Advances in human biology and continued epidemiologic ...



The Abortion and Breast Cancer Link:

Advances in Human Biology and Continued Epidemiologic Studies Increasingly Point to a Link

I. INTRODUCTION::

In the past 30 years, landmark advances in the developmental and the molecular biology of the breast coupled with the repeated epidemiologic studies from around the world have more and more tightly linked induced abortion as an independent factor in breast cancer. The “independent link” refers to the fact that induced abortion before 32 weeks gestation will change the breast in a way that makes it more likely to develop breast cancer in later years.

Indisputable ways induced abortion increases breast cancer risk

Given the present state of knowledge of breast physiology and reproductive risks described in standard medical texts, it is indisputable that induced abortions cause an increase in breast cancer incidence.

It is well known that different pregnancy outcomes lead to changes in the rates of breast cancer among women:

• Once pregnant, if a woman chooses to maintain her pregnancy and achieves a full term pregnancy, she will lower her risk of breast cancer. This is well known and undisputed in medical circles.

• If she chooses to have an induced abortion she may remain childless, a condition which increases her risk of breast cancer.

• Or if she chooses abortion and then has another pregnancy, the abortion will have delayed this pregnancy which delay also increases her risk of breast cancer.

• If she already had a full term pregnancy and chooses to abort a subsequent pregnancy she loses the risk reduction that an additional full term pregnancy would have afforded her.

• The use of instruments such as dilators during an abortion increases the risk of having a premature delivery in future births. If that premature delivery is before 32 weeks, she will have an increased risk of breast cancer.

The well established research conclusions that underpin this indisputable increase in the risk of breast cancer from induced abortions are:

1) A full term pregnancy decreases the risk of breast cancer.

2) Delay of a full term pregnancy increases breast cancer risk.

3) Every full term pregnancy after the first pregnancy further decreases the risk of breast cancer risk.

4) Induced abortion increases the risk of premature delivery, thus increasing the risk of breast cancer if the delivery is before 32 weeks

Long Established Insight on Childlessness and Breast Cancer:

It has been known for centuries that remaining childless increases a woman’s risk for breast cancer and conversely, is was also known that pregnancy was protective. In 1743, Ramazzini of Padua observed that there was an increased number of breast cancer among nuns. In 1842, a hundred years later, Rigioni-Stern noted a 3 fold increase risk of breast cancer among nuns. Nuns were largely childless whereas the rest of the population had pregnancies early on in their reproductive lives.

Delayed first pregnancy: The longer a woman waits to have her first full term pregnancy (FFTP, used hereafter), the higher her risk of breast cancer as her immature, cancer-vulnerable breast tissue is exposed to carcinogens for a longer duration. A woman who has her FFTP at 20 years of age has a 90% lower risk of breast cancer than a woman who remains childless or waits until she is 30 for her FFTP.[i]

Each year a woman delays pregnancy she has a 5% increase in risk for pre-menopausal breast cancer and a 3% increase in risk for postmenopausal breast cancer.[ii] For example, having an induced abortion at age 20 followed by a full term pregnancy at age 30 would increase her risk of premenopausal breast cancer by 50%. Other studies have shown that breast cancer risk increases 0.7% for each year subsequent births were delayed after her first birth.[iii] Yet another study has shown that if a woman has a pregnancy and lactates within five years after an abortion, she will decrease her risk by 20% compared to if she waits 10 or more years to lactate for the first time.[iv]

Increased number of pregnancies: For each pregnancy she has subsequent to her first, her risk of breast cancer will decrease another 10% [v].

Abortion and Subsequent Premature Births: Two large meta-analyses show that induced abortion increased a woman’s risk of premature delivery.[vi] Also, the more induced abortions a woman has the higher her risk of subsequent premature births.[vii] In 2006 the Institutes of Medicine listed induced abortion as an immutable cause of premature birth in its publication on prematurity.[viii]

No matter the length of her pregnancy, in the first 32 weeks she will have changes in her breast tissue which will increase her risk of breast cancer. When a woman gives birth naturally, it takes many hours to dilate the cervix for birth. During an abortion the cervix is forcibly dilated and subjected to injury. Due to use of instruments such as dilators during an abortion she may have a premature delivery. If the premature delivery is before 32 weeks, she will have an increased risk of breast cancer. Approximately 3% of all premature deliveries are before 32 weeks.[ix]. Approximately 12.5% of all births are before 37 weeks and are considered premature [x]

Breast Tissue Changes in First Pregnancy: When a woman becomes pregnant for the first time, her immature and cancer-vulnerable breast tissue matures into cancer-resistant tissue. Eighty-five percent of her breast will become fully mature type 4 lobules which contain the first milk, colostrum. After weaning, theses lobules regress to type 3 lobules which are also cancer resistant. This biological change accounts for the known fact that never having a full term pregnancy (nulliparity) increases a woman’s risk of breast cancer (as in the case of nuns). After a full term pregnancy, only 15% of her breast tissue remains susceptible to forming cancer.[xi] It is the genetic changes that occur in the breast lobules during a full term pregnancy which gives lifelong protection. [xii] Molecular biologists have also determined that progenitor or stem cells in the breast do not become terminally differentiated (reached their full potential growth or matured) until they have undergone the environment of pregnancy and have lactated. [xiii]

Premature Births and Increased Risk of Breast Cancer: Several studies have shown that prematurity before 32 weeks more than doubles breast cancer risk.[xiv] In fact, the biologic mechanism for premature delivery, induced abortion and second trimester miscarriage as causes for increase risk of breast cancer is the same mechanism for all three: Abortion, premature delivery or 2nd trimester miscarriage all leave the breast with more places for cancers to start when the pregnancy ends. The woman’s breasts have been exposed to the same pregnancy hormones (estrogen, progesterone, human chorionic gonadotropin (hCG) and human placental lactogen hPL) all of which lead to the same breast changes. Elevated levels of estrogen and progesterone, stimulated by hCG, cause more cancer vulnerable breast tissue to form. It is only after 32 weeks of gestation that the elevated levels of hPL, concert with other pregnancy hormones, allow the full maturation to cancer-resistant breast tissue to occur. Therefore, whether the pregnancy ends with a premature birth before 32 weeks, a 2nd trimester miscarriage (which generally have normal hormonal levels) or an induced abortion, breast cancer risk is increased.

II. Developmental Biology Supports The Independent Link Between Breast Cancer And Induced Abortion

Developmental biology concerning the breast changes that occur with puberty and with a normal pregnancy further supports the theory of an independent link between induced abortion and breast cancer.

Immature, cancer-vulnerable Type 1 and 2 lobules mature to cancer-resistant Type 4 lobules after 32 weeks of pregnancy. After weaning, Type 4 lobules regress to Type 3 lobules which have permanent genetic changes that protect against the development of cancer within these Type 3 lobules.

Lobules are units of breast tissue comprised of a milk duct with surrounding mammary (milk) glands, which are in turn composed of individual breast cells. Each breast cell contains a nucleus—a center space that contains DNA, the coded complete blueprint of genetic information that every cell in the body contains.

After puberty a woman has only Type 1 and 2 lobules in her breasts. During the first half of pregnancy of Type 1 and 2 lobules increase in number as the breast doubles in volume. . 85% of breast cancers arise in Type I lobules which form ductal cancers.. 10-15% of all breast cancers arise in Type 2 lobules which form lobular cancers.. Type 1 lobules have a greater number of estrogen and progesterone receptors in their cells’ nuclei than Type 2 lobules. Type 2 lobules have more of thes receptors than Type 3 lobules, which have significantly fewer.

