Use of hormone replacement therapy and risk of breast cancer: nested case-control studies using the QResearch and CPRD databases
BMJ 2020; 371 doi: https://doi.org/10.1136/bmj.m3873 (Published 28 October 2020) Cite this as: BMJ 2020;371:m3873Linked Practice
Risk of breast cancer with HRT depends on therapy type and duration
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Dear Editor
As stated in our response to earlier comments, the aim of this study was to provide more detailed information about excess breast cancer risks associated with all of the commonly used HRT treatments. This type of information has been noted as needed in the current NICE guidelines on treatment of menopause symptoms.[1, 2]
The relationships between HRT use and excess breast cancer risks are complex, and depend on the type of treatment, duration and recency. Our study has shown that excess risks increase with duration of exposure at different rates for different HRT treatments, and then disappear over time after discontinuation, fitting the causality principle.[3] It is true that observational studies do not directly measure causality. Citing one rather low overall odds ratio (for ever use) from the abstract is not, however, a sustainable argument against a potential causal relationship. We also report larger odds ratios for recent use of oestrogen-progesterone combinations taken for longer periods of time. Although there are clearly other risk factor/disease combinations with higher levels of association, this also does not invalidate our study.
The case-control study design is a generally-accepted alternative to a cohort study.[4] For the nested case-control study design with incidence density sampling of controls, which we used, it is well established that the odds ratio directly estimates – and can be interpreted as – a rate ratio which might be obtained from a cohort study.[5] In this study the rate ratio is for breast cancer in women exposed to HRT compared with women who were not exposed. So Dr Manyonda is incorrect in saying that the study cannot measure the increased risks of breast cancer. Confidence intervals are added to show the precision of the estimated odds ratios.
The study was based on an open cohort of patients, who were followed from the latest of: date of patient registration with the general practice; date from which the practice contributed records to the database; the start date of the study period. Patients were followed until they had a diagnosis of breast cancer or left the study for any other reason. This also fits the definition of a prospective study.
Selecting cases and matching them with selected controls is a standard methodological approach. The study was based on information reflecting exposure to HRT hormonal treatments for different durations and at different times. It is not designed to assess average lifetime risks – instead it provides estimates of current risk for a woman who, for a specific duration, has used, or is still using, a particular HRT treatment.
In relation to randomised trials, there are many different types of HRT treatment in the modern world, each using different hormones and having specific side effects. So, for practical reasons, it would infeasible to use a randomised control trial for a comprehensive study of a rare outcome and include all commonly used HRT formulations. Only one RCT on this topic has sustained the required numbers of participants (Women's Health Initiative),[6] but it investigated only two hormones.
Press releases were provided by Nottingham University and The BMJ, both highlighting that our study did not confirm ‘the higher than expected’ findings of the 2019 Oxford meta-analysis, which was widely reported and followed by a health warning. We are not responsible for journalists’ errors based on their own attempted comparisons, particularly when not checked with us prior to publication. The Telegraph online article has, however, been corrected[7] and the error acknowledged in both printed and online editions.
We do our best to explain in lay terms our research results and their correct interpretation, using podcasts, working with journalists and responding to queries. We have stressed the relatively low magnitudes of most excess risks on many occasions, and we always advise women to discuss any concerns with their doctor. We agree that life style factors can be important with respect to risk of developing breast cancer. It is also, however, useful for women, when considering the benefits and risks of HRT treatments, to have the best possible and most detailed information on excess risk levels for the same outcome.
1. National Collaborating Centre for Women’s and Children’s Health, Menopause. Full guideline. National Institute for Clinical Excellence, 2015(Version 1.5).
2. National Institute for Clinical Excellence, Menopause: diagnosis and management. NICE guideline NG23, 2015.
3. Hill, A., The Environment and Disease: Association or Causation? Proc R Soc Med, 1965. 58(5): p. 295-300.
4. Coggon, D., Geoffrey Rose, and D. Barkey. Chapter 8. Case-control and cross sectional studies. In Epidemiology for the uninitiated. 2020; Available from: https://www.bmj.com/about-bmj/resources-readers/publications/epidemiolog....
5. Pearce, N., What does the odds ratio estimate in a case-control study? Int J Epidemiol, 1993. 22(6): p. 1189-92.
6. Chlebowski, R.T., et al., Breast Cancer After Use of Estrogen Plus Progestin and Estrogen Alone: Analyses of Data From 2 Women’s Health Initiative Randomized Clinical Trials. JAMA Oncology, 2015. 1(3): p. 296-305.
7. Roberts, L. Breast cancer risk from HRT lower than previously feared, The Telegraph 29/10/2020. Available from: https://www.telegraph.co.uk/news/2020/10/29/women-hrt-far-likely-develop....
Competing interests: No competing interests
Dear Editor
Advice on avoiding menopausal symptoms without using potentially carcinogenic HRT
‘Pill’ or menopausal or HRT withdrawal symptoms can be prevented by avoiding alcohol, tobacco smoking and eating a stone age rotation diet of lean meat (such as lamb, beef, venison, pork, chicken or other poultry) or fish with vegetables and salads. Eat meals three times a day. The commonest foods causing allergies are wheat, yeast, corn, eggs, milk , oranges and coffee, tea or chocolate. Rotate different fruits unless thrush or gut candida needs to be treated first with antifungal medication. Probiotics can help with absorption of food from the gut and preventing gut candida.
