Novelty Means Severity: The Key To the Pandemic
Nothing in a pandemic makes sense except in the light of novelty: A guest essay.
This edition of Insight is a guest essay by Dylan H. Morris, a Postdoctoral Researcher at UCLA who studies how the evolution of RNA viruses is shaped by ecological processes within and between hosts. Dr. Morris brilliantly unpacks something we hear all the time but is rarely explained in its full and profound implications: for humans, SARS-CoV-2 is a novel coronavirus. And that distinction means so much.
Understanding novelty in the context of the immune system is related to most everything we’re discussing and debating. What do variants mean for vaccines? Should children get immunized? How can we think about vaccine breakthrough cases in terms of severity or Long Covid? What might endemic COVID look like? What about waning immunity and reinfections? And why is vaccinating globally an urgency?
Geneticist Theodosius Dobzhansky had once famously said, “Nothing in biology makes sense except in the light of evolution.” This essay makes something else clear: Nothing in this pandemic makes sense except in the light of novelty.
Without further ado, here’s Dr. Morris.
Novelty Means Severity
by Dylan. H. Morris, PhD
SARS-CoV-2 is new to our immune systems. That makes it very dangerous. Viruses that are new to us spread faster and are more lethal than old familiar ones.
Some scientists are tempted to chalk this up to evolution. The argument is that a virus that leaves its host alive will outcompete one that kills its host. Viruses do sometimes become less deadly as they adapt to a new host species (like us), but they also sometimes become more deadly. But whether wrong or right for a given virus, this tempting just-so story can be a distraction.
Novelty is bad regardless of virus evolution.
When a virus is new, nobody possesses acquired immune protection against it. Acquired immune protection is a different kind of adaptation: not virus evolution, but our own learned—adaptive—immunity. We build it over our lifetimes as we encounter new pathogens and learn how to fend them off.
If nobody has adaptive immune protection, a virus spreads faster. Even a few immune individuals in a population can meaningfully slow the rate of virus spread, since they are less likely to become infectious and infect others. If there are enough immune individuals, the virus may not be able to spread at all. This is the logic of population immunity and herd immunity. It is important. We talk about it a lot.
If nobody has adaptive immune protection, a virus causes severe disease in more of the people it infects. This is also important. We don't talk about it enough.
Unless we eradicate SARS-CoV-2—possible but not likely, especially in the short term—just about everyone is going to encounter the virus sooner or later. But those who have adaptive immunity from infection or vaccination may not get sick at all. Even if they do, they will be less likely to get very sick or die.
Now that we have safe, effective vaccines, we can give people immunity without causing dangerous disease. That puts us into a global race against the virus. The more people who see the vaccine before they see SARS-CoV-2, the fewer severe cases, long-term health problems, and deaths. Faster worldwide rollout will save lives. It really is that simple.
* * *
But why is it that simple? Can't immunity wane over time? Can't the virus evolve, allowing it to infect the immune? Can't it infect or reinfect some immune individuals even without evolving? And if it does, won't we be back in March 2020, facing a fast-spreading virus that causes severe disease?
Vaccines are great, but are they the answer if they fail to prevent infections?
It is tempting to think of adaptive antiviral immunity as a castle wall. It might keep out some marauding raiders, but if they breach it, the defenseless townsfolk will be slaughtered just as easily as if there had been no wall in the first place.
That's the wrong mental model. Some features of immune protection are indeed wall-like: mucus and antibodies in our noses can restrain infectious respiratory viruses before they ever reach one of our cells and start replicating.
But we don't just have that wall. We also have a castle guard, who can fight and kill the marauders even if they successfully breach the wall, protecting our townsfolk. And if they're already familiar with the marauders' particular weapons and tactics, they'll have practiced counter-tactics that can give them the upper hand.
What's more, this is where the analogy—already strained, I admit—breaks completely: outside of fantasy novels, medieval castles could not produce an exponentially growing quantity of guardsmen in time to counter an exponentially growing horde of marauders.
With that in mind, I'll now drop the Arthurian analogies and explain the actual mechanisms, but remember this: adaptive immunity gives you a head start against a virus infection that you didn't have before, when you were immunologically "naive" to it. That head start could be the difference between having a mild fever and ending up on a ventilator.
* * *
When you're infected with a virus you're never seen before, you first mount an innate immune response. This is largely non-virus-specific; it acts to limit virus replication in ways that are more or less the same regardless of the virus. For example, the innate immune response kills infected cells, preventing them from producing new infectious virus particles (virions).
Later in this first infection, you mount an adaptive immune response. This is a targeted response to the specific new invading virus. Like the innate response, it has many facets ("the immune system is complicated", as immunologists love to remind us). But there's one key concept for our purposes here: antigenic recognition. In essence, your body gets more effective at fighting the virus by learning how to recognize some of the proteins that make up a virion. Through a process of trial-and-error, you produce immune cells and antibodies that match those proteins. This gives you both a targeted antiviral arsenal and an early warning system in the case of reinfection.
