Could Perimenopausal Estrogen Prevent Breast Cancer? Exploring the Differential Effects of Estrogen-Only Versus Combined Hormone Replacement Therapy

Table of Contents

HRT and Breast Cancer: The Evidence From Observational Studies

Probably the three most influential observational studies on breast cancer risk in association with HRT are the “Collaborative Re-analysis” [6] and the “Million Women” [7] studies and the meta-analysis by the Collaborative Group on Hormonal Factors in Breast Cancer in 2019 [8] (Table 1 [6-10]). The Collaborative Re-analysis of data from 51 epidemiological studies of 52,705 women with breast cancer and 108,411 women without breast cancer [6] concluded that the risk of having breast cancer diagnosed is increased in women using HRT and increases with increasing duration of use. This effect is reduced after cessation of use of HRT and has largely, if not wholly, disappeared after about 5 years. The general implication was that HRT caused breast cancer. This re-analysis of observational studies included retrospective and prospective studies. Shapiro et al (2011) [11] tested the re-analysis on the basis of standard criteria when an observational study asserts causality. They examined the following factors: time order, bias, confounding, statistical stability, strength of association, dose-duration response, internal and external consistency and biologic plausibility. They concluded that the causality link reached by the Collaborative Re-analysis was defective. This independent report means that the Collaborative Re-analysis has low scientific validity because of serious significant epidemiological faults.

The Million Women’s Study [7] was a prospective cohort of UK women aged 50 to 64 years invited to undergo screening mammography at 3-yearly intervals. Among 828,923 postmenopausal women who were current users of HRT and followed for an average of 2.6 years, the study concluded that current use of HRT was associated with increased breast cancer incidence and mortality, and the effect was substantially greater for estrogen-progestogen combinations (cHRT) than for other types of HRT. Again, the general implication was that HRT caused breast cancer. This prospective study had many methodological shortcomings and the most cogent was that it did not exclude breast cancers that appeared within 1 year, as they were most likely to have been present at baseline. Shapiro et al (2011) [12] also tested the Million Women’s Study on the basis of standard criteria when an observational study asserts causality. They examined the following factors: time order, bias, confounding, statistical stability, strength of association, dose-duration response, internal and external consistency and biologic plausibility. They concluded that HRT may or may not increase the risk of breast cancer, but the Million Women’s Study did not establish that it does. The causality link was unreliable because of defects in quality of design, execution, analysis and interpretation. They commented that size alone did not guarantee that the findings are reliable. This independent report means that the Million Women’s Study again has little scientific validity because of serious epidemiological faults. However, the Million Women’s Study differentiated that there was a lower risk of incident breast cancer between women on ERT or tibolone compared to women on cHRT [7].

The recent meta-analysis from the Collaborative Group on Hormonal Factors in Breast Cancer in 2019 [8] again reviewed the evidence from epidemiological rather than interventional studies, and results are not in keeping with the randomized Women’s Health Initiative (WHI) trials which have recently reported that women receiving estrogen post hysterectomy had a lower long-term risk of breast cancer.