Impressions of "Pegylated Interferon Lambda," a new Covid-19 treatment which performed well in a major study.

A ball-and-stick model of a human interferon. Getty Images.

Earlier this week, a new study published in the New England Journal of Medicine found that patients who received one injection of “Pegylated Interferon Lambda” did better than those who received placebo. Interferons like this one work by helping cells keep viruses from invading and taking up shop.

The new study compared outcomes in mostly vaccinated participants across several periods, including Omicron. Interferon recipients had lower rates of the study’s “primary outcome” of interest; Participants were deemed as having experienced the primary outcome if either a Covid-related hospitalization or an emergency department stay longer than 6 hours occurred.

I’ll need to spend more time with this paper before I finalize my thinking on this. But there are some interesting things to say now, which I’ll share.

In favor of the drug, not only did the trial succeed in lowering rates of its primary outcome by about half but also in a slew of secondary outcomes. When everything agrees, bending in the same direction, that can be a sign that a drug is working.

Here, both components of the primary outcome—Covid-related hospitalizations and emergency department stays of longer than 6 hours—were statistically lower in the interferon recipients.

That said, some key secondary outcomes were not as impressive. For example, Covid-19 deaths, all-cause deaths, and all-cause hospitalizations were not statistically lower than that of the placebo group, even though the numerical rates were lower (that is, the raw rates of these events were lower, but the findings were not statistically significant). Now, that could be because the drug doesn’t work as well as billed, or it could be simply due because the study was not designed to have enough patients to be able to make such determinations. That said, even for secondary outcomes not reaching statistical significance, it’s conspicuous and impressive how many had non-statistically significant decreases in outcomes like death. That gives me hope that this thing works.

The skeptic in me notes that a few findings were less impressive than the headline. For example, when researchers measured all-cause death or Covid-related hospitalization (factoring out longer ER stays), about half of that signal of benefit vanished (i.e., there was a 25% reduction in those outcomes as opposed to a 51% reduction in the primary outcome). When researchers measured all-cause ER visits, all-cause hospitalizations, or all-cause deaths, the signal of benefit almost vanished altogether; there was just a 6% reduction in these outcomes combined, a that finding was not distinguishable from noise/chance. This matters. As a patient, I don’t really care whether I die of Covid or of something else. If a drug lowers my Covid death rate, but not my overall death rate, I sort of lose interest. The point is to live, not to die in the same time window but not have the word “Covid-19” on my death certificate.

Nevertheless, as I mentioned, a bunch of secondary outcomes found either a statistically significant sign of benefit or had promising results that did not clear statistical thresholds. So, this all looks encouraging for interferon. And if this works even a fraction as well as these data suggest, it would still be good news. What’s more, there’s no reason this treatment should only work on Covid-19. So, this might have broader applications.

And yet, as the New York Times reported, the FDA is unlikely to authorize this drug for use in the United States. The reasons for this are actually kind of silly. First, the study was done in Brazil and Canada—and the FDA likes US data. It seems to me that in a pandemic, science is science, regardless of its locale. What matters is the quality of the endeavor, not the country code where it took place. Japan made a mistake by not accepting US data on mRNA vaccines back in 2021, which meant that it was slow to get its vaccination campaign up and running. We should not make the same mistake on therapeutics. Second, the FDA doesn’t like that the study was conducted by academics, rather than the pharmaceutical company that makes the drug. This is downright bizarre. If anything, industry-run studies are more likely to have been designed to detect a benefit than studies done by academic researchers. (That said, the company that makes the drug donated the drug to the study, so it’s not as if this study was truly independent anyway.) Punishing the study for not being substantially biased by virtue of being run by the company that makes the drug seems backwards.

Something I hear few people discussing is how expensive this drug is. I can’t quite figure out what the drug costs per dose, but it seems like it’s many thousands of dollars. Considering that hundreds of patients would need to get the drug to save one life, scaling this drug up may not be feasible in many places. In a global pandemic, we’re looking for drugs that are both effective and affordable.

It will be interesting to see if these findings are replicated. I’m also looking for other experts (skeptics in particular) to chime in. If these data are too good to be true, it would be great to know that before we start spending billions of dollars on this particular interferon.