Drink only spring water from a reputable commercial source. Use glass bottles to avoid contamination with oestrogenic substances from plastic bottles. Taking ‘Pill’ or HRT steroids increases the risk of essential nutrient deficiencies and food allergies which may be overt or masked - which is usually a consequence of frequent repeated ingestion of the same food. It is important to take a good mineral and vitamin supplement to replace common post ‘pill’ deficiencies which often include zinc, magnesium, copper, selenium, chromium, B vitamins, vitamin C, Vitamin D, Vitamin E complex, omega-3 and omega-6 essential fatty acids. Some private laboratories can measure mineral, vitamin and essential fatty acid levels which is very helpful for the best preconception care.
Competing interests: No competing interests
Dear Editor
HRT are highest level carcinogens
The sole aim of Vinogradova and colleagues was to deliver comprehensive information to help individual women and their doctors make better-informed treatment choices by providing risk estimates about additional risks of breast cancer with all commonly-used HRT treatments. Unfortunately, it has been a mistake to believe that progesterones and oestrogens, are needed to treat a long-evolved physiological event as HRT hormones are highest level carcinogens.[1]
In reality, a normal menopause protects older women from the up and down regulation of thousands of genes by high hormones levels in normal cycles or in pregnancies. Menopausal symptoms are abnormal and usually due to essential nutrient deficiencies, food or chemical allergies, alcohol drinking or the auto-brewery syndrome (increased fungal gut dysbiosis). Also increases in toxic DNA adducts are common due to cadmium from smoking, mercury from dental amalgams and nickel from stainless steel cookware or cheap jewellery and commonly used aluminium.
Dr Peter Lewis refers to outdated observation studies. He seems reluctant to accept that The Collaborative Group on Hormonal Factors in Breast Cancer reviewed all the available relevant "randomised" evidence and found risks did not generally differ between different progestagenic constituents. The Group found that micronized progesterone and oestrogen HRT significantly doubled the risk of breast cancer (RR 2.05 (1.38-3.56). [2,3]
Doctors should be welcoming the falls in HRT use (from 36 million to 12 million women in Westernised countries) after the early terminations of the Women’s Health Initiative HRT Studies. The consequent large decrease in both breast cancer and ovarian cancer incidence and mortality is great news. Now is not the time to keep flogging the disgraced HRT band wagon!
1 IARC. Combined estrogen-progestogen contraceptives and combined estrogen-progestogen menopausal therapy. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans 2007; Volume 91.
2 Collaborative Group on Hormonal Factors in Breast Cancer (2019) Type and timing of menopausal therapy and breast cancer risk individual participant meta-analysis of the worldwide epidemiological evidence. Lancet 394: 1159-1168.
3 Beral V, Peto R, Pirie K, Reeves G. Menopausal hormone therapy and 20-year breast cancer mortality. Lancet Published Online August 29, 2019 http://dx.doi.org/10.1016/ S0140-6736(19)32033-12
Competing interests: No competing interests
Dear Editor
We welcome the responses to our article on HRT and risk of breast cancer.
This study addressed research needs highlighted in current NICE guidance for menopause.[1, 2] As in our earlier paper on risks of VTE associated with HRT,[3] no other issues related to HRT use were studied. Our sole aim was to deliver comprehensive information to help individual women and their doctors make better-informed treatment choices. The findings provide estimates for additional risks of breast cancer associated with all commonly-used HRT treatments in UK NHS primary care settings over the past 20 years. We know many women will welcome this information when weighing up the benefits and risks of different treatment options.
Specific responses:
Michael Baum considers our paper ‘alarmist’. In fact, we report findings in terms of absolute numbers in order to make magnitudes of additional risk clearer. Many past studies have identified increased risks of breast cancer associated with HRT – we highlight that our study did not confirm the ‘higher than expected’ findings of the 2019 meta-analysis[4] (which caused a health warning[5]) and report faster risk declines once HRT has stopped. He describes us as dismissive of RCTs when we merely noted the impossibility of using RCT design for this study – in some circumstances RCTs are the only appropriate research design, and we consider RCT and observational studies to be complementary rather than in opposition.
He also incorrectly suggests: A) we failed to take account of the 2017 Women’s Health Initiative findings)[6] (these related to mortality outcomes, so no comparison would be useful); B) we failed to report on mammography despite it having been identified as a confounder (mammography findings are in Table 1 and Supplementary eTable1); and C) we failed to report on mortality despite having said ‘that mortality was included amongst the outcome measures’ (this was not a mortality study, so mortality was not an outcome measure).
Finally, he contrasts the low excess risks we report with the severity of some menopausal symptoms. This is precisely why our estimates of absolute excess risk should be useful for women considering HRT. They provide non-polemical and potentially reassuring information for women who may have severe menopausal symptoms but are concerned about developing breast cancer.
