header icons

Is Abortion Related to Breast Cancer? Too Political to Address

Ellen covered the international economics con-ference in Puebla, Mexico, for the Houston Catholic Worker several years ago. A graduate of the University of Notre Dame, she has an MBA from Indiana University.

Angela Lanfranchi, M.D., a New Jersey breast surgeon, recently described under oath, in a California court, the medical community’s approach to the abortion/ breast cancer link:

“In September 1999 I wrote a letter to the president and each of the board members of my medical society, the American Society of Breast Surgeons. My letter … said that doctors… need to get this information (about abortion and breast cancer) to the public, and asked that an expert be invited to address the Society on this issue. Some time later I called the president, Dr. Rachel Simmons, and she told me, apologetically, that she presented it to the board but they felt it was ‘too political.’

“In March 2000 I attended the Miami Breast Cancer Conference…. I asked the conference director, Dan Osman, M.D. if he knew there was a link between abortion and breast cancer. I was stunned when he said that he did. I asked him why there couldn’t be a presentation about it at the meeting. He said it was ‘too political.’

“Over the past three or four years, I have spoken with many authorities and people in a position to be well informed. Some have been straightforward and said they know it is a risk factor but felt it was ‘too political’ to speak about. Others have been evasive…. Some have been openly hostile…. Some initially hostile doctors … debated it with me and have changed their minds. Some pro-choice doctors have come to agree it is true and do tell their patients about the risk. Some doctors who were initially skeptical have started obtaining a complete reproductive history on their patients and found as I did that … cases of breast cancer in young women are associated with an abortion history….” (Bernardo et al v. PPFA)

The American Medical Association, New England Journal of Medicine, American Cancer Society (ACS), Susan B. Komen Foundation, Planned Parenthood, the National Abortion Rights Action League, Health and Human Services’ National Cancer Institute (NCI), and the National Breast Cancer Coalition (NBCC) have denied or ignore claims of a link between abortion and breast cancer. Is that link the concoction of “anti-choice zealots”, as Planned Parenthood Federation of America (PPFA) claims? Or could Dr. Lanfranchi be speaking the truth?

Breast cancer is a major public health issue in the United States. The NCI states: “The incidence of breast cancer has been rising for the past two decades….” The 2002 ACS Annual Report on the Status of Cancer stated that between 1995 and 1999, 139 out of 1000 white women and 121.5 out of 1000 black women were diagnosed with breast cancer annually. The NCI estimates 1 in every 8 American women will get breast cancer in her lifetime.

Joel Brind, Ph.D, is a professor of biology at Baruch College of the City of New York. In his 1996 meta-analysis (an analysis of all the relevant studies to date) in the Journal of Epidemiology and Community Health, he found a 30% risk increase of breast cancer among all women who had had an abortion, and a 50% risk increase among women who had aborted before their first full-term pregnancy. This surprised many people. Chris Kahlenborn, M.D., a Pennsylvania internist, recalls that when he first heard this idea at a pro-life meeting, he thought it sounded “goofy,” and that the evidence for it seemed to be “an emotional argument.” To satisfy his medical curiosity, he read the studies to make up his own mind. He became convinced, and wrote a book, Breast Cancer: Its Link to Abortion and the Birth Control Pill.

What convinced scientists like Dr. Kahlenborn and Dr. Brind that the medical establishment is wrong here? Elizabeth Shadagian, M.D., clinical assistant professor of obstetrics and gynecology at the University of Michigan School of Medicine, declared under oath in the California court:

“Based upon my professional education, training, study of the literature, and my own independent meta-analysis of spontaneous and induced abortion in pre and post-menopausal women, it is my opinion that induced abortion causes an increased risk of breast cancer as follows: (1) by abrogation of the protective effect of a first full term pregnancy (fftp) on breast cancer risk via the increased risk that comes from delaying the fftp to a later time in a woman’s life; and (2) the independent effect of increased breast cancer risk apart from the delay of fttp, which may be even greater if: A) the induced abortion is before a first full term pregnancy, … B) the woman is a teen, … C) the woman is black, … D) the woman is over the age of 30, … E)the pregnancy is terminated at >12 weeks gestation … or F)the woman has a family history of breast cancer….” (Bernardo et al v. PPFA)

