Why whales should be riddled with cancer, and possibly are not
Cancer starts when a cell picks up the wrong combination of mutations and stops obeying the rules about when to divide. Every division is a small roll of the dice. That suggests a blunt prediction: an animal built from vastly more cells, living vastly longer, should roll the dice far more often and get far more cancer.
A blue whale has on the order of a thousand times more cells than a human and can live for decades. By the simple arithmetic it should be riddled with tumours. The epidemiologist Richard Peto pointed out in the 1970s that the expected scaling does not seem to show up in nature, and the puzzle now carries his name: Peto’s paradox.
The idea
If the paradox is real, big long-lived species must have evolved extra defences that small short-lived ones never needed. A mouse can afford sloppy cancer suppression because something else will probably kill it within two years. An elephant cannot.
The most cited example comes from elephants. Lisa Abegglen, Joshua Schiffman, Carlo Maley and colleagues reported in JAMA in 2015 that the African elephant genome carries at least 20 copies of TP53, a tumour suppressor genetumour suppressor gene A gene whose normal job is to stop cells dividing out of control, so losing it raises cancer risk. that humans carry only once. Nineteen of those copies are retrogenesretrogene An extra copy of a gene made when its message was copied back into the DNA, which may or may not still work., and some are switched on. In cell experiments, elephant white blood cells exposed to radiation were roughly twice as likely as healthy human cells to undergo apoptosisapoptosis A built-in process by which a cell deliberately kills itself, often when it is damaged., killing themselves rather than living on with damaged DNA. The rate of DNA repair, by contrast, looked similar. The proposed mechanism is not better repair but a lower threshold for self-destruction. From 644 recorded elephant deaths, they estimated cancer mortality at about 5%, against 11 to 25% in humans.
Why it might change a decision
The paradox makes a prediction you can check against any new claim: if it holds, lineages that grew large should show signs of extra cancer suppression, and that suppression should be visible as lower cancer rates than their size alone predicts. When a headline announces that some animal “holds the key” to cancer, the testable question is whether anyone has compared its cancer rate with what its size and lifespan would predict, across many individuals, or whether the claim rests on a mechanism seen in cells from one or two animals.
How solid is this?
The honest answer is that two large, recent comparative studiescomparative study Research that looks for patterns across many species at once rather than experimenting on one. disagree about whether the paradox exists at all.
For the paradox. Orsolya Vincze and colleagues, writing in Nature in 2022, used zoo records for 110,148 adult mammals across 191 species, with pathology reports for 11,840 of them. Cancer was common across mammals and varied enormously, from no recorded cases in 47 species to 57% of deaths in the kowari, a small marsupial. Mammal-eating carnivores had the highest rates and ruminants the lowest. Body mass explained less than 1% of the variation between species and life expectancy under 3%, and neither relationship was statistically significant. The authors read this as support for the paradox, and stress that zoo data call for caution because diet and environment are controlled by people.
Against the paradox. George Butler, Chris Venditti and colleagues, in PNAS in 2025, analysed neoplasianeoplasia Abnormal tissue growth, which includes both harmless tumours and cancers. prevalence in 263 species of amphibians, birds, mammals and reptiles. They found that larger species do have more cancer, in all four classes, and titled the paper “No evidence for Peto’s paradox in terrestrial vertebrates.” They added a twist: in birds and mammals, lineages that went through bursts of rapid size change had less cancer than their current size predicted, which they suggest may reflect evolved defences. They describe that mechanism as speculative.
These results are less contradictory than the headlines make them sound. Both find that cancer rises far less with size than the cell arithmetic would predict. They disagree on whether the remaining slope is zero or modestly positive, using different datasets and different outcome measures. Both are correlationalcorrelational Showing that two things vary together, without establishing that either one causes the other. comparisons across species, and both lean heavily on captive animals, whose lifespans, diets and stress differ from wild ones.
The elephant work is one study, and its parts differ in strength. The gene count is a property of the genome. The cell-death experiments are thinner: the main radiation experiments used cells from a single African elephant, and it is not established that the extra copies make working protein. The authors also disclosed a pending patent on elephant TP53, which is not evidence against the result but is worth knowing.
The popular version, that big animals are immune to cancer and elephants explain why, outruns the data. The defensible version is narrower: cancer risk scales much more weakly with size than naive arithmetic predicts, elephants have at least one plausible defence, and whether size matters at all is still argued.
Sources
- Abegglen, L. M., Caulin, A. F., Chan, A., et al. (2015). Potential mechanisms for cancer resistance in elephants and comparative cellular response to DNA damage in humans. JAMA, 314(17), 1850–1860. Link
- Vincze, O., Colchero, F., Lemaître, J.-F., et al. (2022). Cancer risk across mammals. Nature, 601, 263–267. Link
- Butler, G., Baker, J., Amend, S. R., Pienta, K. J., & Venditti, C. (2025). No evidence for Peto’s paradox in terrestrial vertebrates. PNAS, 122(9), e2422861122. PDF