The source of any cancer that develops in a body results from a mutation or damage done to the blueprint of the cell’s DNA. Unless the mutation is a critical one, such as in a gene that fixes errors in DNA when it is copied during cell division, most cancers form after several mutations have built up in a cell over a number of years, When cancer cells arise in the milk ducts but do not penetrate the basement membrane (outer layer of the duct) it is said to be in-situ cancers. These cancers are curable because they haven’t penetrated or invaded the basement membrane where the lymphatic channels on blood vessels are located, so they can’t spread to other parts of the body. Invasive cancers have penetrated the basement membrane and can spread throughout the body becoming metastatic and life-threatening.

Most invasive cancers start as in-situ cancers.

Pregnancy causes Type 1 lobules to increase the number of ductules (which become mammary glands) from an average of eleven ductules per lobule to forty-seven, becoming Type 2 lobules. Type 2 lobules mature still more fully into Type 3 lobules at which stage there is an average of eighty ductules in each lobule. Type 3 lobules have very few estrogen and progesterone receptors and do not quickly copy their DNA, thereby decreasing the possibility of mutations and carcinogenesis.[xv] By 32 weeks these Type 3 lobules start to produce colostrum, the first milk, thereby becoming Type 4 cancer-resistant lobules.

We know exactly which genes have been turned off and on (down-regulated and up-regulated) throughout a full-term of pregnancy.[xvi]

During this time of maternal breast maturation, in the womb at 32 weeks gestation, a parallel development is occurring in the fetus: the solid cords of epithelial cells on the fetal chest wall become canaliculized (become hollow), thereby developing the milk ducts and glands of the newly forming fetal breasts.[xvii]

A mother’s breasts enlarge very soon after conception, making sore and tender breasts one of the first signs of pregnancy. Even before the embryo (or blastocyst) implants in its mother’s womb, a chemical signal, hCG (human chorionic gonadotropin) produced by the embryo, causes its mother’s ovaries to increase production of estrogen and progesterone in order to sustain the pregnancy. The maturation process that protects a woman from breast cancer happens only because the child in her womb produces the hormones hCG and hPL which prepare the mother to breast feed. In the first half of pregnancy, hCG stimulates estrogen and progesterone levels which cause the breast to enlarge with increased numbers of Type 1 and Type 2 lobules. In the later half of the pregnancy, hPL rises three times higher than the mother’s prolactin levels by the end of pregnancy, and enables full differentiation to Type 4 lobules which produce colostrum (the first milk). Thus the baby causes the mother’s breasts to get ready to feed him.

To summarize: Infants are born with Type 1 lobules in their breast tissue. During puberty in girls some Type 1 become Type 2 lobules, at which point she has a mixture of approximately 75%Type 1 and 25%Type 2 lobules. Pregnancy after 32 weeks causes Type 4 to develop with the milk the baby will need. After weaning these Type 4 lobules regress to Type 3 lobules but with the genetic changes of Type 4 remaining in the new Type 3 lobules (regressed Type4 lobules). After menopause, the Type 3 morph yet again into what appear to be Type 1 lobules, however the genetic changes which have afforded cancer resistance remain. Therefore the woman who has a full term pregnancy with Type 4 lobules gets lifelong benefits from these genetic changes, even without breastfeeding.

After about eleven weeks gestation it is the fetus and placenta --- not the mother --- which produces most of the needed estrogen and progesterone to sustain the pregnancy. Fetal developmental abnormalities that prevent adequate production of those hormones cause miscarriage (spontaneous abortion) in the first trimester. The levels of the pregnancy hormones (estrogen, progesterone, and hCG) during an abnormal pregnancy that result in a spontaneous abortion (miscarriage) in the first trimester were insufficient to stimulate breast development and thus leave the mother’s breasts unchanged. Therefore, following a first-trimester spontaneous abortion, the mother typically has no change in breast-cancer risk as her breasts were never stimulated to grow. Often a mother who spontaneously aborts (miscarries) in the first trimester will often remark that she never “felt” pregnant before she miscarried; she had no morning sickness nor sore and tender breasts that she may have experienced in prior pregnancies. Thirty-one percent of all conceptions will end in a spontaneous abortion.[xviii] Thus such miscarriages are not cancer-risk-raising events, because there is something wrong with the embryo so that hCG is not stimulating the mother’s ovaries to produce more estrogen and progesterone, or something is wrong with the mother’s ovaries which can’t produce enough estrogen/progesterone, resulting in the spontaneous termination of the pregnancy (miscarriage). After 11 weeks the fetus produces enough estrogen/progesterone to maintain the pregnancy. Most miscarriages are in the 1st trimester.

2nd trimester miscarriages are usually due to physical problems, such as the cord is around the neck leading to fetal death or the placenta tears. These 2nd trimester miscarriages occur in pregnancies which have the normal amount of estrogen and progesterone and therefore the breast undergo those changes which increase the risk of breast cancer.

During a normal pregnancy, by the end of the 3rd trimester 85% of the mother’s breast is composed of cancer resistant Type 4 lobules which contain colostrum.

However, if the mother’s pregnancy does not continue past 32 weeks, either due to premature delivery or an induced abortion, she will not get the protective effect of a full term pregnancy because her breast tissue will not have developed Type 4 cancer resistant lobules. Assuming a premature delivery or an abortion, the longer a mother is pregnant up to 32 weeks the greater the numbers of Type 1 and 2 lobules she will have formed, leaving her breasts with more cells which are at risk of developing into cancer cells.[xix] The Melbye 1997 study showed that for every week of gestation before an induced abortion, the risk of breast cancer increased 3%.[xx] Type 1 lobules are where ductal cancers start (85% of breast cancers are ductal cancers). Type 2 lobules are where lobular cancers start (10-15% of all breast cancers). Thus there will be more sites for cancers to start, following a premature delivery or an induced abortion up to 32 weeks.

There is about a 3 percent increased risk in her chance of cancer for each week of gestation before the induced abortion. [xxi] With each passing week more Type 1 and 2 lobules are formed, and the breast doubles in volume by 20 weeks. The volume is composed of more immature, cancer-vulnerable breast lobules, leaving the mother with increased places for cancers to start and a higher breast cancer risk.

If the pregnancy is a normal healthy one that goes to past 32 weeks, even though she delivers prematurely, she will have partial protection for breast cancer. Over the weeks between 32 and 40 weeks she gains a proportional 11% reduction in risk [xxii] If she delivers at 40 weeks which is “full-term,” there will be near complete (about 85 percent) maturation of the mother’s mammary glands into Type 4 lobules. Type 4 lobules have progressed through a complete maturation process.[xxiii] Type 4 are completely mature, but not all the breast tissue matures: 15% remain as Types 1 and 2 and are thus cancer-susceptible. This is why there is a known significant protective effect against breast cancer when a woman has a full-term pregnancy.

