Background:
Breast cancer (BC) encompasses biologically distinct molecular subtypes—Luminal A (LA), Luminal B (LB), HER2?enriched (HER2+), and Triple?Negative (TN)—each characterized by unique incidence patterns, clinical trajectories, and treatment responses. These subtype?specific differences reflect divergent underlying aetiological mechanisms, suggesting that classical risk factors may not influence all subtypes in the same way. However, despite substantial epidemiological research on BC risk factors, no comprehensive evaluation has systematically examined how these established factors relate to the full spectrum of molecular subtypes. As a result, important subtype?specific risk profiles may remain obscured, underscoring the need for analyses that disentangle heterogeneous associations to inform more tailored prevention and risk?stratification strategies.

Objectives:
This study examined how established BC risk factors relate to specific molecular subtypes within a large cohort of BC patients from Quebec City, Canada.
 
Methods:
We analyzed two complementary cohorts of patients diagnosed with primary invasive BC between 2000–2019 (n = 13,356) and 2000–2012 (n = 6,775) at the Centre des maladies du sein, one of Canada’s largest specialized breast disease centers. Molecular subtypes were determined based on Hormone Receptors (HR) and Human Epidermal Growth Factor Receptor 2 (HER2), assessed through immunohistochemistry. Adjusted prevalence ratios (PRs) and 95% confidence intervals (CIs) were estimated using multivariable log-binomial regression models. Bonferroni correction was applied to reduce the risk of type I error across multiple comparisons.

Results:
Risk factor associations differed markedly across BC subtypes. Women aged 60–69 displayed a higher prevalence of LA tumors (PR = 1.30; 95% CI: 1.22–1.37) but a lower prevalence of LB, HER2+, and TN subtypes. Underweight women were more likely to develop TN tumors (PR = 1.49; 95% CI: 1.11–2.00) and less likely to develop LA tumors (PR = 0.86; 95% CI: 0.79–0.95), while obesity was more strongly associated with LB tumors (PR = 1.45; 95% CI: 1.18–1.80). Past oral contraceptive use was associated with an increased the prevalence of LA and LB tumors, whereas current use was inversely related to LA. Both being an ex-smoker (PR = 1.09; 95% CI: 1.06–1.12) and having a first-degree family history of BC (PR = 1.08; 95% CI: 1.03–1.12) were modestly associated with LA prevalence. Postmenopausal status (PR = 1.68; 95% CI: 1.35–2.10) and prior hormone replacement therapy use (PR = 1.61; 95% CI: 1.25–2.03) showed positive associations with TN tumors.

Conclusions/ Implications:
Our results suggest that traditional breast cancer risk factors exert distinct effects across molecular subtypes, reinforcing the heterogeneous etiologic pathways underlying the disease. Patterns involving age, BMI, hormonal exposures, smoking, and family history varied substantially by subtype, particularly for LB and TN tumors. These findings suggest that meaningful advances in risk prediction and prevention will require subtype?specific approaches rather than global strategies. Integrating these differential associations into clinical and public health frameworks could accelerate progress toward more precise breast cancer prevention

Summary of associations of Breast Cancer Risk Factors with Molecular Subtypes