Now that we’re seeing that coronavirus vaccines are indeed possible (and are on their way), let’s talk about the remaining unanswered questions and the things that we will be getting more data on. Here are some of the big issues – it’ll be good to see this stuff coming into focus. I’ll put these into the form of questions (think of it as a tribute to the late Alex Trebek, whom I was glad to help remember in this article, and more here). Each one will have a summary answer at the end of the section, if you just want to skip to that part.
How long will the vaccine protection last?
This one can’t be answered with total confidence by any other way than just waiting and watching. But we will be able to give a meaningful answer well before that, fortunately. Here, just out in the last couple of days, is the most long-term and comprehensive look at the duration of immunity in recovered coronavirus patients. In fact, it appears to be the largest and most detailed study of post-viral-infection immunity in the entire medical literature (!) It’s from a multi-center team at the La Jolla Institute for Immunology, UCSD, and Mt. Sinai, and it looks at 185 patients who had a range of infection experiences, from asymptomatic to severe. 38 of the subjects provided longitudinal blood samples across six months.
We’ve already seen from the convalescent plasma comparison samples in the various vaccine Phase I trials that the antibody response to coronavirus infection can be quite variable, and that was the case in this study as well. That gives you wide error bars when you try to calculate half-lives, and it’s not even clear what kind of decay curve the antibody levels will best fit to (it might well be different in different patients). But one figure to take home is that 90% of the subjects were still seropositive for neutralizing antibodies at the 6 to 8 month time points. The authors point out that in primate studies, even low titers (>1:20) of such neutralizing antibodies were still largely protective, so if humans work similarly, that’s a good sign. An even better sign, though, are the numbers for memory B cells, which are the long-term antibody producers that help to provide immunological memory. B-cells specific to the Spike and to the nucleocapsid coronavirus proteins actually increased over a five-month period post-symptom-onset, thus with no apparent half-life at all. These had interesting variations in antibody type (by the end of the period, they were strongly IgG, the others having dropped off), but as the paper notes, we really don’t have many viral infection profiles in humans to compare these results to. B-cell memory overall, though, looks to be long-lasting, and is expected by these results to stretch into years. For what it’s worth, there are patients who survived the 1918 influenza pandemic who had B cells that still responded with fresh neutralizing antibodies after over 90 years, so they can be rather hardy.
What about the other immune (and immune memory) component, T cells? The news there is good as well. CD4+ and CD8+ memory T cells appear to have half-lives of at least five or six months in these patients, and helper T cells (crucial for those memory B cells to respond later on) were completely stable over the entire period studied. Again, there are very few viral infection studies to compare this one to, but these numbers look consistent with long-term protection via reactivated immune memory.
Looking over the whole set of patients, it was clear that the immune system’s famously individual character was on full display here. That heterogeneity could well be the reason that there are real cases of re-infection, although it still seems to be rare. Different components of the immune response (both in antibodies and T cells) varied widely among patients, and these differences only became more pronounced over time. Nevertheless, at the five-month time point in a measure of five components of immune response and memory, 96% of patients were still positive on at least three of them (the categories were IgG antibodies against the Spike receptor-binding domain (RBD), IgA antibodies against the same Spike RBD, memory B cells aimed at the RBD, total SARS-CoV-2-specific CD8+ T cells, and total SARS-CoV-2-specific CD4+ T cells).
Bottom line: Taken together, this study, several others over the past few months, and this recent work all paint a consistent picture of a strong, normal, lasting immune response in the great majority of patients. Add in the results we’re seeing from the two vaccines that have reported interim data so far, and I think that the prospects for lasting immunity from vaccination are also very good. Remember, the early vaccine data suggested antibody responses at least as strong as those found in naturally infected cases. There seems (so far) every reason to think that vaccine-based immunity will be as good or better than that conferred by actual coronavirus infection. I very much look forward to more data to shore up this conclusion, but that’s how it looks to me at the moment.
How effective are these vaccines? Will they provide total protection or not?
We’re just starting to get numbers on this, and we are definitely going to know more as the various trials read out interim data and then reach their conclusions. So far, though, the efficacies we’re seeing have been more than I had really hoped for. I thought that they would work, and I didn’t think that meant just the FDA’s floor of 50% efficacy, but I sure didn’t have the nerve to predict that the first two readouts would be 95% (Pfizer just reported their final readout this morning). I can’t overemphasize how good that news is, especially when you compare it to some earlier worries that a useful coronavirus vaccine might not even be possible at all. Cross that one off the list!