If the mother then breast feeds, many studies have shown that breast feeding decreases breast cancer risk in proportion to the length of time she breast feeds. Women who breast feed so that all of the infant’s calories come solely from her breast milk will also lose their menstrual cycles for up to 2 years. The fewer menstrual cycles a woman has in her lifetime, the lower her risk of breast cancer. Many of the cycles a woman initially regains may be anovulatory (without an egg produced). These cycles are lower in estrogen and therefore do not increase the mother’s risk for breast cancer as much as normal ovulatory cycles. For each successive pregnancy, the mother further reduces her risk by 10 percent each time.[xxiv]

Now we will transition from this review of what we know at the micro, biological level to look at the patterns at the macro, demographic level.

III. Epidemiologic Studies support the Abortion Breast Cancer Link

An Overview

Over 50 years ago, the first epidemiologic study looking at breast cancer and abortion was done in Japan which had legalized abortion before few other countries had done so. Published in 1957, the study showed a three fold increase in the risk of breast cancer in women with a history of induced abortion.[xxv] This was followed by another Japanese study in 1968 which also showed an increase risk of breast cancer with induced abortion.[xxvi] Ten years later in 1978, a Russian study also showed an increase risk of breast cancer with abortion.[xxvii] Three years later in 1981 the first U.S. study was done and it also found an increased risk of breast cancer with abortion.

Public attention was not raised on the issue until 1994 when another U.S. study, commissioned by the National Cancer Institute (NCI), was reported in Time magazine and the popular press. That Time article interviewed the study’s author Janet Daling who disclosed, that although the overall increase in breast cancer risk in women with abortion history was 50%, a sub group of teenagers who had aborted experienced an 800% increase in breast cancer, a very alarming statistic, even if the number of teenagers in this sub group was low.

Two years later in 1996, a British medical journal published the first quantitative meta-analysis of all the studies which looked specifically at induced abortion.[xxviii] It excluded studies which did not differentiate between induced and spontaneous abortion. It showed that there were 23 studies,17 of which showed a positive association. There were 10 which were statistically significant, meaning that there was a 95% certainty that those studies did not show the association by chance. This meta-analysis created a fire storm in the English press when it was published. There was such an outcry that Stuart Donan, the editor of the journal in which it was published, felt a need to write an editorial in the following issue stating that although he was pro-choice, he was also “pro-information”.

Within 3 months of the Brind meta-analysis’ publication, in January 1997, the Melbye “Danish” study was published in the prestigious New England Journal of Medicine. It was hailed by a N.C.I. epidemiologist, Patricia Hartge as being “flawless” and proclaimed as a study that settled the question of whether induced abortion increased the risk of breast cancer. Its conclusion was that it did not, despite the fact that the study found a statistically significant 38% increase risk of breast cancer with second trimester abortions. And, despite Hartge’s proclamation, the Melbye study did have several significant flaws which will be discussed in a later section of this paper.

By the year 2000, there were 27 studies which showed an increase risk of breast cancer, 17 of which were statistically significant[xxix]. The U.S. Congress has political oversight of the N.C.I. In 2002, 28 Congressmen signed a letter asking the NCI to amend its web site concerning breast cancer and induced abortion as the preponderance of the data demonstrated a risk. The web site reported that the data was “inconclusive” and “inconsistent”.

This letter resulted in its web page being removed pending a February 2003 Workshop on “Early Reproductive Events and Breast Cancer Risk” conducted by the NCI). One hundred scientists participated in a three day workshop which concluded that induced abortion was not a risk and did not merit further study. However premature delivery was considered an “epidemiologic gap” requiring more study. [xxx]

In 2004, a large “reanalysis” of data from 53 epidemiological studies, including 83,000 women with breast cancer from 16 countries was published in the British journal Lancet.[xxxi] This study concluded that based on its prospective data, there was no association between induced abortion and breast cancer. It could not be termed a meta-analysis as it did not confine itself to peer-reviewed published studies but it used unpublished data sets as well. This study was widely reported in the public press. It also showed there was a small 10% increase in risk when retrospective data was used. The study described this increase as “misleading” in its conclusion citing recall bias as altering the data. Recall bias will be described in a following section of this paper. However, as discussed below, this study used 28 sets of data that had been collected but not previously published. Therefore, there was no peer review process looking at how the data was collected and analyzed --- which would have occurred if the data was published individually as a study in a peer reviewed journal. This study also did not include ten of the known seventeen, previously published, peer reviewed studies (which normally would happen in such a meta-analysis), thus further tilting the data and thus the outcome.

It is important that each data set be published in a peer reviewed study so that one could determine if the population studied reflected the population. For instance, most abortions in the US are done on women less than 25 years old who are childless. Being childless at the time of an abortion increases the risk of breast cancer. Many of these women are teenagers, another high risk group. If the study population of a study concerning abortion and breast cancer included mostly parous women (those with children) at the time of their abortion, this would influence the outcome to show a very weak association compared to childless teenagers. It is also important to know other subcategories of patients in order to evaluate the reliability of the outcome of the study. There is also no way to know how long the patients were followed; the minimum length of time would need to be 10 years for an accurate outcome to be found.

Given the confusion this has caused the best that can be done, unhappily, is to disregard this piece of research. It does not contribute to the steady march of scholarship, to the steady accumulation of clear data and research, with a gradually evolving clarity of patterns and thus data-based, well-reasoned conclusions that can be safely promulgated to the public (even with the ever-ongoing caveats that the public now appreciates) that further research may change the conclusions of the day.

Between the publication of the meta-analysis in 1996 and 2005 there were ten epidemiology studies published which were based on prospective data and distinguished between induced and spontaneous abortion, all of which found “no link” between procured abortion and breast cancer. However for reasons that will be outlined below, these studies have to be described as “incomplete” at best, and seriously deficient in their method.

These ten studies were analyzed in detail in an article by Brind [xxxii] who had written letters to the editor about these studies, most of which letters were published. Major flaws affecting the studies’ outcomes were addressed. One major problem common to many of these was that, for many of the women in the studies, their reproductive years covered a time when abortion was illegal, therefore unrecorded and underreported. A further confounding issue is the natural reluctance to admit to having had an illegal procedure.

Another criticism common to many of these studies is that the studies did not allow for adequate post-abortion time to pass, as it takes at least 8 to 10 years to develop a clinically detectable breast cancer. The 1997 Melbye study in which 25% of patients were less than 25 years old at the conclusion of the study and the 2007 Harvard Nurses study which followed most of its patients for less than 10 years are examples of this flaw.

Yet another weakness: Although many of the studies had large enough numbers of women which would normally result in statistical strength, the samples studied were not representative of the general population of the country. For instance the 2001 Goldacre study [xxxiii] reported that only 1% of patients had an induced abortion over a 30 year period despite the fact that in the UK, the abortion rate was 1% per year over that same 30 year period. The Goldacre study used only abortions that were done in the hospital while the vast majority of abortions in the UK are done in outpatient facilities. This means that over 90% of those unknowns within the sample were misclassified as having no abortion when they had had one. In a 2005 Scottish study by Brewster et al[xxxiv], the sample studied, which found no association of abortion with breast cancer, was also significantly skewed. Brewster’s study population did not accurately reflect the general Scottish population studied. Brewster’s study had only 5.% of patients who were childless at the time of abortion, while in Scotland 58% of abortions are done on childless women, mostly teenagers, a ten fold difference Childless teens are very high risk for breast cancer as was demonstrated by the 1994 Daling study. By having a study group that did not accurately reflect the Scottish population, the results were skewed. As we have seen in Section I above, induced abortion greatly increases a woman’s risk of breast cancers if she is childless more than a woman who has had several pregnancies and therefore has less cancer-vulnerable breast tissue. I refer the reader to the Brind analysis for a complete analysis of the flaws in these papers.[xxxv]

