Background: The effects of use of different types of hormone therapy (HT) on breast cancer risk according to prognostic factors are largely unknown. Methods: We linked data from the Norwegian Prescription Database and the Cancer Registry of Norway during 2004-2009 on all women aged 45-79 (N=686,614). We estimated rate ratios (RR) and 95% confidence intervals (CI) for breast cancer in relation to HT using Poisson regression. Results: During an average 4.8 years of follow-up, 7,910 invasive breast cancers were diagnosed. Compared with non-users of HT, users of estradiol and tibolone were more likely to be diagnosed with grade I, lymph node negative, and ER+/PR+ tumors. However, compared with non-users, users of the most common estrogen and progestin combinations (estradiol-norethisterone (NETA) preparations (Kliogest®, Activelle® or Trisekvens®)), were at a 4 to 5-fold elevated risk of grade I tumors, 3-fold elevated risk of lymph node negative tumors, and 3 to 4-fold elevated risk of ER+/PR+ tumors. Importantly, estradiol-NETA users were also at a 2 to 3-fold increased risk of medium differentiated (grade II) tumors and tumors with lymph node involvement. Conclusions: Use of oral estradiol, tibolone as well as estradiol-NETA predominantly increases risk of breast cancer with favorable prognosis characteristics. However, use of estradiol-NETA preparations also increases the risk of breast cancers with less favorable characteristics. Impact: The HT preparations most commonly used in the Nordic countries are associated with both breast cancers with good and less favorable prognosis characteristics.
- Received March 24, 2016.
- Revision received July 6, 2016.
- Accepted July 16, 2016.
- Copyright ©2016, American Association for Cancer Research.