So far, 28 studies on abortion and breast cancer show an increase of risk of breast cancer from abortion. The oldest studies were done in Japan in the 1950’s. In 1980, the landmark Russo and Russo study tracked the incidence of breast cancer in rats which had undergone abortions versus that in control populations. 100% of the rats which had abortions “developed benign lesions,” and 77% of these developed carcinomas. The Russo and Russo study also observed that breast cells in rats reached maturity only after a full-term pregnancy. This study underpins a biological theory tested in later studies of humans. According to this theory, estrogen blood levels rise by 2000% through the first trimester of pregnancy, and a form of estrogen, estradiol, causes breast cells to multiply. During the third trimester, differentiation occurs, shutting off cell multiplication and preparing women to lactate. If a woman has an abortion she has more undifferentiated, and therefore cancer vulnerable, cells than she had before the pregnancy. These cells are also exposed to carcinogenic estrogen levels 2000% higher than usual. Planned Parenthood’s fact sheet “Anti-Choice Claims about Abortion and Breast Cancer,” however, argues that this biological theory has “never been proven.”

What would make abortion a dangerous risk factor? Dr. Joel Brind points out, “while … cigarette smoking require(s) thousands of exposures to produce detectible increases in cancer incidence, the induced abortion patient’s risk of breast cancer later in life is measurably increased after a single exposure.” Even though estrogen is a substance natural to the body, sizeable exposure is a known risk factor. Breast cancer organizations agree that early menarche (puberty) and late menopause are known risk factors, thus implicating estrogen exposure. Moreover, full breast maturation is known to protect against breast cancer; therefore, experts acknowledge that childbearing mitigates against cancer risk. The controversy here arises from the evidence that abortion, which according to this theory exposes many undifferentiated cells to high levels of estrogen, is itself a risk factor for breast cancer.

The court records in the recent California case, Bernardo, Colip & Duffy-Hawkins v. Planned Parenthood Federation of America and Planned Parenthood of San Diego and Riverside Counties shed light on this question. Three women sued their local Planned Parenthood branch and PPFA because, when they obtained abortions, they were not warned about the risk of contracting breast cancer. The Planned Parenthood fact sheet denying “anti-choice claims” was argued to be misleading. The court, however, decided in favor of Planned Parenthood because a California law protected Planned Parenthood’s free speech rights. Whether a medical provider’s right to free speech should override the patient’s right to informed consent would be worth extended discussion. But our concern here with the Bernardo suit is the expert testimony which appears to make the medical community’s, and particularly the NCI’s, position untenable. The testimony 1) mentions clinical studies that point to a definitive biological mechanism which accounts for abortion-caused breast cancer; 2) exposes the medical establishment’s inability to dispute key facts that support a link between abortion and breast cancer; 3) exposes an obvious flaw in a watershed study that claimed to disprove a statistical link; and 4) indicates a politically motivated silence on this issue within the medical community.

The NBCC acknowledges that clinical studies are “the future of breast cancer research.” Oncologist Dr. Jose Bufill, a clinical professor at Indiana University Medical School, declared in the Bernardo case:

“There is clinical research data suggesting that …. induced abortion may predispose women to a specific form of breast cancer unique to a particular carcinogenic pathway…. Ownby …. found that patients with a history of induced abortion had tumors with a more aggressive histological appearance… and with a more aggressive clinical course. Breast cancers in women with prior abortions recurred after initial treatment twice as often as those affecting women without a history of abortion (Ownby, 1983). Olsson and colleagues made similar observations in 1991. They found that young women who had abortions and subsequently developed breast cancer had a statistically higher incidence of cancers with ….factors … associated with an adverse prognosis (Olsson, 1991).

Clinical studies like Ownby’s and Ollson’s point to progress towards discovering the biological mechanism through which abortion causes breast cancer. It makes sense that a single sudden carcinogenic exposure could create a form of cancer that, as Bufill states, has “a more aggressive clinical course.”