After 2005, there were only two papers published --- in 2007 and 2008 --- that concerned only abortion and breast cancer, the Harvard Nurses Study[xxxvi] and the California Teachers Study[xxxvii]. Both studies concluded that there was no increased risk of breast cancer with abortion. Disturbingly, both studies did not include the development of ductal in-situ breast cancers which account for over 60,000 cases of breast cancer a year. In situ breast cancer is treated with surgery, radiation and drugs. It has increased in incidence in premenopausal women 400% over the last 40 years in the US and 600% worldwide. Furthermore, most of these cancers develop into invasive breast cancers, which are the only cancers these two studies considered. If a cancer is serious enough that a woman might need to undergo mastectomy in order to be cured, as is so often the case with in situ cancers, it is strange to have kept the count of such cancers out of the study. By disregarding what the public and all women consider to be serious threats to their lives (in-situ breast cancers) the results are not reflective of a woman’s true cancer risk, nor of women’s desire to know the truth, nor of the true rate of breast cancer.

The California’s Teachers Study not only disregarded in situ cancers that develop into invasive breast cancers, but they removed women from the study if they did develop in situ cancer. This means that if these women with in situ cancers went on to develop invasive cancers during the study’s duration they would not be included as they had already been removed.

The California Teachers Study collected data in 1995 and ended in 2004, a 9 year period. Only the total number of women with induced abortion in the study is reported and not when they had an abortion. If most of the women in the study had their abortion close to 1995, there is no way to know if an adequate follow-up time occurred, a minimum of 10 years. The authors chose to use nulliparous or childless women as a control group. As this group of women already has an increased risk of breast cancer, any increase risk in the abortion group would appear less than if the comparable group had had a pregnancy. After all, the purpose of the study is to know if pregnant women who have an abortion are at increased risk of breast cancer compared to pregnant women who do not have an abortion and give birth. By choosing an inappropriate comparable group the authors influenced the outcome to show less of an effect. It is not surprising that these biases were found in this study as one of the authors, Leslie Bernstien, was reported as saying in an interview in 2003, “I don’t want the issue relating induced abortion to breast cancer risk to be part of the mix of the discussion of induced abortion, its legality, its continued availability.”[xxxviii]

The Harvard Nurses study also had inadequate follow up times in which to observe patients after an abortion. The study time was from 1993 to 2003. Patients were sent questionnaires every two years. Only the patients who had an abortion to report in 1993 questionnaire were followed for 10 years, . All the abortions occurring after 1993 were inadequately followed i.e. the majority of the patients who had an abortion during the study period. The minimum time needed for a breast cancer to grow enough to become clinically detectable is 8-10 years. As the questionnaires were sent to patients every 2 years, even abortions done in 2001 were counted. That is similar to expecting a lung cancer to develop 2 years after smoking cigarettes. It is not credible to design a study that would include such a short time for follow up. Mesotheliomas (lung cancers) take 20 years to occur after exposure to asbestos. Studies for cancer must allow adequate time for follow-up to be valid. A detailed analysis demonstrating other flaws of this study was published in 2007[xxxix]. Following a large number of patients for less than 10 years would greatly underestimate resulting cancers

Another aspect grossly underestimating the resulting cancers from abortion was excluding, ductal carcinoma in situ (DCIS) which are early non-invasive cancers. It may take ten or more years for DCIS to become invasive. There were 339 cases of non-invasive DCIS cancers in this study while the number of cases of invasive cancers was 233. In other words, most of the women who developed cancer during the study period were not counted! Both of these types of cancer can result in partial or complete mastectomies. No reason was given for excluding the non-invasive cancers. A published analysis of the flaws of the Harvard Nurses Study was done by Brind and can be referred to for a more complete analysis.[xl]

Bias was shown in the California Teachers study when in the introduction it is stated that no studies that collected prospective data showed a link between abortion and breast cancer. The Howe study[xli] which showed a 90% increase risk of breast cancer with abortion and a record linkage study not subject to recall bias was disregarded.

Since 2007, there have been many studies concerning breast cancer risks which have shown induced abortion as a risk for breast cancer. These studies are from all over the world. They are interesting because these papers are studying all risks of breast cancer and acknowledge that induced abortion is a risk for breast cancer. In fact, in the discussion section, they will acknowledge that their results on induced abortion are consistent with the rest of the world’s literature, and by routinely controlling for abortion are implicitly acknowledging what scientists across the rest of the world know. A study from Iran, published in 2007, showed a statistically significant 62% increase risk of breast cancer with abortion.[xlii] . In 2008, a study from China showed a 64 % statistically significant elevated risk of breast cancer. In 2009, three additional studies from the US (2009 Dolle study), Turkey and China,[xliii] found that abortion statistically significant increased breast cancer risk by 40%, 31 % and 26% respectively. All three risk rates were close in number to the 1996 Brind quantitative meta-analysis which showed a 30% overall increase in risk of breast cancer with induced abortion. In 2011, an Albanian study showed a 186% statistically significant increased risk of breast cancer with induced abortion.

Ecologic Epidemiological Studies Confirm the Abortion Breast Cancer Link:

What is most notable about the epidemiologic studies concerning the abortion breast cancer link is that all the ecological studies show a strong association. Ecological epidemiological studies examine trends in large populations based upon comparisons using statistical records kept by governmental agencies.

In 2007, Patrick Carroll, an actuary, published his paper, “The Breast Cancer Epidemic: Modeling and Forecast Based on Abortion and other Risk Factors” in the Journal of American Physicians and Surgeons[xliv]. Carroll found that abortion was the greatest predictor of breast cancer incidence rates in nine European countries: England, Wales, Scotland, Northern Ireland, the Irish Republic, Sweden, the Czech Republic, Finland and Denmark. Using computerized abortion and breast cancer registries, he found that the greatest predictor of future breast cancer incidence was a nation’s abortion rate. Furthermore, within the United Kingdom, its constituent nations which have the highest abortion rates also have the highest breast cancer rates. England, with the highest abortion rates, has an incidence of breast cancer of 116 per 100,000, while Northern Ireland where abortion is much less prevalent, has an incidence of 97 per 100,000. Scotland lies between England and Northern Ireland in both breast cancer and abortion rates. There has been a 70 per cent increase in the risk of breast cancer in the United Kingdom between 1971 and 2002.

A 1989 study of breast and cervical cancers showed that three countries in the USSR (Russia, Estonia and Soviet Georgia) increased their breast cancer rates from 270 to 330 per cent as the abortion rates increased.[xlv] The author stated that in the USSR, the majority of women have used abortion as the principal method of birth control and that ever since its legalization in 1955 the number of abortions has been exceeding the number of live births. In the discussion she also declared that, “all four abortion indicators used in the study … appeared to be the most significant correlates of the incidence of both cancers.” The four independent variables (‘indicators’) she referred to were, frequency of induced abortion, abortions to live birth ratio, frequency of out-of-hospital abortions and frequency of the termination of the first pregnancy. We already know from Carroll that’s true for European countries; and we cited the Remmenick paper 1989 [xlvi]

By contrast, Romania had one of Europe’s lowest breast cancer rates while abortion was illegal under Ceausescu. Since its legalization in 1989 breast cancer rates appear to have risen dramatically based on an analysis of the experience of this one county. Breast cancer incidence increased there from 25 per 100,000 women in 1988 to 40 in 1996 and 51 in 2006.[xlvii] The total number of abortions reported in Romania jumped from 193,084 in 1989 to 992, 265 in 1990.[xlviii] In 1960 breast cancer comprised 7 per cent of all malignant tumors; by 1996 it had risen to 23 per cent, ranking first in cancers among women. Given the explosion in the abortion rate, we may expect it to rise even more in coming years.

