What we know about Omicron


By Dylan Barry, GCPPP staff, February 1st 2022

In the last two months, the world has been overrun by the Omicron variant of the coronavirus (B.1.1.529). In that time, scientists have worked frantically to try to understand what makes the new strain different from previous variants of concern—and what those differences imply for its course. Much of the evidence has been reassuring, both in comparison with previous variants and for the protection afforded by vaccines. But the impact of Omicron has nevertheless been strikingly varied around the world, even between countries with similar rates of vaccination, making any rush to conclusions still premature.

In this ANALYSIS, we summarise the best available evidence on the Omicron variant. The important issues are:

  1. What do Omicron’s origins tell us about the prospects of future variants?
  2. To what extent does Omicron evade pre-existing immunity and how good have existing vaccines proved to be against it?
  3. How transmissible is Omicron?
  4. How severe is Omicron?
  5. Why do countries’ experiences with Omicron seem so varied?

1. What do Omicron’s origins tell us?

On November 26th, the World Health Organisation (WHO) announced that the coronavirus strain B.1.1.529 would henceforth be considered a “variant of concern”—giving it the Greek name Omicron. The variant is a highly mutated strain of the coronavirus, with 60 mutations relative to the ancestral Wuhan strain of the virus.

The consensus amongst virologists is that Omicron started spreading somewhere in sub-Saharan Africa between late-September and early-October last year, based on the level of genetic diversity in early Omicron samples. As our ANALYSIS from South Africa shows, the first full genome sequences of the variant were reported in Botswana and South Africa on November 22nd and 23rd, respectively—with the earliest of the South African samples dating from November 8th. In retrospect, however, the earliest sign of a new wave of infections in South Africa came in the form of a rising concentration of SARS-CoV-2 RNA fragments in wastewater samples taken at sites outside the city of Pretoria, South Africa’s capital, on October 25th.

Nevertheless, although South Africa saw the first major Omicron outbreak, virologists consider it likely that the variant actually emerged elsewhere in sub-Saharan Africa. This reflects South Africa’s world-class genomic sequencing capabilities, which would have picked up the variant sooner if it had been circulating in that country from late-September or early-October.

That is the “when” and “where”—but what of the “how”? This matters, because the answer weighs on the issue of whether Omicron can be seen as a culmination of sorts, as an indicative phase in the virus’s overall evolution, or whether it should be seen more as stand-alone.

When trying to uncover the origins of a new virus variant, virologists start by trying to construct its evolutionary tree. Essentially, they compare the variant’s genome with the genomes of all other known variants, past and present, to see where the newcomer fits. Omicron, however, presents a puzzle. It is only distantly related to any of the other major strains like the Delta, Beta or Alpha variants. Remarkably, Omicron’s closest cousins appear to be from back in mid-2020—over a year ago. That is highly.

There are three plausible explanations. The first is that Omicron may have been diverging steadily since 2020 in a part of the world with little to no genomic sequencing, well before its recent explosive spread. There is certainly precedent for this. In early 2021, scientists in South Africa detected what was at the time the most highly mutated strain of the coronavirus in a handful of travellers from Tanzania. That variant never took off elsewhere—but it proved that highly divergent strains can emerge seemingly from nowhere in places with little to no ongoing sampling of the coronavirus.

This hypothesis is less plausible for Omicron. The variant has so many different adaptations for transmissibility and immunity-evasion that it is difficult to imagine how it would not have begun its rapid spread at an earlier point in its development. The variant is also clearly identifiable in most SARS-CoV-2 PCR tests—the tests fail to pick up one of its genes—meaning it ought to have been conspicuous even in places with limited testing capacity and no genomic sequencing capabilities.

The second explanation is that Omicron may have developed for a period of time in a secondary animal species. The premise is that a mid-2020 strain of the coronavirus might have made the jump to an unknown animal population, evolving rapidly in response to new evolutionary pressures, before making the jump back to humans. In a recent paper published in the Journal of Genetics and Genomics, a team of researchers from the Chinese Academy of Sciences in Beijing make the case for this scenario. Since Omicron diverged from its closest known cousins, it has picked up 45 so-called “point mutations”—single substitutions made to the chemical letters, known as bases, that make up its genetic code.