In the Bernardo case, defense witness Dr. Polly Newcomb, Senior Scientist at the University of Wisconsin, claimed to have found weaknesses in the studies showing a link between abortion and breast cancer. But Dr. Jane Anderson, clinical professor of pediatrics at the University of California San Francisco, declared to the court that:

Dr. Newcomb’s Declaration does not and cannot dispute that there is indeed a statistical association between induced abortion and increased breast cancer risk observed in the epidemiological literature that cannot be explained by chance…. She does not dispute the fact that 28 out of the 37 studies … report an increase in the risk of breast cancer overall among women who have one or more induced abortions, and that 17 of these 28 positive studies achieve statistical significance (i.e., greater than 95% certainty that the finding is not due to chance). The dispute, rather, is over whether this observed association is more likely explained by the hormonal disruption that is known to occur when a woman’s pregnancy is artificially interrupted, or by “reporting bias.”

Dr. Newcomb did not propose an alternative biological theory but attributed the statistical link between abortion and breast cancer to reporting bias, also known as “response bias.”
Response Bias and the Melbye Study

Response bias is a term for false information obtained from study participants. If present, it skews study results. Some scientists argue that women will not admit to having an abortion unless they believe they have suffered an adverse consequence, such as breast cancer. The response bias charge has been used by the medical establishment to discredit earlier studies linking abortion and breast cancer, including the Brind meta-analysis. The Melbye study, described in PPFA’s fact sheet “Anti-Choice Claims about Abortion and Breast Cancer,” claimed that it avoided the response bias problem:

“One of the most highly regarded studies on abortion and breast cancer was published in the New England Journal of Medicine in 1997. This study of 1.5 million women found no overall connection between the two (Melbye et al., 1997). This study benefited from its size – 1.5 million women – and by linking data from the National Registry of Induced Abortions and the Danish Cancer Registry, thereby avoiding one of the pitfalls observed in some case-control studies – that women with breast cancer were more likely to recall having had an induced abortion than women without breast cancer, particularly because abortion had been illegal (Brody, 1997; Westhoff, 1997).”

Many cancer organizations tout this study, by Danish epidemiologist Mads Melbye, as important in disproving an abortion/ breast cancer link. The ACS fact sheet “What Causes Breast Cancer?” reassures women that “A large, recent study indicated that induced abortions do not increase the risk of breast cancer.”

Until June 2002, the NCI fact sheet stated: “Recent large studies, particularly cohort studies, show no association between breast cancer and previously recorded spontaneous or induced abortions. In a large-scale epidemiologic study reported in The New England Journal of Medicine in 1997, researchers compared data from Danish health registries that included 1.5 million women and more that 10,000 cases of breast cancer. …The authors found that ‘induced abortions have no overall effect on the risk of breast cancer.'” This fact sheet has been removed. The NCI’s current physician guidelines sheet now bases its position against a link on the findings of an NCI-funded recent study performed in Shanghai, China. This new study has not yet been subject to critical examination.

Melbye’s study’s claim to avoid response bias should not distract attention from fatal flaws in the study’s design. According to the testimony of Jane Anderson, M.D., 50% of the subjects Melbye studied were not appropriate for the study cohort, that is, a group followed forward in time from a medically significant event:

“Specifically, fully one-fourth of the Melbye cohort was born between the years 1968-1978, all of whom were under age 25 at the end of the study. Hence they were too young to account for more than 8 cases of breast cancer, while accounting for some 40,000 abortions. In like manner, the oldest one-fourth of the Melbye cohort, born between 1935-1945, contained most of the breast cancer patients, yet they were too old to have most of their abortion histories – which registry only commenced in 1973 – recorded in the computer registry.”

So one-fourth of the Melbye sample were too old to have had their abortion histories reliably recorded. Another fourth was too young to be followed forward through middle-age, when women typically get breast cancer. This huge flaw in 50% of the Melbye cohort should have been obvious to anyone who has studied high school mathematics. However, the peer review of the Melbye proposal approved it for US government funding. On May 31, 2002, just before the NCI fact sheet disappeared, Patricia Hartge, an NCI senior scientist, stated that the Melbye study is “exactly the study you’d want to do…. as clear as a bell.” When asked if Melbye’s cohort design was proper, she answered, “Yes…It is a standard and sound design.”

