Oh, OAS- why can't I quit you?

OAS appears to emerged once again through a rather poor hamster study.

Original Antigenic Sin (OAS) has come up once again as a talking point for forever-Wuhan recognition, this time from a recent hamster study¹ which came out in late June.

OAS, in very loose terms, suggests that repeat exposure to an antigen would rely on recall of prior exposure to said antigen rather than the formation of a novel, specific response in dealing with the foreign substance. This hypothesis has been argued to have been substantiated when it comes to vaccination, in particular with flu vaccines, suggesting that vaccination with old strains of a flu virus mounts an immune response targeting that older strain more than the newer one.

As a model for how our immune systems work, OAS shouldn’t be considered an inherent negative- why would the body mount a different response to a similar antigen when recall may work just as well? After all, this is what is intended to occur with the adaptive immune system that first undergoes sensitization and then memory recall when dealing with pathogens. However, OAS takes it one step further and implicates the immune system in focusing only on old, first-ever antigen that it becomes sensitized to.

I guess we always remember our first-ever…anything, huh?

I have raised criticisms over the use of OAS because it tended to rely on misinterpretations. For instance, the first instances of OAS, for some reason, looked at UK data of nucleocapsid seroprevalence among vaccinated citizens as an indicator that OAS was occurring. This shouldn’t make sense, because these two are distinct antigens (spike vs nucleocapsid) which OAS is not modeled to examine. This tangential interpretation of OAS would suggest that your body can’t deal with a bacterial infection because it’s caught up looking at SARS-COV2 spike- we’d essentially be killed off by any other pathogen that doesn’t look like the very first thing we were ever exposed to.

This, on the surface, should have raised some eyebrows right off the bat, but OAS nonetheless continued to persist while also someone including this misinterpretation into the already questionable model. A bit ironic that staunch support of OAS itself appears to be committed to OAS as it now appears to have become an infallible model. It’s hard to argue against such a hypothesis after it has become widely adopted.

I’ll leave those interested in deeper OAS criticism to take a look at Brian Mowrey’s posts in regards to OAS (there are many, but this one in particular is part of a series looking at the lost-in-translation process that led to modern interpretations of OAS):

The Mess Francis Made

Now, the point in raising criticisms of OAS is not to outright dismiss it even as someone who has been highly critical of it, but to raise questions about what data is being looked at to make such claims, hence why this recent study from Fujita, et al. is worthy of scrutiny.

This study isn’t any different than what has come around, and in this case it’s actually a lot more reductive in it’s approach compared to other studies such as the famous immune imprinting study (which used the term immune imprinting rather than OAS).

---

The layout is rather simple- two groups of hamsters were examined. In one group, naïve hamsters were naturally infected with different variants including the initial D614G-carrying strain of the virus which is alleged to be the first one to have made it’s way to the US (I leave room open to criticisms of the first virus to hit the US) labeled as B.1.1.

Researchers also looked at infection with Omicron subvariants BA.2, BQ.1.1, and XBB.1, and strangely included a section for XBB.1.5 even though hamsters were not challenged with this variant in particular.

The second group of hamsters were given 3 doses of Moderna’s mRNA vaccine over the course of a month, followed by challenge with one of the 4 above variants.

From Fujita, et al. Figure 1A. The layout of the OAS study. Unvaccinated hamsters were challenged with one of the 4 variants and had serum collected 16 days after exposure. In vaccinated hamsters, Moderna’s vaccine was provided 3 times in 28-day intervals followed by infection with one of 4 variants. Hamsters then had serum collected at the 16 day mark.

There’s a few things to point out here, including the mismatch between hamsters assigned to variants in the unvaccinated group relative to the vaccinated group (why are there more hamsters in the B.1.1 and BA.2 group?).

It’s also important to remember that the vaccine schedule these hamsters were put on don’t exactly match the 2-dose+booster timeline that humans were exposed to, so the results may not be directly comparable.

But more importantly, and an issue that has come about in prior OAS studies, is the fact that unvaccinated animals are never exposed to another variant. That is, these sorts of studies use a group which does nothing more than to show that the animal model in question (in this case hamsters) produce an immune response to the variant they are exposed to. It doesn’t provide any refutation that OAS cannot occur through natural infection because this study is not designed to test that.

Figure 1C. Neutralization assays using serum taken from hamsters challenged with the specific variant noted at the top of the figure. No surprise, hamster antibodies neutralize pseudovirus expressing the spike antigen that they were challenged with more than other variants.

So do hamsters naturally infected with B.1.1 then challenged with BQ.1.1 show a response that primarily targets B.1.1? We don’t know, at least from this study, so this doesn’t provide us any information to suggest that OAS may not occur through natural infection. Again, this setup just tells us that hamsters respond best to the variant they were exposed to.

However, this may not stop people from interpreting the study in this way in order to argue that natural immunity is better than vaccination. Now, I should remind readers that I believe natural immunity is much better, but if I were to make that claim I wouldn’t use OAS as an explanation, and certainly not this study since this can’t provide us that information.

This is a strange predicament that has plagued OAS studies where the unvaccinated animals aren’t properly compared to vaccinated animals.

Why does this inconsistency exist?

If we look at the results for the vaccinated hamsters, we get a rather strange picture:

As expected, hamsters vaccinated but not infected show a response that predominately targets the B.1.1 variant. Note that these results would be more comparable to the unvaccinated group than the other 3 vaccinated + infected hamster groups.

