Why imitation progesterones, including MPA which is in Prempro, cause breast cancer but
Prometrium® (oral micronized progesterone) does not:
A Summary of the Scientific Evidence
Written by Michael L. Williams, update of November 24, 2009.
1. It is generally accepted by the medical community that Prempro causes breast cancer.
It is now generally accepted that Prempro causes breast cancer, and that it (or its two pill predecessor, Premarin plus Provera) caused more than 300,000 American women to get breast cancer in the past twenty-five years. The best evidence that this causal relationship between Prempro and breast cancer is generally accepted are the recent public statements of the principal investigators of the Women's Health Initiative made to numerous media outlets on the publication of the latest WHI study data, about the medical implications of the combined new study of the Prempro trial and observational study (NEJM Feb 5, 2009).
In this new WHI study, the investigators followed the 15,000 women in the Prempro randomized trial for an additional two and one half years beyond the first WHI Prempro trial report, for a total follow-up of 8 years. In a subset of the Observational Study, they followed another 41,449 women who had not been assigned treatment but took it voluntarily. They report data here on 11 years of follow-up since enrollment in the OS began. The cut off for data available for this analysis was December 31, 2005. 8
1.1 Public statements by the WHI breast cancer investigators.
Rowan Chlebowski, a medical oncologist at Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, and lead author of the new WHI study, told the press the following (the investigators' statements to the press are italicized):
In the last decade in which it was still widely used (1992--2002), long-term hormone therapy probably caused breast cancer in 200,000 women. 3
A National Cancer Institute sponsored study published in the Journal of the National Cancer Institute in 2007 had examined the rate of drop in breast cancer among menopausal women in several groups of women following the dramatic drop in HRT use that began in mid-2002. The investigators estimated that based on their data and the public national data, "the impact of the declining use of postmenopausal hormone therapy could account for an estimated 17,500 fewer ER-positive invasive breast cancers annually among women aged 50-69." 10 What is tragic is that prescriptions of Prempro did not fall to zero, but fell only by about 2/3.; If we could eliminate the other 1/3, we could prevent another 9,000 or more breast cancers every year. Unfortunately, millions of women are still being prescribed these carcinogenic synthetic progesterone imitations, and are at substantially increased risk of breast cancer. At trial, we have been able to prove to jury after jury that the drug companies have lulled many doctors into thinking these drugs are safe, and that there are no good alternatives.
Chlebowski statements continue in italics:
For longer duration therapy, this provides new evidence that the risk is higher than we previously thought," said -- He said that a 55-year-old woman whose mother developed breast cancer at 62 would have a 2 percent chance of developing breast cancer in five years. Doubling that risk with long-term hormone use translates to a 4 percent chance of getting cancer, or 1 in 25. 3
Estrogen-progestin appears to act as a growth factor in the body, stimulating cancer cells. But the effect seems to dissipate quickly once hormones are withdrawn.