Prior research on similar RNA viruses suggests that the spectrum of bases most likely to pick up point mutations is related to the animal the virus is replicating in. The Beijing team argues that the spectrum of point mutations in Omicron is characteristic of virus evolution in mice. In particular, several changes to the Omicron spike protein overlap with SARS-CoV-2 mutations known to help the virus bind to the entry receptors on mouse cells. Nevertheless, most virologists think that this is an unlikely origin story for Omicron. This is because RNA viruses tend to absorb snippets of their host species’ RNA—a more damning genetic fingerprint. Omicron has no such snippets from an animal host, mouse or otherwise, but it does have an insertion of human RNA—a strong hint that the strain developed exclusively in humans.

The third, and most persuasive, explanation for Omicron’s origins is that it may have developed over the course of more than a year in a single immunocompromised patient with a chronic coronavirus infection. That such a never-ending infection is possible is well-established. Starting in March 2020, a 47-year-old lymphoma survivor with a suppressed immune system had a single coronavirus infection last for over a year—only beating the virus in April 2021. These chronic infections have proven to be a near-perfect environment for rapid coronavirus evolution. Starting in August 2020, a 45-year-old patient with severe Hughes syndrome, an immune disorder, fought a 154-day battle with SARS-CoV-2. In that time, the virus strain circulating in his body picked up 34 mutations—including 12 to the spike protein.

That stellar rate of virus evolution has been witnessed in other immunocompromised patients too—especially in those with leukaemia or lymphoma. That is thought by some virologists to be how the Alpha variant emerged in November 2020 in the United Kingdom. In sub-Saharan Africa, however, Omicron is most likely to have emerged in an HIV/AIDS patient. That is because the region plays host to the world’s worst HIV/AIDS epidemic—with a shocking 20.4% of South Africans living with the virus. Before the emergence of Omicron there was already at least one well-attested case of a 216 day SARS-CoV-2 infection leading to concerning SARS-CoV-2 evolution in a patient with advanced HIV and antiretroviral treatment failure. It is has long been suggested that the same mechanism may have been responsible for the Beta variant too.

2. Omicron versus pre-existing immunity

However Omicron may have emerged, its extensive mutation profile has given it a serious leg-up on previous variants of the coronavirus. Omicron contains over 30 mutations to its spike protein, with 15 of those mutations located on the receptor binding domain (RBD), one of the main targets of neutralising antibodies. The key question has been: how good is Omicron at escaping pre-existing immunity, whether acquired through infection or, crucially, vaccination?

The evidence so far offers both bad and good news. The bad news is that mutations do appear to be enough to flummox most of the neutralising antibodies in patients with pre-existing immunity. This is why there have been so many re-infections and breakthroughs despite vaccination. The good news is that, once infected, the more powerful immune responses by T-cells prompted by the best vaccines are proving to be effective.

The early evidence from South Africa suggests that Omicron is 3.37 times more likely than Delta to result in a breakthrough infection—which is consistent with more recent data from Denmark. In a study published in Nature, a team from Columbia University found that Omicron is remarkably resistant to neutralisation by serum (the part of blood that does not clot) taken from those previously infected with the coronavirus. The same is true of serum samples taken from individuals fully vaccinated with any of the four major Western vaccines—Pfizer/BioNTech, Moderna, Oxford/AstraZeneca, and Johnson & Johnson—with even the serum of boosted patients showing substantially diminished neutralising ability against Omicron.

Ultimately, the team from Columbia showed that 17 of 19 monoclonal antibodies tested (including ones currently approved for use in patients) either saw their neutralising ability significantly impaired or abolished altogether—a significant blow to the body’s defences against the coronavirus. That means that almost all monoclonal antibody treatments against the coronavirus have been rendered ineffectual by Omicron—narrowing the range of tools available to medical professionals in treating serious cases of COVID-19.

This evidence is consistent with recent research focused specifically on vaccine efficacy. In a laboratory study performed by Pfizer/BioNTech, the Pfizer/BioNTech vaccine saw its efficacy (measured in antibody titres) reduced 25-fold against Omicron, relative to the original wild strain of the coronavirus. Fortunately, a single booster dose was enough to raise immunity against Omicron back up to comparable levels. This set of results has since been backed up by findings from a study from Harvard University—which confirmed the same for the Moderna vaccine. Three new studies by the Centres for Disease Control and Prevention (CDC) in the United States suggest that the receipt of either mRNA booster is 90% effective at preventing hospitalisation due to infection with Omicron. Nevertheless, even a fourth dose of an mRNA vaccine appears not to be enough to stop significant numbers of Omicron breakthrough infections, according to data from Sheba Medical Centre in Israel.

