Breast Cancer, Its Link to Abortion and the Birth Control Pill

The Progestins and Breast Cancer

Chris Kahlenborn, MD
Reproduced with permission
(chapter twelve)
Breast Cancer:Its link to Abortion and
the Birth Control Pill

Do progestins cause breast cancer when used by women at a young age?

Progestins are a class of female sexual hormones that have been used by millions of women around the world over the past 25 years. They come in several forms: 1) the all-progestin oral contraceptive pill (ie, the minipill which women may know as Ovrette®, Nor–QD®, or Micronor®); 2) the injectable contraceptives such as Depo–Provera®, which is injected into a woman's muscle; or 3) an implantable contraceptive (ie, Norplant®) which is surgically implanted under a woman's skin. These progestins work by diminishing a woman's frequency of ovulation, by changing the lining of the uterus, and by thickening cervical mucus. The minipill was introduced in 1973 in the U.S., whereas Depo–Provera was approved by the Food and Drug Administration in 1992. Norplant was approved in 1991 and was used by about 1 million Americans in 1995 1. The central question of this chapter is: Do progestins cause breast cancer when used by women at a young age?

Q–12A: Is there any evidence that progestins cause breast cancer in vitro (ie, in a laboratory setting)?

Yes. Progestins began to receive more attention when it was noted that the rate of breast cell division was highest during the luteal phase of a woman's menstrual cycle at which time progesterone levels are highest. Further studies by researchers Pike 2 and White 3 noted that oral contraceptive pills (OCPs) with potent progestins carried higher risks for causing breast cancer than combination OCPs that contained less potent progestins. But do progestins actually affect breast tumor cells when one studies this in vitro (ie, in a laboratory setting)?

Anderson et al 4 noted that women who were given OCPs that contained only progestin, had the highest rate of breast cell division, even higher than standard combination OCPs. In addition, Schoonen et al 5 noted that pregnanes, which are a specific class of progestin (medroxyprogesterone acetate, found in Depo–Provera, is a type of pregnane), enhanced the growth of human breast cancer cells. Catherino et al 6 also noted that norgestrel — which is the hormonal agent in Norplant and which is also found in Ovral and Ovrette, stimulated MCF–7 cells, another type of human breast cancer cell.

Q–12B: Does use of progestins cause early abortions, and if so could this be playing a role?

Progestin use appears to cause more early abortions than use of combination OCPs (ie, those that contain both an estrogen and a progestin), primarily because they allow such a high rate of breakthrough ovulation. William's Obstetrics comments specifically on Norplant: “Up to one third of cycles may be ovulatory based upon serum progesterone determinations.” 7 (ie, a woman may be having three or four abortions per year while on Norplant if she conceives during those cycles). (For details concerning how Norplant or OCPs work, see Appendix 5 of book). Although the studies regarding breakthrough ovulation have been done on Norplant, the same phenomenon probably occurs with all of the other progestins in addition to Depo–Provera. It was noted earlier in Chapter 6 (“Breast Cancer, Its link to Abortion and the Birth Control Pill” by Chris Kahlenborn) that the effect of a very early abortion (ie, within the first week of gestation) regarding breast cancer is difficult to know, but that a very early “hormonal blow” may or may not have an effect on the risk of breast cancer.

Q–12C: Could you explain the history of Depo–Provera and why it has caused so much controversy?

Depo–Provera is a type of progestin (ie, medroxyprogesterone acetate) which is given to women via an intra-muscular injection. It has been used by over 30 million women in more than 90 countries 8 and by about 2 million in the U.S. annually [1]. It is manufactured by the drug company Pharmacia–Upjohn.

The original New Drtug Application for Depo–Provera was filed in 1967. Controversy arose in the mid 1970s, after it was noted that Depo–Provera led to a notable increase in breast cancer in dogs and uterine cancer in monkeys. In spite of evidence that Depo–Provera caused an increase in the breast cancer rate in mice and dogs, and its rejection by the FDA in 1978, the World Health Organization (WHO) initiated a worldwide trial, experimenting on women in the Third World!

There have been three fairly large “trials” including the WHO study, each of which has showed a marked increase in the rate of breast cancer in younger aged women who have used Depo–Provera for a number of years. In spite of this, on October 29, 1992, in what can only be described as a gravely irresponsible decision, the FDA formally approved Depo–Provera as a contraceptive for use in the U.S.! [8]

Q–12D: What did the animal trials show regarding Depo–Provera?

