Researchers around the world are working frantically to create a vaccine for COVID-19. More than 150 vaccines are in some phase of development, with 23 now in human trials.
This week brought a spot of good news from one of the frontrunners: According to a study published in the New England Journal of Medicine, in a “Phase 2” trial, Moderna’s vaccine led 45 patients, ages 18-55, to produce antibodies able to neutralize the virus — and in doing so, caused only minor side effects, such as headaches and fatigue.
That vaccine is based in part on the work of Jason McClellan, an associate professor of bioscience at the University of Texas. He spoke with us from his office.
Lots of us are holding our breaths waiting for a COVID-19 vaccine that would let us resume our normal lives. But we know that vaccines usually take 10 years to develop. Is it possible to get one faster?
I'm optimistic about the current vaccine development for SARS-CoV-2/COVID-19. I think we are likely to have one or more approved for use by the end of this year. That would be historically fast: The fastest vaccine ever developed took four years.
Wow. Let's back up to December, when the new coronavirus appeared. How did your lab get involved so fast? Did you receive some sort of Bat Signal?
Information was circulating on Twitter. A lot of us scientists all follow each other, so at the end of December 2019, we were aware that these pneumonia outbreaks were occurring in Wuhan. That suggested either an influenza virus or a coronavirus.
Early in January 2020, it was confirmed that it was a coronavirus — a beta coronavirus, similar to the first SARS. I was snowboarding in Park City, Utah, when I got a call from Dr. Barney Graham at the National Institute of Health’s vaccine research center. I’ve worked with him since 2009; we researched spike proteins on other coronaviruses.
Barney said, “It looks like it's a beta coronavirus. We want to race on this; try and get a vaccine out quickly. Do you want to collaborate?”
I immediately texted my lab: “We're gonna get going on this.”
On January 10, Chinese researchers made the sequence of the SARS-CoV-2 genome available online. So that weekend, we started designing exactly what we wanted the vaccine antigen to look like.
Barney got in contact with Moderna. And then from there, it's been a sprint.
So what is an antigen? This thing you designed?
An antigen is a portion of the virus, or even the whole virus, that's injected as part of a vaccine, so that our immune systems learn to recognize the virus.
There are different ways to make a vaccine; we're seeing that now. One is, you use the entire virus. It can’t be the full infectious virus because then you'd just be infecting people. So you attenuate it in some way — inactivate it or kill it. Several of those whole-virus vaccines are being developed. But in China, the data on those haven't looked so good.
The other way is to use portions of the virus that our immune system can learn to recognize. If you've seen pictures of the coronavirus, you’ve seen these large proteins on the surface sticking out — the spikes. That is what our immune system recognizes. When people are infected with coronaviruses, we make antibodies against the spike protein.
For the virus to infect us, the spike protein is critical. It has two functions. The spike binds to receptors on the surface of our cells, and that attaches the virus to our cells.
Then, once bound, the spike undergoes this major rearrangement, kind of like a Transformer, and it fuses the viral membrane with our host cell’s membrane — and that fusion of the two membranes allows the viral genome to enter the cell. So then the cell is infected.
We want the vaccine to lead to antibodies that can stop those two processes — that can prevent attachment and that prevent the membrane-fusion process. So we know to use the spike or a portion of the spike as our vaccine antigen. That’s what most of the vaccines are doing.
We hear a lot about different kinds of vaccines — RNA vaccines, DNA, and so on. What are the differences? How do those work?
Great question. A traditional one would be a sub-unit vaccine, a protein-based vaccine. In that case, you would make the viral protein — the spike protein — purify it, and then inject that purified protein into the arm. Your body recognizes that it's foreign, makes antibodies against it and hopefully protects you.
Making those vaccines takes awhile. You first have to make a stable cell line that can produce the protein in high concentrations, and scale it up into large bioreactors — which has to be done under Good Manufacturing Practice standards. So sub-unit vaccines can take a while, but it's a tried-and-true approach.
I think the farthest along for that approach is Novavax. They got $1.6 billion last week from Operation Warp Speed. They're using a recombinant spike protein that contains stabilizing mutations that my lab developed, so we're rooting for them.
Well, actually, we're rooting for everybody.
Other kinds of vaccines ask your body to do more than just recognize the antigen. Those other kinds — DNA vaccines, RNA vaccines, or viral vector vaccines — all involve some sort of blood nucleic acid.
