In November South Africa officially announced that a new strain of coronavirus, later named Omicron, had been discovered in the country. On November 26, the WHO convened an extraordinary meeting devoted to it. On December 8, four laboratories simultaneously confirmed the experts' worst fears: Omicron had learned to evade a significant portion of the antibodies produced after vaccination or disease caused by previous variants of SARS-CoV-2, if not all of them. Molecular biologist and science journalist, the author of The Virus That Broke the Planet Irina Yakutenko explains what this data means, whether a new pandemic awaits us or, as some hope, Omicron will prove to be a milder version of coronavirus than its predecessors.
The virus that can do it all
Omicron became famous as soon as the first decoded genomes of this variant appeared online. Virologists were drawn to its mutations. First of all, it had a lot of them: more than 50 significant changes (that is, changes in the sequence of not only the nucleotides in the gene but also the amino acids in the protein encoded by the gene), including 30 in the spike protein gene. Secondly, these mutations did not look random: scientists knew many of them from other coronavirus variants, where they performed specific functions. For example, the N501Y spike substitution was carried by the alpha, beta, and gamma variants, and, as virologists believe, it helped them better infect cells and more efficiently transmit the disease from one infected cell to another. The K417N mutation was found in the Beta variant and some Delta varieties and apparently helped them evade some of the antibodies. The P681H substitution makes the spike protein of the virus easier to be cut by the cellular enzyme furin, a step necessary for the coronavirus to enter the cell. Before that, we saw a very similar P681R mutation in the Delta and Alpha variants.
In short, «Omicron» has gathered a whole bunch of changes in its spike protein gene that help it hide better from antibodies or infect cells more efficiently. But it is not certain that all the mutations we have seen before in other strains work the same way in Omicron: the protein can be thought of as a tangled skein of wool, and each amino acid substitution slightly changes the shape of that skein, and often in some fairly distant part of it: the changed shape of the thread cluster shifts all the other parts of the skein as well. And when there are many such substitutions at once, the final configuration may not be the sum of small changes, but something completely new. In this situation, the spike protein will no longer recognize not only the antibodies targeting the places where the changes occurred, but also those that were bound to other fragments of the spike, whose sequence has stayed the same, but whose shape has changed.
Another point that distinguishes Omicron from the other varieties of SARS-CoV-2 is its place on the evolutionary tree of coronaviruses. To begin with, B.1.1.529 (as Omicron is called according to the generally accepted PANGO coronavirus classification, based precisely on the genetic similarity of variants), is not a descendant of Delta. This fact is in itself quite remarkable: in November 2021, the Delta strain was the absolute hegemon around the world, displacing, thanks to its increased contagiousness, all other variants. The most logical scenario seemed to be a further development of Delta in the direction of evading antibodies. However, Omicron grew out of a branch called B.1, which appeared much earlier than the branch leading to Delta. And the length of the Omicron branch is much greater than the average branch lengths between the ancestral and descendant variants. In other words, the Omicron variant acquired all its numerous mutations as if in a flash. But this does not happen in nature: changes always occur gradually, and a sudden abundance of changes in one genome suggests that the Omicron variant had been evolving for a long time somewhere under the radar, unnoticed by scientists.
And the most likely place of hidden evolution is in the body of a person with immunodeficiency. Such a person's immune system cannot get rid of the virus definitively, but nevertheless constantly tries to attack it. As a result, the virus multiplies for months; virus particles that are nevertheless recognized by antibodies are being destroyed, and those that manage to evade the antibodies survive. When faced with new varieties of the virus, the immune system adapts its antibodies, and the cycle of selection repeats. Over many rounds of selection, a virus can accumulate a huge number of changes - and, if such an updated virus jumps over to someone else, that variant will be completely unfamiliar to the new victim's antibodies. It means that the virus will be able to multiply unhindered, as if the new host had no antibodies against SARS-CoV-2 at all.
This is what was assumed by the scientists when they first saw the Omicron genome. But without actual experimental data, all assumptions are merely theories: even computer models are unable to reliably predict how the relationship between the Omicron spike-protein and antibodies will change with the acquisition of so many mutations. So, for the first two weeks, the media and experts tried to figure out how well B.1.1.529 evades antibodies based on circumstantial data: the rate of spread and the proportion of those vaccinated among those infected.
