The dominant idea about how cancer gets started is called the "two-hit hypothesis." First proposed by Alfred Knudson in 1971, it holds that a cancer starts when one cell gets a mutation in both of its copies of a gene that normally blocks cancer formation (two hits). These two mutations disable the tumor-suppressing function in that cell, which then becomes cancerous. Eventually, the idea was expanded to include two hits not necessarily in the same gene but, rather, in genes controlling the same tumor-suppressing pathway.
But a new idea is challenging the two-hit hypothesis, shifting the focus to the role of the immune system in suppressing cancers. It's an idea that could have big implications for treatments.
Taking a hit
Getting two hits in one cell was considered to be a random and unlucky event. Since mutations occur each time a cell divides, the more times each cell divides, the greater the chances that it would happen. This was why, it was thought, cancer incidence increases with age; the longer a cell has been around, and the more times it has divided, the more opportunities it has had to accrue the two requisite mutations in the same tumor-suppressor pathway.
Evidence for the two-hit hypothesis came primarily from children with retinoblastoma, who have a germline mutation in the RB1 gene (named for the disease it causes) and are therefore born with one hit in every cell already. These kids usually end up with tumors in their eyes by the time they turn five.
Personalized medicine has been focused on the two-hit model. The idea is to identify the key mutations in a given cancer, then target and nullify them. It has been touted as the wave of the future for a while, but its successes have been mixed. Not every cancer has an obvious target gene, and many tumors can evolve resistance to targeted drugs.
Immunotherapy, by contrast, has achieved some striking successes. Much of it relies on engineered T cells designed and synthesized to kill specific tumor cells. But it also involves awakening the body's existing T cells, which would go on to help fight the tumor. Tumors generally have proteins on their surface that can activate T cells, but they also have mechanisms to suppress the immune system. Cancer immunotherapy relieves this suppression, freeing the T cells to fight the tumor.
A new analysis suggests that the relationship between the immune system and tumor cells provides not only the basis for this new therapeutic approach but also the explanation for increased cancer incidence as we age.
T cells arise in the thymus (that’s why they’re called T cells), but the thymus starts to atrophy around the time we turn one and the number of viable T cells it churns out drops continually over time. Mathematical modeling suggests that cancers do not primarily arise because getting two hits in one cell becomes more likely as we age. Instead, cancer-causing mutations seem to occur at roughly the same rate over the course of our lives, but our T cells wipe out these proto-cancer cells before they become clinically problematic.
It is only as our reservoir of T cells declines as we age that one of these continuously produced cancer cells can overcome immune surveillance and blossom into disease. The same immune system decline would explain the rising incidence of infectious diseases with age.
The authors cite a couple of observations supporting their model. One is that women get fewer cancers than men, since they have more circulating T cells and their T cell levels drop at lower rates. Another is that sharks, which have notoriously low cancer rates, do not experience this thymic atrophy as they age.
They also make a few practical recommendations. Nine out of ten of the cancers that best fit this new model have rates that spike in the late fifties, so they suggest that this might be a good age for more stringent cancer screening. And especially given the success that immunotherapies have already had in fighting some types of cancers, they suggest that more therapies that shore up T cell production or alleviate T cell exhaustion might be a better bet than trying to counteract or even prevent specific cancer-causing mutations.
It's important to emphasize that the two models aren't completely exclusive—mutations are still important for a cancer's development and progression, and they can still be targeted with treatments. The new proposal just drives home that, even if a cell picks up damaging mutations, it won't go on to form a cancer if the immune system kills it.
PNAS, 2018. DOI: 10.1073.pnas.1714478115 (About DOIs).