Background:  HIV/AIDS increases the risk of certain cancers. Previous studies have reported that age at cancer diagnosis is far younger among persons with HIV/AIDS than in the general population, suggesting that cancer development is “accelerated” in persons with HIV/AIDS. However, these estimates were influenced by differences in the age structure of these populations, ie, people with HIV/AIDS are generally younger. We assessed differences in the age at cancer diagnosis between the AIDS and general populations, after adjusting for population age structure.

Methods:  The HIV/AIDS Cancer Match study links 15 US HIV/AIDS and cancer registries. Using data from 338,349 persons with AIDS from 1980 to 2006, we compared observed cancers in persons with AIDS and the general population, and expected cases calculated by applying general population rates from the cancer registry to follow-up time of persons with AIDS. Expected cancers represent cases that would occur if the general population had the same structure as the AIDS population, defined by age, sex, race, year, and registry. Median age at cancer diagnosis was estimated for each cancer.

Results:  The proportion of older persons (at greatest cancer risk) was far smaller among people with AIDS than the general population (3% vs 17% over age 60 years). Reflecting this difference, the observed median age at diagnosis for most cancers was around 15 to 30 years younger among people with AIDS than in the general population. After accounting for the age structure, we found no difference in observed and expected ages at diagnosis for most cancers. However, the observed median age at diagnosis for persons with AIDS was 3 to 4 years younger than expected for non-Hodgkin lymphoma, lung cancer and anal cancer (all P <0.001 and significant after adjustment for multiple comparisons).

Conclusions:  For most cancers, age at diagnosis did not differ between persons with AIDS and the general population. Previous reports that did not account for large differences in underlying age structure incorrectly concluded that persons with AIDS develop cancer at a younger age. After controlling for differences in population structure, age at diagnosis remained slightly younger for only a few cancers among people with AIDS. These differences may reflect the effects of HIV/AIDS in accelerating development of some cancers, differences in cancer surveillance, or differences in the age at first exposure to other cancer risk factors (eg, HPV infection, tobacco).