But what’s strange here is the fact that vaccinated + exposed hamsters appear to show a rather strong response against BA.2 across all 3 of these hamsters. This isn’t seen with the unvaccinated hamsters, aside from the strong cross-reactivity that appeared between XBB.1 mice and XBB.1.5 pseudovirus.

So what exactly do these results tell us? Well, the results here are explained in the following way by Fujita, et al. (emphasis mine):

We then assessed whether the breakthrough infections of three Omicron subvariants, BA.2, BQ.1.1 and XBB.1, induced the neutralization activity against the variant infected, as observed in unvaccinated hamsters above. However, although anti-B.1.1 (ancestral SARS-CoV2) neutralization activity was observed, the sera obtained from breakthrough infection hamsters did not exhibit prominent antiviral effects against the variant infected (Figure 1D). Although the 50% neutralization titer (NT50) of the sera obtained from BA.2 breakthrough infection hamsters against BA.2 was significantly lower than that against B.1.1 (4.3-fold), all sera obtained from eight breakthrough infection hamsters with BA.2 exhibited anti-BA.2 activity (Figure 1D). However, the NT50 values of the sera obtained from BQ.1.1 and XBB.1 breakthrough infection hamsters against the variant infected were profoundly lower than that against B.1.1 (Figure 1D). Moreover, the NT50 values of the sera of three and five hamsters’ breakthrough infection with BQ.1.1 and XBB.1 respectively were below detection limit (Figure 1D). These observations suggest that the induction of antiviral humoral immunity against a SARS-CoV-2 variant of infection is attenuated by comparatively-ancestral vaccine inoculation prior to infection.

I am emphasizing the above sentence because it highlights again that hamsters aren’t being properly compared given that the naturally infected hamsters were never re-exposed to a different variant.

Note that the last sentence suggests that 3 hamsters in the BQ.1.1 group, as well as 5 hamsters in the XBB.1 group don’t show any detectable neutralization, which means that 4 and 3 hamsters, respectively, do.

The problem is that a lot of this language is relative. What exactly does “profoundly lower” mean? Does it mean no neutralization? The language here appears to be intentionally qualitative as it makes it harder to discern the actual effects from repeat vaccination. Note that measures of statistical significance appear to compare the variant in question to the other groups, and are not comparing neutralization titers between unvaccinated and vaccinated hamsters.

For example, neutralization titers of unvaccinated BQ.1.1. hamsters are not being compared to vaccinated BQ.1.1. hamsters, so we don’t know whether there are statistically significant differences between these groups.

---

There’s more to this study that are worth examining but I’ll leave it at that for now. There are a lot of issues that raise more questions than answers.

The main question readers may have is to ask why the results for vaccinated hamsters seem so wonky. It may be that these results are indicative of cross-reactivity, which explains the weird response towards BA.2 in particular. Again, something that OAS proponents seem to overlook is the fact that we should expect the body to rely on recall before committing a novel response. Otherwise, what would be the point of forming an adaptive immune system?

However, one caveat to this argument is the 16-day time-lapse, which is likely there to provide time for possible differentiation against the breakthrough variants. The authors may suggest that this is evidence of immune imprinting, but again without a group of hamsters who are repeatedly exposed we don’t exactly know what the response in naturally immunized hamsters should be, and whether it may show indications of OAS.

But if we were also to argue that immune imprinting may be happening, why not see if vaccinated hamsters were sick for longer than their unvaccinated counterparts, or experience worse symptoms? That would at least provide evidence that the lack of a proper antibody response is harmful to the hamsters.

It seems like not much comparison is actually done between the vaccinated and unvaccinated groups anyways. This is one of the reasons why I find it difficult to look at this study as validation of OAS occurring.

---

Suppose that one looks at this study and infers that a more virulent, more pathogenic variant is likely to come at some point by way of OAS occurring in vaccinated people.

It’s an argument that we have heard for well over a year, and yet the evidence doesn’t appear to corroborate this assumption. Arguments in support of this assumption tended to rely on weird readings of data or anecdotal evidence. If it doesn’t appear that a new variant wave has led to further deaths or illness, then not much is said, the hysteria gets swept under the rug, and will appear once again with another OAS study or a new variant emerging.

Now, could the vaccines lead to more dangerous variants in the future? Possibly since many things are possible, but an argument that something may occur due to non-zero probabilities doesn’t mean that OAS would be the cause of this occurrence, especially given that OAS doesn’t provide much predictability outside of speculation.

And yet OAS continues to live on, and serves as an example of something I have referred to as an infallible hypothesis (maybe there’s another term that people have used)- the hypothesis must be correct, and therefore information must be interpreted in a way that validates the hypothesis.

I’m not fully arguing against OAS, but am making the point that if one were to argue in favor of OAS we should hope that they study they use is one that properly creates such a scenario. This study, unfortunately, isn’t it.

---

Substack is my main source of income and all support helps to support me in my daily life. If you enjoyed this post and other works please consider supporting me through a paid Substack subscription or through my Ko-fi. Any bit helps, and it encourages independent creators and journalists such as myself to provide work outside of the mainstream narrative.

Click on the image above or visit the website at https://ko-fi.com/moderndiscontent

1. Shigeru Fujita and others, Impact of imprinted immunity induced by mRNA vaccination in an experimental animal model, The Journal of Infectious Diseases, 2023;, jiad230, https://doi.org/10.1093/infdis/jiad230