Original Antigenic Sin, Antibody-Dependent Enhancement, and COVID-19's Future

Original antigenic sin(OAS) has not yet been demonstrated in SARS-CoV-2. However, I strongly believe that OAS will manifest itself, and it will be a bad and worsening problem over time. If I am correct, it may mean that getting vaccinated will become actively dangerous, depending on antibody-dependent enhancement(ADE) and its degree.

OAS occurs when an immune response is produced by the body to a closely related pathogen or vaccine(such as a different strain of a virus) instead of generation of a novel immune response. This is because of homology between the two intruders' epitopes and resulting cross-reactive immune activity. SARS-CoV-2 has clearly demonstrated induction of variably cross-reactive antibodies with strain-variable neutralization potential.

It has been demonstrated in many viruses ranging from Dengue Fever to influenza to HIV-1. The set of viruses is not arbitrary: OAS abuse is a particularly effective strategy in viruses that mutate quickly, like influenza, HIV-1, and potentially SARS-CoV-2.

It is a relic of immunological memory, the very thing vaccines and prior infection use to protect you, being outpaced by viral evolution. Its significance in SARS-CoV-2 depends on many factors discussed below. I believe it has not yet been observed because, until recently, SARS-CoV-2 simply didn't have enough genetic diversity to exhibit OAS. It hadn't had enough time to mutate.

OAS is probable

Here, I disagree with SAGE.

Although rare, reinfections of SARS-CoV-2 are more severe than novel infections. This indicates that SARS-CoV-2 has more evolutionary work to do to get around antibodies and T-cells better, but that the full OAS framework is already in place, so making it more serious should not be too hard. The alternative explanation would be simply lingering damage from the prior infection.

ADE and OAS are entirely separate phenomena, and either one can present itself on its own. However, they are synergistic, and thus sometimes found together in the same virus. I believe that COVID-19's ADE is another indication that OAS is also at work, much like in Dengue Fever.

Dengue Fever has the most notorious OAS, which was (also) unanticipated. This ended in tragedy for so many young people that researchers were jailed in response to the development and administration of suboptimal vaccines.

Dengue Fever accomplishes ADE through afuscolation of IgG antibodies and promoting FcγRIIIa expression, which makes the disease more severe by increasing antibody-dependent cellular cytoxicity. This causes worse cytokine storms overall, a significant cause of death from COVID-19.

https://www.cell.com/cell-reports/pdf/S2211-1247(20)30595-7.pdf

High neutralizing antibody titers prevent ADE in Dengue Fever, and may do the same in SARS-CoV-2. That would be a good sign if we can maintain consistently high neutralizing titers across the population, but rapid evolution and variable and rapidly decreasing antibody titers in response to vaccination make this unlikely to occur widely.

COVID-19 exhibits similar afuscolation and changes in Fcγ functionality, and while early evidence indicated that this was coincidental, not causal, more recent findings indicate that afuscolation increases with disease severity. This implies that SARS-CoV-2 is itself deliberately triggering this, which would be consistent with the large potential evolutionary advantage OAS can confer, although it does not rule out other causes.

Some coronaviruses have displayed significant OAS. Perhaps the most notable example is FCoV, an alphacoronavirus, where vaccination usually causes worse clinical outcomes. Curiously, this was only observed in matching serotypes, which is at odds with the expected pattern in Dengue Fever. It's reasonable to infer that this is because type II FCoV antibodies are unlikely to be highly cross-reactive with type I FCoV, as they utilize different entry receptors.

SARS, the original, also exhibits ADE and infects monocytes, but it never had the chance to develop multiple serotypes. SARS-CoV-2 also infects monocytes. It uses a different entry mechanism than ACE2, which is quite feasible given the number and importance of integrins and the example accessory protein ORF7a, with LFA-1 implicated in that case.

OAS' observable debut would be delayed by our current vaccines, which often induce very high antibody titers in younger people, and future vaccines could be made even more immunogenic. This comes at a cost in terms of side effects, so it can't be simply ratcheted up indefinitely over the long term.

