Pre- and Post-Menopausal Hormone Therapy Has Differential Effect on Breast Cancer Subtypes

Current use of combined estrogen-progestin therapy (EPT) is associated with significantly higher odds of developing luminal-like breast cancer subtypes in postmenopausal women, particularly among those with healthy body weight, according to a large pooled analysis published in JAMA Network Open.

Investigators found that among women with healthy body mass index (BMI; 18.5 to 25 kg m2 or less), current EPT use was associated with more than double the odds of luminal A–like tumors (OR, 2.51; 95% CI, 2.26–2.80), 1.5 times the odds of luminal B–like tumors (OR, 1.47; 95% CI, 1.17–1.86), and nearly double the odds of luminal B–ERBB2-like tumors (OR, 1.95; 95% CI, 1.61–2.37). In contrast, no significant associations were found for the ERBB2-enriched–like or triple-negative subtypes. Estrogen-only therapy (ET) and oral contraceptive (OC) use showed weaker and less consistent associations with breast cancer subtypes, though some heterogeneity was observed by BMI and duration of use.

Study authors set out to explore persistent questions about whether exogenous hormones influence the risk of breast cancer differently depending on tumor biology. Although published evidence has established that menopausal hormone therapy increases overall breast cancer risk, there is much more limited information about associations with molecular subtypes, the researchers wrote. Their analysis focused specifically on whether associations differ by intrinsic-like subtype, including luminal A–like, luminal B–like, luminal B–ERBB2-like, ERBB2-enriched–like, and triple-negative, and to determine if body mass index (BMI) modifies those associations.

For the study, Charlotte Le Cornet, PhD, an epidemiologist in the division of cancer epidemiology, at the German Cancer Research Center, Heidelberg, Germany, and colleagues pooled data from 31 case-control studies within the Breast Cancer Association Consortium, encompassing 42,269 women with breast cancer and 71,072 controls. Of the cancer cases, 23,353 had intrinsic-like subtype data. Le Cornet and the large research team used polytomous logistic regression models, adjusted for age, BMI, and hormone use history, to estimate subtype-specific odds ratios for EPT, ET, and OC exposure.

Risk Diminished with Higher BMI

The findings indicated clear subtype-specific risks associated with EPT use, and the associations diminished with increasing BMI. Among women with overweight (BMI between 25 and 30 kg/m2), current EPT use was still linked with elevated risk of luminal A–like (OR, 2.18; 95% CI, 1.83–2.61) and luminal B–ERBB2-like (OR, 1.83; 95% CI, 1.32–2.55) tumors, though the effect sizes were smaller than among healthy-weight women. In women with obesity (BMI of 30 or greater, associations persisted for luminal A–like (OR, 1.40; 95% CI, 1.02–1.92) and luminal B–ERBB2-like (OR, 1.68; 95% CI, 1.01–2.78) tumors but heterogeneity was weaker, according to the study.

Associations with ET-only showed weaker associations and only in women with healthy weight. Current ET use was linked with modestly increased odds of luminal A–like tumors (OR, 1.16; 95% CI, 1.01–1.32), with longer duration of therapy strengthening the association (OR, 1.29; 95% CI, 1.09–1.52). The investigators noted with interest that among women with overweight or obesity, ET use was associated with decreased odds of several subtypes, including triple-negative and ERBB2-enriched–like disease.

Premenopausal OC Use

Current OC use among premenopausal women was associated with luminal A–like (OR, 1.27; 95% CI, 1.05–1.54) and luminal B–like (OR, 1.46; 95% CI, 1.07–1.98) tumors. Recent OC use (less than 5 years since cessation) was linked to higher odds of triple-negative (OR, 1.75; 95% CI, 1.33–2.29), ERBB2-enriched–like (OR, 2.33; 95% CI, 1.55–3.51), and luminal B–ERBB2-like (OR, 1.50; 95% CI, 1.09–2.08) tumors. Le Cornet and colleagues noted, however, that heterogeneity by subtype was modest, and associations were not modified by BMI.

Among the study's limitations the researchers acknowledge reliance on self-reported hormone use, potential misclassification of tumor subtypes based on immunohistochemistry, and the absence of data on contemporary hormone formulations or routes of administration. Nevertheless, the large sample size and stratification by BMI are notable strengths supporting the findings, they wrote.

"Clear differences were observed in associations between current EPT use and luminal-like breast cancer subtypes compared with other subtypes, particularly for women with healthy weight," the authors summarized. "Less apparent subtype heterogeneity was observed in associations between ET use and OC use.

"These results provide further evidence of etiologic heterogeneity in breast carcinogenesis. Future studies on contemporary formulations, patterns of use, and routes of administration of exogenous hormone usage are warranted," they concluded.