Cancer prevalence across vertebrate species decreases with gestation time, may increase with adult mass

In 1977, Sir Richard Peto, FRS, FAACR, hypothesized that cancer prevalence in animals should correlate linearly with body size and lifespan, as larger animals have more cells in which to accumulate damage, and long-lived animals have more time to acquire mutations. He observed, however, that this did not seem to be the case based on available data, a finding known as Peto's paradox.

Since then, comparative oncologists have investigated what other factors may dictate why some species get more cancer than others. "We want to learn more about what makes humans unique in their cancer susceptibility," said Zachary Compton, PhD, a postdoctoral fellow in the University of Arizona Cancer Center's NCI T32 fellowship program and first author of the study. "There's some innate cancer risk that's just part of our species identity, and comparative oncology can help us learn why."

Comparative oncology studies have historically been hindered by sample size and species diversity, which has limited researchers' understanding of how growth and reproduction characteristics may impact cancer development, Compton explained. He and his colleagues collected a total of 16,049 necropsy records from 292 species at 99 animal care institutions in the United States and London; all species included in the analysis had at least 20 available necropsies. The researchers assessed whether neoplasia (uncontrolled growth) was present and in what tissue, and they developed a terminology dictionary based on the neoplasia descriptions to predict whether neoplasms were benign or malignant.

Across all species, the median prevalence of neoplasia was 4.89%, and the median prevalence of malignancy was 3.2%. Mammals had the highest rates of both neoplasia (12%) and malignancy (7%), and amphibians had the lowest rates (1.2% and 0%, respectively). By comparison, approximately 39.3% of Americans will be diagnosed with cancer during their lifetime, according to federal statistics; but Compton warned that we may not know the exact prevalence of malignancy in humans, as most natural deaths do not warrant autopsies, and there is currently no robust way to estimate benign neoplasia at the population level.

Larger body mass was significantly associated with a higher prevalence of neoplasia; for every tenfold increase in body mass, the risk of neoplasia increased by 2.1%, making this one of few studies to contradict Peto's paradox, Compton explained. Other factors associated with malignancy and/or neoplasia included increased maximum longevity, larger litter size in mammals, and, in a subset analysis of 15 species, somatic mutation rate, or the rate at which new mutations occur in the body. Longer gestation time was significantly associated with a lower prevalence of both neoplasia (5.3% decrease in neoplasia risk per tenfold increase in months of gestation) and malignancy (5.65% decrease in malignancy risk per tenfold increase in months of gestation).

Because both longevity and gestation time are generally associated with body mass, the researchers performed additional analyses in which they normalized data based on these factors. Body mass was more strongly associated with neoplasia and malignancy prevalence when accounting for gestation time, and vice versa.

Compton noted that the complex interactions between traits that don't evolve independently, such as body mass and gestation time, will require further research to tease apart. "Peto's paradox is typically discussed as the singular relationship between body mass and cancer," he said. "But if we want to truly appreciate Peto's paradox, we have to appreciate the entanglement between various life history traits."

Regarding longevity, some researchers have hypothesized that rates of neoplasia may be artificially inflated in human-managed populations, because animals in captivity often live far beyond their natural lifespans, Compton explained. In their study, however, longevity did not independently correlate with neoplasia or malignancy rates. The vast majority of species typically developed tumors within their natural lifespan, and age at death only correlated with higher rates of malignancy in amphibians.

"We not only controlled for natural lifespan, but we discovered that we didn't need to control for it," Compton said. "The rates we were seeing weren't because animals live longer in zoos."

Finally, the analysis identified animals with exceptionally high and exceptionally low rates of neoplasia. Ferrets and opossums had the highest rates, with 63% and 56% having neoplasia at necropsy, respectively. Species with low rates of neoplasia included the Nubian ibex, the tammar wallaby, and several species of bats. Many species had no malignancies observed, which may be strong candidates to study cancer suppression mechanisms, Compton explained. He added that species with high rates of neoplasia may serve as models of spontaneous cancer development, and of some human cancer predisposition syndromes.

"This research gives us a wide range of hypotheses to pursue, including what molecular pathways and exposures are promoting or preventing cancer development across these species," Compton said.

Limitations of the study include the fact that most analyzed animals lived in the care of humans rather than in their natural habitat. Further, the necropsy records did not include information about animals' environmental exposures or infection with cancer-predisposing viruses, and some cancers, including blood cancers, may be difficult to identify in a necropsy.

Funding for this study was provided by the National Institutes of Health, the CDMRP Breast Cancer Research Program, the Arizona Biomedical Research Commission, Hyundai Hope on Wheels, the Soccer for Hope Foundation, the Li-Fraumeni Syndrome Association, Kneaders Bakery & Café's Hope Fights Childhood Cancer, the Helen Clise Presidential Endowed Chair in Li-Fraumeni Syndrome Research, the Primary Children's Hospital Foundation, the University of Utah Department of Pediatrics Research Enterprise, the National Scientific Research Fund, and the Agence Nationale de la Recherche and Région Nouvelle-Aquitaine -- Chaire d'excellence 'Cancer et Biodiversité', both in France.