In China, the enforcement of a one-child policy, which includes using involuntary abortion, correlated with a subsequent substantial increase in breast cancer rates. Since 1983, there has been a 30 per cent increase in the incidence of breast cancer in China.[xlix] Over the last ten years the incidence of breast cancer increased 31 per cent to 55 per 100,000 women per year in Shanghai and increased 23 per cent to 45 per 100,000 women per year in Beijing. [l] Even more alarming, a study published in 2008 reported that China was on the “cusp of a breast cancer epidemic” forecasting an incidence of 100 new cases of breast cancer per 100,000 women aged between 55 and 69 by 2021.[li]

In the United States, following the legalization of abortion in 1973, there has been a marked increase in the risk of non-invasive and invasive breast cancers from 1975 to 2000.[lii] Invasive cancer incidence went from 105 to 136 per 100,000 women. Non-invasive cancer in pre menopausal women went from 3 to 12 per 100,000 women, a nearly 400 per cent increase in incidence. On a smaller scale Washington State breast cancer rates in black women rose after abortions became state funded (i.e. after increasing abortion availability to poor black women).[liii]

Discussion of Noteworthy studies denying an abortion breast cancer link

Many times, deniers of the abortion breast cancer link will cite a well publicized study that has been reported upon in the popular press such as the Melbye Danish study and the Breal reanalysis. The popular press is then reported upon by the media and the public remains misinformed. However, careful scientific scrutiny shows these studies to be egregiously flawed.

1997 Melbye “Danish” Study

When it was published, the Melbye study was hailed as the definitive answer to the question, does abortion increase breast cancer risk?[liv]. However, it misclassified 60,000 women who had legal abortions as not having had abortions because the authors used abortion registries starting in 1973 instead of 1940. It also violated the Bradford Hill criterion of temporality by collecting breast cancer cases for study starting in 1968 while collecting data on abortions using records that started in 1973. Another factor that contributed to the study’s methodological flaws relates to the biology of breast cancer. It takes an average of eight years for a cancer cell to grow into a clinically detectable cancer of 1 cm. diameter. If an abortion in an 18-year-old causes a breast cancer cell to form, it is not likely to be detectable until she is at least 26 years old. Fully one-quarter of the patients in the Melbye study were under 25 years old when the study ended, accounting for only eight cases of breast cancer. Because of what is known about the development of breast cancer, these young women should not have been included in the study at all.[lv]

Yet even with this and other major flaws, the study showed a statistically significant increase of 89 per cent in breast cancer risk in abortions performed over 18 weeks’ gestation . This fact was not mentioned in the conclusion of the paper which instead stated that there was no link at all between abortion and breast cancer. This paper is still cited as large and conclusive, as if a single study could be conclusive. It is often used in major textbooks to show there is no link between abortion and breast cancer. For example, one standard breast text has a table of risk factors in a chapter which states that induced abortion has no effect on breast cancer risk, belying the fact that in text in the same chapter acknowledges a 38% increase in risk for abortions done after 12 weeks. [lvi]. If only the students would read the original research (or the editors of the texts would fact check) there would be a much more accurate impression of the validity of the abortion breast cancer link both in academic and popular publications. This is not an uncommon situation when abstracts which are bolded on the front page of the paper belie what is in tables and text which are readily available in the paper itself. For example, in the 1997 Melbye study, the abstract on the front page concluded there was no breast cancer risk with induced abortion while the paper showed a statistically significant increase in the risk of breast cancer for second trimester abortions.[lvii]

2005 Beral “reanalysis” of 53 studies

In March 2004, The Lancet published a study [lviii]hailed by its researchers, Valerie Beral and colleagues, as the definitive analysis that puts to rest the claim that abortion increases breast cancer risk. Beral was quoted as saying, “Scientifically, this is really a full analysis of the current data.”[lix]

However, a review of the study reveals that it is not a “full analysis”. Indeed, serious methodological flaws—especially in the selection of studies to be included—render the Beral study unreliable. The authors were guilty of several errors in selecting the studies to include and to exclude in their analysis.

First, they eliminated 11 studies for reasons having nothing to do with science (e.g., “principal investigators could not be found,” or “researchers declined to take part in collaboration”), and four other studies’ worth of data were simply not mentioned at all.[lx] Thus, in the end they included only 24 of the 41 studies that, from 1957 to 2005, contain data on induced abortion and breast cancer. To supplement these 24 studies, the researchers added a further 28 unpublished studies. This means that the majority of the studies included in their analysis could not be consulted by other researchers as they had never been published, nor had they stood the test of peer review, though the studies they eliminated had.

A further close look reveals that many of the statistically significant studies that demonstrate a link between abortion and breast cancer are among those excluded. To be precise, of the 41 studies published up to 2004, 29 actually show increased risk of breast cancer among women who have chosen abortion (epidemiologists call this a “positive association”.) Seventeen of these 29 studies are statistically significant. Yet ten of the 17 significantly positive studies in the literature were excluded by Beral and her colleagues. In fact, if the results of the 15 studies excluded for supposedly unscientific reasons are averaged, they show an increase in breast cancer risk of 80 per cent among women who had abortions.[lxi]

Beral and colleagues also divided the studies into two separate categories: those that used retrospective methods of data collection (i.e., interviews of breast cancer patients rather than control subjects), and those that used prospective methods (i.e., medical records taken long before the breast cancer diagnosis). The 39 studies that used retrospective methods showed a significant overall 11 per cent increase in risk with abortion. The 13 studies that used prospective methodology showed a significant 7 per cent decrease in risk with abortion.

Instead of reporting the results of her findings accurately (that is, that the retrospective studies showed that abortion increases the risk of breast cancer and prospective showed that it reduces the risk of cancer), they simply declared that the retrospective studies were unreliable—because of ”recall bias”. Despite the theoretical possibility that recall bias exists, tests for such bias have proven negative,[lxii] as we will demonstrate below.

In regards to the 13 studies that Beral used showing a decrease incidence in breast cancer with abortion, or a protective effect, the first one listed in Figure 2 of the paper was the 1997 Melbye study that was discussed above in detail as to the reasons it was not a credible study having many flaws. She also listed a study from Scotland showing a 20% decrease in risk, but the data was unpublished and therefore can not be assessed independently. It is possible that this data was published the following year as the 2005 Brewster study which was criticized in the previous section. The 2001 Goldacre study, the 2000 Tang study and the 2003 Erlandsson study which were also used by Beral were negatively criticized by Brind in 2005 [lxiii] and in this paper for very credible reasons. Another study listed was the Ye 2002 Shanghai study which did show a positive correlation: 116% increase if the abortion was done before a full term pregnancy and a 95% increase if the abortion was greater than or by 13 weeks. It is possible that Beral’s finding that induced abortion was protective of developing breast cancer was not promoted as a risk reduction strategy is because of the obvious weakness of that finding when the data is analyzed.