Lanari et al 9 gave 40 mice an injection of DMPA (depot–medroxyprogesterone acetate: the hormone in Depo–Provera) and 40 mice did not receive it (ie, the controls). Breast cancer developed in 16 out of 40 of the DMPA treated mice and 0 out of 40 in the controls. Another classic study by Larsson 10 showed that 5 out of 20 beagles that received DMPA developed malignant breast cancer (ie, 25%) whereas none of the 40 control dogs did.

Q–12E: When an animal study shows a significant risk, is that not a clear warning that the drug carries a high risk in humans?

Absolutely. It is virtually impossible to conceive of how the WHO went on to test this drug on an experimental basis in women in the Third World after noting that the FDA had rejected its approval in 1978. Researchers began to argue that the beagle was a poor model on which to perform clinical trials with DMPA. In a thorough analysis of the subject, Rutteman, a veterinarian researcher from the Netherlands, disputed the claims that the beagle was a poor model: “It has been claimed that the dog is unique in its sensitivity to the mammary tumor promoting effect of progestins and that this tumorigenic effect results from progestin–induced growth hormone (GH) induction. A thorough review of the literature does not support these claims” 11. In regard to the FDA's later approval of DMPA in 1992, a high ranking person who had direct access to the FDA proceedings told this author (confidentially) that there was nothing wrong with the trials in the dogs and that it was inappropriate for the FDA to have approved Depo–Provera. In addition to these comments, one must also wonder why so few were alarmed about DMPA's effects in mice.

Q–12F: What did the WHO experiments on Third World women show?

There were two large trials of women who used DMPA and one smaller one:
  1. The WHO trial 12: A study of 869 women with breast cancer, 377 of whom were under the age of 45. It acquired data from four different countries from 1979 to 1988. The WHO study found that women who had used DMPA at least 3 years before the age of 26 had a 141% increased risk: [RR = 2.41 (0.59–9.87)] of getting breast cancer. Overall risk for any use was 1.21 (0.961.52) and use before a first full–term pregnancy (FFTP) was 1.35 (0.3–6.12).
    The WHO study suffered from some obvious problems. First, it had a huge stack effect (13.5% of “cases” vs 27.6% of “controls” were under the age of 35). Second; very few women took DMPA before their FFTP (first full–term pregnancy) so it was difficult to measure its effect on this select population. Third, women who had taken DMPA for a period of time at a young age would barely have experienced a 10–year latent period. The effects of DMPA in young women who took it early in life might not have shown up because too short of a latent period was used. 
  2. The New Zealand study 13: A study of 891 women, 388 of whom were under 45. Overall risk for all women and any use was 1.0 (0.8–1.3). Women who used DMPA for 2 to 6 years before the age of 25 had a 4.6 (1.4–15.1)–fold risk and women who used DMPA for 2 to 5 years before their FFTP had a 3.2 (0.41–24.3)–fold risk.
    The New Zealand study suffered from the same problems as the WHO study, including an egregious stack effect (ie, 7.2% “cases” vs. 21.7% “controls” under the age of 35) as well as the death effect. It too examined few young women who had taken DMPA before their FFTP.
  3. The Costa Rican study 14: A study of 171 women, 60 of whom were under the age of 45. Overall DMPA use conferred a 2.6–fold risk (1.4–4.7). The risk of use at a young age was elevated but the authors did not give details regarding specific risks. The risk of using DMPA after 10 years since first use was 4.0 (1.5–10.3)
    The Costa Rican trial suffered from a severe stack effect. In addition, it was a rather small trial. Its greatest strength is that because Costa Ricans began using DMPA earlier than other countries, it had a statistic for a longer latent period (ie, a 300% increase in breast cancer after 10 years since first use). The study was funded by USAID and Family Health International — hardly unbiased observers.

Q–12G: Did the stack effect have a bearing on the studies?

Yes. If the stack effect had been properly avoided in the design of the trials, the risks for DMPA use in young women would almost certainly have been significantly higher.

Q–12H: Could you summarize the most worrisome finding of the trials?

The most concerning finding is the fact that each study found an increased risk of at least 2.0 in women who had taken DMPA for more than 3 years before the age of 25 (Costa Rica's results were not printed, but a 2–fold risk deduced from their data is most likely a conservative estimate).

These findings were summarized nicely by Skegg et al 15 who pooled the results of the largest two studies — the WHO and the New Zealand studies. After pooling the data, he found that women who had taken DMPA for between 2 and 3 years before the age of 25 had a 310% statistically significant risk of getting breast cancer [RR=4.1 (1.6–10.90)] whereas women who had taken DMPA for more than 3 years prior to the age of 25 had a 190% increased risk that was also significant [RR=2.9 (1.2–7.1)]. Because women around the world are now taking contraceptives at an earlier age, these findings should have stunned the WHO into stopping the trials immediately.

