Immunity does not disappear after COVID-19 infection!

I’m in a particularly unpleasant mood tonight because once again media reports are throwing out the idea that immunity to coronavirus is transient and won’t last long enough to protect the population. This is a recurrent theme that seems to come out almost every week, either based on some single patient whose doctor reports a patient who had a “recurrence of their COVID-19” without data. Also, the media has highlighted a number of poorly written and unpublished reports that have suggested that antibody titers, i.e. concentrations, do not persist after COVID-19 infection. 

First let me say, unequivocally, that no one survives a COVID-19 infection without an intact immune system. Without both cellular immunity that clears infected cells and antibodies that neutralize free virus no one survives this infection. In fact, many of the young people who are dying from COVID-19 have compromised immune systems, either because of immunosuppressive drugs, or cancer, or genetic problems with immunity. 

Almost all immunity does dissipate to some degree over time. The reason you get a booster shot for a vaccine like tetanus and diphtheria every 10 years is because the immunity does decline. This does not mean it disappears entirely, and even with lower antibody levels, people remain protected. The reason for this is there are memory cells in the immune system that recognize infectious agents when they reappear and provide an accelerated response to the infection.

So, if your immunity has declined a few years after an initial infection and you encounter the infection again, you may feel ill for a few days, but you then clear the infection and are unlikely to transmit it to others. This is why your immune response to some childhood diseases, like measles, may drop over time, but you never need a booster vaccine.

The same is true for COVID-19. The initial reports out of Korea that suggested people could get re-infected with COVID-19 were based on PCR assays that showed evidence of the virus at time points after the acute infection. This was later shown to be due to remnants of the virus and none of these individuals had any evidence of live virus or a true reinfection. Several manuscripts have suggested antibody titers decrease over time, but have used questionable assays or have interpreted their data in a way that enhances the hypothesis that immunity goes away. 

Another recent report has been repeatedly cited by a number of news outlets, in part because it comes from some well-respected hospitals in London. Despite this, it is again an unpublished account of only 65 individuals with COVID-19.

There are several unusual aspects to this patient group as 77.2% were male with average age of 55.2 years (range 23-95 years), and half of these patients were deathly ill patients and hospitalized for months, including patients who had been on ECMO (extra-corporeal-membrane-oxygenation; an artificial lung machine).

Having this serious a chronic and prolonged illness after COVID-19 will suppress immunity regardless of the primary disease. Importantly, the findings in these patients likely have little implication for otherwise healthy individuals who have recovered from COVID-19.

Importantly, the data from this study do not show a real decline in antibody over time. The first figure shows that all patients develop antibody and the titers persist over 60 days. The manuscript then suggests that titers drop from 75 to 100 day, but these results are based on only a few serum samples from what appears to be 5 subjects. The confidence limits of this curve fit are in grey and become so wide as to be uninterpretable in terms of a fall in titer.

The same is true for figure 2A, which shows antibodies capable of “neutralizing” the virus. Again, these antibodies appear in high titer after infection and only fall after day 75. But this fall again appears to be based on only 6 samples, and the confidence limits for these “declines” in antibody are so wide they are meaningless.

The reason for such few samples in the crucial, later time points of the study is not stated.  The authors do report “sequential serum samples were collected from individuals at time-points between 1- and 94-days post onset of symptoms (POS) and were based upon availability of discarded samples taken as part of routine clinical care, or as part of a heath care worker study (6 subjects). Thus, this was not really a prospective, timed study of antibodies titers, but just assays run on discarded blood!

In summary, this unreviewed manuscript does not show a fall in antibody after COVID-19 infection that would indicate a loss of immunity. 

Again, there is no definitive data that shows that people who have been infected with COVID-19 are at risk for reinfection.