Another major flaw of the Beral study was that an inappropriate comparison group was chosen. The authors compared apples and oranges when the effects of having had a pregnancy that ended in abortion were compared with the effect of “not having had that pregnancy.” Once a woman has had a healthy pregnancy, however long, her breasts are different than before that pregnancy started. Pregnancy forever alters the breast and physiologically these women are as different as pre- and post-menopausal women. Just as the effect of hormone replacement for post-menopausal women is studied in relation to other post menopausal women who have no exposure to hormones, pregnant women who undergo abortion need to be compared to pregnant women who do not undergo induced abortion. The comparison group used was useless.

There have been several published criticisms of the Beral study[lxiv]. In his letter, Carroll, who has done ecological studies of abortion and breast cancer, said the Lancet study's credibility is "impaired by the failure throughout the world to fully record induced abortions."[lxv]  Referring to a prospective study on British women by Goldacre et al.,[lxvi] Carroll asserted that the authors "found only a small fraction of the women that could have been expected to have had abortions from what is known from national incidence data." Carrroll added that if a British study can have so many unrecorded abortions, then under-recording must be even more prevalent in countries that legalized abortion after Great Britain did in 1967.  Carroll said many of the abortions relevant to the Goldacre paper, which were performed in other countries, were illegal. Women in such countries can naturally be expected to be more reluctant to admit to having had abortions than British women will be.

Recall Bias

“Recall bias” is the most widely and oft-reported argument used against the abortion-breast cancer link. This is the hypothesis that women who have developed breast cancer will be more likely to admit that they have had abortions than women who are well. The theory is based on the assumption that healthy women are more likely to conceal what could be embarrassing behavior but are more likely to tell the truth should they become ill, seeking a reason for their illness. Recall bias thus supposes that many women who do not have cancer will not report their abortions while those who do have cancer will report them. Case control studies in which researchers rely on interviews for their data are thought to be those potentially susceptible to recall bias. This is because researchers assume interviewees will not admit in an interview to “socially unacceptable behavior,” such as abortion -- unless they are sick. However, recall bias has not posed any such problem in other areas of medical research where case control studies have been used to gather data of other socially unacceptable behavior. For instance, in case control studies testing for a link between alcohol consumption and liver damage, interviewees were assumed to report their alcohol consumption accurately. The same is true for interviews in which interviewees were asked how many sexual partners they had, when inquiring into connections with cervical cancer, and whether they were involved in anal intercourse, when inquiring into HIV. Abortion would not seem to be a more socially unacceptable act than any of these. Why then should recall bias be thought to taint research about abortion but not the others?

Recall bias is a hypothesis worth testing. Despite the theoretical possibility that recall bias exists, tests for such bias have proven negative,[lxvii] as we will demonstrate below.

Often the “Swedish” study is cited when using the argument of recall bias; yet it was convincingly refuted in a subsequent letter published by the journal in question.[lxviii] Yet, studies that have confirmed the ABC link internally controlled for recall bias in their study populations.[lxix] Moreover, a study conducted specifically to test for recall bias in abortion-breast-cancer research reported having found evidence of it; however, methodological problems with the study acknowledged after publication revealed that it actually failed to show that recall bias taints such research. Instead, the study’s results supported what is true to clinical experience: almost equal numbers of women with cancer and without cancer concealed their abortions.[lxx] Researchers in the Lindefors-Harris study had before them both cancer and abortion computer registries in order to verify the responses of the women who were interviewed. Two groups of women were interviewed: those with cancer and those without. The researchers hypothesized that more of those without cancer would deny their abortions while more of those with cancer would admit to them. Such a result would be evidence of recall bias. Instead, they found no statistically significant difference between the responses of the two groups of women. Women with cancer and women without cancer both underreported their abortion in similar percentages (5 out of 24 women or 21 per cent, and 16 out of 59 or 27 per cent, respectively). In short, most healthy women and sick women admitted to the abortions officially documented in the abortion registry while some healthy women and some sick women lied. There was a statistically insignificant difference of barely six per cent between the two groups, which even in large studies would not greatly affect the results. On the other hand, researchers did find women—both healthy and sick—who admitted to abortions that were not documented in the abortion computer registry. The researchers labeled this phenomenon “over reporting,” claiming that women who told the researchers that they had had abortions that had not been reported in the computer registry were mistaken or lying. The researchers would have been better advised to assume a mistake in the registry or that the women had their abortions in another country. Only with this wrongheaded assumption of over reporting did the authors then conclude that they had significant evidence of recall bias. Over reporting, of course, does not exist. The researchers were forced to acknowledge their error in a subsequent exchange of letters to the editor. Unfortunately, since most doctors read only the abstract of a paper and do not follow letters to the editor, those reading the study today still are presented with the false results.

The Bradford Hill Criteria for determining causality in epidemiological studies.

Before any causal statements may be made that a risk factor is a cause of a disease, and not just merely associated with it, strict criteria must be largely met. Just because a study shows a positive association of a factor with a disease, does not necessarily mean that factor is the cause. The same holds for the abortion-breast cancer link.

To use a simple for example, large, statistically significant and reproducible studies might show that people who carry matches in their pockets have a higher risk of lung cancer. Without the additional criteria of a plausible biological theory of how the matches cause lung cancer, these studies, no matter how many are done, show only a positive association between matches and lung cancer. Knowing that matches were associated with lung cancer might lead scientists to investigate how the matches were used to light cigarettes and then discover the true cause of lung cancer. It was the Bradford Hill criteria, published in 1964, that brought the United States Surgeon General to place warnings on cigarette packages to the effect that they increased the risk of lung cancer.

The following nine criteria were established by Sir Austin Bradford Hill in 1964 and were used to show the causal link between cigarettes and lung cancer. Epidemiological studies done concerning the abortion-breast cancer link do show that they meet the criteria for classifying abortion as a causal risk for breast cancer.

Criterion 1: Strength of Association: Studies must show a large relative risk, greater than 3.0 An RR of 3.0 means there is a 200% increase in risk. (1,0 is null. 0.5 is a 50% decrease in risk)

If there is only a ten per cent increase in risk, it is difficult to say the risk is causal. There are subsets of women that show a greater than 200 per cent increased risk in breast cancer with abortion, for instance in the 1994 Daling study, which showed risks in some sub-groups of women to be much more than 200 per cent.[lxxi] For example, abortions done on women less than 18 years old between 9 and 24 weeks gestation had a relative risk of 9.0 or an 800% increase in risk. If a woman had abortion at age 30 or later between 9 and 24 weeks gestation her risk of breast cancer increased 230%.

Statistically significant studies require 95% probability that the study results were not obtained by chance alone and are thus rated as statistically significant at the 95% level. There are now twenty-five statistically significant studies which show the abortion-breast-cancer (ABC) link.

Criterion 2: Temporality: the hypothesized cause precedes the outcome. The exposure to the risk must occur before the disease is detected, meaning that the abortion must occur before the breast cancers form.

This may seem so obvious that it need not be mentioned. However, a well-known study, that by Melbye and colleagues (1997) violated this rule when it collected breast cancer cases from a registry starting in 1968, but abortions only from 1973.[lxxii] The cancer cases between 1968 and 1972 had no place in the study. By including them the authors reduced the link between breast cancer and abortion.

Criterion 3: Consistency: The preponderance of studies must show a positive association.