The findings should help allay lingering skepticism among some physicians about hormone therapy's role in triggering breast cancer. 4
If you stop hormones, the risk of breast cancer [associated with hormone use] rapidly declines. The risk decreased rapidly in both groups after they stopped the pills, even though both groups had mammograms with similar frequency. That fact weakens the argument that the drop in breast cancer cases was due to fewer women getting mammograms. There was a rapid decline in breast cancer incidence after stopping hormones, while mammography use didn't change between the groups. 2
The rapid decline in cancer rates was due not only to an overall drop in breast-cancer risk, but also to the withdrawal of excess estrogen, which may actually have served as a treatment for tiny, preclinical breast cancers. When you change from a high- to a low-estrogen environment, it's like giving breast cancer treatment. These are preclinical cancers that are below the level of detection, and that accounts for why biologically we can see such a quick effect in stopping hormone therapy. 7
The press interviewed another major breast cancer expert about this explanation:
"That is a reasonable and biologically plausible explanation for why we might be seeing a more precipitous drop in breast cancer than we might expect from the normal lead time for reduction of malignancies," says Dr. Jonathan Berek, chair of obstetrics and gynecology at Stanford University School of Medicine, who was not involved in the study. 7
Chlebowski commented on another new large study of women on Prempro conducted by the new American Cancer Study that was published in the journal Cancer, on line, just a week before this combined WHI study. (Cancer. 2009 Mar 1; 115(5): 936-45). 11
Eugenia Calle, former vice president of epidemiology for the American Cancer Society, and her colleagues looked at about 68,000 women who were cancer free at the study's start in 1992, evaluating their HRT use and risk of getting breast cancer, and following them until mid-2005. They found that the risk of breast cancer increased substantially after three years of using HRT-for lobular cancer, the therapy doubled the risk at three years-but there was no increased risk for those who used it for less than two years. Chlebowski said that cancers may simply be harder to detect on mammograms during initial hormone use. 2
A truly safe interval can't be defined. 2
Marcia Stefanick, co-author of the study, and a breast cancer epidemiologist and professor of medicine at Stanford, told the press:
For both groups, the risk of cancer was much more pronounced than suggested by previous data. This is very strong evidence that estrogen plus progestin causes breast cancer. 1
The good news is that if you discontinue hormones, your risk drops pretty quickly. You start women on hormones and within five years, their risk for breast cancer is clearly elevated. You stop the hormones and within one year, their risk is essentially back to normal. It's reasonably convincing cause-and-effect data.
We know nationwide, breast cancer rates dropped within a year of the initial publication. We now have good evidence that, yes, this happened absolutely in association with the drop of the hormones. 5
Asked to comment on how this new WHI study's findings, Dr. Donald Berry, who had published in 2007 in the NEJM a study of population trends in prescriptions of HRT (fn 9) and the nearly simultaneous drop in breast cancer in menopausal women, he said: "The great thing about this report is that it had individual women [his study had not had the ability to track individuals], and the randomized element. That adds credibility. The current [WHI] report didn't examine whether the cancers were estrogen-receptor positive or negative. [My] research found that much of the growth of breast cancer was due to tumors fueled by estrogen." 5
Joanne Manson, another co-author of the new WHI study, and a professor at Harvard Medical School said:
There was a controversy about the decline after WHI. [In July 2002, there was a large outcry in the press about the surprise news that the Prempro trial had been stopped prematurely due to an increase in breast cancers in the women taking Prempro.]
After disagreements emerged over the cause of the decline, the WHI researchers scrutinized women in two groups. In both study populations, the risk decreased rapidly after stopping hormone therapy, and in neither case was it explained by changes in mammogram screening. 5
1.2 The breast cancer risk from Prempro is greater in newly menopausal women than in women many years past menopause when they start the drug.
In May, 2009, the WHI investigators published yet another analysis of the breast cancer risk as well as the overall harm/benefit balance of both Premarin and Prempro, using combined data from both the two WHI trials (Premarin and Prempro) and the Observational Study. The point of this new WHI study was to see if the "timing hypothesis" of HRT advocates was confirmed or refuted by the WHI data. It was clearly refuted.
The "timing hypothesis" is this: the women in the WHI trials were much older than the typical HRT user; instead of being newly menopausal, the vast majority of women in the trials were more than ten years past menopause when they first started HRT. The notion was that although HERS and WHI trials found a cardiovascular risk (heart attacks and strokes), it was because both studies included mostly older menopausal women, not newly menopausal, and that if a study were done on newly menopausal women, the cardiovascular benefits found in the observational studies would be confirmed in that subset of women.
So the WHI investigators analyzed all the data they had on newly menopuausal women in the trials and the OS. What they found was that the cardiovascular risk was still present in newly menopausal women (in Prempro users), but more important, that the risk of breast cancer was much higher in these newly menopausal women, too---higher than in the older women that made up most subjects in the trial.