The behaviour of the Oxford/AstraZeneca and Johnson & Johnson vaccines is similar. In an independent studyfrom the University of Oxford, the Oxford/AstraZeneca vaccine saw its efficacy (measured in antibody titres) reduced 34.3-fold against Omicron, relative to the original strain of the virus. That was improved to a 12.7-fold reduction by a third dose of the vaccine—comparable to the performance of the Pfizer/BioNTech booster in the same study. There is sparser data for the Johnson & Johnson vaccine, but results from the ongoing Sisonke II trial in South Africa suggests that a Johnson & Johnson booster dose is 85% effective at preventing hospitalisation due to infection with Omicron.

Those are the major Western vaccines—but what of the non-Western vaccines? In particular, how do the major inactivated-virus vaccines—Sinovac, Sinopharm and Bharat Biotech—fair against Omicron? Not well, apparently. In a study at Shanghai Jiao Tong University in China, for example, researchers identified neutralizing antibodies against Omicron in only 8 out of 292 people tested after two doses of the Sinopharm vaccine. That number rose to 228 after a booster dose, however. The same appears to be true of the Sinovac and Bharat Biotech vaccines, based on a handful of smaller studies. Worryingly, in one study the Sinovac vaccine produced no neutralising antibodies against Omicron at all—a result Sinovac disputes.

How much worse inactivated-virus vaccines perform than the major Western vaccines remains unclear—but the answer appears to be at least somewhat. If it turns out to be significantly worse, that will be worrying. On their own, China’s Sinovac and Sinopharm vaccines account for nearly 5 billion of the more than 11 billion vaccine doses delivered worldwide so far. Those vaccines have played an especially important role in inoculating large swathes of the Middle East, Central Asia, the Asia Pacific region and Latin America—as well as many countries in Africa. The five largest beneficiaries of Chinese vaccine exports have been Indonesia, Iran, Pakistan, Brazil and the Philippines.

The gist of all this is that prior neutralising antibody immunity does not go a long way against Omicron, at least without a booster. On the bright side, however, it appears that T-cells induced by prior infection or vaccination continue to retain close to their full potency against Omicron. That is excellent news because T-cells constitute the immune system’s heavy artillery, offering strong protection against severe SARS-CoV-2 infection.Nevertheless, because the body’s T-cell response takes longer to get up and running than its antibody response, Omicron has resulted in greater numbers of both breakthrough infections and reinfections with the virus. You’re likelier to get infected by Omicron, therefore, but not very likely to have to be hospitalised or die.

3. How transmissible is Omicron?

The remarkably rapid spread of Omicron across the globe appears, at least superficially, to be consistent with it being a variant that is significantly more infectious than any before it. But the truth is not quite so cut and dried. This is because disentangling what proportion of Omicron’s rapid spread is due to its (potentially) elevated transmissibility and what proportion is due simply to its ability to infect those with pre-existing immunity (and hence more of the people it comes into contact with) is a near-impossible statistical feat to perform with real-world data.

Nevertheless, it does appear that Omicron is at least somewhat more infectious than the Delta variant, the previous infectiousness champion. There are several lines of evidence for this. In a study in mid-December, scientists from the University of Hong Kong found that Omicron replicates roughly 70 times faster in upper respiratory tract tissue than the Delta variant—suggesting that it may require lower exposure to the virus for an infection to establish itself. That dramatic rate of upper respiratory tract replication may also lead to greater viral shedding into the environment when infected patients breathe, cough or sneeze—meaning that any contact with an infected patient is more likely to result in an infection.

The combination of those two effects may result in substantially greater transmissibility. That is certainly backed up by anecdotal evidence of Omicron’s spread in hotel quarantine facilities. The aforementioned results are also consistent with the findings of a study that used pseudovirus particles fitted with Omicron spike proteins to investigate differences between Omicron and Delta. That study found that the Omicron spike protein was more effective at entering human cells than any previous variant—suggesting that Omicron is four times more infectious than the original coronavirus strain and twice as infectious as the Delta variant. In a recent study, researchers in France also estimated that Omicron is roughly twice as transmissible as the Delta variant.

4. How severe is Omicron?

While Omicron appears to be both better at evading the human immune system and more easily spread, it is also proving to be less virulent—presenting less severe disease than previous variants. The clearest evidence of this comes from hospitalisation rates. The early evidence from South Africa suggested those infected with Omicron were only 20% as likely to be admitted to hospital as those infected with the Delta variant—a change mirrored in other key metrics of severity.