Breasts, not Bombs

Few Americans know that the Department of Defense (DoD) funds cancer research, and fewer know about its program’s peculiar origins. The National Breast Cancer Coalition website states, “Congress… budgeted $25 million to the DOD to screen female military personnel for breast cancer in FY1992. Because of continued lobbying in FY93 by … organizations, such as the National Breast Cancer Coalition, Congress boosted its commitment … by appropriating $210 million to the DOD for scientific research. Within the DOD, the Congressionally Directed Medical Research Programs (CDMRP) were created to disburse research monies to qualified scientists.” Total funding through the DoD Breast Cancer Research Program (BCRP) in FY 1993-2001 totalled $ 1.145 billion, which would buy 26 M1A1 battle tanks. Apart from the DoD, total funding for breast cancer research through the NIH in FY 1992-2001 was $4.22 billion. Lots of federal money, one fifth of it defense money, is going for breast cancer research.

The Melbye study was awarded $486,801 through the BCRP in fiscal year 1995. A technical abstract accompanying the award, DAMD award 17-96-1-6321, states “A database containing complete parity information on all women in Denmark born since 1935 will be created based on information from the Central Person Registry in Denmark.” But, as Dr. Anderson testified, women born between 1935 and 1945 were too old to have been properly accounted for in the Danish abortion registry begun in 1973. The abstract’s reference to the year 1935 should have alerted statisticians to a problem in Melbye’s cohort selection.

Furthermore, Gail Whitehead, assistant to the Director of the BCRP, states that the Melbye study is the only abortion/ breast cancer study the BCRP ever funded. The BCRP’s panel of experts must have considered the case closed with that study. If Jane Anderson is correct about the obvious inadequacy of the cohort in the Melbye study, those experts should be fired.

Time for a congressional investigation

“The informing function of Congress,” wrote Woodrow Wilson, “should be preferred even to the legislative function.” Rep. Chris Smith (R-NJ) is trying to get a congressional investigation here. During debate on an abortion bill in 1999, Smith alleged a “coverup”:

Rep. Smith (NJ): My friend from New York Mrs. Lowey says there is no linkage of abortion and breast cancer. Yet 10 out of 11 studies on American women report an increase in breast cancer when a woman undergoes an abortion. The “denial” people remind me of the tobacco institute … who year after year said that there is no connection between smoking and lung cancer …. Dr. Janet Daling, with a National Cancer Institute funded study, found that after just one abortion there is an increase in the aggregate of all women of about 50 percent in the propensity to get breast cancer. She is not a pro-lifer. She does not agree with my position…. Why the coverup?….

Rep. Lowey (NY): ….In 1996 the National Cancer Institute, concerned that some anti-abortion groups were misrepresenting the science on the subject issued a statement: “….There is no evidence of a direct relationship between breast cancer and either induced or spontaneous abortion.”

Note Rep. Lowey’s reference to the NCI’s 1996 statement. In 1996, the Melbye report had not yet been published. On what basis could the NCI have coherently defended a position that Planned Parenthood justifies based on a study not published until 1997?

The Congress has conducted hearings on public health issues such as methamphetimines and date rape drugs, and armed services issues such as morale, welfare, and recreation. Why not look into abortion and breast cancer, and its defense funding history?

Epidemiologist Janet Daling, described by Rep. Chris Smith as “not a pro-lifer,” was quoted in the L.A. Daily News on September 28, 1997: “I would have loved to have found no association between breast cancer and abortion, but our research is rock solid, and our data is accurate. It’s not a matter of believing, it’s a matter of what is.”

Is Dr. Angela Lanfranchi right? Are women increasing their risk of breast cancer because it is “too political” to insist that they be provided accurate and truthful information? We need full and open Congressional hearings on this issue. Since one of eight American women gets breast cancer, it is time to discover “what is.”

Houston Catholic Worker, Vol. XXII, No. 6, November 2002.