Reeves · Breast cancer research : BCR 2025 · prospective cohort study · n=1,228,671

The aetiology of breast cancer subtypes: results from the Million Women Study.

Cited 1 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective non-randomized cohort study

PubMed 41457233 · doi:10.1186/s13058-025-02197-1 · record verified 2026-08-26

What was done

Researchers analyzed data from 1,228,671 UK postmenopausal women aged 50–64 recruited into the Million Women Study between 1996 and 2001, followed for an average of 19.8 years (SD 6.5). The study examined associations between established breast cancer risk factors and specific breast cancer subtypes defined by estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) status.

What was found

During follow-up, 58,134 incident breast cancers with known ER status and 40,627 with complete surrogate molecular subtype status were documented. Most established risk factors were primarily associated with ER-positive disease (heterogeneity p <= 0.002 for each): age at first birth, age at menopause, BMI, height, alcohol intake, and menopausal hormone therapy were positively associated, whereas parity was inversely associated. Prior oral contraceptive use showed a significantly stronger association with ER-negative than ER-positive cancer (p = 0.002). For surrogate molecular subtypes, parity had a modest positive association and breastfeeding had an inverse association with basal-like cancer (exact numerical effect sizes were not reported in the abstract).

Why it matters

These findings demonstrate that major breast cancer risk factors do not act uniformly across all disease forms, providing distinct etiological clues that can guide targeted prevention for aggressive subtypes like basal-like and ER-negative cancers.

Limits

The abstract reports statistical significance levels but omits specific effect sizes (such as relative risks or hazard ratios) and confidence intervals. Subtypes were classified using surrogate clinical receptor markers rather than direct genomic profiling. Receptor status was unavailable for a portion of incident cases, and the study cohort was limited to UK postmenopausal women.

Cited by