unpublished data from a Phase 2 trial of an experimental IFN drug, nebulized IFN-beta, that appeared to reduce the risk of hospitalized COVID-19 patients developing severe disease by 79 percent.

Any new leads on drugs are welcome in this pandemic. Only the antiviral agent remdesivir has been granted emergency use authorization by the FDA for treating patients with severe COVID-19. On June 25, encouraging preliminary results of the UK RECOVERY trial prompted the US National Institutes of Health (NIH) to recommend dexamethasone, a corticosteroid, for treating COVID-19 patients on mechanical ventilation or those requiring supplemental oxygen. But there are no drugs approved yet to be given early in the infection or preventively to those at high risk.

In such a bleak therapeutic scenario, mounting evidence from laboratory studies and a couple of clinical trials on IFNs, either alone or in combination with other antivirals, suggests that synthetic IFNs might be able to fill in this gap.

Viruses and interferons are mutually antagonistic

Confocal microscopy images of primary human hepatocytes from liver donors with (left) and without (right) IFN-lambda production, labeled in red.

There are three types of natural IFNs, each secreted by their own suite of cell types. Both type I (IFN-alpha and IFN-beta) and type III (IFN-lambda) are antiviral proteins produced early on in infected cells to protect other cells in the vicinity. “Once the genetic material [of the virus] is exposed, [IFNs] will chew it up,” says Fish. At the same time, IFNs recruit to the site of infection immune cells that will also help clear the virus.

IFNs I and III, for the most part, are part and parcel of the innate immune response of the body, with type III specifically involved in the localized immune response at sites where pathogens enter the body, such as mucosal barriers of the lungs, intestines, and liver. But the body’s natural IFN response doesn’t appear to be enough to thwart SARS-CoV-2 infection.

A paper published in Cell on May 28 by Benjamin tenOever, a virologist at the Icahn School of Medicine at Mount Sinai in New York, and colleagues found that SARS-CoV-2 infection induces low levels of type I and type III IFNs in ferrets. And Fish and Wack note that all viruses have numerous mechanisms to dampen the IFN response.

It stands to reason, therefore, that giving synthetic IFN drugs would boost the antiviral immunity of the body, perhaps giving the host the upper hand.

Type I or type III: Which is better to fight COVID-19?

While IFN specialists posit that type I and III IFN drugs are attractive therapeutic options in COVID-19, they disagree about which of the two would be most suitable in COVID-19, and experimental evidence indicates that both types have potential.

A study published June 19 demonstrates that the type I IFN-beta-1a, currently approved for the treatment of multiple sclerosis, was highly effective in inhibiting SARS-CoV-2 replication in vitro. A preprint posted May 7 shows that treatment with the clinical candidate IFN-lambda-1a, a type III IFN, inhibits SARS-CoV-2 replication both in vitro and in mouse models. Another preprint posted May 20 demonstrates that both type I and type III IFNs reduce SARS-CoV-2 replication in primary human airway epithelial cultures.

Although no synthetic IFN-lambda is yet FDA-approved, preliminary results in a clinical trial in hepatitis D patients show that it has the same antiviral potency as type I IFN but is much safer and better tolerated. A study in mice found that synthetic IFN-lambda was as effective as IFN-alpha in treating influenza but didn’t have the same proinflammatory effects.

The limited side effects of IFN-lambda could be due to the distribution pattern of receptors, which appear exclusively on barrier surfaces, such as the epithelium of the lung, intestine, and liver, whereas receptors for type I IFNs are ubiquitous, says Jeffrey Glenn, a professor at Stanford University and the founder of Eiger BioPharmaceuticals, a company that manufactures an IFN-lambda drug. “The benefit-to-risk is really compelling [for IFN-lambda]. IFN-alpha is a very tough drug to take as it has lots of side effects.” These include flu-like symptoms, fatigue, low blood counts, insomnia, and mood disturbances.

Ludmila Prokunina-Olsson, a senior investigator at the NIH who discovered and cloned a novel IFN-lambda gene in 2013 and is not associated with any clinical trials on IFNs for COVID-19, tells The Scientist in an email that type I IFN is a very potent antiviral but its mode of action is more inflammatory. “There is enough inflammation in the COVID lungs and avoiding any additional inflammation is critical,” she says. “If using any interferons, we suggest using IFN-lambda.”

Michael Gale Jr., an immunologist at the University of Washington who is currently working on how host cells recognize SARS-CoV-2 and launch the innate immune response but is not involved in IFN-lambda trials, also supports the clinical use of IFN-lambda. “With SARS-CoV-2, it’s important not to activate the lymphocytes as they are already activated in the inflammatory process.” As the receptors for IFN-lambda are not expressed on lymphocytes, immune cells are not activated.

IFN-lambda trialists say they worry that type I IFN, which is known to be proinflammatory, could precipitate a cytokine storm in COVID-19 patients. A study published July 10 in Science Immunology confirms that the type I IFN response is indeed proinflammatory in coronavirus infections and plays a pivotal role in the development of severe COVID-19.

“I would be very hesitant to use type I IFN,” says Thomas Marron, a cancer immunologist at the Icahn School of Medicine at Mount Sinai who is leading an IFN-lambda trial.