The paper is Prentice et al.,: "Benefits and Risks of Postmenopausal HT When It Is Initiated Soon After Menopuase." It was accompanied by an "Invited Commentary" by Emily Banks and Karen Canfell from Australia, "Hormone Therapy Risks and Benefits---The WHI findings and the Postmenopausal Estrogen Timing Hypothesis". 22
The new paper's conclusion: Prempro's risk benefit balance is negative, that is, the risks clearly outweigh any benefit, whether a woman starts years or weeks after menopause. "In summary, the results presented here suggest that the unfavorable balance of benefits and risks observed in the CEE/MPA (Prempro) trial as a whole also applies to recently menopausal women." (from the last paragraph of the study).
Note that Dr. Robert Langer is a co author of this new WHI paper, even though he is a consultant and expert witness for Wyeth in the Prempro litigation (which he properly discloses in the paper).
This WHI finding, that the risk of breast cancer is significantly higher in women who start the drug at or early in menopause, was confirmed by a French study published on September 14, 2009. Fournier et al., Estrogen-Progestagen Menopausal Hormone Therapy and Breast Cancer: Does Delay from Menopause Onset to Treatment Initiation Influence Risks? 26
The conclusion of this paper states: ". . .our results indicate that, contrary to what is currently hypothesized regarding the risk of heart disease, initiation of MHT [menopausal hormone therapy] close to menopause onset rather than later may not limit the increased risk of breast cancer. Instead, even short duration of use of some EP-MHT's were associated with substantially elevated risks of breast cancer when treatment was initiated close to menopause. Our finding that, for short durations of use around menopause, progesterone in EP-MHT may be safer regarding breast cancer risk than other progestagens needs to be confirmed in other settings." In another section of this Summary, I lay out the corroborating evidence that bioidentical progesterone in MHT is truly much safer than the synthetic, imitation progesterones used in Prempro, Activella, Femhrt, Combipatch, and Climara Pro. One major goal of this summary is to empower women to demand that their doctors familiarize themselves with this evidence, and to empower women and their doctors to demand that until these carcinogenic combination hormone therapy drugs are taken off the market, that the warnings be changed to advise women not to use the imitation progesterones as their first choice of a progestogen in MHT.
A brief summary of the most recent epidemiological evidence that provides additional support for Dr. Chlebowski's and Stefanick's conclusion that Prempro caused (and is still causing) breast cancer in thousands of women annually for the past twenty five years.
In addition to the studies discussed above, there are a few other large observational studies published recently that provide rich and detailed support for the proposition that Prempro causes ER+ breast cancer, and that the risk doubles for women who take it for as short a time as 18 months, perhaps less.
Million Women Study
The largest cohort study on the planet with relevant data is the Million Women Study in the UK. The most popular form of HRT in the UK was Wyeth's Prem-pak, a combination of Premarin plus levonorgestrel (LNG), another androgenic, synthetic and patentable imitation of natural progesterone. Also, some women in the UK used another synthetic progestin, norethindrone acetate, or NETA. NETA was the most popular progestin in most of the Scandinavian countries, and has had some use in the USA as an alternative to MPA, such as in the combination hormone therapies Activella or Combipatch. But thousands of women in the UK study also used MPA as their progestin, so this large cohort study had data on all three synthetic progestins combined with estrogen in HRT.
None of the UK women used oral micronized progesterone, the most popular progestogen used in France. (The largest cohort study in France, the "E3N" teachers study, has found that oral micronized progesterone, unlike the non-bioidentical progestins, does not increase breast cancer risk.) 18, 19
There have been two main breast cancer reports from the MWS. The first, published in
2003, analyzed the association between the three main types of EPRT in use in the UK and breast cancer, not differentiated by histology or receptor subtype. Fn 12 They had 9,364 breast cancer cases at the time when data collection stopped for this analysis. Nevertheless, the study found that in current users of EPRT the risk of breast cancer was doubled (RR=2.0). This is exactly the same overall effect, averaging durations, found by the WHI observation study above: current users of Prempro have doubled their chances of getting breast cancer.