One or two studies are never enough to support a causal link. However, out of 67 world-wide studies done to date, 53 show a positive association, and 31 are statistically significant while a total of 17 studies show no link. There are many reasons studies may not show a link. If the population studied had abortions followed shortly by a full term pregnancy in many instances, a very small effect would be expected and might not be observed. If the abortions were used to limit enlarging a family with many children, a small effect is expected as the mothers’ breast tissue. This circumstance occurs in countries such as Italy where the effect of induced abortion on breast cancer risk has been found to be less than other European countries. There may be bias in selection of patients so that very few young women are included who would have the highest risk with abortion. The study may not follow most of the patients long enough for cancers to form which needs to be at least 8 – 10 years for breast cancers to be detected based upon the average doubling time for breast cancer cells. The study may choose to not look at all breast cancers and consider only some, such as invasive breast cancers, thereby eliminating 30-50% of cancers. For instance, in the U.S. there were approximately 230,000 cases of invasive breast cancers and almost 57,000 non-invasive breast cancers in 2011.

Criterion 4: Theoretical Plausibility: The biological mechanism that explains the reason for the risk association must be biologically plausible.

The breast physiology which explains the risk of breast cancer with induced abortion has been thoroughly explained in a previous section, and is supported by standard medical texts. Elevated levels of estrogen during pregnancy leave the breast with increased numbers of Type 1 and 2 lobules where breast cancers arise. If the pregnancy does not proceed past 32 weeks, the mother’s breast is left with more places for cancers to start.. If the pregnancy continues past 32 weeks, her breast tissue matures into cancer resistant Type 4 lobules with contain the early milk, colostrums. This same physiology can account for other well accepted reproductive risks such as nulliparity (no births) and premature delivery before 32 weeks. Furthermore It has been shown that the longer a woman is pregnant before an abortion, the higher her risk of breast cancer.[lxxiii]

Criterion 5: Coherence: The hypothesis, when proven, does not do violence to related sets of scientific findings but, instead, fits in with them.

The biological hypothesis of the abortion-breast-cancer link( abortion causes an increase in the number of cancer susceptible lobules and cells and therefore risk) is consistent with all other reproductive risk factors concerning pregnancy, such as: the protective effect of a full-term pregnancy while nulliparity (no births) increases risk, early age at full-term pregnancy decreasing risk, late age at full-term pregnancy increases risk, the transient increase in breast cancer risk in older women who have their first pregnancy late in their reproductive life, lower risk after a full-term pregnancy because of an increase in sHBG (serum hormone binding globulin) causes decreased bioavailability of estrogen, lower prolactin levels in parous women (who have given birth) decreases risk, lower risk with each full-term pregnancy and higher risk with premature delivery before 32 weeks.

Lung cancer does not form the day following smoking a pack of cigarettes; it takes years of smoking..

The association of breast cancer and abortion is in accord with the known natural history and biology of breast cancer. It takes an average of eight to ten years for one breast cancer cell to keep doubling so that it forms a tumor of clinically detectable size, about one centimeter. The time periods in studies showing the increase in breast cancers occur within the time frame appropriate for the development of breast cancer, namely at least eight to ten years after exposure.

Criterion 6: Specificity of cause: Factor x leads to outcome Y

There are ecological epidemiological studies that show induced abortion is the best predictor of breast cancer rates in a country. Induced abortion was found to be the greatest predictor of breast cancer rates in nine European countries.[lxxiv] Another study done in a tenth country, the USSR, also showed a similar outcome.[lxxv]

Criterion 7: A Dose effect is observed.

If a factor is causal then, based on biological mechanisms, the more one is exposed to the risk, the higher will be the risk of the disease. With cigarettes we now know that the more cigarettes one smokes, the higher the risk of lung cancer.

The longer one is pregnant before an abortion the more immature breast tissue would be formed up to 32 weeks, and the higher will be the risk of breast cancer. Despite its flaws, the 1997 Melbye study showed a three per cent increase in the risk of breast cancer with each week of gestation.[lxxvi] The Melbye study also found a statistically significant increase of 89% in risk of 2nd trimester abortions. The Melbye study did not study late term abortions in detail except to examine a subset that occurred at greater than 18 weeks.

There is also evidence that increasing numbers of abortions also increase breast cancer risk.. called a dose-effect outcome.[lxxvii] Within this dose-effect phenomenon is the fact that just as one exposure to asbestos can cause a mesothelioma to form, one abortion may induce the formation of a breast cancer.

Criterion 8: Experimental studies: Variables are experimentally controlled-for and yield predicted results consistently, first in animal studies, later in human (if ethically permissible, which, of course, is not the case with abortion).

Two pathologists studied the effect of a breast carcinogen (DMBA) given to groups of rats. The rats were either virgin, had a litter of pups, or had been aborted. The aborting rats developed breast cancers at a much higher rate when given DMBA than virgins, or those which had pups. [lxxviii] This is all quite in line with the micro biology of human breast cancer findings covered above.

Criterion 9: By Analogy: Similar exposures should result in similar effects. For example heavy exposure to cigarette smoking causes bladder cancer as well as lung cancer.

Premature deliveries before 32 weeks also double breast cancer risk because the breasts are left with more lobules where breast cancers can start. An abortion can be thought of as premature delivery by an abortionist.[lxxix]

It is to be noted that the National Cancer Institute, surprisingly, in coming to its conclusion that there is no link between abortion and breast cancer, does not subject its conclusion to these standard criteria

IV. Role of the National Cancer Institute (NCI): Information given to professionals and to the public

At the present time egregious errors occur on the NCI’s website when it explains the changes which occur in a women’s breast during pregnancy and her levels of estrogen with that pregnancy. In some cases they explain that the levels of estrogen are lower during pregnancy, such as early pregnancies, and therefore early pregnancy lowers breast cancer risk. The NCI web site also states that when a woman becomes pregnant late in life she is exposed to more estrogen and therefore has a higher risk of breast cancer. However, in a normal pregnancy estrogen levels are always elevated.[lxxx]

It is not credible for the NCI to maintain that there is no increase in breast cancer risk with induced abortion based upon the workshop they conducted in 2003.

• It is wrong because that assertion flies in the face of normal breast physiology and known reproductive risks for breast cancer acknowledged in standard texts;

• It is wrong because it flies in the face of world wide epidemiologic studies of which 53 show a positive association, 31 of which are statistically significant, while only 14 show a negative association.

• It is wrong that the NCI obfuscates the afore-detailed breast physiology and pregnancy changes on their web site which misleads both the public and professionals by not illustrating the changes in lobule types before and after pregnancy, critical issues in the development of breast cancer.

• It is wrong that a leader of the 2003 NCI Workshop on Early Reproductive Events and Breast Cancer Risk (which denied that induced abortion was a risk) would in 2009 be a co-author of a study[lxxxi] which acknowledges that induce abortion is a known risk factor for breast cancer and yet never correct the record on this, even when asked directly by an international journalist.[lxxxii]

The lack of scientific integrity at high scientific levels is a growing problem and, on the face of it, seems to affect the abortion-breast cancer field of inquiry.

A study done in 2005 on US scientists and published in the British journal Nature found that there was a statistically significant 15.5% incidence of researchers admitting to “Changing the design, methodology, or results of a study in response to pressure from a funding source”.[lxxxiii] The scientists studied were those that had received funding from the United States government, including the National Institutes of Health (NHI).