And finally, on December 8, four laboratories at once presented the first results of experimental testing of Omicron's resistance to antibodies from the blood of individuals vaccinated and/or exposed to other variants of the coronavirus. The first preprint (that is, an article that has not been reviewed and has not yet been published in a scientific journal) was presented by the laboratory of Alex Segal from the Africa Health Research Institute in South Africa. The scientists performed the so-called neutralization reaction: in such experiments, scientists infect cells that are sensitive to the virus with it, while simultaneously adding blood serum with antibodies. If the antibodies recognize the viral particles well, the serum will protect the cells from infection (by neutralizing the virus). Since more antibodies are needed to effectively fight the virus in the body than to neutralize it in vitro, researchers add serum with various dilution factors: when antibodies work, these dilution factors can be in the hundreds or thousands. As long as neutralization works, the greater the dilution factor, the better the immune system can be said to function.
Scientists in South Africa have shown that the effectiveness of antibodies from the serum of vaccinated people in neutralizing Omicron is 41 times less compared to Delta. For people who had had the disease before or after vaccination, the drop in efficacy was slightly less pronounced, but also very significant. That is, the antibodies produced after encountering viruses of other strains or their fragments did not react to Omicron.
The antibodies produced after encountering viruses of other strains or their fragments did not react to Omicron
On the same morning, quite similar results were posted by German experts led by Sandra Zizek, director of the Institute of Medical Virology in Frankfurt am Main. Serum from people who had received a second dose of mRNA vaccines six months earlier could not, in principle, neutralize the virus. Two weeks after the booster dose, Omicron was 11 times less effective at neutralizing the virus compared to Delta; three months after the booster, it was 37 times less effective.
In two other papers, scientists also found a decrease in the neutralization efficacy of vaccinated sera, but less than in studies from South Africa and Germany. In a preprint by biologists at Karolinska University in Stockholm, the maximum decrease factor ranged from 1 to 23, while the Pfizer paper mentioned the factor of 25. There may be several reasons for the difference. The Germans and Africans worked with a live virus, while the authors of the other two studies worked with a so-called pseudo virus. This term refers to a hybrid, which is a specially attenuated virus, with the spike-protein gene of the coronavirus inserted into its genome. The pseudo virus synthesizes the coronavirus spike protein and displays it on its surface. Using the spike, the pseudo virus can cling to the same receptors as the real SARS-CoV-2 virus and infect cells.
Pseudo virus experimental systems are very popular because the pseudo virus can be handled under normal conditions, whereas manipulation of SARS-CoV-2 is only allowed in Biosafety Class 3 (BSL-3) laboratories. However, such models do not always adequately reflect the actual processes occurring during infection. To prove that the results obtained with pseudo viruses can be transferred to the actual virus, special tests need to be performed, which have not yet been done in the case of Omicron. And in any event, if B.1.1.529 is able to infect cells more easily due to mutations not in the spike-protein but in other proteins, it will not be possible to prove it using pseudo viruses.
The second reason the results are different from one another is the time that has elapsed since the vaccination. Zizek and colleagues worked with the sera of people vaccinated with two doses six months ago, while Pfizer researchers used the sera of people who received the second vaccination three weeks ago. Obviously, in the second case, the antibody count would be higher, and therefore the neutralization would be more effective. In Pfizer's paper, volunteers received booster shots just one month before the blood tests, and, according to the press release, these sera were able to neutralize Omicron as efficiently as the serum of those who had received two shots of the vaccine was able to neutralize Delta.
Due to the different design of the experiments, we can tentatively conclude that two doses of the vaccine completely lose the ability to neutralize B.1.1.529 within the period between three weeks and three months. The booster increases the efficacy, but this effect also diminishes considerably after three months.
Another consequence of the new properties of Omicron is the pointlessness of measuring the antibody count. Even for old strains we didn't know how many we needed for good protection, nevertheless many people tried to outline the efficiency corridor. Now all this work will have to start all over again - and only after new Omicron adapted vaccines become available. At this point, many experts believe this step cannot be avoided, with manufacturers promising to release the first batches of updated vaccines by March 2022.
The results of the laboratory experiments look very alarming, but it's important to understand that a 40-fold reduction in neutralization efficiency does not equal a 40-fold reduction in vaccine effectiveness. Humans and their immune systems are much more complex than cell cultures, so you can't directly translate the results of in vitro studies to humans. Obviously, such a significant drop in neutralization efficacy will affect the actual protection afforded by vaccines, but epidemiological data must also be collected in order to assess it.