In the MWS, in women exposed for one year or less to EPRT, the risk was statistically significantly increased at 1.45, meaning that these drugs can cause breast cancer in some women in less than a year of exposure.
The risk of breast cancer steadily increased with years of exposure. The investigators compared the relative carcinogenicity of the three different synthetic progestins and found no statistically significant difference in their ability to increase the rate of breast cancer. See Fig 5 and text on p. 426. The authors concluded that EPRT had probably caused 15,000 extra breast cancers in the UK from 1992-2002, just in women aged 50-64.
The second report from the MWS on EPRT and breast cancer analyzed the associations with different histological types of breast cancer, but no receptor status data were reported.13 The million-plus women had now been followed for another year and there were now 14,102 breast cancer cases for analysis. These researchers found a stronger causal effect on lobular and tubular types of breast cancer than on ductal, but even ductal tumors (again, including ER- ductals as a diluting factor) had almost doubled in frequency in current users exposed less than five years. For lobular and tubular subtypes, the risk was more than doubled in current users exposed less than five years. See Figure 3.
The MarketScan health insurance claims data base study
In January, 2009, a group of epidemiologists from Boston University published a case control study using a large private insurance claim data base that allowed them to track HRT use and breast cancer outcomes (though not by receptor or histology subtype-they had to use "all" breast cancer, including the ER- tumor types not related to HRT use, which of course will lead to an underestimate of the true relative risk). 14 They had 4,515 breast cancer cases and 18,058 matching controls.
For women who used combination EPRT for four years or more, the risk of breast cancer more than tripled (OR=3.10). Obviously, for ER+ tumors the doubling time would be significantly less than 4 years. Most interesting, they recruited the bulk of the breast cancer cases and controls after the appearance on the market of new "low-dose" Prempo (see Table 1), and so they looked to see if they could detect an effect of dose of drug. They could not, meaning low dose Prempro did not apparently lower the risk of breast cancer.
It is important to remember two things in reading the papers summarized above:
1. They are assessing risk of "all" breast cancers; if they would ever publish their receptor and histology data, we know that the RR's for ER+ breast cancers would be higher and appear sooner than for "all" breast cancers. Since ER- tumors make up roughly 25-33% of all tumors, any assessment of "all" breast cancers will systematically underestimate the risk for ER+ tumors by a factor of 30% or more; that is, if the RR for all breast cancer at 3 years is 1.7, the RR for ER+ breast cancers at 3 years would be well over 2.0.
The same thing is true for the main histological subtypes of breast cancer. Studies that have enough total breast cancers to separately analyze the risk of ductal, lobular, tubular and other subtypes of breast cancer consistently find a higher and earlier risk of tublar and lobular breast cancers compared to ductal. Again, when all types are combined, the overall risk estimate will be lower than when the subtypes are separately analyzed.
2. They do not distinguish AT ALL between types of progestogens: there is no mention that progesterone has been found much less dangerous than MPA, or any acknowledgment that the WHI results apply only to Prempro, not to Premarin plus Prometrium, not to estradiol plus Prometrium.
The American Cancer Society Nutrition Cohort Study
Dr. Chlebowski briefly discussed this March 2009 study by Dr. Eugenia Calle and colleagues from ACS in his public remarks above. 11 These scientists followed a group of 68,000 American women from 1992-2005. All of them were cancer free at enrollment. There were eventually 3,139 breast cancers. They were able to analyze the date by histology (ductal or lobular, though not by receptor status) and by one year at a time of exposure. Although they found no increased risk in either type of breast cancer in the first two years, "we observed a significant increase in risk for both types of breast cancer at 2 to 3 years of use (for ductal cancer: RR of 1.91 95% CI 1.39-2.61, and for lobular cancer: RR of 1.95, 95% CI 1.04-3.65) and for all years thereafter."
They also found that the increased risk lasted no more than two years after a woman stopped taking the drugs.
