New Estimates for Breast Cancer Risk With HRT

Veronica Hackethal, MD

October 29, 2020

A new, large study has confirmed that different types of hormone replacement therapy (HRT) are associated with an increased risk for breast cancer and has provided additional information on factors associated with that increased risk.

The study was published online on October 28 in The BMJ.

"The study confirms increased risk of breast cancer in patients taking HRT, but shows that the magnitude of risk depends on a number of factors," first author Yana Vinogradova, PhD, told Medscape Medical News via email. Vinogradava is a medical statistician at the University of Nottingham in the United Kingdom.

The study also suggests that the risk may be lower than was estimated in a large meta-analysis of 24 trials that was published in 2019 in The Lancet. In that study, researchers suggested that the risk for breast cancer with HRT was higher and persisted longer than had been thought.

This conclusion from the meta-analysis was widely reported in the lay press and led to the UK Medicine and Healthcare Products Regulatory Agency issuing a safety alert for HRT regarding breast cancer. Experts in the field questioned the alert and said it caused undue anxiety. The European Medicines Agency also issued a safety alert because of the study.

This new study was begun before publication of the meta-analysis. Although the results are broadly similar in suggesting increased risk for breast cancer with HRT use, findings from the new study suggest that the risk is lower than had been estimated in the meta-analysis and that the risk diminishes more rapidly after stopping HRT than was suggested by the meta-analysis.

"The publicity surrounding publication of the meta-analysis highlighted unexpectedly high risks and led to a heightened level of concern in some quarters," Vinogradova commented. "Our study, based on general population data, has not confirmed any such findings. In general, it showed lower levels of risk and clarified the variability of magnitude within them."

Vinogradova said the discrepancy could be related to the fact that the studies were designed differently. The meta-analysis relied on results from 24 studies that were conducted around the world at different periods and included women of different ages and backgrounds. The studies in the meta-analysis used different methods, including questionnaires that relied on women's memories and so could have been biased, she said.

In contrast, the new study analyzed electronic medical record data collected prospectively by general practices in the United Kingdom. The data came from the QResearch and Clinical Practice Research Datalink (CPRD) databases, the two largest primary care databases in the United Kingdom, which were linked to hospital, mortality, and cancer registries.

Because this study used a "consistent design" and "consistent data sources," these new results "are likely to be more accurate and reliable for assessing risks among HRT users," Vinogradova commented.

This study used an observational design and so cannot prove that HRT causes breast cancer. These results may better represent women in the general UK population compared to the earlier meta-analysis, she added.

Commenting on the new study, Michael Jones, PhD, senior staff scientist in genetics and epidemiology at the Institute of Cancer Research, London, also emphasized that the study was large and that the data came from general practioner medical records, "so the strong statistical associations are unlikely to be due to chance.

"The results of this study generally confirm what has been seen before and is well established ― that the use of combined estrogen plus progestogen HRT is associated with increased risk of breast cancer, and this risk increases with duration of use. But reassuringly, after stopping HRT, the raised risk of breast cancer mostly returns to that seen in nonusers of HRT," he said.

"It's important to note that no one study should be considered in isolation," he added. "Even though some risks were found to be slightly smaller than those reported in another meta-analysis of the worldwide epidemiological evidence recently published in 2019, women considering use of HRT should still follow advice given to them by their GPs."

Study Details

In the study, researchers evaluated all types of HRT commonly prescribed in the United Kingdom over the past 20 years, including topical estrogen, vaginal pessaries, and creams. They grouped HRT use by recent (within the past 5 years) and past (5 or more years ago) and HRT duration as short term (less than 5 years) and long term (5 years or longer). Results were adjusted for a range of factors that could affect breast cancer risk, including lifestyle, smoking, alcohol consumption, other medical conditions, family history, and use of other prescribed drugs.

The analysis included 98,611 women aged 50 to 79 years who were first diagnosed with breast cancer between 1998 and 2019. These women were matched by age and general practice to 457,498 women who were not diagnosed with breast cancer over these years. HRT use was reported in 34% (33,703) of women with breast cancer and in 31% (134,391) of women without breast cancer.

Overall, the risk for breast cancer was increased with use of most HRT drugs (adjusted odds ratio [OR], 1.21; 95% CI, 1.19 – 1.23), compared to not using HRT drugs. The highest risk was tied to combined estrogen-progestogen HRT (adjusted OR, 1.26; 95% CI, 1.24 – 1.29). The lowest risk was tied to estrogen-only HRT (adjusted OR, 1.06; 95% CI, 1.03 – 1.10). Estrogen cream and vaginal estrogen were not associated with increased breast cancer risk.

In general, breast cancer risk was higher among recent HRT users and those receiving long-term therapy. HRT-associated breast cancer risk increased with age and declined after discontinuing treatment. Therapy of less than 1 year was not associated with increased breast cancer risk.

Women who had recently been receiving long-term combined estrogen-progestogen HRT had a 79% increased risk for breast cancer (adjusted OR, 1.79; 95% CI, 1.73 – 1.85) compared to never-users. Among recent long-term users of combined HRT, breast cancer risk was highest for norethisterone (adjusted OR, 1.88; 95% CI, 1.79 – 1.99) and lowest for dydrogesterone (adjusted OR, 1.24; 95% CI, 1.03 – 1.48).

Women who had recently been receiving long-term estrogen-only HRT had a 15% increased risk for breast cancer compared to never-users (adjusted OR, 1.15; 95% CI, 1.09 – 1.21).

Among women who discontinued HRT 5 or more years ago, risk for breast cancer was no longer increased for long-term estrogen-only therapy and short-term estrogen-progestogen therapy. However, breast cancer risk remained elevated 5 years after discontinuing long-term estrogen-progestogen (adjusted OR, 1.16; 95% CI, 1.11 – 1.21).

HRT-associated risk for breast cancer increased with age across all durations of therapy.

Compared to never-use, recent long-term estrogen-only therapy was associated with 0 extra breast cancer cases per 10,000 women years among women aged 50 to 59 years and 8 extra cases per 10,000 women years among women aged 70 to 79.

Recent long-term estrogen-progestogen use was associated with 15 extra breast cancer cases among women aged 50 to 59 and 36 extra cases per 10,000 women years among women aged 70 to 79.

Past long-term estrogen-progestogen use was associated with 0 extra breast cancer cases among women aged 50 to 59 and 8 extra cases per 10,000 women years among women aged 70 to 79.

Summarizing, Vinogradova said that the increased risk for breast cancer with HRT appears to be "relatively small, particularly for younger women and for any women who use HRT only for a restricted period."

Decisions about whether to use HRT and which type to use should depend on symptom severity, patient factors, and suitability of other treatment options, she commented.

"Particularly for those women who our study has shown to be most at risk, these decisions should be made through discussions between the patient and her doctor," she concluded. "We hope that the new and more detailed information provided by our study will facilitate such prescribing decisions."

The study was partially funded by the School for Primary Care Research of the National Institute for Health Research, by Cancer Research UIK, and by the Cancer Research UK Oxford Center. Vinogradova has disclosed no relevant financial relationships. Senior author Julia Hippisley-Cox is an unpaid director of QResearch and was a paid director of ClinRisk until 2019. The other authors have disclosed no relevant financial relationships.

BMJ. Published online October 28, 2020. Full text

For more from Medscape Oncology, join us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: