more_vert
VARIANTS of concern have had a dramatic impact on how the COVID-19 pandemic has unfolded. The emergence of the Alpha variant, detected in December 2020, demonstrated the remarkable ability of SARS-CoV-2 to become more infectious. Subsequent variants, such as Beta and Gamma, rendered some vaccines less effective, notably leading to South Africa halting the rollout of the AstraZeneca vaccine in early 2021.
The Delta variant, detected in late 2020, had a profound impact on public health systems and human health worldwide. Not only was it more infectious than other variants but it also increased disease severity. Indeed, Delta overtook most of the other strains of the virus to become the most common variant by mid-2021. Recently, the Omicron variant, with its immune evasion but reduced disease severity compared with Delta, has rapidly overtaken Delta (here and here).
The detection of variants of concern would have been impossible without genomic surveillance, a toolkit that scientists use to sequence the genomes of circulating strains and share them with peers through online databases. The COVID-19 pandemic is the first global crisis that has benefited from these efforts, and Australia has made a key contribution here, with about 80% of all positive cases having the viral genome sequenced during the winter of 2020 in Victoria. Moving forward, the Communicable Diseases Genomics Network will prioritise genome sequencing of infections detected at the border, those with increased disease severity, and those from vaccinated individuals.
Genomic surveillance has also helped us understand how the virus evolves. Viruses change constantly, mostly by accumulating small mutations in their genomes. The very first genomes of the virus, reported in early 2020, enabled scientists to calculate how rapidly the virus could accumulate mutations, at about one to two mutations per month.
This “evolutionary rate” is roughly constant and can be useful to ascertain how rapidly we can expect to see different viruses emerging. Viruses with many mutations can behave very differently from ancestral variants, so understanding the evolutionary rate is also useful in assessing whether vaccines will protect against future strains. For comparison, influenza viruses accumulate mutation four times faster than SARS-CoV-2 and vaccines are usually only effective for a year, although there are other factors at play, including waning immunity, which can shorten the time over which a vaccine is effective.
Theoretically, viruses have an upper limit in how rapidly they can accrue mutations. If they mutate very rapidly, there is a high chance some mutations will be detrimental and the virus will not survive, known as the “error catastrophe”. The upper limit on the evolutionary rate is determined by the kind of genome of the virus and other molecular characteristics, but it is difficult to determine precisely.
Paradoxically, the first variant detected, Alpha, had accumulated at least 20 mutations within about 3 months, a finding that appeared to violate the notion of a predictable evolutionary pace of the virus. This finding was not a fluke. Subsequent variants also displayed unusually large numbers of mutations: around nine for Beta, 15 for Gamma, 12 for Delta, and over 30 for Omicron.
The large number of mutations that define variants raised the question of whether the virus was accelerating its evolutionary pace, a worrying possibility due to the unpredictable nature of how mutated viruses may behave.
Most SARS-CoV-2 genomes are deposited in the GISAID (Global Initiative on Sharing All Influenza Data) database, with over 9 million genomes available for research purposes. In a recent study published in Molecular Biology and Evolution, we mined this database to obtain a sample of variants Alpha, Beta, Gamma and Delta, and did a range of computational analyses to understand why variants have such large numbers of mutations. These analyses are possible due to a substantial groundwork in theoretical virus evolution, biostatistics and bioinformatics in the past decade.
In our analyses we assessed different hypotheses to explain the emergence of variants. We considered scenarios where variants simply mutate faster than their progenitor viruses, those where the mutations process was random, and those where the evolutionary rate was accelerated for a short period of time. The latter had overwhelmingly high support, indicating that the SARS-CoV-2 virus has the ability to increase its evolutionary pace for a few weeks in order to accumulate many mutations and ultimately give rise to a variant. However, the pace at which variants mutate is indistinguishable from that of the original strains.
The mechanism under which variants arise is akin to a runner doing a marathon. The runner has a comfortable speed they can maintain for the course. If they increase their average pace too much for too long, they are likely to break down, but they can benefit from short sprints to overtake other runners and perform better than maintaining a constant pace.
A fundamental lingering question relates to the conditions under which accelerated evolution occurs. Some of the mutations in the Alpha, Delta and Omicron variants were also found in patients who were immunosuppressed and with long term SARS-CoV-2 infections, which has led some to suggest that chronic infections are responsible for the emergence of variants. Other theories involve the virus, and particularly Omicron, infecting other host species, such as rodents, where replication and evolution can occur faster.
The answer may well depend on the particular variant, but regardless of the precise mechanism, the virus is almost certain to continue “sprinting” to generate new variants. Genomic surveillance will be our most useful weapon to detect, track, and limit their spread.
Sebastian Duchene is an Australian Research Council DECRA Fellow, in the Department of Microbiology and Immunology, Peter Doherty Institute for Infection and Immunity at the University of Melbourne.
The statements or opinions expressed in this article reflect the views of the authors and do not represent the official policy of the AMA, the MJA or InSight+ unless so stated.