COVID Vaccines: Letter to a Friend

An old friend recently contacted me with some questions about the COVID vaccines. I spent enough time composing my response that I thought I would post it here in case anyone finds it useful. Below is a lightly edited copy of my emailed reply.

---------------------------------------

These are good questions. Instead of taking them point by point, I think it makes sense to address them in the course of a longer discussion. (Probably too long, but I can’t help it.) First, I should state that I am only giving general information and not specific medical advice. Ultimately, I would follow whatever your kids' doctors advise.

Messenger RNA (mRNA) is an intermediary between the genome and the cellular machinery that makes proteins. It essentially tells cells which amino acids to link together in order to make a protein. All proteins in the body (specifically in cells) are made using mRNA. When a virus infects a cell, it starts making its own viral mRNAs, which causes the cell to start making viral proteins. The spike protein is one of a number of proteins that the coronavirus makes. The spike protein is important because it is responsible for attachment to cells, and then entry into the cells. From an immunology and vaccine perspective, it is the most important protein because if you can make antibodies that bind to it in the right spot, the spike protein is unable to attach to or enter new cells. Thus, the virus is neutralized.

Right now there are two basic classes of approved coronavirus vaccines.

1. Moderna and Pfizer use mRNA technology. They take chemically synthesized mRNA (coding for spike) and formulate it with some lipids and cholesterol to form a nanoparticle that can be taken up by cells. mRNA is easily degraded, which is why it can't be simply injected by itself. It needs that lipid nanoparticle to protect it until it is delivered into cells. Cells have various mechanisms of sensing RNA that is out of place, so to speak, so the vaccine mRNA has slight chemical modifications to help avoid alerting cells to its presence. This is done in order to help maximize the amount of protein produced; otherwise the cell would try to shut down protein production, which would negatively impact the ability to generate a good immune response. Currently, two doses are considered fully vaccinated.

2. Johnson & Johnson and AstraZeneca (Europe) take a different approach. They use a different kind of virus called adenovirus as a delivery vehicle (a 'vector'). (They actually use different adenoviruses, but the principles are the same.) The adenovirus is modified in two basic ways. First, it is missing some key parts of its genome such that it can only replicate in special laboratory cells, but not in people. Second, the gene for the coronavirus spike protein has been added into its genome. When the adenovirus is injected, it can "infect" cells, but it can't make copies of itself. However, it does make spike mRNAs, which are then turned into the spike protein by the cell. Currently one dose is considered fully vaccinated.

Aside from a small mutation in the spike protein that helps keep it in the optimal shape for an immune response, there is no difference between the spike protein made by the vaccine vs spike made by the virus. The production of spike by both vaccines is transient. Ultimately, the mRNA and/or adenovirus are degraded, as is the spike protein that was produced. How long does the spike protein stick around? One study of some people who got the Moderna vaccine detected spike in blood for about 7-10 days after the first vaccination, and then barely at all after the second vaccination--presumably because the immune system was poised to deal with it quickly. I don't know whether comparable data has been generated for the adenovirus vaccines, but it would probably be something similar (but only one dose). There could be some trace amounts retained in lymph nodes for longer, which is normal and allows further honing of antibodies, but for all practical purposes I think it's safe to say that vaccine-produced spike is gone by 2-3 weeks. There is nothing about the vaccines that would make spike a permanent part of your body any more than natural infection would.

Any new vaccine or drug is tested in animals (if possible) and then gradually introduced into humans. First, 10s of people are tested for safety. Then a few hundred or thousand are tested for efficacy. Any obvious problems should be apparent at this point. However, in a genetically and biologically diverse population, it is impossible to predict whether more rare problems will arise. This is further complicated by the fact that weird and unexpected medical events happen every day. Sorting out coincidences from causal connections is a challenge, and the more rare the adverse event, the larger the population is needed in order to reach statistical certainty. Even when there is statistical certainty, the adverse events need to be weighed against the benefits.

I'm not intimately familiar with the FDA licensing process, but several factors came together that enabled the quick development of the COVID vaccines. First, the delivery systems (mRNA or adenovirus) already had a lot of research behind them. Second, previous coronavirus research made the choice of targeting the spike protein a natural one. Third, the expanding pandemic made it easier to generate the needed data to compare vaccinated vs non-vaccinated. It takes a lot longer to do this when a disease is less common. Finally, the federal government took on a lot of the financial risk. Ordinarily vaccine companies would wait until near the end of the process to invest in manufacturing facilities. In this case, the federal government essentially said, "Go ahead and start preparing for manufacturing while collecting the data from the early phases of evaluation. If it turns out that the vaccine fails safety or efficacy standards, we will cover the cost of having prematurely invested in manufacturing."

