RESUMEN
BACKGROUND: Advanced age is a consistent risk factor for cancer; nonetheless, cancer incidence typically declines after age 75-85 for most solid tumors. METHODS: To delineate the true cancer age-incidence pattern, we performed a population-based cohort study using Swedish Cancer Register data from 1970-2014 on nine common, adult (age 20-99) cancers categorized as requiring high (pancreatic, lung, non-meningioma brain), medium (anorectal, urinary bladder, non-Hodgkin lymphoma), and low (melanoma skin, breast, prostate) diagnostic invasiveness based on the perceived risk of complications associated with histopathologic verification. We estimated the reported incidence and the proportion of autopsy-detected cancers by age but also projected a corrected incidence assuming the same proportion of unexpected cancer findings if all deaths underwent autopsy. RESULTS: The registered cancer incidence dropped after peak age around 65-84, with the exception of melanoma skin. This pattern was attenuated when exploring the proportion of incident, unexpected cancer findings in autopsy material by age. The "total" cancer incidence, reported plus projected incident autopsy cases, increased monotonously with age. CONCLUSIONS: The long-established cancer incidence decline in elderly is most probably an artifact due to reduced diagnostic intensity. IMPACT: Biological drivers to the cancer incidence decline in elderly are unlikely and resources are better allocated to prepare for the anticipated cancer pandemic when numbers of healthy elderly increase. Cancer alarm symptoms in elderly fit for cancer therapy should be investigated promptly and clinical cancer trials focus to also include elderly to set updated standards for cancer therapy in the dominating age group.