[4] A 25-year longitudinal study of 17,738 male British civil servants, published in 1998, showed a positive correlation between height and cancer incidence with a high degree of statistical confidence, even after risk factors like smoking were controlled for.
[5] A similar 2011 study of more than one million British women found strong statistical evidence of a relationship between cancer and height, even after controlling for a number of socioeconomic and behavioral risk factors.
[6] A 2011 analysis of the causes of death of 74,556 domesticated North American dogs found that cancer incidence was lowest in the smaller breeds, confirming the results of earlier studies.
In a 2015 study, the San Diego Zoo surveyed results from 36 different mammalian species, ranging in size from the 51-gram striped grass mouse to the nearly 100,000 times larger 4,800-kilogram elephant.
The study found no statistically significant relationship between body size and cancer incidence, offering empirical support for Peto's initial observation.
[citation needed] A 2014 paper in Evolutionary Applications by Maciak and Michalak emphasized what they termed "a largely underappreciated relation of cell size to both metabolism and cell-division rates across species" as key factors underlying the paradox, and concluded that "larger organisms have bigger and slowly dividing cells with lower energy turnover, all significantly reducing the risk of cancer initiation.
[21] Additionally, the basal metabolic rates of larger animals are generally lower, following a well-defined inverse logarithmic relationship, which is typically associated with reduced oxidative stress.
[13] In one experiment, laboratory mice were genetically altered to express "always-on" (always on meaning it doesn't get deactivated by the MDM2 gene) active TP53 tumor antigens,[clarification needed] similar to the ones found in elephants.