Background: It is unclear whether hormone replacement therapy (HRT), in addition to increasing risk for breast cancer, affects the type of breast cancer diagnosed. We conducted this investigation to assess whether the type of hormone used (none, estrogen, progesterone, or combined) and duration of use influences subsequent breast cancer histology.
Methods: We performed a retrospective cohort analysis among women listed as incident cases of breast malignancy in the Kaiser Permanente Northern California Cancer Registry during 2003 (n = 2830). Type and duration of hormone used (none, estrogen, progesterone, or combined) before breast cancer diagnosis was obtained from electronic pharmacy records. The association between type and duration of hormone use with characteristics of subsequent breast cancers was examined.
Results: Among women aged >50 years (n = 1701), any use of estrogen, progesterone, or combination therapy was not associated with an increased risk of estrogen receptor (ER)-positive disease. However, >6 months' use of combined HRT increased the odds of ER-positive tumors (odds ratio, 1.65; 95% confidence interval, 1.07-2.5; P = .02). Estrogen HRT patients were more likely than nonusers to present with low-grade (P = .05), and early-stage tumors (P = .03). This trend was not seen in combined HRT users.
Conclusions: Short-duration HRT did not increase the likelihood of ER-positive breast cancer. However, prolonged duration of combined HRT, but not estrogen or progesterone alone, resulted in a marked increase in ER-positive disease. Our findings suggest that the effect of combined HRT on breast cancer incidence or progression is not immediate and that long-term use is more likely to affect breast cancer histology.