German chancellor Angela Merkel said it best: “We need to get to a doubling time of longer than ten days.” The number of hospital admissions and deaths caused by COVID-19, those doubling intervals, is more than ten days. For example, on March 26 we had one thousand deaths; on March 28, two thousand. That’s a doubling time of two days. By April 1 we had four thousand deaths. That means we had a doubling time of four days. If you can get to a ten-day doubling time, it is likely that hospital discharges will exceed admissions and you can say that you accomplished what you wanted—no longer overwhelming the healthcare system. That’s what you are worried about—not that you can’t take care of these patients, but that you can’t take care of any patients.
Once you get there, things can loosen up and you can go back to work even though there still may be cases and deaths. But that is not the goal. The goal is to not overwhelm the health system and allow us to go back to work. When do we get there? I’m going to predict that by the end of April, things start to look better and we start to see a gradual increase in our doubling times.
Heilbrunn: Do you think we are likely to reach the high end of the predictions of deaths? Or is that just unknowable?
Offit: It’s surprising, isn’t it? We know that lockdowns, or sheltering in place, works. China, Singapore, Japan, and South Korea were all able to pretty much end the spread of this virus while knowing that there had to be tens of millions of people in those countries who were still susceptible. Yet still, they were able to end it—that’s a good sign.
Now Vice President Mike Pence has said that he thinks we are mimicking Italy. Italy has 13,000 deaths right now and they are roughly one-fifth our size. That would come out to 65,000 deaths, but they are predicting as many as 240,000 deaths. Which means that Italy would have to have 50,000 deaths, for us to be Italy. It’s just surprising that it could get that bad; even though Italy is still suffering deaths, it looks like their doubling time is about eight days. So it should be coming off this curve and I don’t see this 240,000 death estimate. I don’t see that. But they are also close to the data, so they could be seeing something I’m not.
We did many things wrong. We were slow to ban travel from China. We were very slow to do testing—South Korea had already done 150,000 tests by the time we had done fewer than 500. We were slow to lockdown. We had a president who, if anything, tried to play this down: “We’ve got it under control, it’s only a few cases, don’t worry about it.” And we were very slow to supply protective gear, ventilators, etc. so people were not getting the care they need or being protected in the manner they need to be. So we’ve done a lot of things wrong. As a consequence, we are doing worse than most countries.
Heilbrunn: So you would have banned travel from China earlier? President Donald Trump keeps bragging that he did it quickly.
Offit: Right, but if you look a little more closely at how that played out, he had to be convinced to do that. He didn’t want to offend his friend President Xi [Jinping]. China was not a good actor here. We should not have had to have a whistleblower to tell us that there was a novel virus that was killing people in Wuhan. They should have been quick to tell the world that this was going on so the world could prepare for it. They didn’t do that. But the moment it did become clear, Trump was very slow to ban travel. But once the virus was here, once there was community spread, it didn’t really matter.
You can make the same case for what is happening in New York. We were pretty slow to ban travel from Europe, when clearly there were problems in Spain, Italy, and France. I suspect that may be the reason that New York suffers so much now—there was a lot of travel from COVID-heavy territories, regions, or countries, coming into LaGuardia, coming into JFK.
Heilbrunn: You were the co-inventor of the rotavirus vaccine. How critical do you think it is that we develop a vaccine against the coronavirus? Or do you think the virus will largely burn itself out?
Offit: We certainly should not assume that it will burn itself out. We should make a vaccine as quickly, efficiently, and safely as possible. There is some good news about the vaccine. Human trial studies done decades ago show that if you are inoculated with one of the four strains of human coronavirus that circulates in our country every year and are challenged with that virus a year later, you are protected. Good. That means that there is protection—it’s probably years, not decades—but that’s a good sign. You also know which protein you are interested in. You are interested in that spike protein—the glycoprotein—that’s the protein that attaches to cells. If you can prevent the virus from attaching to cells, then you can’t get infected. And we live in an age of recombinant DNA technology where we can make that protein itself in a manner similar to the Hepatitis-B or HPV vaccine, or we can use either messenger RNA or DNA vaccines that express that protein.
So the messenger RNA approach is the one being used by Moderna and that is already in the human trials that Dr. Anthony Fauci has talked about. But they are really at the beginning of this process. You still don’t have an immunological correlate of protection; you still don’t really have a dose. The forty-five people who are being tested in Washington state have been divided into three groups of fifteen and each given a different dose. They started with a low dose, then they moved to a middle dose, and now they are moving to a higher dose. But that’s only fifteen people in each group. Really, to do this right, you would need thousands of people to make sure you are giving the right dose—not too much or too little. I presume that they will move to that, but it takes time.
Heilbrunn: What is a realistic timeline for a vaccine?
Offit: You want to make sure that a vaccine induces an immune response that you think will be safe. I’m assuming that they are not doing animal-model studies because it seems that they moved very quickly to human trials. It would have been nice to do animal-model studies. While mice aren’t bad, they do give you hints to what could be potential safety problems. But it looks like that is not happening.
When you are where they are now—which is about forty-five people tested—you do need to gradually expand that to thousands of people at the dose you think you’re looking at and make sure a significant percentage of the population is receiving that, and that that population represents the U.S. population. You want those people to develop an immune response that would protect them, even though you don’t, right now, have immunological correlate protection. We don’t know that, because we aren’t doing animal studies and we obviously aren’t doing human trial studies with this virus. Then you would need an efficacy trial, and that would best be done on healthcare workers.
Healthcare workers are the ones who are most likely to come in contact with COVID-19 patients, both by frequently being in contact and in close contact with them because they have to examine them. That would probably be the group with which to do an ethical trial, but that takes time. That takes years. For us, in the development of the Rotavirus vaccine, that took sixteen years before we got to the large, definitive, phase three trial. That was a placebo-controlled, prospective, eleven-country, four-year, $350 million trial, on 70,000 babies. That was a definitive phase three trial.
My sense is that they are moving along in this “break the glass mode,” which is to say that they will use fewer people in these studies and may bypass the FDA. They might just move to offer this. That said, to put this in perspective, Dr. Fauci is certainly right. The mRNA approach is very quickly scale-uppable. You can make very large numbers, remember you are talking about making billions, hundreds of millions of doses. But the mRNA still needs to be delivered in this complex lipid delivery system, which is not so easy to scale up. It could take a year to just scale up that part of it. And the filling takes time. Even if they use multi-dose vials, the filling alone could take a year. So, I don’t see eighteen months—Dr. Fauci could see something I don’t here, he is certainly closer to it—but I think eighteen months is a very, very, very optimistic timeline.