One interesting finding reported in this study: they looked back at national breast cancer trends in the 1980's, when EPRT was first becoming popular, and found that the proportion of all breast cancer cases with a lobular component increased from 9.5% in 1987 to 16.8% in 2002. The authors remark: "This trend is consistent with a stronger effect of combination hormone therapy, which gained common use in the mid- to late-1980s, on lobular carcinoma than ductal carcinoma." See p 937.
The MARIE Study in Germany
This is a large case control study from Germany. The investigators recruited 3,464 breast cancer cases and 6,657 matched controls. They have so far published two reports. The first, fn 16, contains an analysis of the associations between different types of HRT and breast tumors. Unfortunately, they lumped MPA together with natural progesterone as "one type" of HRT, a design they would not use today, given the French cohort studies that have found no increased risk of breast cancer in women using oral micronized progesterone in EPRT. 18, 19 But even so, for current users of combination HRT, the risk of ductal tumors was almost double (OR=1.75), for lobular it was 3.04, for ductal/lobular mixed, 3.51, and for tubular 4.19. See Figure 2a. Obviously, had the analysis been confined to ER+ ductals, the OR would have probably have been higher.
In a second published report on the same group of women, the authors did analyze receptor status and other tumor characteristics. 17 Some of the strongest causal evidence comes from these subtype analyses. For example, if they looked only at ductal tumors of grade 1, the OR= 2.60 for current users of all HRT, including ERT users in the mix. The same was true for smaller tumors, and for ER+ tumors. These data are consistent with the WHI investigators' explanation of the promotion effect of Prempro: it feeds tiny, occult tumors and causes them to grow rapidly to detectable size. The effect is most seen when confined to small, ER+ tumors of low grade, and then the effect is quite strong.
A Brief Primer on Types of Epidemiological Evidence
Randomized Clinical Trials:
If a drug is thought to be relatively safe, but its benefits are in doubt, researchers will conduct randomized clinical trials on women to "prove" the benefits. However, this is literally an experiment on the women subjects, so if there is suspicion that the drug will cause serious harm to some of the subjects, clinical trials cannot be done-it would be unethical to experiment with potentially dangerous drugs on women. Instead, researchers study the hazard in question by observing women who voluntarily, in their normal lives, decide to take the drug. These studies are "observational" as opposed to "experimental", but they are the only way to ethically assess the risk of breast cancer with Prempro.
Observational Studies:
There are three main types of observational studies used to assess breast cancer risk from Prempro and other forms of HRT.
1. Case Control Studies. The most common type of observational study is the case control study, where a large number of women with breast cancer are collected (the cases), then women without breast cancer are found who match the cases in age and other characteristics (the controls). The investigators then determine what percentage of women in each group used Prempro. If the percentage of women with breast cancer on Prempro is significantly higher than the percentage of women without breast cancer, the study finds an association between Prempro and breast cancer expressed in a "relative risk" or "hazard ratio" or "odds ratio" (RR, HR or OR). Case-control studies are the quickest and cheapest of the observational studies to do. A good case control study can be conducted in two to three years.
2. Cohort Studies. The second major type of observational study is a cohort study. Here, the investigators enroll a large number of women and then follow them medically for many years. The follow up can be retrospective (reviewing past medical history) as well as prospective. These studies are much more expensive than case control studies and take much longer to complete. Nevertheless, there are several very large cohorts of menopausal women being followed in the USA and other countries that report important Prempo and breast cancer data from time to time in the peer reviewed literature. These studies are designed to follow the same group of women for many years, even decades.
3. Ecological Studies. Finally, the third main type of observational study relevant here is the "ecological" study, where the investigators analyze trends of Prempro use and breast cancer incidence in large populations of women, but without individual data. In these populations it is possible to see "intervention effects" of drugs when they cause cancer rates to go up when prescriptions go up and cause cancer rates to go down when prescriptions go down. There are several useful such studies on Prempro and breast cancer.