But for many researchers, type I IFN is not off the table for treating COVID-19. While Feld, an IFN-lambda proponent, is likewise concerned about type I IFNs’ safety in this context, he says what is appealing about type I IFNs is that being approved drugs, they can be mass-produced and given right away. “Obviously, even if we prove that IFN-lambda is highly effective, getting it mass produced will be a challenge.” Additionally, advocates of type I IFN drugs posit that the concerns about IFN-alpha’s tolerability are not so much an issue with short-term treatment in an acute viral infection such as COVID-19. And as SARS-CoV-2 spreads systemically, IFN-lambda won’t protect all cell types.

IFN-lambda has another downside. If given for too long or at too high a dose, Gale cautions, “we could run into problems where it would impede tissue regeneration,” particularly in the lungs. In a study published in June, Wack’s team found that type I and III IFNs activate antiproliferative and cell death pathways in airway epithelial cells of mice infected with influenza virus and that excessive or prolonged IFN production disrupted lung epithelial repair during recovery from the viral infection.

Harvard Medical School’s Ivan Zanoni led a second study also published last month that generated similar results regarding IFN-lambda using a synthetic version of SARS-CoV-2. He tells The Scientist that IFNs might be very good when given early in COVID-19 to limit the spread of infection. “When you move to severe COVID-19, ICU patients or hospitalized patients, one has to very careful as we have found that [both type I and type III] IFNs have a dark side,” he says.

“The general consensus in the field is that any antiviral in COVID should be used early, even preventively,” says Prokunina-Olsson. “In later disease stages, when the infection spreads to the lungs, any antiviral could be dangerous” because of the risk of a cytokine storm or damage to lung tissue by immune cells.

Both Wack and Zanoni emphasize to The Scientist the need to exercise caution with such potent antivirals as IFNs. “It’s complicated,” says Wack, “and I would be very, very, very cautious about this.”

Putting IFNs to the clinical test

Both type I and III IFNs are being tested in clinical trials globally. In a Phase 2 multicenter randomized trial in Hong Kong that was published in The Lancet on May 8, triple antiviral therapy with injectable IFN-beta-1b, lopinavir-ritonavir (an oral protease inhibitor), and ribavirin (an oral nucleoside analog) was safe and superior to lopinavir-ritonavir alone in reducing the duration of virus shedding, diminishing symptoms, and enabling patients with mild to moderate COVID-19 to go home from the hospital sooner.

The ongoing multicenter randomized controlled trial from Synairgen in the UK is testing inhaled IFN-beta-1a against a placebo in hospitalized patients with confirmed or suspected COVID-19, non-hospitalized patients with chronic health conditions, and frontline health workers who are at a high risk of coronavirus infections. Evangelos Andreakos, an immunologist based in Athens, Greece, says using inhaled type I IFN is an interesting approach that may sidestep the drug’s systemic side effects by delivering it directly to the lungs.

Other trials include the WHO SOLIDARITY trial, launched on March 20, that includes testing an IFN-beta in conjunction with other antiviral drugs, and an open-label trial in China, which reported May 7 in a preprint that IFN-alpha nose drops appeared to prevent coronavirus infections in close to 3,000 medical workers. In addition, Fish is actively seeking partners to conduct clinical trials on IFN-alpha in COVID-19 patients in India and elsewhere.

For type III IFNs, there are currently six clinical trials in the US and Canada that are testing Eiger BioPharmaceuticals’s IFN-lambda on a wide spectrum of COVID-19 patients.

Upinder Singh, an infectious disease specialist at Stanford University who is running one of these trials on IFN-lambda, says that her study is focusing exclusively on COVID-19 outpatients. The rationale is that 80 percent of patients who get infected will never get admitted. “But they are the ones who are spreading it in the community. Our goal is to identify patients who are very early in their disease, the idea being that if you can get to them early, you could reduce viral shedding and decrease the time in quarantine,” she says. About 60 patients have been recruited in the Stanford trial so far.

A trial Marron is running at Mount Sinai and another at Massachusetts General Hospital Boston are testing Eiger’s IFN-lambda in hospitalized COVID-19 patients, whereas a trial at John Hopkins University is testing the preventive use of the drug in people who are at high risk of getting infected. Meanwhile, preliminary results from a trial of inpatients and outpatients that Feld is leading show that the drug is safe and has minimal side effects, he tells The Scientist.

Clifford Lane, the deputy director for clinical research and special projects at the National Institute of Allergy and Infectious Diseases, points to the NIH treatment guidelines, which recommend against the use of type I IFNs in COVID-19 except in the context of a clinical trial, the rationale being that these IFNs generally showed no benefit when they were used in patients with other coronavirus infections and that the significant toxicities of type I IFNs outweigh the potential benefits.

The NIH guidelines also state that of the two type I IFNs, IFN-beta is better tolerated than IFN-alpha, but the results of a recent randomized controlled trial do not support the use of IFN-beta-1a in the treatment of acute respiratory distress syndrome.

Remarking on the enthusiasm surrounding the use of IFN-lambda in COVID-19, Lane says, “I am open-minded about any and all new approaches as long as they are tested in a rigorous fashion.”