Somatic mutation rate and cancer risk may be linked

Somatic mutation rate and cancer risk may be linked in a way that is not captured simply by asking which cancer-associated mutations a person carries. An analysis of normal human tissues found striking variation between individuals: even after accounting for age, mutation burdens could differ by roughly eightfold. One possible explanation is that some people accumulate somatic mutations substantially faster than others.

Somatic mutations are DNA changes acquired by cells during a person's lifetime. They arise as cells divide, through environmental damage and because DNA copying and repair are not perfect. Most have little or no biological consequence. But accumulating more mutations creates more opportunities for the combinations of changes that can eventually contribute to tumor formation.

A study by Sophie Pénisson and colleagues in Cancer Research shifts attention from the tumor itself to this variation in apparently normal cells.

An eightfold difference within normal tissue

The researchers integrated previously published whole-genome and whole-exome sequencing datasets from normal and cancerous human tissues. Their broader analysis included the bladder, colon, esophagus, liver, lung, prostate and stomach, with systematic direct comparisons between normal and cancer cells possible for five organs.

One result was unsurprising: cancer cells carried far more somatic mutations. Across the directly compared tissues, their median mutation burden was roughly four times that of normal cells.

The more unusual finding was the spread within the normal samples.

After age was taken into account, mutation burdens still varied dramatically between individuals, with differences reaching approximately eightfold. Some normal samples even fell within mutation-burden ranges seen among cancer samples.

Another observation became crucial to the authors' interpretation. Normal tissues from people with cancer tended to contain more mutations than normal tissues from people without cancer.

That raises a different question from the familiar one of why tumors become highly mutated: why should an excess of mutations also be detectable outside the tumor?

Three explanations for the same pattern

The researchers considered three broad hypotheses.

Under the classical model, cancer-driving mutations alter cellular behavior and initiate tumor evolution, after which additional mutations accumulate as the expanding population of cancer cells continues to divide.

A second possibility is a catastrophic model, in which rare extreme genomic events rapidly produce large numbers of alterations.

The third is what the authors call the "tail hypothesis." It proposes that baseline mutation rates differ substantially between people. Most individuals would cluster around an average rate, while a minority would occupy the high end, or tail, of the distribution and accumulate mutations faster.

If cancer generally requires a particular combination of alterations, a person whose cells acquire mutations more rapidly would have more chances to reach such a combination during a lifetime.

The authors compared these explanations with several features of the observed mutation-burden distributions. More than one model could account for aspects of the cancer data. The tail hypothesis, however, provided the most straightforward explanation for an additional observation: the higher mutation burden found in normal tissues from people with cancer compared with normal tissues from people without cancer.

That makes the hypothesis plausible, but it does not establish it as the mechanism.

The study did not follow the same healthy individuals prospectively for decades and measure how quickly new mutations appeared before cancer developed. Instead, it analyzed mutation burdens already present in large genomic datasets. Individual differences in mutation rate are therefore inferred from the distributions rather than directly measured over time.

Mutation burden is also not synonymous with cancer risk. The identity of the mutations, the genes and tissues involved, clonal expansion and many other biological factors influence whether a cell ever becomes malignant.

From which mutations to how quickly they arise

If the proposed model holds up in independent and especially prospective studies, it could add another dimension to cancer-risk assessment.

Cancer genomics often focuses on the identity of mutations already present in a person's cells. This study points to another potentially informative variable: the rate at which an individual's normal tissues generate and retain somatic mutations in the first place.

In principle, such a measure might eventually complement established risk factors and help identify people who could benefit from intensified surveillance or prevention. The current work does not, however, provide a clinical test that can predict an individual's cancer risk from mutation rate.

Its immediate implication is more fundamental. Two people of the same age can apparently arrive at that age carrying very different numbers of acquired mutations in otherwise normal tissues.

If those differences do reflect persistent person-to-person variation in mutational accumulation, cancer biology may need to ask not only which mutations appeared, but also how quickly each person's cells generate opportunities for dangerous combinations to emerge.