How study design of Prempro Breast Cancer studies leads to a systematic underestimate of true risk.
In order to see a causal effect between a drug and cancer, it is important to precisely define what drug exposure means, and what type of cancer one is suspicious is connected. With Prempro, it is obvious now that its effect on breast cancer is primarily through promotion of microscopic, occult and harmless tumors into clinically significant breast cancers. The effect wears off after about two years, so studies that combine past users with current users will dilute the effect, that is, will under estimate the true effect. Many studies in the published literature are of "ever-never users", and they all have this systemic design defect leading to underestimate.
Similarly, we know now that Prempro acts as a growth factor for ER+ tumors, but not ER- tumors. Any study that combines both types of breast cancer into one analysis will dilute the effect on the subgroup of ER+ tumors, and will systematically underestimate the size of the causal effect. Even more obvious are studies that combine both ERT and EPRT into one analysis of "HRT"-these will also systematically underestimate the risk of Prempro, because ERT carries a much lower risk. The lower ERT effect dilutes the stronger Prempo effect.
Finally, there is a human selection bias that affects all the observational studies of HRT that recruit voluntary controls: it is known that women who take HRT are much more willing to volunteer to be controls in medical studies than women who don't use HRT. This causes a dilution of the true effect because the controls are not representative of a true random sample of all menopausal women. They are a non-random sample who use HRT much more often. The recent German studies (MARIE, dicussed above) actually not only acknowledged this bias, but did a substudy to estimate its effect, finding that it lead to a ten percent underestimate of true risk. 16
European Consensus that Oral Micronized Progesterone has a much lower, or even no, risk of breast cancer when used in combined menopausal hormone therapy.
There is now a substantial body of scientific evidence that women who use oral micronized progesterone, or its nearly identical mirror image cousin dydrogesterone, instead of MPA, NETA, or LNG, are at much lower, perhaps no, increased risk of breast cancer, at least for the first five years of use.
First, there are three prospective cohort observational studies comparing women on estrogen plus oral micronized progesterone or dydrogesterone to women taking estrogen only or estrogen plus one of the synthetic imitation progesterones like MPA.
The largest cohort and the one with the longest follow up data is a cohort of French teachers or teachers wives, the E3N cohort. The investigators of that study have published three reports on breast cancer outcomes so far.
In 2005, they published their initial results, with x years of follow up. 27 There were 54,548 women, who had not taken any kind of MHT before study entry, followed for an average of 5.8 years. The mean duration of MHT use was 2.8 years. They found a statistically significant increase in risk of breast cancer, but only in those women taking some form of imitation progesterone. The paper stated: "Micronized progesterone may be preferred to synthetic progestins in short-term HRT." It is this published finding in 2005 which I contend that women were entitled to be informed about when contemplating starting or continuing MHT. Subsequent reports from this French teachers study have strengthened this finding.
In 2007, the same group of investigators published an update. 18 This time the title of the paper carried the main message: Unequal risks for breast cancer associated with different hormone replacement therapies. Now, with 8.1 years of follow up, the original finding was confirmed: all the imitation progestgerones save one, the one closest in molecular structure to natural progesterone, called dydrosgesterone, caused an increased risk of breast cancer, but oral micronized progesterone and its close cousin drydrogesterone, did not. The paper concluded: These findings suggest that the choice of the progestagen component in combined HRT is of importance regarding breast cancer risk: it could be preferable to use progesterone or dydrogesterone.
More recently, the French investigators have published an analysis of the relative effects on different types of breast cancer. 19 They found that the imitation progesterones caused an increased risk of ER+ tumors, including ductal tumors, and especially lobular tumors. In the discussion section, they review and summarize studies by other scientists on women's and monkeys' breast tissue and conclude: "These studies support our findings suggesting that, when combined with an estrogen, progesterone may have a safer risk profile in the breast than some other progestagens."
