Hacker News Reader

Comment by dnautics | Hacker News Reader

> The consensus I’m hearing from epidemiologists is the mRNA vaccines should offer better immunity than infection

Serious question: How can we possibly know this? I'm all for vaccinations (I'm going in for my second shot this weekend), but I just worry about commentary backed by intuition that doesn't carry sufficient disclaiming.

View on HN

I had this exact question and asked around before finding out a friend's brother specializes in immunology. Apparently two separate exposures triggers a much longer lasting immune response, and because of the nature of the vaccine it's much more targeted to one specific aspect of the virus, and consequently more potent in a way. Moderna also made a statement that they expected their vaccine to last for 2 years based on their ability to measure the number of antibodies and its decline over time [1].

[1] https://www.reuters.com/article/us-health-coronavirus-modern...

Yes. So long as no mutation takes place on the protein spike.

Mind you, it is a protein. So you're banking on the virus encoding a different/new protein that still binds to ACE2.

And this part I'm less confident in, but my understanding is the protein spike is unlikely to mutate because it's exactly what allows the virus to bind where it does. If that mutated and that mutation was still somehow successful, I believe it likely wouldn't be as bad a virus.

My understanding is that virologists and immunologists can't really estimate the chance of a more virulent strain that escapes antibodies analytically, because the probability space is too big. A mutation might escape antibodies but also greatly inhibit how fast it spreads, and so on.

Its been passed around on TWiV and immunologist twitter awhile back that there's ~20 different epitopes on the spike protein that would need to mutate in order to achieve immune escape.

The read on that is that immune escape should come at a cost to fitness and less virulence and transmissibility. The virus will have to "pay" with fitness in order to buy stealth.

At the same time though there have very obviously been multiple different mutations on the spike protein which have enhanced binding / viral load / transmissibility / virulence. So the virus clearly has some room to maneuver in that space.

Depends on the significance of the mutation. The vaccines have shown to be pretty good at protecting at the mutations we have observed so far. However, it is right to assume, that eventually a mutation comes around which evades any existing immunity. That is one the biggest reasons to reduce infection count globally. The higher the infection count, the larger the risk of a mutation that resets everything back to the beginning immunity-wise.

My original intuition was it would be the opposite.

This actually makes a lot of sense.

Thanks for sharing and correcting my line of thinking on it.

All the data and research I've seen so far show that they are roughly equivalent. Here are some more articles:

https://www.latimes.com/science/story/2021-02-24/coronavirus... https://www.medrxiv.org/content/10.1101/2021.02.14.21251715v

It's also consistent with other vaccines - for example, HPV vaccination confers better immune protection than a natural infection. See https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4813740/ vs. https://pubmed.ncbi.nlm.nih.gov/29029053/

They also measured significantly higher density of antibodies at the six week mark in people who had their second vaccine dose vs people who recovered from an infection.

Source?

https://www.nejm.org/doi/full/10.1056/NEJMc2032195

> They also measured significantly higher density of antibodies at the six week mark in people who had their second vaccine dose vs people who recovered from an infection.

I am not seeing evidence of this from the source you provided. It only mentions that natural infection "produces variable antibody longevity and may induce robust memory B-cell responses despite low plasma neutralizing activity."

Sure, there's mechanistic reason to believe that this is the case, but to I'd still like to see the actual data that this is the case.

From my understanding, there are two main reasons. 1) The two spaced exposures. Repeated spaced exposures build a significantly stronger response. 2) Isolation of the spike protein. Just like ML might pick up on a correlated part of the image while learning to recognize things (many examples, no good links), your immune system may learn to react to different parts of the virus. The spike protein is the most unchanging part, and all the vaccine will teach your body to attack.

> The spike protein is the most unchanging part, and all the vaccine will teach your body to attack.

Does that mean that a natural immunity may offer protection against other coronaviruses that have a different spike protein structure? The logic being that the less targeted defenses built through a natural immunity would latch onto some other commonality?

Not necessarily. They will likely try to key on something very specific, as this reduces the risk of it triggering on you (creating an autoimmune disorder). So natural infection may be less effective. We assume that mutations to the virus that hide it from the natural defenses against a random part, but have no impact on the virility, are generally more frequent and numerous than mutations to the important spike protein.

As far as I understand it is less likely for the spike protein to have a radical mutation and keep the ability to attach to cells at the same time. That's why the variants we are seeing have the same spike protein as other mutations of the virus tend to quickly disappear.

If a new variant with a different spike emerges, I guess we'll call it SARS-CoV-3 and produce a new mRNA vaccine for it.

Worse, it’s actually commentary backed by politics and virtue signalling to encourage people who have had COVID to still get vaccinated.

Reply on news.ycombinator.com