The debate over post-menopausal hormone replacement and breast cancer has raged on for some time now, and won’t likely abate any time soon. That said, a recent French study has turned up provocative findings suggesting that the type of progesterone used in HRT may be the decisive factor as far as breast cancer is concerned.

The main conclusion from the massive E3N-EPIC study, involving nearly 55,000 women, is that the association between HRT and the risk of breast cancer varies according to the type of progestin used. No increase was observed with micronized progesterone (bio-identical progesterone), while synthetic progestins did increase risk when combined with estrogen.

E3N was a prospective population study of 54,548 women born between 1925 and 1950. The average age at the outset was 52.8, and ranged from 40 to 66. The average follow-up was 5.8 years, but ranged from 0.1 to 10.6 years. During this time, there were 948 primary invasive breast cancers in the study cohort. Data on the influence of HRT on breast cancer incidence was derived from questionnaires identifying the types of hormonal therapies the women used.

Overall, there was a small but definite increased risk of breast cancer among the women who took hormones compared with non-users. The relative risk was 1.2 for users versus non-users (CI = 1.1–1.4). However, subanalysis of the type of HRT showed an interesting pattern.

There were 30 cases of breast cancer among women using estrogen alone, and the adjusted relative risk was estimated at 1.1. Among those using estrogen plus micronized progesterone (bio-identical progesterone), there were 55 cases, and the adjusted relative risk was 0.9, indicating a small decrease in relative risk.

Among women using estrogens and synthetic progestins, there were 268 breast cancers, and the relative risk was 1.4, a significant increase (Fournier A, et al. Int J Cancer. 2005; 114: 448–454). The risk associated with synthetic progestins increased with duration of use (from less than 2 years to 4 or more years). This was not seen for bioidentical progesterone or estrogen alone.

There were no differences in breast cancer risk between oral and non-oral (transdermal) delivery of HRT; both increased the relative risk. The authors concluded that estrogen (bio-identical estradiol) alone did not increase the risk of breast cancer, but estrogens, orally or non-orally, with progestins, did increase the risk. This seemed more pronounced with non-oral delivery. More specifically, risk of breast cancer was not increased with estrogens and micronized progesterone. Non-oral estrogen increased the risk of breast cancer only when it was combined with synthetic progestins.

In interpreting this study, keep in mind that HRT practices and preferences in France are somewhat different than in the US. French women use oral HRT far less commonly than women do here. In the Fournier study, 55% of women used transdermal gels and 45% used transdermal patches.

Further, the progestational agents used were mainly oral micronized progesterone or an array of progesterone derivatives. Some women in the French study used progestins from the 19-nortestosterone family (norethindrone acetate or lynestrenol). Not all of these are available in the U.S. Of note, on the estrogen side, hardly anyone in the study used conjugated equine estrogens, which are still widely used in the US.

While HRT in general seemed to increase risk, there were baseline differences between HRT users and the non-users. HRT users had earlier menarche, earlier menopause, less nulliparity, more children born before age 30, more benign breast disease, less obesity, higher education, and a greater previous use of oral contraceptives. The authors insist these differences do not in any way alter the fundamental findings.

One of the surprises of this study is the rapid pace at which breast cancers were identified. Fournier calculated a relative risk of 1.7 associated with less than one year of exposure to non-oral estrogens combined with synthetic progestins.

This trial raises very provocative questions: Do the data show differences among various progestational agents? Are the doses administered different in their effects? What does the rapid appearance of breast cancer in all groups mean? Is it possible that the combination of estrogen with specific progestins leads to earlier detection of preexisting tumors while bio-identical progesterone either does not have this effect or takes longer to show the effect?

Is it indeed possible that estrogen and synthetic progestins does increase the risk of breast cancer and that estrogen plus bio-identical progesterone does not? Is bio-identical progesterone indeed a preferred progestational agent for short and long term hormone therapy? It would appear to be so.

Dr. Fournier and the E3N-EPIC team have clarified a very important issue in post-menopausal hormone management, and have gone a long way in making the case that bio-identical progesterone, when combined with estrogen (largely bio-identical estradiol) does not increase breast cancer risk. They also show that estradiol by itself does not increase risk. This confirms similar findings from the Women’s Health Initiative that seem to exonerate estrogen when given by itself.

Mammography Goes Digital, Improves Accuracy in Young Women

One of the long-standing problems in conventional film mammography is that breast density impairs sensitivity. Digital mammography allows for enhanced contrast in areas of dense tissue. A massive multicenter study indicates this greatly improves detection of neoplasia in younger, premenopausal women.

Radiologists at 33 mammography sites in North America recruited almost 50,000 women presenting for screening mammography. All women had both digital and film mammography in random order. The radiologists used a seven-point malignancy scale as well as the traditional BI-RADS scale to rate mammograms and determine the need for further evaluation. If breast cancer was confirmed on pathology within 455 days, women were classified as positive for cancer. A total of 42,760 women were ultimately evaluated.

Overall, the diagnostic accuracy of digital and film imaging were similar. However, the digital mammography was significantly better than film for women younger than age 50, as well as for those with heterogeneously dense or very dense breasts, and for those who were pre- or perimenopausal (Pisano ED, et al. N Engl J Med. 2005; 353: 1773–1783).

This study provides good evidence that digital imaging improves the accuracy of mammography in younger women and those with dense breasts. Since HRT increases breast density, it makes sense that peri- and postmenopausal women on HRT would also benefit from digital methods.

Digital mammograms have additional benefits: facilitation of image access, transmission, and storage; lower radiation dose than conventional film methods; and computer-aided diagnosis. The bad news is the cost. Digital systems are up to four times more expensive than conventional film set-ups. Nonetheless, the findings of this study will likely increase the rate at which breast health centers and radiology facilities utilize digital mammography.