Two other, smaller cohorts of French women have been reported to show the same effect: oral micronized progesterone is safer for the breast. The first of these is the MISSION study which is following about 5,000 women over time to see the effects of MHT. 28
About half the women taking MHT were on oral micronized progesterone and half were using one of the imitation progesterones. In the preliminary report in 2007, the risk of breast cancer was lower in the oral micronized progesterone group, although the difference was not statistically significant. The second study, with 3,175 women followed for an average of 8.9 years. 30
Human randomized clinical trials support the finding of the epidemiological studies that progesterone does not stimulate the breast as does MPA.
There are three published randomize clinical trials that analyzed breast tissue from women exposed to estrogen plus either natural progesterone or the imitation progesterone, MPA. Hofseth fn 31 exposed menopausal women to either estrogen alone, estrogen plus MPA, or no hormones, and then analyzed breast tissue samples. Menopausal women scheduled for breast biopsies were categorized by their intake of hormone pills in the three months before surgery: some took estrogen only, some estrogen plus MPA, and some no hormones at all. The study found that women taking the combination had much more breast cell proliferation than women in the other two groups, and that the increase in proliferation was concentrated in the terminal duct lobular unit of the breast, where most breast cancers arise.
In the Foidart study, 32 investigators using blinded techniques exposed women to hormones by having them spread a gel on their breasts every day for two weeks, before they were scheduled for breast biopsies. Neither the doctors nor the women knew whether they were using a gel containing estrogen only, progesterone only, estrogen plus progesterone, or a placebo. Breast cell proliferation was highest in the women who received estrogen only. The women who received progesterone alone or progesterone with estrodial had significantly lower proliferation than the women getting estrogen alone. They concluded: "progesterone administration dramatically limited the estradiol-induced proliferation of normal human breast epithelial cells." (p 967).
Chang's study, 33 was similar in design to Foidart's, and resulted in similar findings.
Randomized clinical trials in menopausal monkeys confirm that MPA stimulates breast tissue proliferation and activates many more genetic growth pathways than does natural progesterone when combined with estrogen.
In a series of papers published by a group of investigators at Wake Forest University, menopausal monkeys were exposed to human equivalent doses of estrogen only, estrogen plus MPA, and estrogen plus oral micronized progesterone. In the study which analyzed risk markers for breast cancer in these monkeys, the authors concluded: "Compared to placebo, E2 (estradiol) + MPA resulted in significantly greater breast proliferation in lobular and ductal epithelium, while E2 + P4 (micronized progesterone) did not . . .These findings suggest that oral micronized progesterone has a more favorable effect on risk biomarkers for postmenopausal breast cancer than medroxyprogesterone acetate (MPA
In the most recent report from this group, the investigators measured how many and what kind of genes were turned on in the breasts of menopausal monkeys given the same three regimens as in the first study. "Treatment with E2 alone induced modest effects on select genes relating to epidermal growth factor receptor (EGFR) activity, which were augmented by MPA but not P4, consistent with patterns of epithelial cell proliferation. . . . These findings suggest that a standard dose of oral E2 + MPA a more pronounced effect on gene expression in the breast compared to E2 alone . . ." 35
Review papers which, while containing no new data, support the findings from these French studies that oral micronized progesterone is significantly safer than MPA, NETA, or LNG when used as the progestational component of MHT.