Q–12I: What happened instead?

In spite of the findings alluded to above, Thomas and Noonan, in summarizing the WHO findings wrote: “These results provide reassurance that women who have used DMPA for a long time and who initiated use many years previously are not at increased risk of breast cancer.” [12, p.833]. In contrast, Paul et al (author of the New Zealand study) at least admitted in the abstract of his 1989 paper that: “Despite the lack of an overall association, these findings suggest that medroxyprogesterone may increase the risk of breast cancer in young women.” [13, p.759]. In addition, in a paper written in March 1992 and sponsored by Upjohn (a pharmaceutical company) 16, Staffa et al noted that “There is need for further study, particularly of patients in potentially high–risk groups, including those with (1) extended hormone exposure before age 26 and⁄or first full–term pregnancy and (2) exposure in the post menopausal period.” Even with these warnings, the FDA went on to approve Depo–Provera for women in the U.S. on October 29, 1992!

Q–12J: Which parties bear the largest responsibility for the failure to warn women of Depo–Provera's noted risks in young women who take it for extended periods of time!

The drug manufacturer of Depo–Provera (ie, Pharmacia–Upjohn), the WHO, and the FDA bear direct responsibility. One's physician most certainly also bears a great share of responsibility because there are only three main studies which he or she should be aware of.

Q–12K: Are black women in the U.S. at especially high risk of developing breast cancer from Depo–Provera?

It would appear so. It has already been noted by Skegg et al [15] that women who take Depo–Provera (DMPA) for 2 years or more before the age of 25 have at least a 190% increased risk of developing breast cancer. But according to a recent article in the Wall Street Journal, Depo–Provera accounted for 19% of all contraceptive use in black women aged 15 to 19 years old, but for only 8% of all contraceptive use in young white women aged 15 to 19 years old 17.

Q–12L: Are black women in South Africa at especially high risk of developing breast cancer from Depo–Proveral?

Yes, it would appear so. A study performed on South African women in 1997 found that 72% of black women had used an injectable progestin contraceptive and that 30% of women had used one for 5 years or more 18. The progestin used most often was either DMPA or norethisterone and these injectable hormones have been used there since the mid 1960s. The study noted that white South African women had a far lower use of these progestin hormones.

Q–12M: In regard to the risks of breast cancer, why has Depo–Provera received so much attention whereas Norplant and the other progestins have received so little?

Depo–Provera was developed long before Norplant and thus has received the most study. Whatever results apply to it, may apply to the other progestins.

Q–12N: Would use of Norplant (which is made of levonorgestrel) and other progestins carry a risk as high as DMPA (Depo–Provera)?

White et al [3] noted that use of norgestrel containing oral contraceptives (eg, Ovral, which is made by Wyeth–Ayerst), gave rise to a 50% increased risk of breast cancer. Pike et al [2] lists Ovral among those OCPs containing the “highest potency progestins.” In his study, high potency progestins carried an infinitely increased risk in women who took them for more than 6 years when under the age of 25. Thus, there is no reason to believe that Norplant is any less dangerous than Depo–Provera, and when one considers the fact that women keep this hormone in their bodies for 5 years at a time, it could cause a significantly higher risk than Depo–Provera.

Q–120: How dangerous are the progestin “minipills?”

No one is really sure, although they may contain norgestrel which was noted to stimulate the growth of breast cancer cells by two different researchers 19,20 The Oxford study noted an overall increased risk of 19% (ie, 1.19 [0.89–1.49]) in women who had taken progestin only OCPs (ie, minipills) for 4 or more years, but the Oxford study said nothing about extended use in young women, especially before their FFTP [21, p.98S]. The New Zealand study 22 noted that women who had taken progestin only OCPs before age 25 had a 1.5–fold risk (0.73-2.9), although the result did not achieve statistical significance.

Q–12P: Does DMPA use cause any other changes?

Yes. “Prolonged DMPA use may be associated with reversible reduction in bone density, probably related to suppression of endogenous production of estrogen” [8]. In addition, women who used DMPA had a 1.2 non–significant relative risk of developing cervical cancer in two large trials [23, p.673]: (ie, [RR=1.2 (0.3–4.5)] in the Costa Rican trial; [RR= 1.2 (0.84–1.72)] in the WHO trial). Last, “although significant findings differed among centers, overall DMPA users had higher low–density lipoprotein (ie, the ‘bad cholesterol’) and lower high–density lipoprotein levels (ie, the ‘good cholesterol’).” [8, p.1546]

Q–12Q: Has anyone studied the effect of the progestins on cervical cancer in humans?