Those efficacy numbers, though, are measured for symptomatic coronavirus cases. The vaccine trial participants are not being pulled in at regular intervals for testing to see if they’ve gone positive-though-asymptomatic. We may get controlled data of that sort eventually, but for now, we know from the Moderna trial that the few people who came down with symptoms at all had very mild cases. The antibody levels that we’re seeing would argue for a low probability of having a significant number of vaccinated people walking around asymptomatically shedding coronavirus, and for anyone who does to be shedding a lot less of it for a shorter period of time.
From a public health standpoint, that’s what you need. Epidemics are a matter of probabilities, and you can lower the chances of spread for a virus like this in any number of ways. They surely vary in efficacy, but include keeping distant from other people and avoiding any crowding in general, wearing masks, avoiding indoor situations with people that you haven’t been exposed to (such as going to the grocery store when it’s not so crowded), minimizing the time you spend in any higher-risk situation in general (getting those groceries in an organized fashion and getting back outside), and more. The fewer people there are around shedding infectious particles, the better (obviously), but the worst case for a weakly effective vaccine might be that it could actually raise that number for a while by creating more asymptomatic cases rather than having the infection make people aware that they need to stay the hell inside. But I don’t think we’re going to see that. I think that the efficacy levels we’re seeing are indeed going to be epidemic-breaking if we can get sufficient numbers of people vaccinated. Right now we’re up around the efficacy of the measles vaccine, which is very effective against a virus that is far more infectious than SARS-Cov-2. . .if enough people take it. (Believe it, if the current coronavirus were as infectious as measles is, we would be hosed).
Bottom line: the results we have so far indicate that these vaccines will indeed provide strong protection in the great majority of patients. The number of asymptomatic cases among the vaccinated population will be a harder number to pin down, but I believe that we should be in good enough shape there as well, based on antibody levels in the primate studies and what we’re seeing in humans.
What about coronavirus mutations? Will the virus move out from under the vaccine’s targeting?
The SARS-Cov-2 virus has indeed been throwing off mutations, but all viruses do. They replicate quickly, and errors pile up. Fortunately, though, none of these have proven to be a problem so far. There’s been a lot of talk about the D614G mutation being more infectious, but the difficulty of proving that shows that it’s certainly not way more infectious, if it is at all. And it doesn’t seem to have a noticeable effect on disease severity – so far, no mutation has.
The recent news from Denmark about a multi-residue mutant (“Cluster 5”) that might be less susceptible to the antibodies raised by the current vaccines is a real concern, but the news there, thus far, is also reassuring. The vaccine efficacy warning might be true, but it was also based on a small amount of preliminary data. And the Cluster-5 variant has not been detected since September, which suggests that (if anything) this combination of mutations actually might make the virus less likely to spread. From what we’ve been seeing with the Spike protein, evading the current antibodies looks like it’s going to be difficult to do while retaining infectiousness at the same time. We already know from a Pfizer analysis that many of the common mutations are just as susceptible to neutralizing antibodies raised by their vaccine. Update: here’s an effort to monitor the developing mutations to keep an eye out for just these issues.
I know that many people are wondering about the similarity to influenza, and to the yearly (and not always incredibly effective) flu vaccines. Flu viruses, though, change their proteins far more easily and thoroughly than the coronavirus does, which is why we need a new vaccine every year to start with. SARS-Cov-2 doesn’t have anything like that mix-and-match mechanism, and it’s a damn good thing.
Bottom line: the coronavirus can’t undergo the wholesale changes that we see with the influenza viruses. And the mutations we’re seeing so far appear to still be under the umbrella of the antibody protection we’ll be raising with vaccination, which argues that it’s difficult to escape it.
What about efficacy in different groups of people? Where will the vaccines work the best, and where might there be gaps?
This is another area that is definitely going to come into better focus as the current trials go on. For the moment, we know that the results we have seen so far come from participants in a range of ages and ethnic backgrounds. There’s not much expectation that things will vary much (if any at all) across the latter, although it’s always good to know that for sure, and not least so you can point to hard evidence that it’s so. Age, though, can definitely be a factor. Older people are quite likely more susceptible to coronavirus infection in the first place, and are absolutely, positively at higher risk of severe disease or death if they do get infected. The immune response changes with aging, and it is very reasonable to wonder if the response to vaccination changes in a meaningful way, too.
But as mentioned above, we have more data from the Pfizer vaccine effort just this morning. The overall efficacy was 95%, and the efficacy in patients 65 and older was all the way down to 94%. This is excellent news. No numbers yet for people with pre-exisiting conditions and risk factors, but I’m definitely encouraged by what we’re seeing so far.
