Aspirin Use and Incidence of Colorectal Cancer According to Lifestyle Risk

Daniel R Sikavi,Kai Wang,Wenjie Ma,David A Drew,Shuji Ogino,Edward L Giovannucci,Yin Cao,Mingyang Song,Long H Nguyen,Andrew T Chan
DOI: https://doi.org/10.1001/jamaoncol.2024.2503
2024-08-01
Abstract:Importance: Aspirin reduces the risk of colorectal cancer (CRC). Identifying individuals more likely to benefit from regular aspirin use for CRC prevention is a high priority. Objective: To assess whether aspirin use is associated with the risk of CRC across different lifestyle risk factors. Design, setting, and participants: A prospective cohort study among women in the Nurses' Health Study (1980-2018) and men in the Health Professionals Follow-Up Study (1986-2018) was conducted. Data analysis was performed from October 1, 2021, to May 22, 2023. Exposures: A healthy lifestyle score was calculated based on body mass index, alcohol intake, physical activity, diet, and smoking with scores ranging from 0 to 5 (higher values corresponding to a healthier lifestyle). Regular aspirin use was defined as 2 or more standard tablets (325 mg) per week. Main outcome and measures: Outcomes included multivariable-adjusted 10-year cumulative incidence of CRC, absolute risk reduction (ARR), and number needed to treat associated with regular aspirin use by lifestyle score and multivariable-adjusted hazard ratios for incident CRC across lifestyle scores. Results: The mean (SD) baseline age of the 107 655 study participants (63 957 women from the Nurses' Health Study and 43 698 men from the Health Professionals Follow-Up Study) was 49.4 (9.0) years. During 3 038 215 person-years of follow-up, 2544 incident cases of CRC were documented. The 10-year cumulative CRC incidence was 1.98% (95% CI, 1.44%-2.51%) among participants who regularly used aspirin compared with 2.95% (95% CI, 2.31%-3.58%) among those who did not use aspirin, corresponding to an ARR of 0.97%. The ARR associated with aspirin use was greatest among those with the unhealthiest lifestyle scores and progressively decreased with healthier lifestyle scores (P < .001 for additive interaction). The 10-year ARR for lifestyle scores 0 to 1 (unhealthiest) was 1.28%. In contrast, the 10-year ARR for lifestyle scores 4 to 5 (healthiest) was 0.11%. The 10-year number needed to treat with aspirin was 78 for participants with lifestyle scores 0 to 1, 164 for score 2, 154 for score 3, and 909 for scores 4 to 5. Among the components of the healthy lifestyle score, the greatest differences in ARR associated with aspirin use were observed for body mass index and smoking. Conclusions and relevance: In this cohort study, aspirin use was associated with a greater absolute reduction in risk of CRC among individuals with less healthy lifestyles. The findings of the study suggest that lifestyle risk factors may be useful to identify individuals who may have a more favorable risk-benefit profile for cancer prevention with aspirin.
What problem does this paper attempt to address?