Yes, in a large study Herrero et al 24 found that women who had received injectable progestins (ie, usually DMPA [depot–medroxyprogesterone] or norethisterone enanthate) for at least 5 years and who had used them at least 5 years ago, suffered a 430% increased risk of developing cervical cancer [RR=5.3 (1.1–10.0)].

Q–12R: Has any progestin been withdrawn by its manufacturer?

Yes, Deladroxate was pulled by its original manufacturer because “this drug induced a high number of breast cancers in dogs...” 25. In spite of this, other smaller manufacturers have now started making it, and it is used in Latin American countries.

Q–12S: Are there any countries where women may be at particular risk from the injectable progestin contraceptives?

Yes, China, New Zealand, South Africa, and some Latin American countries such as Costa Rica have extensive use of the injectable progestin contraceptives and their women, especially those who used these hormones for extended periods of time (ie, over 2 years) at a young age, are at increased risk for breast cancer [16, 18].

Q–12T: Why has no one said anything to women who get Norplant or Depo–Provera injections?

Actually, some people have tried, but the monetary, political, and media forces have blocked any headway. Staffa, commenting on the studies on DMPA wrote: “...two potential high–risk groups were suggested: women who used Depo–MPA (ie, DMPA) at least 2 years before age 25 and long–term users of Depo–MPA before first full–term pregnancy” [16]. In addition to this, a special panel advised the FDA not to approve Depo–Provera at all: “In the end, Weisz and Stolley concluded that the FDA should not sanction contraceptive uses of the drug at all. Ross, however, advised the agency to approve it for women who are mentally retarded or drug addicts.” 26

Q–12U: Does use of the artificial progestins increase the ease of HIV transmission?

Preston Marx, a virologist, exposed two groups of monkeys to an immune virus called SIV (Simian immunodeficiency virus), which is similar to HIV. He gave 10 of these monkeys a placebo and 18 other monkeys received progesterone. Only one of the placebo group contracted SIV whereas 14 out of the 18 monkeys who had received progesterone contracted the virus 27. “Three of the progesterone group went on to be rapid progressors, developing AIDS in a few weeks and dying within three to four months. Normal disease course in monkeys is two years,” according to Marx [27]. In 1996, researchers found that giving progesterone to female monkeys increased their risk of getting AIDS [27]. This makes sense biologically because progesterone is known to thin the vaginal mucosal lining which may reduce the barriers to viruses or bacteria.

Preston's work is supported by studies of women in Thailand, Rwanda, and Kenya. Ungchusak et al. have noted that prostitutes who used injectable contraceptives (which consist of progestins) had a 240% increased risk [RR = 3.4 (1.2–13.2)] of contracting HIV when compared to those prostitutes who did not use it 28. Allen et al 29 noted that women in the general population of Rwanda who used injectable contraceptives, had about a 25% higher HIV infection rate than women who did not use them (ie, 38% vs. 30% repectively). Plourde 30 noted that Kenyan women who were positive for HIV, had used medroxprogesterone (ie, the ingredient in Depo–Provera) over 5 times longer, as a group, than women who did not have HIV.

Another important finding by Mostad et al in the 1997 Lancet 31 reported that in a group of 318 Kenyan women who had the human immunodeficiency virus, those who were users of depot medroxyprogesterone were 2.9 times more likely than women who did not use hormonal contraceptives to have HIV–1 cells in their cervical secretions. These women would theoretically be more likely to spread the virus.

These findings should shock Africans and Asians because the injectable progestins are widely used in such countries as Indonesia, Thailand, Kenya, Botswana, and Rwanda, where the current ramifications and future implications are especially serious because these countries are among the hardest hit by AIDS [1].


1 Garrett L. Contraceptive linked to AIDS risk. Pittsburgh Post-Ga-zette. May 7, 1996. [Back]

2 Pike MC, Henderson BE, et al. Breast cancer in young women and use of oral contraceptives: possible modifying effect of formulation and age at use. The Lancet. October 22, 1983: 926-929. [Back]

3 White E, Malone K, Weiss N, Daling J. Breast cancer among young U.S. women in relation to oral contraceptive use. J Natl Cancer Inst. 1994; 86: 505-514. [Back]

4 Anderson TJ, Battersby S, et al. Oral contraceptive use influences resting breast proliferation. Hum Pathol. 1989; 20: 1139-1144. [Back]

5 Schoonen W, et al. Effects of two classes of progestagens, pregnane and 19-nortestosterone derivatives, on cell growth of human breast tumor cells: II. T47D cell lines. J Steroid Biochem Mol Biol. 1995; 55: 439-444. [Back]