So that's providing purified antigen, not asking your body to do very much other than recognize it. The other forms — a DNA-based vaccine, an RNA-based vaccine, or a viral-vector vaccine — all involve some sort of nucleic acid.
For the DNA-based vaccines, you have DNA that encodes for an antigen — let's say it's a spike protein. You have to get the DNA inside your cells — like, inside your muscle cells. That can be a little tricky. Sometimes they have to use this like gun-type device to get the plasmid with the DNA into the cells. The plasmid then goes into the nucleus of your cells. It gets transcribed, translated, protein is expressed — and then your own cells are making the viral proteins.
Wait: There's a tiny gun that can shoot spike DNA into my cells?
It’s not that tiny. There are different ways of doing it, but yeah, one is an electroporating gun that can get the plasmid in.
Okay. Back to vaccine types.
Another type is the mRNA-based vaccines. Moderna’s is one of the farthest along there, and another is a collaboration between BioNTech and Pfizer. Messenger RNA, mRNA, is what carries the genetic code for making proteins. So you select an mRNA molecule that encodes for the antigen, for the spike protein. That gets injected into your cells, the mRNA goes into the cytoplasm, and it immediately just gets transcribed by the ribosomes, and they making the viral proteins.
So your own muscle cells are making the viral spike protein, which your immune system recognizes as foreign and makes antibodies against it.
So that’s similar the way a virus hijacks our cells and gets them to churn out copies of itself? Except that it’s not the full-on virus?
Yeah. There’s a lot of benefits to having your own body make the protein rather than just injecting the protein directly. It involves T-cell responses and other types of immune responses, so there could be a lot of benefit.
And also, this type allows us to move really quickly. You can immediately just start synthesizing the RNA molecules. I think the Moderna Phase One mRNA vaccine trial started 66 days after the genome sequence was available — which is just incredible.
Phase One: Those are the very first trials in human beings? To be sure the vaccine doesn’t outright kill people?
Yeah. Phase One is in human beings. Moderna’s is done, and the information on it just came online this week in the New England Journal of Medicine.
Phase Ones are generally done with tens of people — 30 or 40 people. You just want to make sure that it's safe, that you're not getting a lot of adverse events. It generally involves dose escalation. The first couple people will get a relatively small dose, then you wait a week or two to see what happens. If that looks okay, then the next set of people get injected with maybe five times as much, etc.
Phase One can also give you some idea of the vaccine’s efficacy. You can draw blood and look to see: Did the people make antibodies that can neutralize the coronavirus?
What does it mean, to neutralize a virus?
That means that in a lab, the antibody can prevent virus from entering cells — like, in a lab’s cell culture dish. Neutralizing antibodies correlate very well with protection for coronaviruses. So that that's really sort of one of the things that people are looking for: After vaccination, are you generating robust levels of antibodies? The Moderna vaccine looks pretty good, as does the Pfizer/BioNTech one. They're generating antibodies in people at levels similar to people who have survived COVID-19 infections. That's basically what you want: to mimic an infection without all the deleterious events.
We keep hearing that it's possible that some people who've had COVID-19 don't have antibodies later, or that maybe their antibody levels drop off to the point that they're not effective. Could that happen with a vaccine? Would we have to have booster shots?
It's a good question. We don't know yet.
We're concerned that people who’ve been infected and recover can maybe then can get reinfected. There have been some reports.
Not all infections are the same either. Some people might get a very minor infection and be asymptomatic. They would be less likely to produce a robust immune response than somebody who's been infected and hospitalized, because a very sick person is exposed to really high viral loads, so they generate more antibodies.
When different people get infected, they generate different antibody titers — different concentrations of antibodies — and those concentrations wane over time. That's natural. That happens for almost all infections. We can't have all antibodies be maximally ramped up all the time.
So there's a recall response. We have memory cells, and when the immune system recognizes the virus again, those memory cells are able to quickly expand and make more antibodies. That's good. That’s what we're trying to get with a vaccine: If you're vaccinated, hopefully that protects you in the short term from sickness.
Long-term? You know, we're just trying to prevent against severe disease. A vaccination might not prevent you from ever getting infected again, but maybe the new infection is only like a mild cold for a couple of days. It doesn't lead to hospitalizations.
We hear a lot about the immune system overresponding to COVID — about cytokine storms that come late in a serious infection. Could that be a problem with vaccines? Would we have to worry that in some people, you might get an overresponse?
