31 May 2022 Blog Post: Making Sense of Antibody Levels
As BA.12.2.1 begins to cut through a wider swath of the population, we are in clinic noticing a significant uptick in cases. This May we have had 27 new cases of COVID-19 in the practice, as opposed to only 4 in April. This has led to a number of questions about booster vaccinations and their role in neutralizing the virus – i.e. stopping its entry into cells and therefore preventing illness.
Two studies were published in 2021 based on vaccine trials which connecting neutralizing antibody responses to SARS-CoV-2 with vaccine efficacy allowing for a ‘correlate’ of protection for vaccines (and boosters) against COVID-19. We have discussed these studies in prior posts (Khoury et al. and Earle et al.).
Khoury link: Khoury, D.S. et al. Nat. Med. https://doi.org/10.1038/s41591-021-01377-8 (2021).
Earle link: Earle, K.A., et al. Vaccine https://doi.org/10.1016/j.vaccine.2021.05.063 (2021).
But before launching into the specifics, let’s take a step back. In general, testing for antibodies is used as a marker of prior infection and recovery. In some cases, such as HIV antibody testing, it can be used to identify an infection (HIV viral load can also be tested to quantify viral activity, the goal being an ‘undetectable’ viral level with treatment). Antibodies are generally long lasting but their interpretation is more complicated than it first appears. They can provide clues about the following:
1. When was the person exposed?
2. Has immunity waned or never formed?
3. Was the person exposed or vaccinated?
4. Did the person have the disease or just infection?
When antibody testing for SARS-CoV-2 is considered, two main antibody classes are considered:
1. 𝐍𝐮𝐜𝐥𝐞𝐨𝐜𝐚𝐩𝐬𝐢𝐝: The viral “coat” that packages the SARS-CoV-2 RNA genome in a protective covering. Antibodies to this protein motif indicate prior community acquired exposure and likely clinical infection with the virus. I have had rare cases of individuals who do not recall having COVID-19 but have positive nucleocapsid antibodies but this is quite unusual. These individuals likely had very mild symptoms and, therefore, went unnoticed. These antibodies do seem to be very long lived as I can still detect nucleocapsid antibodies among those with documented clinical infection as early as March 2020.
2. 𝐒𝐩𝐢𝐤𝐞: This antibody test is used primarily to detect an adaptive immune response to the SARS-CoV-2 vaccine. In some instances, this antibody level may return detectable levels as a result of community acquired infection alone (I have seen this in some patients who were unvaccinated but had the infection). However, SARS-CoV-2 vaccines preferentially target the spike protein of the virus and typically generate a robust quantitative response beyond that seen from infection alone.
Using both the nucleocapsid and spike protein antibody results can distinguish between vaccination alone, community-acquired infection alone or infection (primary or breakthrough) and vaccination together.
Adding further complexity to this picture is the fact that the spike protein antibody returns a numeric value which can be different based on the performing laboratory. LabCorp for instance returned values up to a maximum of 2500 from the test inception (Summer 2021) through early 2022. However, in 2022 the test parameters changed to accommodate values up to 25,000. Theoretically higher levels of antibody should provide “better” protection against infection but, again, there is significant nuance to the role of these neutralizing antibodies. What we do know is that they protect against severe disease, hospitalization, need for mechanical ventilation and death.
Understanding a correlate of protection and, more specifically, an ‘absolute correlate’ (meaning a protective threshold) has significant downstream implications. Firstly, new vaccine candidates (ones that can be made more cheaply and distributed more widely) would not necessarily need to go through expensive clinical trials if they could simply demonstrate that they generate a sufficient immunity threshold. Similarly, those that are immunosuppressed could also quantitatively determine if their vaccine response was sufficient for protection (and if not would be clear candidates for Evushield). Lastly, serosurveys could be undertaken to determine the proportion of the population that has achieved adequate protection.
Results for the two studies referenced above showed a significant correlation between vaccine efficacy and vaccine-induced neutralizing antibody activity. But getting at these numbers (spike protein antibody levels), is no easy task. Digging through Supplementary material, Khoury et al suggest the following levels are adequate for protection against symptomatic disease:
⊛ Pfizer: 19
⊛ Moderna: 32
⊛ J&J: 105
Earle et al. hedge their bets a bit more but suggest the following:
⊟ Pfizer: A mean value of 361 (CI: 235, 541) provides 94.6% protection
⊟ Moderna: A mean value of 360 (CI: 273, 476) provides 94.1% protection
⊟ J&J: A mean value of 224 (CI: 158, 318) provides 66.5% protection
But here’s the real challenge – these neutralization antibody studies were done during vaccine development, when alpha was the primary circulating variant. As the figure below shows, there has been a steady progression in variants – each seemingly more infectious than the last. That would, logically, mean that a level of 361 providing 94.6% protection against alpha would provide far less protection against Delta or Omicron. But how much less?
But wait – more bad news. Also, remember too that there is a well described a decline in neutralization titer with time for up to 8 months after SARS-CoV-2 infection or vaccination (perhaps even less with repeated boosters). Frequent re-testing of antibody levels is simply not practical, both at the individual and population levels.
So how do we interpret spike protein antibody levels and, more practically, how can we contextualize these levels with protection against infection? Of course, there is no actual answer but here’s my approach.
In discussing spike protein antibody levels with patients, I first “benchmark” their results against what might be expected. Although these data are not great, our expectations can be set based on the NIH “Mix and Match” study (which I have also previously discussed). After a primary series and one booster, levels taken two weeks after that second booster are in the 3,000-6,000 range (for the mRNA series which are most typical in the US). Those who have had J&J and add an mRNA booster (one shot, as recommended) can expect levels in the 2,500-3,200 range (Table below). There are no available data on levels after the 2nd booster.
In our practice, we are seeing levels anywhere from 3,000 to above 25,000 (this is the upper range). I think at this point, 5,000 should be about the minimum level we should be seeing.
And why do I think that should be the minimum? Well let’s look at the data above showing 94% protection against even mild symptomatic disease for Pfizer and Moderna at levels of 360. But Beta was about 50% more transmissible than Alpha. Delta was about twice as transmissible as Beta and Omicron has been estimated to be 5 times more transmissible than Delta. So: 1.5x2x5 = 15.
Revising upwards, 360 multiplied by a factor of 15 gives us 5,400.
This isn’t terribly scientific and, admittedly, is a decent amount of statistical hand waving. But it does attempt to draw on what we know from the vaccine trials and what we have seen in terms of transmissibility across the subsequent variants.
Happy to hear thoughts and rebuttals! Comment below!
𝗦𝗶𝗴𝗻 𝗨𝗽 𝗳𝗼𝗿 𝗢𝘂𝗿 𝗡𝗲𝘄𝘀𝗹𝗲𝘁𝘁𝗲𝗿