A certain piece of popular mythology about viral evolution seems to be unkillable: that it is the nature of viruses to, over time, evolve to be less pathogenic and more contagious. The reasoning, the less sick the host is, the more likely they are to go out and make contact with other potential hosts, resulting in mutations that cause less severe disease to be selected for over time.

Unfortunately, this is not exactly the case. Evolution does not plan in this manner. It happens that the Omicron variant is less pathogenic than its ancestors, but this is an aberration, not a trend. Before Omicron, the real trend was for each new variant to grow more pathogenic, not less. 

The mutations that evolution primarily selects for are not the mutations that spread the virus to as many hosts as possible, but rather what allows the virus to most rapidly replicate within a host during an infection. Natural selection first selects traits that produce short-term competitive advantage, not long-term herd persistence. Mutations that increase replication rate will rapidly replace ancestral strains within a host, resulting in them being selected for. Mutations that cause increased transmissibility by other means, such as reduced severity, will not as rapidly outcompete ancestors during an infection course. Most mutations which make a virus less virulent are also likely to make the virus weaker against immune defenses, reducing the likelihood of finding a new host. 

But viruses do get less fatal, at least sometimes, as Omicron has proven. 

In the case of respiratory viruses, pathogenicity and transmissibility are most influenced by 2 factors: where in the lungs a virus likes to replicate, and how quickly a virus replicates. The faster the rate of replication, the more pathogenic and the more transmissible. This is the Delta variant’s trick. Compared to the Alpha variant, infection with Delta in the unvaccinated was 2.35x more likely to require hospitalization and 1.3x more likely to result in death. 

Site of infection is an even more important factor, and it is the main trait influencing Omicron’s characteristics. Compared to its ancestors, Omicron likes replicating higher in the lungs;  about 70x faster in the nose and bronchi, and 10x slower in the alveoli compared to Delta. Animal challenge trials indicate similar behavior. The higher up in the lungs a virus infects, the less pathogenic and more transmissible it will be. A virus higher in the lungs does not have to go as far during an exhale, and does not have to get as far in their next host to find a place to replicate, resulting in higher chances of a host infecting a contact. 

The reason for the reduced pathogenicity has to do with some very basic lung physiology. Oxygen is absorbed in the lowest level of the lungs, the alveoli. Even temporary disruption of this absorption is very, very bad. Naturally, the body goes nuts if it sees anything obstructing oxygen capture, and the heightened inflammatory response can lead to life-threatening secondary issues, like pneumonia and ARDS.

Higher up in the respiratory system, though, is not nearly as sensitive. Cells lining the bronchi and throat do not need to exchange gasses, they just need to be smooth enough for air to flow over them. The immune system is not as worried about infections of these cells either, which results in a tamer inflammatory response and a milder disease course. Common cold viruses tend to replicate exclusively in the nose and upper airways. Secondary symptoms of an inflammatory response in these regions are mild annoyances like runny nose, sneezing, and congestion, not lung failure and death.

So it is not “myth busted,” but it still is not accurate to state that viruses lose their teeth with age. At best, this is a simplification, and at worst, a false sense of security. And still, no matter what traits a virus may have, no matter what evolution may do to it, the most certain way to decrease its pathogenicity is to teach the immune system its epitopes. For this, there are two options: get the virus, or get vaccinated against it.