Studies of body size evolution, and life-history theory in general, are conducted without taking into account cancer as a factor that can end an organism’s reproductive lifespan. This reflects a tacit assumption that predation, parasitism and starvation are of overriding importance in the wild. We argue here that even if deaths directly attributable to cancer are a rarity in studies of natural populations, it remains incorrect to infer that cancer has not been of importance in shaping observed life histories. We present first steps towards a cancer-aware life history theory, by quantifying the decrease in the length of the expected reproductively active lifespan that follows from an attempt to grow larger than conspecific competitors. If all else is equal, a larger organism is more likely to develop cancer, but, importantly, many factors are unlikely to be equal. Variations in extrinsic mortality as well as in the pace of life — larger organisms are often near the slow end of the fast-slow life history continuum — can make realized cancer incidences more equal across species than what would be observed in the absence of adaptive responses to cancer risk (alleviating the socalled Peto’s paradox). We also discuss reasons why patterns across species can differ from within-species predictions. Even if natural selection diminishes cancer susceptibility differences between a species, within-species differences can remain. In many sexually dimorphic cases we predict males to be more cancer-prone than females, forming an understudied component of sexual conflict.