Where is pi?

Last year, the World Health Organization began assigning Greek letters to worrying new variants of the coronavirus. The organization started with alpha and swiftly worked its way through the Greek alphabet in the months that followed. When omicron arrived in November, it was the 13th named variant in less than a year.

But 10 months have passed since omicron's debut, and the next letter in line, pi, has yet to arrive.

That does not mean SARS-CoV-2, the coronavirus that causes COVID-19, has stopped evolving. But it may have entered a new stage. Last year, more than a dozen ordinary viruses independently transformed into major new public health threats. But now, all of the virus's most significant variations are descending from a single lineage: omicron.

"Based on what's being detected at the moment, it's looking like future SARS-CoV-2 will evolve from omicron," said David Robertson, a virus expert at the University of Glasgow.

It's also looking like omicron has a remarkable capacity for more evolution. One of the newest subvariants, called BA.2.75.2, can evade immune responses better than all earlier forms of omicron.

For now, BA.2.75.2 is extremely rare, making up just 0.05% of the coronaviruses that have been sequenced worldwide in the past three months. But that was once true of other omicron subvariants that later came to dominate the world. If BA.2.75.2 becomes widespread this winter, it may blunt the effectiveness of the newly authorized boosters from Moderna and Pfizer.

Every time SARS-CoV-2 replicates inside of a cell, it might mutate. On rare occasions, a mutation might help SARS-CoV-2 replicate faster. Or it might help the virus evade antibodies from previous bouts of COVID-19.

Such a beneficial mutation might become more common in a single country before fading away. Or it might take over the world.

At first, SARS-CoV-2 followed the slow and steady course that scientists had expected based on other coronaviruses. Its evolutionary tree gradually split into branches, each gaining a few mutations. Evolutionary biologists kept track of them with codes that were useful but obscure. No one else paid much attention to the codes, because they made little difference to how sick the viruses made people.

But then one lineage, initially known as B.1.1.7, defied expectations. When British scientists discovered it, in December 2020, they were surprised to find it bore a unique sequence of 23 mutations. Those mutations allowed it to spread much faster than other versions of the virus.

Within a few months, several other worrying variants came to light around the world — each with its own combination of mutations, each with the potential to spread quickly and cause a surge of deaths. To make it easier to communicate about them, the WHO came up with its Greek system. B.1.1.7 became alpha.

Different variants experienced varying levels of success. Alpha came to dominate the world, whereas beta took over only in South Africa and a few other countries before petering out.

What made the variants even more puzzling was that they arose independently. Beta did not descend from alpha. Instead, it arose with its own set of new mutations from a different branch of the SARS-CoV-2 tree. The same held true for all the Greek-named variants, up to omicron.

It's likely that most of these variants got their mutations by going into hiding. Instead of jumping from one host to another, they created chronic infections in people with weakened immune systems.

Unable to mount a strong attack, these victims harbored the virus for months, allowing it to accumulate mutations. When it eventually emerged from its host, the virus had a startling range of new abilities — finding new ways to invade cells, weaken the immune system and evade antibodies.

"When it gets out, it's like an invasive species," said Ben Murrell, a computational biologist at the Karolinska Institute in Stockholm.

Omicron did particularly well in this genetic lottery, gaining more than 50 new mutations that helped it find new routes into cells and to infect people who had been vaccinated or previously infected. As it spread around the world and caused an unprecedented spike in cases, it drove most other variants to extinction.

"The genetic innovations seen in omicron were far more profound, as if it was a new species rather than just a new strain," said Darren Martin, a virus expert at the University of Cape Town.

But it soon became clear that the name "omicron" hid a complex reality. After the original omicron virus evolved in the fall of 2021, its descendants split into at least five branches, known as BA.1 through BA.5.

Over the next few months, the subvariants took turns rising to dominance. BA.1 went first, but it was soon outcompeted by BA.2. Each one was distinct enough from the others to evade some of the immunity of its predecessors. By this summer, BA.5 was on the rise.

The U.S. Food and Drug Administration responded by inviting vaccine makers to produce booster shots that included a BA.5 protein along with one from the original version of the virus. Those boosters are now rolling out to the public, at a time when BA.5 is causing 85% of all COVID-19 cases in the United States.

But BA.5 could be fading in the rearview mirror by winter, scientists said. Omicron has continued to evolve — likely by sometimes jumping among hosts, and sometimes hiding for months in one of them.

Since these new lineages belong to omicron, they haven't gotten a Greek letter of their own. But that doesn't mean they're just a slight twist on the original. Antibodies that could latch onto earlier forms of omicron fare poorly against the newer ones.

"They could arguably have been given different Greek letters," Robertson said.

BA.2.75.2 is among the newest of omicron's grandchildren, identified just last month. It's also the most evasive omicron yet, according to Murrell. In lab experiments, he and his colleagues tested BA.2.75.2 against 13 monoclonal antibodies that are in clinical use or in development. It evaded all but one of them, bebtelovimab, made by Eli Lilly.

They also tested the antibodies from recent blood donors in Sweden. BA.2.75.2 did substantially better at escaping those defenses than other omicron subvariants did.

The researchers posted their study online Sept. 16. Researchers at Peking University reached similar conclusions in a study posted the same day. Both have yet to be published in a scientific journal.

Murrell cautioned that scientists have yet to run experiments that will show the effectiveness of BA.5 booster shots against BA.2.75.2. He suspected that getting a big supply of BA.5 antibodies would provide some protection, especially against severe disease.

"It's still important, but we'll have to wait for the data to come out to see exactly what the magnitude of the boosting effect is," Murrell said.

There's no reason to expect that BA.2.75.2 will be the end of the evolutionary line. As immunity builds to previous versions of omicron, new versions will be able to evolve that can evade it.

"I don't think it's going to hit a wall in the mutational space," said Daniel Sheward, a postdoctoral researcher at the Karolinska Institute and co-author on the new study.

Lorenzo Subissi, an infectious disease expert with the WHO, said the organization was not giving Greek letters to lineages like BA.2.75.2 because they are much like the original omicron viruses. For example, it appears that all omicron lineages use a distinctive route to get into cells. As a result, it is less likely to lead to severe infections but possibly better able to spread than previous variants.

"WHO only names a variant when it is concerned that additional risks are being created that require new public health action," Subissi said. But he did not rule out a pi in our future.

"This virus still remains largely unpredictable," he said.