Avrum Bluming argues that we report low increased risks of breast cancer associated with HRT but fail to mention ‘an estimated 90% cure rate for newly diagnosed breast cancer’ or to discuss HRT’s benefits for menopause symptoms and other morbidities. He suggests our purpose was to ‘decrease anxiety regarding HRT promoted by [the 2019 meta-analysis]’,[4] but argues that the meta-analysis and our study represent ‘small and challengeable increases of risk of a largely curable tumor’, and ‘cause unnecessary alarm,’ citing some spurious factors having ‘stronger reported correlation’ with breast cancer.
Our paper provides information on HRT associations with breast cancer incidence and the introduction addresses benefits with respect to menopausal symptoms. It was not a risk-benefit assessment of HRT. Likening large, well-designed, peer reviewed studies in major journals to reports of statistically invalid random associations does not advance serious discussion. We are also of the opinion that a diagnosis of breast cancer – curable or not – is a significant threat to a woman’s wellbeing.
Ellen Grant questions whether menopause lasts ‘from age 50 to 79 years’. The age range of our study, however, reflects the age range over which some women are using HRT as recorded in NHS primary care data, not a view either on how long menopause lasts or for how long HRT should be used. She also suggests our aim was to find ‘which progestogen caused the least breast cancer’. Although an objective was to investigate differences between treatments, our aim was as described earlier. Finally, she questions the exclusion of women ‘who may have used HRT in the past year which would lower breast cancer numbers.’ This introduced lag is, however, a mechanism to handle protopathic bias, a known issue in drug-cancer association studies.[7]
Peter Lewis notes, importantly, that ‘progestogen’ includes both synthetic progestogens (progestins) and natural progesterone, but that the combined HRT treatments in our study included only progestins. He cites studies showing that combinations using natural (micronised) progesterone have much lower excess breast cancer risks.
Our study, however, was data driven. To deliver information about comparative risks, we could include only sufficiently prescribed treatments. The numbers of prescriptions and cases relating to micronised progesterone were simply insufficient. Our data reflects historical very low usage of micronised progesterone in the context of NHS primary care, with no evidence of steady increase. Why, and what benefits its wider use might bring, were not within our study scope, and further investigation seems appropriate.
Ellen Løkkegaard suggests the study data lacked ‘quite a lot’ of exposure information ‘prior to 1998’. A small proportion of women had missing historical data for HRT use because it was recorded only from the late 1980s. This would chiefly affect estimates for past usage, but an analysis including only women having complete 10-year record histories yielded essentially the same results.
She also comments that we define HRT exposure differently from other studies – specifically because we describe use within 5 years as ‘recent/current’ rather than ‘current’ and because we include a sensitivity analysis with a redefinition of recent/current as ‘within 2 years’. She suggests that analyses based on the 2 year definition provide comparable estimates to those in the 2019 meta-analysis,[4] but were not included in eTable 16.
Use of ‘recent/current’ rather than ‘current’ simply reflects our non-inclusion of prescriptions within the year prior to the index date to avoid protopathic bias. The analysis of use ‘within 2 years’ was included both to demonstrate risk declines following treatment cessation and to provide comparable figures for studies with similar definitions. Regarding valid risk estimate comparisons with other studies, treatment formulation, exposure recency and duration, and sub-cohort age all have to be considered. The meta-analysis[4] main findings were for women, 55 and older, who had used HRT within the last 5 years, so our comparisons used that context.
References
1. National Collaborating Centre for Women’s and Children’s Health. Menopause. Full guideline. National Institute for Clinical Excellence 2015(Version 1.5)
2. National Institute for Clinical Excellence. Menopause: diagnosis and management. NICE guidline NG23 2015
3. Vinogradova Y, Coupland C, Hippisley-Cox J. Use of hormone replacement therapy and risk of venous thromboembolism: nested case-control studies using the QResearch and CPRD databases. BMJ 2019;364:k4810. doi: 10.1136/bmj.k4810
4. Collaborative Group on Hormonal Factors in Breast Cancer. Type and timing of menopausal hormone therapy and breast cancer risk: individual participant meta-analysis of the worldwide epidemiological evidence. The Lancet 2019;394(10204):1159-68. doi: 10.1016/S0140-6736(19)31709-X
5. Rymer J, Brian K, Regan L. HRT and breast cancer risk. BMJ 2019;367:l5928. doi: 10.1136/bmj.l5928
6. Manson JE, Aragaki AK, Rossouw JE, et al. Menopausal Hormone Therapy and Long-term All-Cause and Cause-Specific Mortality: The Women’s Health Initiative Randomized Trials. JAMA 2017;318(10):927-38. doi: 10.1001/jama.2017.11217
7. European Network of Centres for Pharmacoepidemiology and Pharmacovigilance. ENCePP Guide on Methodological Standards in Pharmacoepidemiology. 4.2.2.4. Protopathic bias. Available from: http://www.encepp.eu/standards_and_guidances/methodologicalGuide4_2_2_4.....
Competing interests: No competing interests
Dear Editor
Vinogradova et al’s publication [1] on the use of hormone replacement therapy (HRT) and the risk of breast cancer is another in a long list of periodic publications that cause alarm in thousands of women, often compelling them to abandon a treatment which, in the considered opinion of the National Institute for Health and Care Excellence (NICE), should be offered to most women as the potential benefits far outweigh the risks [2]. Yet this paper has fundamental flaws, especially in its methodology.
