Epidemiological associations between neurodegeneration and cancer - time to refine our research approach.

Cancer and neurodegenerative diseases are devastating aging-related disorders. Epidemiological data suggest a general inverse risk association between them, the mechanisms of which might provide new avenues for developing treatments and prevention strategies. In addition to this observation, meta-analyses have provided some conflicting results, apparently related to disease characterization, data validity, survival bias and unhandled exposure to risk factors. Taking these into account it seems that the association might be specific to certain types of cancers and particular types of neurodegenerative diseases. It should also be noted that, in some cases, the association is positive. Future studies investigating epidemiological association between cancer and neurodegeneration need to take into account several methodological aspects to provide data that can inform basic and translational studies on pathomechanisms. Most importantly, diagnoses must be accurate and validated, disease subtypes should be precisely defined and stratified, sources of possible bias must be explicitly addressed, time-varying trends should be analyzed as potential confounders and null association must always be published.