Many People Lack Protective Antibodies After COVID-19 Infection

This article appeared this week on Medscape. Author F. Perry Wilson interprets the technical data in a very breezy non-intimidating lay-term manner — so that many of us who are not specialists in immunology can understand.


F. Perry Wilson, MD, MSCE
June 24, 2020
Welcome to Impact Factor, your weekly dose of commentary on a new medical study. I’m Dr. F. Perry Wilson.

In what seems like 10 years ago, but was actually just 6 weeks ago, on this very website, I said this:

“This is the COVID that allows us to open up more quickly, assuming that antibodies are protective, which — let’s be honest — if they aren’t, we’re sort of screwed no matter what.”

Cut to a couple of days ago, when I came across this article in Nature — the first deep dive, attempting to answer the question of just how protective those coronavirus antibodies are.

And, at first blush at least, the news isn’t great.

Researchers recruited patients who had recovered from COVID-19 from the Rockefeller University Hospital in New York. The 111 individuals enrolled had to have been asymptomatic for at least 14 days. They also recruited 46 asymptomatic household contacts and some controls who had never had COVID-19.

Now, a brief refresher on antibodies. There are several different types, but we broadly think about immunoglobulin M (IgM) as the short-term antibody, generated in the throes of the illness, and immunoglobulin G (IgG) as the long-term antibody. But here’s the thing: The mere presence of antibodies does not mean that those antibodies are protective. The researchers tease this apart for us.

Source: Wikimedia Commons

They zeroed in on two types of anti-coronavirus antibodies: a group that binds to the spike protein (that’s the crown part of the corona), and more specifically, antibodies that bind to the receptor binding domain of the spike protein. This is the key, if you will, that opens the door of your cells (a receptor called ACE2) to infection. It’s a good bet that if there is an antibody that will shut down the virus, it’s one that will block the receptor binding domain.

Should we start with the good news?

Source: Robbiani DF, et al. Nature. Epub 18 June 2020.

Compared with controls, IgG and IgM levels were higher among those who had recovered from COVID-19. As expected in this convalescent group, a bigger difference was seen in IgG (the long-term antibody) compared with IgM. You can see in this graph that IgM levels seem to go down a bit over time.

Source: Robbiani DF, et al. Nature. Epub 18 June 2020.

And, I’ll note, about 20%-30% of people didn’t have antibody titers significantly above controls. But broadly, okay — the majority of people made antibodies.

But that’s not the key thing here. Were these neutralizing antibodies? Do they stop viral replication?

To figure this out, the researchers genetically engineered a SARS-CoV-2 pseudovirus which expressed the spike protein and let it run amok infecting ACE2-expressing cells in culture.

Source: Robbiani DF, et al. Nature. Epub 18 June 2020.

They then added varying dilutions of patient plasma to the petri dishes to determine how much plasma you would need to shut the virus down by 50%, the so-called “neutralizing titer” 50 (NT50).

The results here were not so encouraging.

Thirty-three percent of the individuals tested had an NT50 of less than 50, which implies essentially no immunity to repeat infection; 79% had an NT50 less than 1000 — they may have partial immunity. Only two people tested had an NT50 greater than 5000.

Higher overall antibody titers were associated with neutralizing ability, as might be expected.

Source: Robbiani DF, et al. Nature. Epub 18 June 2020.

Individuals who had been hospitalized for COVID-19 were more likely to have neutralizing antibodies than those who hadn’t been hospitalized, suggesting that those with more severe illness are more likely to be immune in the future.

Overall, this is fairly concerning. Without neutralizing antibodies, an end to coronavirus transmission seems unlikely. But let’s also remember the empirical data: We don’t yet have any significant numbers of individuals who have been documented to have cleared COVID-19 and then become re-infected. And even without high levels of neutralizing antibodies, a second infection is likely not to be as bad as the first.

There’s another nugget of hope in this study. The researchers didn’t stop by simply measuring how many people had neutralizing antibodies. They actually sequenced 89 different anti-COVID antibodies to determine which specific antibodies were highly neutralizing. They identified 52 that had neutralizing ability and several that had potent neutralizing ability, targeted to specific amino acids on the receptor binding domain.

And here’s the thing: Most of the people in the study had those highly neutralizing antibodies; they just weren’t the main antibodies they were producing. Why is this good news? Because it suggests a pathway for a successful vaccine. We can make these potent neutralizing antibodies; it’s just that many of us don’t. But a vaccine designed to promote that particular antibody response could be highly successful.

All in all, this was a study that suggested that the tunnel we are in now may be a bit longer than we had hoped, but it also shows perhaps a light at the end of the tunnel.

F. Perry Wilson, MD, MSCE, is an associate professor of medicine and director of Yale’s Program of Applied Translational Research. His science communication work can be found in the Huffington Post, on NPR, and here on Medscape

COMMENT: I have a problem with conclusions in this article. First and foremost, the conclusion that “without neutralizing antibodies, an end to Coronavirus transmission seems unlikely”. Pouring dilutions of serum containing neutralizing antibodies onto COVID infected cell cultures in a Petri dish —is not an adequate model for the many ways by which a human is able to fight viral infections.

In the case of a respiratory virus, local mucosal IgA first appears — and then leukocytes and cytokines become available in the area, and then recruitment of cytotoxic T cells, natural killer cells and helper cells, and antiviral macrophages — all play a role in eliminating virus-infected cells. So, to imply that “the lack of sufficient neutralizing serum antibodies means you will get the infection again”, or that “vaccinations work only by that mechanism” — seems misleading to me.

COMMENT: Olga, Judy: In the field of pharmacology, where some political groups have supported (for decades) to stop using animals in drug studies (“because everything can be studied just as easily in a dish of cells in culture, or in silico”), these GEITP pages consider this approach to be naïve. When a drug is administered — it undergoes absorption, distribution, metabolism, and excretion (ADME). These are all complex processes (among many different organs and tissue and cell types) — that there is simply no way that one can duplicate in cell culture, or even embryoid bodies, what happens in the intact animal.

In the field of immunology and viral inflammation, these processes must be at least as complicated. Studying a phenomenon in a dish of cells or in vitro (i.e. in a flask or test tube) usually has little relevance to the intact animal or clinically to the patient.

Hi Dan — I agree with Olga that this article on Medscape is “beyond misleading” — when this is considered:


COMMENT: The Medscape article, and the research about which that article is summarizing, are not “wrong” — but rather simply naïve in only examining/explaining “the antibody function” as a stand-alone description of the immune response against a virus. For example, T cells also develop memory to the proteins presented by the infected cells and T cells can kill cells directly without antibody. However, these cytotoxic T cells are notoriously difficult to study/assess — without knowing the specific peptide presented. Furthermore, the study was definitely not high-throughput, with regard to searching for a wide range of antibody responses. —Michael

This entry was posted in Center for Environmental Genetics. Bookmark the permalink.