The paper addresses the issue of causality of breast cancer by HRT. There are established and defined rules for assessing causality of disease [3] from an exposure including a dose-response relationship. Among other issues, the most important defined rule is that with observational epidemiological research, whether prospective or retrospective, this requires other powerful factors when a cause-effect relationship is evaluated. For example, after statistical significance of an association, the strength of association between HRT use and incidence of breast cancer should be strong. Overall, Vinogadova, Coupland and Hippisley-Cox [1] report that their observational study showed that exposure to HRT was associated with an increased risk of breast cancer (adjusted odds ratio 1.21, 95% confidence interval 1.19 to 1.23). The increased risk was mostly attributable to oestrogen-progestogen therapy (1.26, 1.24 to 1.29), with oestrogen only therapy showing a small increased risk (1.06, 1.03 to 1.10), both compared to never users. While a 21% increase in odds ratio (OR 1.21) appears high when the incidences of breast cancer are compared between HRT users and never-users, this increase is very small in the context of epidemiological studies because they do not directly measure causality because of inherent weaknesses like bias and confounding. In epidemiological studies that have stood the test of time, the magnitude of increases in the risk is much more robust. For example, the increase in mortality from lung cancer in smokers compared to non-smokers was 810% (RR = 9.1) [4] and the increase in mortality from laryngeal cancer in smokers compared to non-smokers was 1900% (RR 20). [5]
In this paper the study design (nested case-control) starts from the disease (breast cancer) compared to non-disease individuals. Therefore, it cannot then measure the increased risks of breast cancer. It can only measure the odds of using combined HRT (cHRT) or Estrogen-only HRT (ERT) or compared to each other. There is also confusion about the concept of the confidence interval (CI). A CI between two comparisons, in this situation, which can only measure the odds ratio (as a measure of risk) of using HRT if a woman was selected on the basis of having breast cancer or not, that odds ratio only measures the risk of using HRT, never the risk of breast cancer. No matter how high the odds ratio is, it is only important in what it is measuring, the odds ratio of using HRT.
Furthermore, a reference to meta-analysis of observational studies is always faulty whenever it is not based on a prospective randomized study. A systematic review is required as a starting point on the assumption that the prospective randomized trial would have minimized bias and confounding factors. These are not factors that are minimized with every prospective study except if they are randomized controlled trials which can only be prospective. A prospective cohort study does not automatically qualify as a systematic review. Then the meta-analysis can be performed as the direction of travel has started from exposure to outcome, on similar measurement variables.
The authors state that their findings were based on prospectively collected electronic health records from the two largest UK primary care databases linked to secondary care data sources. This does not satisfy the definition of prospectively collected data, which must follow the set up of the database before 1998. The authors need to clarify if this study was truly ‘prospective’. They must then go on to say that they did follow the women prospectively from the start of HRT or assessed average lifetime risks. This clearly could not have been the case.
The authors justify their approach in this study with the statement that randomized trials using enrolled participants are now impractical to investigate the risks of breast cancer associated with HRT because of the numbers required. Such a view is simply unsustainable in the light of the publications by Chebloskwi et al, [6] who concluded in a randomized controlled trial that the risk of breast cancer was increased with cHRT after 5 years of use (Hazard ratio (HR 1.26 95% CI 1.02, 1.56, P = 0.04) and also concluded that the risk of breast cancer was not increased after 7.2 years with ERT (Hazard ratio 0.76, 95% CI 0.58-0.94). Similarly, they published that after 16.1 years long-term follow-up [6] the incidence of breast cancer following ERT was significantly reduced (0.76, 95% confidence interval 0.58, 0.98, P = 0.04). With such cogent data, it is difficult to understand why Vinogradova et al have used less rigorous methods to address this very important question pertaining to the commonest cancer in British women.
The increased risk of breast cancer from cHRT is associated with the progestogen component in cHRT, and there are alternative ways of providing progestogens to minimize the risk. [7] It is regrettable when publications fail to present the risks of breast cancer from HRT in context: immediately following Vinogradova et al’s paper, the Telegraph published a piece stating that HRT increases a woman’s breast cancer risk by 80%. The absolute risk from HRT is small, and when seen in the context of risks from other life-style factors such as smoking, obesity and alcohol, women might not be so alarmed [8] (Figure 1). Both women and their healthcare providers are often unaware of these life-style risks, knowledge of which would empower them to make better decisions than the knee-jerk avoidance of their HRT every time an alarmist publication hits the press. The health profession and researchers can and should do better for women.
References
1. Use of hormone replacement therapy and risk of breast cancer: nested case-control studies using the QResearch and CPRD databases. Vinogradova Y, Coupland C, Hippisley-Cox J. BMJ 2020;371:m3873. http://dx.doi.org/10.1136/bmj.m3873.
2. NICE guideline [NG23]. Menopause: diagnosis and management, www.nice.org.uk/guidance/ng 23 (2015, accessed 30 January 2020).
3. Onwude JL. HRT use causes ovarian cancer. BMJ 2015; 350: h840.
4. Doll R and Hill AB. A study of the aetiology of the carcinoma of the lung. Br Med J, 1952: 1271, 2.
5. U.S. D.H.H.S. The Health consequences of smoking: Cancer. A Report of the Surgeon-General. Rockville, MD: Office of Smoking and Health, 1982.