Interestingly, given low South African vaccination rates,  Omicron has not proven quite so mild elsewhere. Nevertheless, the United Kingdom’s Health Security Agency still estimates that those infected with Omicron in the UK are only about a third as likely t­o be admitted to a hospital as those with Delta were, with similar results emerging from an Ontario Public Health study in Canada. The same observation lay behind Denmark's decision to remove all restrictions – masks, vaccine passes and more – on February 1st. At that point, Denmark had one of the highest level of new cases per million population, but also the highest level of booster vaccinations (more than 60% of the population), leading to low and falling ICU admissions and a high level of public confidence in the decision.

As Denmark’s experience shows, the severity of infection with Omicron is significantly reduced by vaccination. The United Kingdom’s Health Security Agency estimates that one vaccine dose reduces the risk of hospitalisation with Omicron by 52%, two doses by 72%, and a booster dose by 88%.

That helps to explain Omicron’s unusual patient profile. In several countries, it has been associated with higher relative levels of hospitalisation among youth and children—the demographics least likely to be vaccinated. In South Africa—where only a quarter of the population is fully vaccinated—the relative rate of hospitalisation amongst individuals with no co-morbidities appears to have been high for the same reason.

As well as the strong T-cell immunity response generated by mRNA and adenovirus vaccines, the fact that Omicron results in less serious disease also appears to be because it replicates less easily in lung tissue than previous variants of the coronavirus. This is the finding of two important studies. The University of Hong Kong study mentioned above found that Omicron replicates 10 times more slowly in human lung tissue than the Delta variant. That finding is consistent with a separate study in which scientists at Washington University infected hamsters and mice with Omicron. After a few days, the researchers found that the concentration of virus particles in the animals’ lungs was 10 times lower than the concentration in rodents infected with previous variants.

5. Why do countries’ experience with Omicron seem so varied?

Finally, while the experience of Omicron in different countries has had some consistent elements, it has also been markedly varied. The wave of infections in South Africa due to Omicron took off spectacularly but stopped almost as quickly—with peak daily case numbers topping out at a level only a little higher than in the preceding Delta wave. The same appears to have been true across Southern Africa. By contrast, in many other places daily case rate figures due to Omicron have shot through the roof. In most other regions, peak Omicron infections have more than tripled the previous peak—usually set by Delta.

Nevertheless, in South Africa, mainland Europe, and the United Kingdom, daily hospitalisation rates due to Omicron have been relatively subdued—not reaching peaks set in previous waves. By contrast, the United States and Canada have seen its hospitalisation rates soar to new heights—bucking the trend. When it comes to coronavirus deaths, the distribution of outcomes is also puzzling, with South Africa seeing far fewer deaths due to Omicron than due to Delta. The opposite has been true in Europe, while the United States has seen similar peak rates of mortality in both waves.

What might explain this divergence of outcomes? The truth is we simply do not know conclusively, but there are some key variables that might be making a difference. South Africa , for example, has been especially hard hit by the pandemic, producing unusually high levels of immunity from previous infection. In particular, South Africa’s exposure to the Beta variant—previously the most vaccine resistant strain, alongside Gamma—might have given residents an advantage on Omicron, reducing hospitalisations and deaths.

The biggest single variable, however, appears to be vaccination, especially in the case of large countries with a wide regional variety in levels of vaccine take-up, both of primary dosing regimes and of boosters.

The United States has a particularly wide range of levels of vaccine take-up, as this chart from the New York Times shows, as well as a low average level of full vaccination, at 63.5%, which can be compared with levels of 75-85% in most European countries. Canada has a higher average level of full vaccination, at 78%, but the booster programme as been slow and varied, with booster rates averaging 37% on January 31st, with provincial rates ranging between 27% and 40%. This compares with over 60% in Denmark and the UK.

These explanations, of course, are speculative. The truth is likely to involve interactions between several important moving parts.

Looking forward

The world’s virologists have learned a lot about the Omicron variant in the last two months. Nevertheless, there is still much we do not know. Most obviously, scientists do not yet have a good sense of whether it results in more or fewer cases of long-COVID, although some of the early data is encouraging. Moreover, sub-variants are emerging that may produce different outcomes. At some point, the move from pandemic to endemic will come as succeeding variants prove milder. However we cannot conclude that Omicron's relative mildness means we are yet entering that phase.  If the pandemic has taught anything it is to expect the unexpected.

We now publish a weekly newsletter to inform friends and supporters of the Global Commission’s progress and to provide updates when new content is published. Please sign up here:

Previous Analysis