There are a few garden-variety side effects that often occur with vaccines. Things like fevers, aches, and so forth--basic flu-like symptoms that are really just automatic parts of an immune response. Other side effects become apparent over time and with larger populations. The CDC and FDA monitor these other side effects and provide updates on the most important ones here: LINK

As you noted, the adenovirus-based vaccines (J&J in the U.S.; Astrazeneca in Europe) have been associated with a rare clotting issue, resulting in a pause in their use earlier this year in both Europe and the U.S. I think the exact mechanism behind this isn't known for sure, but it doesn't seem to be as simple as the presence of the spike protein because the same thing has not been seen with the mRNA vaccines. Clotting issues of this type need to be treated differently than other clotting issues. Part of the reason for the pause was so that the word could get out to physicians that they needed to be aware of this association so that they did not inadvertently use the wrong treatment. The J&J vaccine has also been associated with a rare nerve issue.

The mRNA vaccines (Pfizer and Moderna), on the other hand, have been associated with rare cases of myocarditis, especially in young men after the second dose. The majority of these rare cases have resolved with minimal treatment. It is also worth noting that COVID infection itself can also cause myocarditis.

The duration and effectiveness of natural immunity vs vaccine immunity is still a developing story. It does appear that people with natural immunity have good protection against re-infection. Some studies suggest that they are better protected against re-infection than vaccinated people, and there are good immunological reasons to think that would be the case. However, studies also indicate that natural immunity can be further improved with one dose of vaccine, including better ability to neutralize variants. So as a generalization, I would say that when it comes to re-infection, natural immunity + 1 dose of vaccine is superior to either natural immunity alone or fully vaccinated without prior infection. (For people who have been previously infected, the second dose doesn't seem to add much additional benefit. On the other hand, keeping it at 2 doses (mRNA only) helps to ensure that all people are fully vaccinated.) To be clear, that doesn't mean that natural immunity is preferred, since that implies getting infected and likely having disease, something we want to avoid. It is impossible to predict what future variants may come our way. However, it looks increasingly likely that they will be sub-variants of the current delta variant. Presumably, future booster shots will incorporate updated variant spikes.

But let's come back to natural immunity for a moment. Like most things biological, there is a spectrum. Some people develop fantastic immunity, and others develop mediocre immunity, which is borne out by documented cases of re-infection. Although antibodies are only one part of the immune response, studies suggest that antibody levels correlate pretty well with protection. As far as I know, current commerically available antibody tests cannot be interpreted to indicate the level of antibodies or level of protection. They only indicate that antibodies are present. (The FDA warned about this in May.) If I had to guess where someone who never had symptoms falls on the spectrum of natural immunity, my guess would be on the weak, less durable side. My reasoning is that if the virus didn't cause symptoms, then it probably didn't replicate enough to really get the attention of the immune system. (The scientific literature is a little bit conflicted on this point, but some studies support my view.) Unfortunately, absent a more robust test designed to assess the levels and/or neutralizing capacity of antibodies, there's no way to know where any individual stands. This is part of the reason why people who have recovered from COVID are still encouraged to be vaccinated.

I take your point about long-term safety. (Although I would note that the long-term consequences of COVID aren't known yet either.) We obviously can’t see into the future, but I think this framework might help. Virtually all of the adverse events—even the rare ones—that have been observed have occurred within days to a couple of weeks of vaccination. That is the time when vaccine components are present, spike is being made, and the immune system is most active. So let’s call these acute adverse events. As I explained earlier, the vaccine and spike protein are gone within 2-3 weeks. After that point, any adverse events would have to either be leftover consequences of the acute adverse events, or some aberrant immune activity—like maybe the vaccine provoked the immune system to start attacking some part of the body (autoimmunity). With the possible exception of rare cases of Guillain-Barré Syndrome, I’m not aware of any such thing having been observed. There’s a sense in which anything is possible in nature, but it is hard for me to imagine how there could be any long-term adverse events—not connected to the acute events--that would spontaneously arise months/years after vaccination. Moreover, inasmuch as virus infection and vaccination both stimulate an immune response against the spike protein, it is hard for me to imagine how such a long-term adverse event could be specific to vaccination. That’s not a guarantee of long-term safety, but I think we should have every expectation of long-term safety for the vast majority of vaccinates.

I think I have taxed your patience enough for now. I hope this helps. I am happy to answer any follow-up questions, provide references, or explore other sources of information as you may desire.

Jared