6. Chlebowski RT, Garnet L Anderson GL, Aragaki AK et al. Long-term influence of estrogen plus progestin and estrogen alone use on breast cancer incidence: The Women’s Health Initiative Randomized Trials. San Antonio Breast Cancer Symposium, December 13, 2019.
7. Manyonda I, Talaulikar VS , Pirhadi R, Onwude JL. Progestogens are the problem in hormone replacement therapy: Time to reappraise their use. Post Reproductive Health https://orcid.org/0000-0002- 2058-722X.
8. Women’s Health Concern. Factsheet: Understanding the risks of breast cancer. A comparison of lifestyle risk factors versus Hormone Replacement Therapy (HRT) treatment. March; 2017, Buckinghamshire, UK. https://the bms.org.uk/wp-content/uploads/2016/04/WHC-Understanding Risks of BreastCancer-MARCH2017.pdf.
Competing interests: No competing interests
Dear Editor
Vinogradova and colleagues [1] reflect in their discussion on their different finding from the Lancet meta-analysis from 2019 [2]. Apart from their left censored data omitting quite a lot of prescribed menopausal hormone therapy prior to 1998, they defined menopausal hormone therapy exposure quite differently from the usual categorization as ever, current or past use in the observational studies.
They used the definition recent use within the last 5 years as recent/current use and past use 5 years after the last issued prescription [3]. In a sensitivity analysis they redefine recent use as till within 2 years and past useas till after 2 years (e table 8) [1].
Analyses based on the 2 year definition provide quite comparable estimates in the meta-analysis, however they did not include the estimates in the e table 16, where they compare their result among women aged 55 and older to results from the meta-analysis.
For example, the estimate in the main publication for 5-9 years of combined oestrogen and progestogen is 1,70 (1,64-1,76) with the 5 year definition and 1,99 (1,91-2,08) in the sensitivity analysis with the 2 year definition, quite similar to the meta-analysis estimate 1,97 (1,89-2,04). Also, the past use risk with the 2 year definition is quite comparable to findings from the meta-analysis.
I wonder why the main article is based on the 5 year definition: it confuses the discussion.
1. Vinogradova Y, Coupland C, Hippisley-Cox J.Vinogradova Y, et al. BMJ. 2020 Oct 28;371:m3873. doi: 10.1136/bmj.m3873. Use of hormone replacement therapy and risk of breast cancer: nested case-control studies using the QResearch and CPRD databases.
2. Collaborative Group on Hormonal Factors in Breast Cancer.Collaborative Group on Hormonal Factors in Breast Cancer. Lancet. 2019 Sep 28;394(10204):1159-1168. doi: 10.1016/S0140-6736(19)31709-X. Epub 2019 Aug 29. Type and timing of menopausal hormone therapy and breast cancer risk: individual participant meta-analysis of the worldwide epidemiological evidence.
3. Vinogradova Y, Coupland C, Hippisley-Cox J. Protocol to assess risk of breast cancer associated with use of hormone replacement therapy in real world settings: two nested case-control studies in primary care. 2019. https://nottingham-repository.worktribe.com/output/2309731.
Competing interests: No competing interests
Dear Editor
Dr Peter Lewis responds that progestogens increase the risk of breast cancer but not "natural" progesterone. He ignores the fact that both act by stimulating progesterone receptors and that oral progesterone is micronized to aid alimentary absorption and double its progestogenic potency. Also he ignores the fact that The Collaborative Group on Hormonal Factors in Breast Cancer reviewed "all the available relevant randomised evidence” and found risks did not generally differ between different progestagenic constituents. In fact, micronized progesterone and oestrogen significantly doubled the risk of breast cancer (RR 2.05 (1.38-3.56). [1]
It should never be forgotten that in 1896 Colonel Sir George Thomas Beatson discovered that oophorectomy, which stopped the endogenous secretion of both oestrogens and progesterone from functioning ovaries, remitted metastatic breast cancers. Beatson is considered the father of anti-hormonal treatment of breast cancer as oophorectomy became the standard treatment for advanced breast cancer.
Women are still being overwhelmed by HRT promotions, although there is irrefutable evidence that falls in HRT use lowers both breast cancer incidence and mortality. [1,2]
1 Collaborative Group on Hormonal Factors in Breast Cancer (2019) Type and timing of menopausal therapy and breast cancer risk individual participant meta-analysis of the worldwide epidemiological evidence. Lancet 394: 1159-1168.
2 Beral V, Peto R, Pirie K, Reeves G. Menopausal hormone therapy and 20-year breast cancer mortality. Lancet Published Online August 29, 2019 http://dx.doi.org/10.1016/ S0140-6736(19)32033-12
Competing interests: No competing interests
Dear Editor
Vinogradova and colleagues state that ‘combined oestrogen and progestogen therapy [was] associated with increased risk of breast cancer’ (1). It is important to note that ‘progestogen’ is an umbrella term which includes progestins (synthetic progestogens) and natural progesterone. All of the progestogens in the study were synthetic (medroxyprogesterone, levonorgestrel), norethisterone or dydrogesterone); no combination included progesterone.
