Is Estrogen Carcinogenic

One of the first theories about estrogen and cancer12 was advanced by Henry Lemon, M.D. of the University of Nebraska. Focusing on estriol (the principal circulating estrogen), Dr. Lemon initially argued that greater proportions of estriol are good, and perhaps even anti-carcinogenic. He found that women most likely to survive breast cancer had the largest amounts of estriol. In unpublished work in a small, uncontrolled study, estriol administration appeared to cause remission in a proportion of breast cancers that had metastasized to bone.

Other researchers discovered that Asian women living in Asia, who as a group have lower rates of breast cancer, also had higher proportions of circulating estriol than American women, who have higher rates of breast cancer. Asian women living in Hawaii, who have a breast cancer rate midway between Asian women living in Asia and American women, also have estriol levels midway between those of the other two groups.

Sisters and daughters of women who had had breast cancer were found to have lower proportions of estriol than sisters and daughters of women without breast cancer. Considerable animal research appeared to indicate that estriol was anti-carcinogenic or at least non-carcinogenic. However, other research1 disputed the estriol hypothesis, and present-day researchers tend to disregard it in favor of other theories.

The 2/16a-Hydroxyestrone-Ratio Hypothesis

17b-estradiol (usually just called estradiol) is often called the “principal” estrogen. Even though there’s much more estriol than estradiol normally in circulation, estradiol is considerably more potent, and has been known for decades to be more carcinogenic. H. Leon Bradlow, M.D. and a group at Strang-Cornell Cancer Research Laboratory, New York City, as well as other prominent researchers have developed a body of evidence concerning two metabolites of estradiol and their relative tendencies to promote cancer growth.

Among other things, estradiol is metabolized into estrone, which in turn can be metabolized into either 2-hydroxyestrone or 16-alpha-hydroxyestrone. Importantly, there’s an inverse relationship here: if more 2-hydroxyestrone is made, less 16-alpha-hydroxyestrone is usually made, and vice-versa. In at least one of his publications, Dr. Bradlow2 has termed 2-hydroxyestrone “good estrogen”, and has given us evidence that 16a-hydroxyestrone is “bad estrogen”. He writes: “Evidence from a long series of studies has demonstrated a specific role for 16a-hydroxyestrone as a transforming estrogen, which is more potent than estradiol itself.” (“Transforming” refers to the tendency of 16a-hydroxyestrone to increase cellular growth and proliferation, and even cancerous transformation in estrogen-responsive tissues). He goes on to note that the preponderance of evidence shows that, by contrast, 2-hydroxyestrone is non-carcinogenic or even anti-carcinogenic. He notes that treatment which reduces the 2/16a-hydroxyestrone ratio has been shown to reverse the growth of human larygneal papillomas, caused by the same family of viruses (HPV) implicated in cervical cancers. Other researchers hypothesize that cancer in other tissues – including uterus, prostate, liver and kidney – may be affected by the 2/16a-hydroxyestrone ratio as well as other estrogen metabolites.

There’s considerable and increasing research going on concerning the 2/16a-hydroxyestrone ratio, with a current minority of researchers disputing its meaning and validity. A very recent human study5 states: “2-hydroxyestrone levels and 2/16a-hydroxyestrone ratios were significantly lower (p‹0.05) while 16a-hydroxyestrone levels were higher (p‹0.01) in breast cancer patients. 2/16a-hydroxyestrone ratio was the most significant factor predictive of breast cancer.”

Zumoff1 summarizes the evidence for this hypothesis as follows:

  1. Increased 16a-hydroxyestrone in women with breast cancer.
  2. Increased 16a-hydroxyestrone in women with familial high risk for breast cancer.
  3. Increased 16a-hydroxyestrone in mice with high incidence of breast cancer. The degree of increased risk paralleled the degree of increased 16a-hydroxy- estrone in different strains of mice.
  4. Elevated 16a-hydroxylation inherited as an “auto-somal dominant” gene in mice.
  5. Mouse breast cancer virus (MMTV) is associated with increased 16a-hydroxyestrone. When MMTV-free mice were given the MMTV virus, 16a-hydroxy- estrone went up. After removal of the virus from the animals, 16a-hydroxyestrone went down. (Dr. Zumoff did not note this, but studies have found MMTV in nearly half of human breast cancer tissue).
  6. 16a-hydroxyestrone is “genotoxic” (toxic to DNA) in breast epithelial cells in cell cultures, and induces atypical proliferation.
  7. Indole-3-carbinol (see below) decreases 16a-hydroxy- estrone and largely prevents breast cancer in mice with a high incidence of this disease.