After you’ve gotten rid of all the viable virus and infected cells in your body (“cleared” the infection), the virus-specific B- and T-cells that you developed don't all go away. You keep some of them around to mount a faster and more powerful response if you’re ever reinfected. On reinfection, your cells notice that you're facing a known pathogen. Your memory immune cells then replicate—growing exponentially to counter the virus's own exponential growth within your body.
Sometimes, this "recall response" will crush the infection before you even feel sick. Other times, you'll feel under the weather for a bit as you fight off the virus. But crucially, you have a head start that you didn't have when you were immunologically "naive" to the virus.
* * *
There are many viruses for which we do not have effective vaccines. By the time you reach your teenage years, you've almost certainly been infected by a bunch of respiratory viruses, some relatively nasty, like influenza A, and others comparatively benign, like rhinoviruses and the alphabet soup of seasonal coronavirus (sCoVs): HKU1, 229E, OC43, NL63. We think of the sCoVs as a minor nuisance; in most adults, they cause the common cold.
SARS-CoV-2 is a coronavirus too. So why is it so much nastier than the sCoVs? One reason: it's new. More precisely, it's new to us.
One of the first observations people made about COVID was that it was frighteningly lethal in the elderly, but by and large, children were not getting too sick. Some people were surprised. Conventional wisdom was that influenza hit children and the elderly hardest, while sparing younger adults. Why was SARS-CoV-2 different?
But we need to look a little more closely, because it's hard to reach adulthood without having had the flu. Look at virus severity not by age but by age of first infection, and a pattern emerges: see something for the first time as a kid, and you'll most likely be okay (but only most likely). See it for the first time as an adult, and it can be nasty. The older you get, the worse it becomes to be infected with a virus you've never seen.
Children encounter many viruses to which they have no prior immunity. They compensate with robust innate immune responses that allow them to handle novel infections fairly well.
Robust doesn’t equal invincible. Without widespread childhood vaccination, infectious diseases kill many children, particularly children under five. A first encounter between the immune system and a virus can end tragically, even for a child.1
As you age, you get less good at handling novel viruses. And eventually you get less good at handling any virus, novel or familiar—your immune system ages ("immunosenescence"). The flu, for example, can be very severe in the elderly. But adults, even elderly adults, usually have at least some adaptive immunity to the viruses they face.
Things can get bad if they don't.
Chickenpox is a great example. It's benign in most children, but it's often severe in unlucky adults who make it to adulthood without being infected or vaccinated. When chickenpox vaccines were first rolled out, public health officials worried that the effort could backfire if not enough children got vaccinated. Suppress the virus, but not fully, and unvaccinated people might avoid the virus until adulthood, only to get a severe case.
Because SARS-CoV-2 is a new virus, by definition it's a virus that every adult avoided until adulthood. Small wonder that so many adults are getting severe cases. Almost everyone sees nuisance common cold coronaviruses like sCoV OC43 by the time they turn 10. But see OC43 for the first time at age 40, and you might have a pretty rough time of it. Indeed, OC43 can cause high attack rate, often-lethal outbreaks in nursing homes (where people have prior immunity, but have immunosenced, rendering their overall immune response less robust). Some scientists believe OC43 was the virus behind the 1889-90 global pandemic, which is estimated to have killed around 1 million people.
In an article on OC43, Anthony King writes: "If OC43 was the culprit in the 1889/90 pandemic, it has clearly lost its sting in the past 130 years". Has it? Or do we (almost) all now see it in childhood?
The "almost" may be important. I often wonder about the strong similarity between myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS)—a rare but severe chronic health condition—and many cases of Long COVID. ME/CFS is more common in adults than in children; it often takes hold in adults after a viral infection. What if it is a rare but dangerous consequence of first seeing in your 30s a virus most people first saw in childhood? Evade OC43 or another common virus as a kid, and it could give you post-viral sequelae when it finally hits you in adulthood.
And so while we don't yet have hard data on the efficacy of the vaccines in preventing Long COVID if they fail to prevent infection, the novelty-means-severity principle makes me hopeful. The virus might get you sick, but it won't be new to you. That could matter a lot.2
* * *
These are not new, grand insights. People have made these points throughout the pandemic. But they are easy to forget. And when we forget them, we make mistakes in combating the virus—lethal ones.
In December and January, genomic sequencing efforts in the U.K., South Africa, and Brazil identified virus variants-of-concern: B.1.1.7, B.1.351, and P.1. Vaccines had just begun to be distributed in the USA and the UK. There was a lot of consternation in both countries about B.1.351. Lab studies showing 6-fold reductions in neutralizing antibody titer were breathlessly and misleadingly reported as a "6 times less effective vaccine".
B.1.1.7, meanwhile, was pooh-poohed in some quarters. Sure, UK scientists were reporting that it was substantially more transmissible than the wild-type. But evidence suggested that vaccines would work well against it. Where was the problem? In fact, shouldn't we "root for it" to displace the more antibody-evasive strains?