There is a growing body of editorial or scientific review publications which support the conclusion that women should be advised that oral micronized progesterone is safer for the breast. Here are the best of them:
Mueck et al, "Use of dydrogesterone in hormone replacement therapy", Maturitas (2009) doi.10.1016/j.maturitas.2009.09.013. (published on line ahead of print). 29
Gadducci et al., "Progestagen component in combined hormone replacement therapy in postmenopausal women and breast cancer risk: A debated clinical issue," Gynecological Endocrinology, December 2009; 25(12): 807-815. 38
L'Hermite et al., "Could transdermal estradiol+progesterone be a safer postmenopausal HRT? A review," Maturitas 60 (2008) 185-201. 39
In addition, the plaintiffs in the hormone therapy litigation have commissioned expert reports that analyze the scientific evidence that oral micronized progesterone is a safer choice than MPA for use in menopausal hormone therapy. These reports are from Donald Austin, MD, 40 and a joint report by Australian cancer researchers Dr Wayne Tilley and Dr Steven Birrell. 41
The only two ongoing randomized clinical trials in the USA studying the effects of MHT on menopausal women both rejected synthetic, imitation progestins and instead opted for bio identical progesterone. In the one study that is looking at the potential benefits of long term, early initiated hrt, the investigators are using Premarin, the horse estrogen component of Prempro, but rejected the use of MPA with it, and instead chose Prometrium, oral micronized progesterone. 44
The other trial is sponsored by NIH, specifically the National Institute on Aging. This study is called Early v Late Intervetion with Estrogen Trial (ELITE). The investigators also rejected the use of MPA or any other imitation progesterone and opted to give women with a uterus a vaginal gel containing bio identical progesterone (on the US market under brand names Crinone and Prochieve).
Harvard Medical School Professor Joann Manson, who is one of the investigators for the WHI, and for KEEPS. In her book, Hot Flashes, Hormones & Your Health, she states: :"I think it may be prudent to opt for progesterone over progestin where possible." P. 144. 45
The French regulatory agency for drug licensing states on its website that oral micronized progesterone may be safer than other progestogens in MHT.
Agence Francaise de Securite Sanitaire des Produits de Sante:
"the risk depends on the type of progestogen combined with estrogen and is smaller with micronized progesterone . . ." 46
The International Menopause Society states on its website and in its official MHT guidelines that there is evidence oral micronized progesterone has a lower breast cancer risk than other progestogens.
"micronized progesterone . . . used in association with . . . estradiol may be associated with no increase in risk or lower risk than the use of synthetic progestogens for at least four years, and perhaps even eight years, of treatment . . .." 47
How other risk factors for breast cancer interact with menopausal hormone therapy.
Breast Cancer, Alcohol and HRT: a synergistic carcinogenic effect.
There have been several studies that have found an increased risk of ER+ breast cancer in women who are heavy users of alcohol. But a recent study from
Denmark published in 2008 was able to follow a cohort of 5,000 women enrolled originally in a heart health study for their use of HRT, alcohol and breast cancer from 1981 thru 2002. 15
They found that women who used HRT had a doubled risk of breast cancer in general at baseline and throughout the study. "HR=2.0". And they found a small but not quite statistically significant increase in risk for all alcohol users, HR=1.11, 95% CI 0.99-1.25). For women who did not take hormones, that level of alcohol consumption did not increase their breast cancer risk. But women who used both alcohol and hormones were at a greatly enhanced risk: For women who took hormones and had two drinks a day, the HR was 4.74, 95% CI 2.61-8.59).REFERENCES
1. "Hormone therapy doubles breast cancer risk: study"; AFP Feb 4, 2009.
2. "Drop in breast cancer rates due to drop in HRT use"; HealthDay, Feb 4, 2009.
3. "Study: Hormone therapy caused breast cancer for thousands"; CNN Health, Feb 4, 2009.
4. "Breast cancer risk falls after hormones halted"; The Seattle Times, Feb 4, 2009.
5. "Breast cancer risk drops one year after end of hormone therapy"; Bloomberg News, Feb 4, 2009 .
7. "Halting Hormone Therapy reduces breast cancer risk quickly"; Time Magazine Feb 4, 2009 .
20. Ringa et al., Did the decrease in use of menopausal hormone therapy induce a decrease in the incidence of breast cancer in France (and elsewhere)? Rev Epidemiol Sante Publique 2008 Oct; 56(5): 297-301.
25. Hulley S, Grady D. Postmenopausal Hormone Treatment. JAMA. 2009;301(23):2493-2495.
40. Expert Report of Donald Austin
41. Expert Report of Tilley/Birrell