A large French study which assessed and compared the association between different HRTs and breast cancer risk, followed up 80,377 women for an average of 8.1 post-menopausal years, and found that compared with HRT never-use, there was no increased risk of breast cancer for oestrogen-progesterone (relative risk 1.00), whereas those using oestrogen plus progestins had a 16-69% increased risk of breast cancer (depending on the progestin used) (2).
Another French study also found that breast cancer risk differed by type of progestogen among current users of oestrogen-progestogen therapies. No increased risk was apparent among users of oestrogen + natural micronised progesterone (in fact, there was a 20% lower risk for any duration of use, compared with never-users of HRT) (3).
A meta-analysis of studies of postmenopausal women using progesterone vs. synthetic progestins in combination with oestrogen found that progesterone-oestrogen was associated with a 33% lower risk of breast cancer compared with synthetic progestins (4).
In conclusion, this study confirms an increased risk of breast cancer associated with long term use of oestrogen-only therapy, and an even bigger risk with oestrogen-PROGESTIN therapy. The evidence from other studies demonstrates that oestrogen plus micronised progesterone is much safer than oestrogen-progestin combinations.
1. BMJ 2020; 371 doi: https://doi.org/10.1136/bmj.m3873
2. Breast Cancer Res Treat, 2008;107(1):103-11
3. PLoS ONE, 2013; 8(11): e78016. doi:10.1371/journal.pone.0078016
4. Systematic Reviews, 2016; 5:121; DOI 10.1186/s13643-016-0294-5
Competing interests: No competing interests
Dear Editor
HRT is associated with increased breast cancer and death
Vinogradova and colleagues believe that HRT relieves menopausal symptoms over prolonged periods. Does the menopause really last from age 50 to 79 years? [1]
HRT does not improve the overall quality women’s lives but stopping HRT can be life changing for the better. HRT use fell from 36 million to 12 million women in western countries over the past two decades following the early termination of the Womens’ Health Institute randomized double blind trials of combined and oestrogen-only HRT because of life threatening increases in cancers, heart attacks, strokes and thrombosis. [2,3] The consequent large falls in breast and ovarian cancer incidences and mortality have certainly been better for women’s health. [4-7]
Vinogradova’s case-control study tried to find which progestogen caused the least breast cancer. However, all combination HRTs act predominantly like progesterone to avoid increases in endometrial cancer. Also the authors excluded women who may have used HRT in the past year which would lower breast cancer numbers because risks increase with recent and longer HRT use.
In contrast, in the 2019 Collaborative Groups study of 108 647 postmenopausal women, breast cancer incidences doubled for up 10 years of use of combined HRT to one in every 25 women.[8] This is the kind of large increase which led to breast cancer screening in the first place. Breast cancer mortality also increased.[9]
The menopause is a normal physiological event, which protects older women from having high levels of reproductive hormones. However, menopausal symptoms like flushing can result from sharp falls in steroid hormone levels increasing adverse reactions to foods and chemicals causing vascular over-reactivity. It is important to avoid tea, coffee, smoking tobacco or drinking alcohol. Also, the common auto-brewery syndrome needs antifungal treatment for fungal gut dysbiosis to prevent increased blood alcohol levels, when sugary foods are taken, causing flushing, headaches or migraines.[10-12]
Both hormonal contraceptives and HRT can lower white cell zinc levels, red cell magnesium levels and raise serum copper levels which can reduce copper stores and impair superoxide dismutase function and also increase carcinogenic DNA adducts.[13-15] Zinc is essential for brain function and necessary for hundreds of enzymes. Zinc and magnesium deficiencies can block omega-3 and omega-6 essential fatty acid pathways. Zinc and magnesium deficiencies also increase the risk of headaches, vasomotor symptoms like hot flushes, mood changes, and osteoporosis.[16] Zinc deficiency and dementia brings up 108 reference in PubMed. HRT use can also cause addiction with severe withdrawal symptoms when discontinued.[17]
Women doctors were had a higher risk of suicide than male doctors (2.27 compared with 1.41) in a 2004 meta-analysis of 25 studies.[18] A UK HRT study found that continuous use of HRT for at least one year increased risk of suicide or suspected suicide 2.4 times (CI 1.68– 3.11).[19] In the RCGP oral contraception study attempted suicide risks were 1.42 (1.11 -1.81) for current users and 2.12 (1.66 - 2.71) for former takers compared with controls.[20] How much of the higher risk of suicide among women doctors is due to their training to use progesterone dominant hormonal contraceptives or combined HRT for women who still have a uterus? Why do doctors still not know after 50 years that progesterone/progestogens can increase monoamine oxidase activity and cause depressive mood changes.[21,22]
In 2007 the IARC classified progestogens and oestrogens as Group 1 (highest level of evidence) carcinogens.[23] Rather than still seek the effects of different types and doses of progestogens and oestrogens as I did in the 1960s, it would be helpful if doctors now investigated and treated the common nutritional deficiencies and biochemical perturbations which increase menopausal symptoms, dementia, osteoporosis and cancers.