Bottom line: our first look at efficacy in older patients is very good indeed, and that’s the most significant high-risk patient subgroup taken care of right off the top.
How safe are these vaccines? What do we know about side effects?
As mentioned in the Moderna write-up here the other day, that team saw around 10% of their vaccinated cohort come down with noticeable side effects such as muscle and joint pain, fatigue, pain at the injection site, etc. These were Grade 3 events – basically, enough to send you to bed, but definitely not enough to send you to the hospital – but they were short-lived. For reference, those numbers seem to be very close to those for the current Shingrix vaccine against shingles, from GSK (thanks to their butt-kicking adjuvant mixture of a Salmonella lipopolysaccharide and a natural product from a South American tree). It’s a reasonable trade for coronavirus protection, as far as I’m concerned. And my reading of the Pfizer announcement today makes me think that their side effect profile is even a bit milder. They have fatigue in 3.8% of their patients, and all the other side effects come in lower.
What about lower-incidence side effects? Well, 30,000 patients is a pretty big sample, but on the other hand, the immune system is as idiosyncratic as it can be. There may well be people out there who will have much worse reactions to these vaccines. If you have a literal one in a million, you’re simply not going to see that in a trial this size, or actually in any trial at all. These are about as big as clinical trial numbers ever get. At that point, you’d be looking at such a hypothetical bad outcome in about two or three hundred people if we gave the shot to every single person in the US. And the public health calculation that’s made every time a vaccine is approved is that this is a worthwhile tradeoff. Let’s be honest: if we could instantly vaccinate every person in the country and in doing so killed 200 people on the spot, that is an excellent trade against a disease that has killed off far more Americans than that every single day since the last week of March. Yesterday’s death toll was over 1500 people, and the numbers are climbing.
How about long-term problems, then? These are possible with vaccines, but rare. And unfortunately, there is truly no way to know about them without actually experiencing that long term. We simply don’t know enough immunology to do it any other way. Given the track record over the last century of vaccination, though, this seems to be another deal worth making.
Bottom line: immediate safety looks good so far. Rare side effects and long-term ones are still possible, but based on what we’ve seen with other vaccines, they do not look to be anywhere at all significant compared to the pandemic we have in front of us.
OK, what about the rollout? Who’s getting these things first? When does everyone else get a chance to line up?
Harder questions to answer – there are a lot of variables. Pfizer and Moderna both say that they can make in the range of 20 million doses by the end of the year, but what we don’t know is (1) when the FDA will grant Emergency Use Authorization, (2) how many of these doses can be distributed and how that’s going to happen, (3) what the number of doses available right now might be, (4) how the ramp-up of both production and distribution are going to be coupled in the coming months, (5) what’s going to show up with the other vaccine candidates in testing, and so on.
The person in charge of the “Operation Warp Speed” logistics is Gen. Gustave Perna, who has been in charge of the Army’s Materiel Command (just the sort of background you’d want for an effort this size, I think). We know that manufacturing has already been underway on an “at risk” basis, and it looks like those bets are paying off, given the clinical results. Here’s the rollout strategy that has been announced so far, and it certainly seems sound from what I know about these things. It does leave some questions open, such as what groups are in the initial queue. You would have to think that health care workers would be at the top of the list – these people are risking their health and their lives as they deal with a constant stream of infectious patients, and losing them to illness or death has a severe impact on our ability to deal with the situation.
That situation, it has to be said, is going to be getting worse. It’s been getting worse for weeks, and it looks like it’s going to keep doing that for several weeks more even if we do everything right. And let’s be honest: as a country, as a population, we’re not doing everything right. There are a lot of people taking sensible precautions, but others are letting their guard down when they shouldn’t, and there are of course other people who never put their guards up in the first place and seem to have little intention of doing so. The map says “uncontrolled spread” across most of the US, and they ain’t lying. These vaccines are coming at extraordinary, record-breaking speed, but not fast enough for us to avoid what looks sure to be a 2,000-deaths-a-day situation. Take the worst air crashes in aviation history, and imagine three, four, five, six of them a day. All day Monday. All day Tuesday. No letup. Every single day of the week and all weekend long, a hideous no-survivors crash every few hours. That’s what we’re experiencing right now in terms of the sheer number of deaths.
Bottom line: the very first people to get these new vaccines will almost surely be health care workers, and starting some time on in December. The rollout after that has too many variables to usefully predict, but it’s going to be the biggest thing of its type ever attempted, in people-per-unit-time. And yes, I think it’s going to work, and not a minute too soon.