I think it would be fairly easy to find evidence of directly linked survival -> passing on risk. There haven't been so many generations with modern (cancer) medicine that you couldn't trace whether many folks getting cancer young had recent ancestors who survived getting cancer young because of modern medicine.

On the other hand, if there's significant overlap between all the historical causes of infant/small child mortality that we've eliminated over centuries, and susceptibility to early cancer, that would be much harder to identify.