Probably not. I think a lot of that is being exposed to really high viral loads. The virus has many different genes that it uses to interfere with our immune response. You really wouldn't get that with any of these mRNA or DNA or sub-unit vaccines, where you're just injecting one protein. We're just raising the antibody response against that one thing.
But if that’s an issue, we would see such things in the larger clinical trials.
Is that why we have enormous clinical trials? To look for unusual responses?
Exactly. If there's a side effect that occurs in 10 percent of people, in a 50-person trial, you’d expect to see that in around five people. But what if there is an adverse event that only occurred in one out of 1000 people? To see that two or three times, you’d need a clinical trial that has several thousand people.
So we scale up. Phase Two is in the hundreds of people, and Phase Three is in the thousands or tens of thousands.
Phase Three is not just about safety. It’s also about efficacy, about how well the vaccine works. One group is going to receive a placebo, and one is going to receive the the actual vaccine, and then you can compare outcomes — hospitalizations, duration of hospitalization, deaths, things like that. Hopefully the group that received the vaccine fares better than the group that received the placebo.
That’s the testing phase that Moderna is about to enter.
So once testing shows that we have a vaccine that’s safe and effective, how long does it take to produce and distribute it? How long until it’s in my arm?
Going by what Moderna and others have claimed, I think they would start trying to make tens of millions of doses in the fall, then hopefully tens of millions of doses per month, and then scale it up into the hundred million doses per month.
One of the benefits of having more than one vaccine that works is that you’d have multiple companies that might able to make 10 million doses per month, and that would allow more people to get immunized.
A lot of these plants, they're being scaled up now. They're starting to make a vaccine on the assumption that it’s going to work. They’re not waiting to see whether it works before ramping up.
That's what Operation Warp Speed is doing: providing the money upfront so we can start large-scale manufacturing, so that we’ll be ready right away if a vaccine works.
But we'd love to have multiple options. It’d be great to have viable vaccines in the different modalities. DNA, mRNA, viral vector and protein-based: those are all manufactured differently, so you're not putting all your eggs in one basket. If there's some supply issue with one, you'd still have two or three others.
Speaking of money: What’s the relation of your lab or UT to companies like Moderna, BioNTech and Pfizer? Are you incredibly rich now?
We have intellectual property on some of the stabilizing mutations that two or more companies are using as part of their vaccine. But that all gets complicated with licensing and royalties. It eventually goes to court, probably. Something like that.
But that's not your area.
Yeah, that's not my area. We get a bit of money, and that's exciting. Some of it goes to UT; some of the goes to the scientists. But we haven't received any funds yet that I'm aware of. We'll see how that plays out.
What else do you want Houstonians to know? What should we be watching as we wait for a vaccine?
The Phase Three trial is pivotal. That's really what we're really aiming for. So hopefully there'll be some interim results. It'd be great to know by September if the vaccines are working.
Most people think you'll need a prime. So you'd get vaccinated, and then you'd come back two or four weeks later for the boost, and then a few weeks after that, you'd probably be maximally protected. In the trials they'll be monitoring to see how long it is before people that are vaccinated just once get infected. That'll determine whether you need yearly yearly injections or every other year.
Right now, with COVID, I think the main thing is just trying to keep the cases down. Obviously in Texas we're just exploding with COVID cases from from opening up too early. We just need everybody to social distance, be responsible, wear masks. Try to keep the numbers down. Buy time for the vaccine to be developed and distributed.
One last thing: What’s that toy behind you?
[Picks it up, smiles.] Our lab was the first to determine the structure of the spike protein for this coronavirus -- for SARS-CoV-2. That’s what this is: It’s a 3D print of the spike protein. It’s an atomic representation.
This is a super high-end model. It cost about $1,000. The dean paid for it, and bought himself one. [Laughs.]
[Holds up the model to his computer’s camera.] So in the spike, three identical proteins come together — they’re colored red, green and blue here — so the spike has a little threefold symmetry.
Then those proteins are covered in glycans — see these little multi-colored nubs? Glycans are sugars that come off. They help shield the coronavirus from our immune system. It's really clever. These glycans are also found on many of the proteins in our own body. So when antibodies see them, they don’t bind. They think it’s part of our own body.
Oh, it’s evil!
Yeah. It’s evil.
This interview has been edited for length and clarity.