Adherence to the World Cancer Research Fund/American Institute for Cancer Research Lifestyle Recommendations on...

Early-onset breast cancer (BC) is more common in Hispanic than in non-Hispanic White women. However, little is known about the role of lifestyle factors in the development of early- and later-onset BC in Mexican women.

Our objective was to estimate the potential relationship between lifestyle and BC risk, with particular attention to potential differences between early-onset and later-onset BC.

We used data from 84,534 participants in the Mexican Teachers' Cohort. An 8-item score was constructed to measure adherence to the 2018 World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) cancer prevention recommendations. BC cases were identified through self-reports, linkages with electronic health records, and administrative data. Early-onset BC was defined as diagnosis before age 50, and later-onset BC as diagnosis at age ≥50. To estimate the 10-y risk of early- and later-onset BC, analyses included women <40 y and women aged ≥50 y at baseline, respectively. We used the parametric g-formula to estimate 10-y risks, risk differences (RD), and risk ratios (RR).

Over 10 y of follow-up, 1052 BC cases were identified, including 180 early-onset and 209 later-onset cases. For overall BC, the estimated RR comparing women with scores of 5-8 compared with 0-2 was 0.82 [95% confidence interval (CI): 0.61, 1.15], with a corresponding RD of -0.23 percentage points (95% CI: -0.65, 0.14). The estimated RR comparing scores of 5-8 with 0-2 were 0.75 (95% CI: 0.51, 1.13) for early-onset BC and 1.23 (95% CI: 0.86, 1.83) for later-onset BC. The corresponding RD were -0.20 percentage points (95% CI: -0.48, 0.08) for early-onset BC and 0.34 percentage points (95% CI: -0.25, 0.95) for later-onset BC.

Estimated risks of overall, early-onset, and later-onset BC were broadly similar across levels of adherence to the WCRF/AICR-based lifestyle score. The estimates were imprecise, with CIs including the null value, indicating that chance cannot be ruled out; however, the range of estimates is compatible with both modest decreases and modest increases in risk.