Cancer Incidence and Mortality With Aspirin in Older Adults: Follow-Up of the ASPREE Trial

Prior studies, largely among middle-aged adults, reported aspirin reduces cancer risk after 10 years, particularly for colorectal cancer (CRC). In contrast, the Aspirin in Reducing Events in the Elderly (ASPREE) randomized clinical trial (RCT) reported that low-dose aspirin (LDA) treatment for a median of 4.7 years had no effect on overall cancer incidence but increased risk of incident late-stage cancer and cancer-related mortality.

To assess whether LDA is associated with cancer incidence and mortality in 10 years of follow-up in older adults (aged ≥70 years) and to assess the association with cancer after prior LDA exposure (legacy effects).

This community-based binational (Australian and US) cohort study included community-dwelling older adults (aged ≥70 years for Australian participants and ≥65 years for US minority group participants) free from overt cardiovascular disease, dementia, or independence-limiting physical disability. The cohort was derived from the ASPREE randomized clinical trial conducted from 2010 to 2017, with the observational extension study (ASPREE-XT) following up participants from 2018 to 2024. This study reports data from 2010 through 2022 (long-term outcomes) as well as reports analyses confined to the observation phase only (legacy analyses). Data were analyzed from May to November 2025.

Daily 100-mg aspirin or placebo from randomization until cessation of study drug.

Outcomes were physician-adjudicated incident cancer, type, stage at diagnosis, and cancer mortality.

In 19 114 community-dwelling older adults (mean [SD] age, 75.1 [4.5] years; 56.4% female), a total of 3448 incident cancers and 1173 cancer-related deaths occurred over 10 years of follow-up (median, 8.6 [IQR, 7.4-10.0] years) during ASPREE and ASPREE-XT. LDA was not associated with overall cancer incidence over the long term (hazard ratio [HR] = 0.98; 95% CI, 0.92-1.05), by stage at diagnosis or cancer type, including colorectal cancer (HR = 1.01; 95% CI, 0.84-1.21). However, LDA was associated with increased cancer-related mortality (HR = 1.15; 95% CI, 1.03-1.29). Among 14 907 participants without cancer during the RCT and consented into ASPREE-XT (median age, 78.6 years [IQR, 76.2-82.1]; 57.5% female), 1451 incident cancers and 376 cancer deaths occurred in the post-RCT period, during which original aspirin assignment during the RCT was not associated with differences in cancer incidence (HR = 0.91; 95% CI, 0.82-1.01) or cancer-related mortality (HR = 1.02; 95% CI, 0.83-1.25) compared with original placebo assignment.

In this study, over a median of 8.6 years, LDA was not associated with incident cancer among older adults, but cancer mortality risk was significantly elevated. However, the elevated cancer mortality risk seen with aspirin for participants in the RCT period did not persist into the post-RCT observation period, suggesting no legacy effect.