According to Zumoff: “an impressive and consistent body of studies since 1966 has made it clear that increased 16a-hydroxylation of estradiol is associated with breast cancer and risk for breast cancer in both mice and humans…” He also makes it clear that all the research studies on this hypothesis have been done by a single research group and that “confirmatory results in other laboratories would be reassuring.”

In summary, the 2/16a-hydroxyestrone-ratio theory maintains that this ratio is an important risk factor marker not only for breast cancer, but for any estrogen-related cancer (breast, ovary, uterus, and possibly prostate): a higher “2/16a” ratio is better and a lower one is worse.

Remember, the significance and utility of the 2/16a-hydroxyestrone-ratio is not absolutely established. But since the evidence is strong, and the “risk modification” factors from diet and supplements are safe, it’s wisest to start now to lower our cancer risk.

The 4-Hydroxyestrone Theory

Dr. Ercole Cavalieri of the University of Nebraska is perhaps the most prominent dissenter from the 2/16a-hydroxyestrone ratio enthusiasm. He argues that while 16a-hydroxyestrone indeed promotes abnormally rapid cell growth and proliferation, it is not the cause of the underlying mutation in cellular DNA that initiates tumor cell formation. He reports that 4-hydroxyestrone (another metabolite of estrone; remember that estrone is made from estradiol) when oxidized can react directly with DNA, leaving a “hole” in the DNA that if unrepaired will initiate cancer. He points to supportive evidence from animal research and observations of high levels of 4-hydroxyestrone in breast cancer tissue. He notes that high levels of the enzyme producing 4-hydroxyestrone are found in both benign and malignant breast tumors, and that known environmental carcinogens “turn on” the production of 4-hydroxyestrone.

(Actually, Dr. Cavalieri’s work5 goes well past known sex-hormone-related cancers. In addition to implicating this mechanism in cancer of the breast, prostate, ovary, and endometrium (the lining of the uterus), he says it has a role in brain cancer and many other cancers. He admits that the data isn’t in yet, but he’s confident it will be developed).

Dr. Cavalieri definitely does not attach the same importance to the 2/16a-hydroxyestrone ratio as other researchers, but he does agree that 2-hydroxyestrone is not very dangerous.

Although placing different emphasis on the importance of 16a-hydroxyestrone, both the 2/16a-hydroxyestrone-ratio theory and the 4-hydroxyestrone theory agree that 16a-hydroxyestrone is not a favorable estrogen. As noted above, 16a-hydroxyestrone is made from estrone; what hasn’t been mentioned yet is that estriol, which is less carcinogenic (if carcinogenic at all) than 16a-hydroxyestrone, is itself made directly from 16a-hydroxyestrone. (Put simply: estradiol -> estrone -> 16a-hydroxyestrone -> estriol.) So, to reduce cancer risk, one can employ a strategy which increases 2-hydroxyestrone at the expense of 16a-hydroxyestrone (remember, as one goes up, the other usually goes down), or one can perhaps “drain off” some of the more hazardous 16a-hydroxyestrone and transform it into the safer estriol. These strategies are discussed in our article, Cabbages and Cancer Risk.

Prostate Cancer: The Di-Hydroxytestosterone (DHT) Hypothesis6

For a number of years, the dominant theory in North America about prostate enlargement and subsequent cancer maintained that di-hydrotestosterone (DHT) metabolized from testosterone within the prostate is the cause of the problem. The DHT theory of cancer is supported by the manufacturers of patent medications, such as “Proscar”, which inhibit the production of DHT from testosterone.

If this theory were true, we’d expect that men who take patent medications such as “Proscar” would have lower rates of prostate cancer. Unfortunately, research finds that the rate of prostate cancer is not only not lower among men who take Proscar, but the risk of one particular type of cancer is actually higher. Further weakening the “DHT is bad” theory, is the fact that a French urologist asked men to take DHT; he observed that their prostate glands actually shrank.