1 Vinogradova Y, Coupland C, Hippisley-Cox J. Use of hormone replacement therapy and risk of breast cancer: nested case-control studies using the QResearch and CPRD databases BMJ 2020; 371 doi: https://doi.org/10.1136/bmj.m3873 (Published 28 October 2020)Cite this as: BMJ 2020;371:m3873
2 Rossouw JE, Anderson GL, Prentice RL. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321-33.
3 The Women's Health Initiative Randomized Controlled Trial. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy. JAMA 2004;291: 1701-1712.
4 Grant ECG. Price EH. Reduction in mortality from breast cancer: fall in use of hormones could have reduced breast cancer mortality. BMJ 2005; 330:1024.
5 Colditz GA. Decline in breast cancer incidence due to removal of promoter: combination estrogen plus progestin. Breast Cancer Res 2007; 9:108.
6 Ravdin M, Cronin KA, Howlander N, Berg CD, Chlebowski RT, Feuer EJ, Edwards BK, Berry DA. The decrease in breast cancer incidence in 2003 in the United States. NEJM 2007;356:6.
7 Grant EC. Oral contraceptive progestin and estrogen use and increases in breast, ovarian, and endometrial cancers. JAMA Oncol. 2018 Sep 13. doi: 10.1001/jamaoncol.2018.4146.
8 Collaborative Group on Hormonal Factors in Breast Cancer. Type and Timing of menopausal therapy and breast cancer risk individual participant meta-analysis of the worldwide epidemiological evidence. Lancet 2019;394 1159-68. :10.1016/S0140-6736(19)31474332
9 Beral V, Peto R, Pirie K, Reeves G. Menopausal hormone therapy and 20-year breast cancer mortality. Lancet Published Online August 29, 2019 http://dx.doi.org/10.1016/ S0140-6736(19)32033-12
10 Grant ECG. Food allergies and migraine. Lancet 1979;1:966-969.
11 Grant ECG. The pill, hormone replacement therapy, vascular and mood over-reactivity and mineral imbalance. J Nutr Environ Med 1998 8:789-91. DOI:10.1080/13590849862131
12 Hunnisett A, Howard J, Davies S.Gut fermentation (or the 'Autobrewery') Syndrome: A New Clinical Test with Initial Observations and Discussion of Clinical and Biochemical Implications. J.Nutr.Med. 1990;1:338.
13 Werner S1, Kunz S, Beckurts T, Heidecke CD, Wolff T, Schwarz LR. Formation of DNA adducts by cyproterone acetate and some structural analogues in primary cultures of human hepatocytes. Mutat Res 1997;395(2-3):179-87.
14 Hemminki, K. DNA adducts, mutations and cancer. Carcinogenesis,1993;14,2007–2012.
15 Howard JM. The detection of DNA adducts (risk factors for DNA damage. A method for genomic DNA, the results and some effects of nutritional intervention, J Nutr Environ Med 2002;12:19-31.
16 McLaren-Howard J, Grant ECG, Davies, S. Hormone replacement therapy and osteoporosis: bone enzymes and nutrient imbalances. J Nutr Environ Med. 1998;8:129–138.
17 White M, Grant ECG. Addiction to oestrogen and progesterone. JNutr Environ Med 1998;8:117-120.
18 Schernhammer ES, Colditz GA. Suicide Rates Among Physicians: A Quantitative and Gender Assessment (Meta-Analysis). Am J Psychiatry 2004; 161:2295-2302.
19 Hunt K, Vessey M, McPherson K, Coleman M. Long-term surveillance of mortality and cancer incidence in women receiving hormone replacement therapy. B J Obstet Gynae 1987; 94: 620-635.
20 The Royal College of General Practitioners. Oral Contraceptives and Health. Pitman Medical, London 1974.
21 Grant ECG, Pryce Davies J. Effect of oral contraceptives on depressive mood changes and on endometrial monoamine oxidase and phosphatases. BMJ 1968;3:777-780.
22 Grant EC. Hormonal contraception and its association with depression. JAMA Psychiatry 2017 Mar 1;74(3):301-302. doi: 10.1001/jamapsychiatry.2016.3701.
23 IARC. Combined estrogen-progestogen contraceptives and combined estrogen-progestogen menopausal therapy. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans 2007; Volume 91.
Competing interests: No competing interests
Use of hormone replacement therapy and risk of breast cancer: nested case-control studies using the QResearch and CPRD databases. Does this reflect the current practice?
Dear Editor,
The recent publication by Vinogradova et al. [1] states to provide new estimates of the excess breast cancer risks associated with use of different hormone replacement therapy (HRT) preparations in the UK. In author’s replies to rapid responses it is further highlighted that the publication reflects “all commonly-used HRT treatments in UK NHS primary care settings over the past 20 years”. However, the HRT preparations included in the study do not reflect current prescribing trends.
In their study, Vinogradova and colleagues [1,2] identified 4 types of commonly prescribed progestogens in the UK, which were exclusively synthetic (norethisterone acetate, levonorgestrel, medroxyprogesterone, and dydrogesterone). Similarly, in the previous publication on the risk of venous thromboembolism associated with HRT use the authors included synthetic progestogen data only [3]. In the reply to rapid responses the author highlights that natural progestogen micronised progesterone has not been included in the analysis due to low number of prescriptions and associated cases and states there has been “… very low usage of micronised progesterone in the context of NHS primary care, with no evidence of steady increase”. Furthermore, micronised progesterone has been recently defined as a “rarely used hormone” by the Collaborative group [4]. However, such statements are not in line with current HRT prescribing practices.