OAS can be a good thing or a bad thing. The body will want to make it a good thing by conserving and quickly producing neutralizing antibodies and T-cells; the virus will want to make it a bad thing by making the antibodies non-neutralizing and T-cells dysfunctional. Thus, the long-term success of vaccination would depend on how quickly these cells can be augmented or replaced and how fast the virus changes its immunodominant epitopes. Because reinfections are worse, the antibody modification behavior exhibited by SARS-CoV-2 is similar to that of Dengue Fever, FCoV shows OAS exists in coronaviruses, and SARS-CoV-2 is rapidly evolving new serotypes, I suspect OAS will be bad.

It is a race between how quickly SARS-CoV-2 can evolve and how fast the immune system can reprogram itself and how well it can recycle existing immunity, so it's not like OAS is a light switch. It's more like a knob that the virus will be rotating one way(bad outcomes) and our responses -- existing immune responses or successful revaccination -- will be rotating the other way(good outcomes).

The evolutionary rate of SARS-CoV-2 is extraordinary because of its heavy use of recombination and high global viral loads. Current vaccines already don't do much to lower the pace of evolution because they do little to slow transmission and mutation, particularly by variants of concern, which are the dominant strains and most likely progenitors of future strains.

The most potent SARS-CoV-2 memory B cells display a low rate of somatic hypermutation. This is likely to be an important issue. They are the ones most likely to produce antibodies that emerge successfully from germinal centers because of higher perceived efficacy leading to a greater likelihood of release, so they dominate the immune response twofold. Since they evolve slowly, SARS-CoV-2 finds a relatively easy competitor in the race against natural immunity. Vaccines are a relatively harder competitor more for other reasons.

Because SARS-CoV-2 evolves with such aplomb and anti-S memory B cells and T-cells are long-lived, I strongly believe the virus will turn the knob much harder, and OAS will be found to be significant.

Now that there are multiple serotypes emerging with cross-reactive but differentially neutralizing antibodies, we will have the chance to observe the extent to which OAS is a factor in SARS-CoV-2 and check my hypothesis.

How bad could it be?

Pretty bad; bad enough that SAGE described it as a worst-case scenario. But just as things can go from bad to worse, there are different degrees of worst. Let's dive in.

First, we need to note that most of humanity has probably been immunologically imprinted by SARS-CoV-2 in some shape or form today by either infection or vaccination. Those imprints can be shallow or deep. Because the virus probably evolves faster than our pharmacological and immune responses, shallow would be preferable.

The potential scope of the problem has been made larger by our vaccination drive. Along with being much greater in number, assuming our surveillance is even half-adequate, vaccinees have all been imprinted by very similar S proteins. This is in some contrast to natural infection, where there would be more variability in immune response because people would have been infected by different strains. This will help the virus to hone its evolution on a singular target, giving it an even stronger grip on the knob.

The best case would be that cross-reactive immune activity remains protective, as it does today, making OAS a good thing due to dominance of the pro-immune factors enumerated above. I consider this unlikely.

The worse case would be that cross-reactive immune activity provides little protection due to successful antigenic drift or evolution that doesn't go so far as to make the immune response non-cross-reactive, but does go far enough to make it non-neutralizing. In this case, the virus would face little resistance from the immune system.

The worst case would be if SARS-CoV-2 also exhibits or develops substantial antibody-dependent enhancement(ADE) or gets better at infecting T-cells. This is plausible since SARS-CoV-2 can infect immune cells through the Fcγ complex; in fact, it even intentionally attracts T-cells to infected lung cells.

Ultimately, OAS will likely ending up being a problem for current vaccine recipients, but only a modest one at first. We probably have enough overhead to increase the number of strains targeted and the immunogenicity of even the mRNA vaccines, but it's limited. It's likely to end up being a substantial problem for all of us in the long term. This adds to the pile of science suggesting eradication is impossible and implies the virus will be largely free to dictate its own terms in the end as our ability to vaccinate wanes.