Prostate Cancer: The Testosterone-to-Estrogen Hypothesis

A much more promising theory of prostate cancer builds on the observation that testosterone can be directly metabolized into estradiol! In fact, this pathway is the major source of estrogens for men (both sexes normally have small amounts of opposite sex hormones). If estradiol occurs in excess in the male prostate, then all of the estrogen-and-cancer possibilities discussed above may, and likely do, apply. Indirectly, this is strongly suggested by some of the dietary research reviewed in our article, Cabbages and Cancer Risk.

Breast Cancer and Testosterone

Zumoff combines two groups of studies into an “ovarian dysfunction” hypothesis: one group of studies associates excess testosterone production with higher breast cancer risk, and the other group of studies identifies inadequate progesterone production with higher breast cancer risk. (These latter studies note that women with inadequate progesterone production can still have apparently normal menstrual cycles).

Zumoff writes: “the findings [of testosterone excess] are consistent and impressive, and they are supported by independent findings from many other laboratories: 18 studies to date have reported elevated testosterone levels, although 7 studies have not found such abnormalities.” He also notes that chronic lack of ovulation, accompanied by low progesterone (see below) is often associated with increased testosterone production: “thus, there is no reason to be surprised that a substantial subset of patients with breast cancer show elevated blood and urine levels of testosterone.”

Progesterone and Breast Cancer

Low progesterone can result from poor ovarian function or inadequate stimulation of the ovaries by the pituitary hormone LH or by both. “Luteal inadequacy” is the term used to cover both situations. Zumoff writes: “The evidence for the presence of luteal inadequacy in breast cancer is contradictory but far from negligible.” However, some investigators cited by Zumoff believe that high testosterone accompanying low progesterone is even more important than progesterone alone. Whether low progesterone or high testosterone in women may be considered independent markers of breast cancer risk, or whether they should be considered together, is as yet unclear.

Prolactin and Breast Cancer

According to Zumoff1: “There are four major viable hypotheses concerning hormonal abnormalities in women with breast cancer.” Three of these are discussed elsewhere in this article; the fourth is the “prolactin hypothesis”.

He writes: “The numerous but inconsistent reports of prolactin abnormalities in breast cancer have always seemed like persistent wisps of smoke, suggesting that fire might exist somewhere. The recent exciting findings regarding the long-term effects of pregnancy on lowering serum prolactin may become highly fruitful. The evidence is that the influence of prolactin may be a permissive one, with protection against breast cancer when the levels are lowered, rather than that prolactin excess increases the risk of breast cancer.”

In addition to pregnancies, other ways of lowering prolactin include vitamin B6 (so far, known to be helpful in humans only when prolactin levels are too high), Vitex agnus castus3 (in vitro data only), and nickel4 (animal research only) may also be helpful.

This article was adapted for reprint with permission.
Copyright ® 2000 Agora South, LLC.
For more information on Dr. Wright’s newsletter,
Nutrition & Healing, call 800-851-7100 or 410-223-2611. 


  1. Your editor is greatly indebted to the review by Barnett Zumoff, M.D. (Zumoff B., Hormonal Profiles in Women With Breast Cancer, Obstet Gyn Clin North America 1994;21(4):751-772) for his concise, comprehensive review of hypotheses concerning hormonal abnormalities in breast cancer.
  2. Wright JV, Morgenthaler J. Natural Hormone Replacement for Women over 45, Smart Publications, Petaluma, California, 1997, pages 87-98
  3. Bradlow HL, et al. 2-Hydroxyestrone: the “good” estrogen, J Endocrinol 1996;150:S259-S265
  4. Ho GH, et al. Using 2/16a alpha hydroxyestrone ratio: correlation with serum insulin-like growth factor binding protein 3 and a potential biomarker of breast cancer risk. Ann Acad Med Singapore 1998;27:294-299
  5. For the latest example, see Cavalieri EL, et al. Molecular origin of cancer: Catechol estrogen-3,4-quinones as endogenous tumor initiators. Proc Natl. Acad. Sci. USA 1997;94:10937-10942
  6. A more extensive and detailed discussion can be found in: Wright JV, Lenard L. Maximize Your Vitality and Potency, Smart Publications, Petaluma, California, 1999, pages 153-171