Based on MIMS [5], there are currently three adjunctive progesterone products available in the UK as HRT. These include 5mg medroxyprogesterone tablets (MPA), levonorgestrel (20mcg/24hrs) intrauterine system and micronised progesterone (100mg) capsules. Most recent NHS PCA data demonstrates a steady increase in NHS prescriptions for micronised progesterone in England between February 2008 and November 2020, with nearly 200K prescriptions issued by the NHS between February 2008 and December 2018, and further 486K prescriptions issued between January 2019 and November 2020 [6]. This is further supported by IQVIA data, that demonstrates exponential growth in the defined daily dose sold of micronised progesterone between 2016 and 2020 Q3 in the UK (between 2016 and 2018 Q3 and between 2019 and 2020 Q3 there were over 3.5M and 11.5M defined daily doses sold in the UK, respectively) [7]. This data demonstrates that micronised progesterone use has more than tripled between 2016-2018 and 2019-2020. In comparison, such exponential growth is not observed with adjunctive MPA, which had only 377K defined daily doses sold in the UK between 2016 and 2020 Q3 (data for adjunctive levonogestrel was not available) [7]. Therefore, it is inaccurate to state that micronised progesterone is a rarely used hormone and that there is no evidence of increase in its use.
In comparison to combined HRT preparations, such as transdermal patches and oral tablets containing synthetic progestogens, the number of NHS prescriptions for adjunctive micronised progesterone in England was markedly lower between February 2008 and December 2018 (>1M prescriptions for MPA and dydrogesterone containing preparations each; >6M prescriptions for norethisterone acetate containing preparations; nearly 200K prescriptions for adjunctive micronised progesterone), except for levonorgestrel containing preparations (just over 300K prescriptions) [6]. Based on the NHS PCA data, the number of NHS prescriptions for MPA, norethisterone acetate and levonogestrel containing HRT preparations in England has dropped between 2016-2018 and 2019-2020 [6]. Growth was only observed for preparations containing dydrogesterone (over 30% increase) [6]. This is further supported by IQVIA data on total number of unit packs sold during 2016-2018 Q3 and 2019-2020 Q3 time periods [7]. Despite higher prescription numbers of combined HRT therapies containing synthetic progestogens, the growth in their use has not been as profound as that of adjunctive micronised progesterone based on the above data.
While the recent study concludes that there is an increased risk of breast cancer with a long-term use of combined oestrogen and progestogen therapy [1], it is highly important to indicate that this is based on synthetic progestogen use only and acknowledge the fact that HRT prescribing practices in the UK have changed during the last years. Given that the recent data demonstrates an exponential growth of micronised progesterone use in the UK, we believe this should have been taken into account when reviewing the subject in the recent publication [1], especially since there is evidence demonstrating a difference between the use of synthetic and natural progestogens and associated breast cancer risk, supported by numerous guidelines [8][9][10].
References:
1. Vinogradova Y, et al. Use of hormone replacement therapy and risk of breast cancer: nested case-control studies using the QResearch and CPRD databases. BMJ. 2020 Oct 28;371:m3873. doi: 10.1136/bmj.m3873.
2. Vinogradova Y, et al. Protocol to assess risk of breast cancer associated with use of hormone replacement therapy in real world settings: two nested case-control studies in primary care. 2019. https://nottingham-repository.worktribe.com/output/2309731.
3. Vinogradova Y, et al. Use of hormone replacement therapy and risk of venous thromboembolism: nested case-control studies using the QResearch and CPRD databases. BMJ. 2019 Jan 9;364:k4810. doi: 10.1136/bmj.k4810.
4. Collaborative Group on Hormonal Factors in Breast Cancer. Type and timing of menopausal hormone therapy and breast cancer risk: individual participant meta-analysis of the worldwide epidemiological evidence. Lancet. 2019 Sep 28;394(10204):1159-1168. doi: 10.1016/S0140-6736(19)31709-X.
5. MIMS. Table: Hormone Replacement Therapy (HRT). Available from: https://www.mims.co.uk/table-hormone-replacement-therapy-hrt/womens-heal... Last accessed: February 2021.
6. NHS Digital. Prescription Cost Analysis data, England 2008-2020. Available from: https://www.nhsbsa.nhs.uk/prescription-data/dispensing-data/prescription... Last accessed: February 2021.
7. Analysis from IQVIA-IMS Health, United Kindgom 2016 Q3 – 2020 Q3.
8. Stute P, et al. The impact of micronized progesterone on breast cancer risk: a systematic review. Climacteric. 2018 Apr;21(2):111-122. doi: 10.1080/13697137.2017.1421925.
9. NAMS 2017 Advisory Panel. The 2017 hormone therapy position statement of The North American Menopause Society. Menopause. 2017; 24(7): 728-753.
10. Baber RJ, et al. 2016 IMS Recommendations on women's midlife health and menopause hormone therapy. Climacteric. 2016 Apr;19(2):109-50.
Competing interests: No competing interests