One of these necessary parameters is the dynamics of Omicron distribution. Based on the data related to the change in the number of positive tests in South Africa, experts estimate an increase in new B.1.1.529 infections of 25% per day. That is rather a low estimate, since the proportion of positive tests among all the tests performed is also growing and has already exceeded 26%. The higher this figure, the more infections we have missed: when a tracking system is properly set up, this figure should equal a few percent.
But even a 25% increase per day is comparable to the numbers we saw during the first wave, when the coronavirus began to take over a population that was completely non-immune to it. Omicron's vigor can be partly explained by its ability to evade antibodies: in terms of the immune system of vaccinated people or people who had recovered from the disease, its spike protein is a new formation, against which antibodies are powerless. On the other hand, it cannot be ruled out that Omicron itself is more contagious than Delta, but in order to prove it, we need to compare the rates of spread of the two strains in a non-immune population, which we cannot do for objective reasons.
Another way to estimate contagion requires contact tracing statistics. They are poorly set up in Africa, likewise in Europe after the summer recession, but health systems are still much better at this kind of task. And in the couple of weeks since the virus spread outside Africa, we've already seen at least one episode of super-prevalence: at a Christmas dinner in Oslo, more than 70% of a hundred guests were infected by one person. Another potential case of spread occurred in Sydney: at least 5 of the 140 guests were infected with Omicron at a cruise yacht party where only vaccinated people or people who had recovered from the disease were allowed. The rest are still under quarantine, so, probably, in the end the number of confirmed infections will be higher.
An indirect estimate of the contagiousness of B.1.1.529 can also be made in tests on animals, but it will take some time to conduct and analyze the results of such experiments.
In any case, today we can see that Omicron is not much inferior to Delta in terms of the rate of spread (although statistics are still scarce, all conclusions must be drawn with caution). And now the main question has to do with the course of the disease caused by this strain. The ability to evade antibodies does not mean that vaccines will be much worse at preventing a severe course. Antibodies represent only the first bastion of immunity defense; their role is to prevent infection of the cells. But if it has already happened, T-cells come into play, whose job is to destroy whatever is infected. Vaccines and disease stimulate the formation of not only antibodies, but also T-cells, and this branch of immunity is more resistant to mutations.
Perhaps less dangerous for the vaccinated
A huge volume of statistical information on the vaccinated shows that even after Delta arrived, protection against the severe course in the vaccinated has remained at 90%, although protection against infection has fallen. Compared to the ancestral strains, the sequences of the spike protein, which are recognized by T-cells (they respond to small linear fragments), changed by about 10% in the Alpha and Delta strains. In the case of Omicron, the possible changes are estimated to be in the range of 20-30%. That is, we can expect that T-cells developed against the previous variants of SARS-CoV-2 will retain a strong ability to act against Omicron as well.
So far, it is difficult to say definitively whether people who have T-cells against the other versions of the coronavirus actually retain protection against severe symptoms. In the early days, the number of hospitalizations in South Africa did not increase or increased very slowly, but now that number has increased substantially, as has the number of people in need of artificial respiration. The age composition is also gradually shifting: the number of elderly people in hospitals is increasing every day. This is an expected process, assuming that the pathogenesis of Omicron is similar to that of Delta: older people and those with chronic diseases are more likely to become severely ill, including sometimes after vaccination. Their bodies are not able to fight the virus properly, and even two or three doses of vaccine may not be enough to help.
Another disturbing factor is the high percentage of hospitalized young children. This cohort in Africa is virtually 100% unvaccinated, and the unusual number of children among patients with severe symptoms may indicate that Omicron may be dangerous in its pure form, without antibody and T-cell defense mechanisms. However, we have to wait for better statistics to draw unequivocal conclusions; we cannot rule out that the high number of children is an artifact of the first weeks of the new strain and the increased attention to any manifestations of the disease.
But something about Omicron can already be said unequivocally: according to the results of laboratory experiments, German virologists found that it had become invisible to both monoclonal antibodies included in the Regeneron cocktail. Monoclonal antibodies are a drug that is administered intravenously in the early stages of the disease. It consists of several ultra-high potency antibodies against coronavirus artificially synthesized in a laboratory. Monoclonal antibodies are usually administered to patients at risk who are not likely to fully develop their own immune response. They reduced the risk of bad outcomes several times over, but against Omicron, the cocktail that worked against other strains is apparently useless. The good news is that two other early intervention drugs, Merck's Molnupiravir and Pfizer's Paxlovid, are likely to remain effective because their activity is not linked to the spike protein.
Русская версия статьи