Zeynep understood the error and raised the alarm. "More transmissible" is not benign when a virus is new; it's deadly. As I told her for that article, the problem is the race between vaccinations and the virus. More transmissible variants like B.1.1.7, P.1, and perhaps now B.1.617.2, are harder than the original virus to control with non-pharmaceutical interventions while we vaccinate. They make our race against the virus much harder.
Global vaccine equity is not just a moral duty for wealthy countries like the USA; it's an emergency. A great vaccine in a year's time is worth a lot less than a good vaccine now. There's only so long unvaccinated countries will be able to hold the line against SARS-CoV-2. If the vaccines arrive too late, COVID will already have reached the vulnerable as a novel, severe disease. If the vaccine arrives soon enough, COVID might still spread, as it is in the Seychelles, but people will be protected against severe disease.
* * *
But what about an antibody-evasive variant like B.1.351? What's the point of rushing to vaccinate people if they're just going to get infected by an escape variant?
Novelty means severity.
The seasonal coronaviruses evolve over time and partially evade our prior antibody immunity. So do seasonal influenza viruses. People suffer many flu and coronavirus infections over the course of a lifetime, thanks in part to this antibody-evading evolution ("antigenic evolution"). But until we immunosenesce, these reinfections rarely cause severe to critical illness, even when the virus has evolved partial immune escape.
Rarely is not never. A small number of unlucky individuals fail to mount a meaningful, protective adaptive immune response to infection or vaccination. Immune-compromised or immune-suppressed individuals are particularly at risk of failing to develop this adaptive immune protection, as are the immunosenescent. That is where population immunity comes in: it's up to the rest of us to protect them. We do this by protecting ourselves, thus limiting their exposure (from us).
These unfortunate unprotected individuals account for some fraction of documented reinfections and "breakthrough" infections in the vaccinated. But as the virus evolves or our immunity wanes, we expect to see more successfully protected individuals get infected too. But we also expect that they will retain protection against severe disease. We can already see hints of this effect: a study in the UK found that vaccines reduce the risk of hospitalization and death in elderly individuals with documented infections—they protect the castle even when the wall is breached.
Immune escape is rarely rapid and complete; it's more often gradual. We don't see viruses abruptly change so much that everyone with protective immunity loses all their protection, and the virus regains the full severity associated with full novelty.
B.1.351 is a good example. We now know that several vaccines in fact provide substantial real-world protection against symptoms versus B.1.351. One exception is the Oxford/AstraZeneca (ChAdOx1) vaccine; a small clinical trial in South Africa suggested that ChAdOx1-vaccinated individuals might not be that well protected against B.1.351 symptoms. But ChAdOx1 also induces T-cells that remain robust to B.1.351. That, coupled with animal experiments, suggests that it reduces severity of B.1.351 infection, even if it does not prevent symptoms.
If you want to stay out of the hospital, giving your immune system a preview of the virus is valuable, even if that preview isn't perfectly accurate.
Of course, one should never say never. Suppose a near-complete escape variant suddenly emerged—a SARS-CoV-2 virus that was sufficiently different (sufficiently novel) to cause severe disease in many previously-exposed or vaccinated people. Even in that unlikely nightmare scenario, we would not be back to square one. We have easily updated vaccine platforms now, designed for SARS-CoV-2. We would rush out an updated vaccine matching this new variant.
This is why talking about a "vaccine resistant virus" is misleading—I avoid that terminology. When a virus evolves to evade our immune memory—whether induced by infection or vaccination—that doesn't render the virus impossible to vaccinate against. It means that our immunity needs an update. Exposure to the new version of the virus—whether via infection or, ideally, via an updated vaccine—updates our immunity. The adaptive immune system adapts.
* * *
Remembering that novelty means severity helps us see that the vaccines provide cause for hope, even if SARS-CoV-2 manages to stay with us for years. SARS-CoV-2 might stick around; the COVID-19 pandemic will struggle to do so.
But it also makes clear that those of us in wealthy countries have two choices for how the global pandemic ends: via natural infection or via vaccination. We should choose vaccination. And we must commit to that choice now. We don't have much time.
This is why claims that we should not vaccinate children against SARS-CoV-2 because most pediatric cases are mild miss the point. Most are. Some aren't. And a small percentage of severe cases multiplied by a very large number of total cases can be a large number of severe cases.
People often worry that the vaccines will not prevent Long COVID if they fail to prevent all infection, since some people who have mild COVID-19 cases go on to develop post-infection sequelae. But the novelty-means-severity principle suggests that we should not assume mild first infections are equivalent to mild subsequent infections or mild infections of the vaccinated.
I love essays like this that provide lay intuition to complex expert topics. Thank you Zeynep and Dylan!
Thank you for an amazingly well written article. It answered so many questions.