6 Catherino WH, Jeng MH, et al. Norgestrel and gestodene stimulate breast cancer growth through an oestrogen receptor mediated mechanism. Br J Cancer. 1992: 945-952. [Back]

7 Cunningham FG, et al. William's Obstetrics. Stamford, CT: Appleton & Lange; 1993: 1321-1340. [Back]

8 Kaunitz A. Long-acting injectable Contraception with depot medroxyprogesteroneacetate. Am J Obstet Gynecol. 1994; 170: 1543-1549. [Back]

9 Lanari C, Molinolo AA, et al. Induction of mammary adenocarcinomas by medroxyprogesterone acetate in barb/c female mice. Cancer Letters. 1986; 33: 215-223. [Back]

10 Laarsson KS, et al. Predictability of the Safety of Hormonal Contraceptives from Canine Toxicology Studies. In: Michal, F. ed. Safety Requirements for Contraceptive Steroids. Cambridge, Cambridge University Press; 1989: 203-269. [Back]

11 Rutteman GR. Contraceptive steroids and the mammary gland: is there a hazard? Breast Cancer Research and Treatment. 1992, 23: 29-41. [Back]

12 >Thomas DB, et al. Breast cancer and depot-medroxyprogesterone acetate: a multinational study. The Lancet. 1991; 338: 833-838. [Back]

13 Paul C, Skegg DCG, Spears GFS. Depot medroxyprogesterone (Depo Provera) and risk of breast cancer. Br Med J. 1989; 299: 759-762. [Back]

14 Lee NC, Rosero-Bixby L, et al. A case-control study of breast cancer and hormonal contraception in Costa Rica. J Natl Cancer Inst. 1987; 6: 1247-1254. [Back]

15 Skegg DCG, Noonan EA, et al. Depot medroxyprogesterone acetate and breast cancer [A pooled analysis of the World Health Organization and New Zealand studies]. JAMA. 1995: 799-804. [Back]

16 Staffa JA, Newschaffer CJ, et al. Progestins and breast cancer: an epidemiologic review. Fertility and Sterility. 1992; 57: 473-491. [Back]

17 Freedman AM. Why teenage girls love the shot; Why others aren't too sure. The Wall Street Journal. October 14, 1998. [Back]

18 Bailie R, et al. A case-control study of breast cancer risk and exposure to injectable progestin contraceptives. S Afr Med J. 1987, 87: 302-305. [Back]

19 Jordan C, Jeng MH, et al. The estrogenic activity of synthetic progestins used in oral contraceptives. Cancer Supplement. 1993; 71: 1501-1505. [Back]

20 Jeng MH, Parker CJ, et al. Estrogenic potential of progestins in oral contraceptives to stimulate human breast cancer cell proliferation. Cancer Research. Dec. 1, 1992: 6539-6546. [Back]

21 Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal contraceptives: further results. Contraception. 1996; 34: S1-S106. [Back]

22 Skegg D, et al. Progestogen-only contraceptives and risk of breast cancer in New Zealand. Cancer Causes and Control. 1996; 7: 513-519. [Back]

23 Boyle P, Chilvers C, et al. Depot-medroxyprogesterone acetate (DMPA) and cancer: Memorandum from a WHO meeting. WHO Bulletin OMS. 1993; 71: 669-676. [Back]

24 Herrero, et al. Injectable contraceptives and risk of invasive cervi-cal cancer: evidence of an association. Int J Cancer. 1990; 46: 5-7. [Back]

25 Koetsawang S. Once-a-month injectable contraceptives: efficacy and reasons for discontinuation. Contraception. 1994; 49: 387-398. [Back]

26 Sun M. Panel says Depo-Provera not proved safe. Science. 1984; 226: 950-951. [Back]

27 Marx PA, et al. Progesterone implants enhance SIV vaginal trans-mission and early virus load. Nature Medicine. 1996; 2: 1084-1089. [Back]

28 Ungchusak, et al. Determinants of HIV infection among female commercial sex workers in northeastern Thailand: results from a longitudinal study. J Ac Immune Defic Syn Hum Retro. 1996; 12: 500-507. [Back]

29 Allen S, et al. Human immunodeficiency virus infection in urban Rwanda. JAMA. 1991; 266: 1657-1663. [Back]

30 Plourde, et al. Human immunodeficiency virus type 1 infection in women attending a sexually transmitted disease clinic in Kenya. J Infect Dis. 1992; 166: 86-92. [Back]

31 Mostad SB, et al. Hormonal contraception, vitamin A deficiency and other risk factors for shedding HIV-1 infected cells from the cervix and vagina. The Lancet. 1997; 350: 922-927. [Back]