This lack of scientific rigor and perhaps integrity even in the academic arena has been shown above in the epidemiology section looking at the 1996 Melbye study and the 2004 Beral reanalysis. We are asked to believe that the Melbye authors did not know that induced abortion was legalized in Denmark in 1940 and not 1973 resulting in the misclassification of 60,000 women. We are asked to believe that the authors of the Beral reanalysis did such a poor review of the literature and showed no selection bias while 17 studies that showed a link between abortion and breast cancer were somehow missed or required exclusion even though they had been previously published in major peer reviewed journals.

These are not the only serious distortions and suppressions of data and research. For example, in 1988 Australian epidemiologist Rohan[lxxxiv] published a study on risks for breast cancer but did not include the data he collected on reproductive risks. It was not until 1996 when French researcher Andrieu[lxxxv] used Rohan’s data regarding induced abortion that it was shown that Rohan had found but suppressed a statistically significant 170% increased breast cancer risk following induced abortion. In fact, it turns out he had suppressed the strongest risk factor for breast cancer that he found in his research. Despite the fact that induced abortion had the greatest relative risk of all the risk factors of Rohan’s 1988 collected data , he chose not to publish it, and instead published his findings on other risks with less impact on breast cancer risk.

The reputation of NCI is at stake. Scientific integrity is at stake. Women’s lives are at stake.

V. Future Investigations

Given the afore mentioned research, it would be important to inform the public of the known and undisputed ways induced abortion increases breast cancer risk, reevaluate by meta-analysis the current studies and plan future studies regarding the link between abortion and breast cancer using a standardized data bank. The National Cancer Institute needs to fulfill its mission to educate and protect the public about known risks by changing its inaccurate web site on the physiologic effects of pregnancy on breast cancer risk. There needs to be restoration of scientific integrity that has been disturbingly lacking and allows researchers to have their uncensored data reported.

In light of the foregoing information, it is only reasonable and in the public’s best interests to be informed of the known and undisputed mechanisms that cause induced abortion to be a risk factor for breast cancer, ie, the loss of the protective effect of a full-term pregnancy, the increase risk of delaying a full term pregnancy or losing the benefit of another full term pregnancy and the increase risk of remaining childless. Women also need to be informed of the future risk of premature birth after an induced abortion. Breast cancer rates were significantly reduced after 2002 when women became aware that hormone replacement therapy (HRT) increased breast cancer risk 26% according to the Women’s Health Initiative study. Approximately 15 million women stopped their HRT and breast cancer rates fell by 11% in 2007. There are approximately 1.3 million abortions a year and one could expect women who are at high risk for breast cancer would consider the effect of induced abortion on their breast cancer risk when deciding whether or not to undergo a elective procedure. The National Cancer Institute has as part of its mission to educate the public of cancer risks

The last quantitative meta-analysis of studies that differentiated induced from spontaneous abortion and its impact on breast cancer risk was published in 1996 and found a 30% increase risk of breast cancer with induced abortion. Since that time there has been an additional 28 studies published in the world’s literature. A meta-analysis of these extant studies would help to clarify the independent link between abortion and breast cancer.

In order to clarify the impact of induced abortion and other reproductive risks on the incidence of breast cancer, it would be possible to do prospective studies that would give credible and significant results in a relatively short time period through use of an established group of data banks link to accredited breast centers. Throughout the United States there are nationally accredited breast centers which do mammography screening and non-invasive breast biopsies. The Federal Drug Administration already mandates these centers to carefully follow the results of any breast biopsies done. All breast centers take medical histories from their patients to aid in the mammographer’s interpretation of their mammograms.

A prospective data base could be obtained by having women fill out a complete reproductive, hormonal and breast history form that also included other known risk factors for breast cancer at the time of their annual mammogram. This form would allow researchers to determine the age at which a woman was exposed to a risk in relation to other known risks and to determine if the exposure occurred before or after a full term pregnancy, which is known to be protective. For example, there is a near ten-fold increase in risk for breast cancer if cigarettes are smoked before rather than after a full term birth[lxxxvi].

One such a form used at the Steeplechase Cancer Center in New Jersey can be viewed in the appendix. Each year the form would be updated by the patient at the time of their yearly mammogram screening. Average size breast centers will screen eight to ten thousand women a year. About 3% of theses women will have breast biopsies based on these mammograms and about 1% will be cancers. By collecting the results of theses biopsies and correlating them with their breast histories obtained before screening and biopsy, a robust prospective data base could be obtained in a relatively few years. By asking the age and outcome of each pregnancy it could be readily determined which subsets of women are most impacted by an outcome such as induced abortion. The other known risks factors which were also collected would serve to allow for logistical regression analysis so that the effect of the induced abortion can be separated from other known risk factors.

The constant updating and yearly collection of the same data with each yearly mammogram would also serve as internal controls for reliability of the data as well as being data for a longitudinal study. Women who did not undergo biopsy would serve as control groups. The limitation of this study would be that most of those screened would be at least 40 years old. However, since only 3% of invasive cancers occur in women less than 30 this would not be a serious deficiency.

Since mammography screening breast centers occur throughout the United States and the world, different populations could be easily accessed: urban vs suburban, poor vs wealthy, and all racial groups. As the vast majority of university medical centers have an associated breast center doing mammography screenings, there could easily be associated clinicians and academics to analyze and insure the data was collected properly. By having a standardized data collection history form, a very large data base collected in many locations could be obtained. This data base can be used for specific prospective epidemiologic studies and would be very valuable for longitudinal studies. It would be also useful for doing studies to confirm the results of previous studies. As these breast centers are already in existence and controlled for quality by a national accreditation organization (NSABP), the structure is already in place to start a unified data base collection.

Also, the data base could be used for any other risk factors and the data collection expanded to include other risks as they became known or investigating suspected factors by including them on the history form, for example, environmental exposures to chemicals.

All epidemiologists and clinicians should have access to this data base so that no organization or institute would have control over what studies may be done thus eliminating personal, institutional or funding-source bias. Any bias in collection or analysis of data would be more readily apparent if trends were found in individual breast centers but not in others, assuming that these differences could not be attributed to differences in populations or other local environmental influences.

Despite advances in understanding the genesis of breast cancer and the incidence and the links with induced abortion and spontaneous abortion (miscarriage) there is still many research questions to be answered.

The above data base will allow researchers to look at specific subgroups of women with different risk factors and reproductive outcomes. For instance, previous data[lxxxvii] indicated that the sooner a woman had a pregnancy and lactated after an abortion, the lower her risk for breast cancer as a result of that abortion. This would be important for women to know as there is often a “replacement” pregnancy which follows soon after an induced abortion.

This would also be important information for those most at risk for developing breast cancer after an abortion, such as teenagers and women over 30, especially if they have a family history of abortion.[lxxxviii] Such a study is especially relevant as about 18 % of induced abortions are procured by teenagers.

By using these data bases, there could also be more studies done on the influence of other risk factors on women with induced abortion. For instance, the effect of oral contraceptives on breast cancer risk if taken before or after an abortion could be studied. As induced abortion will leave a childless woman with more undifferentiated breast tissue than a woman who has never been pregnant, one might expect that those women taking oral contraceptives after an induced abortion to be at higher risk for breast cancer than a woman who did not take them after an abortion.

It would also be possible to test more easily for the Bradford Hill “dose effect” criteria regarding the number of abortions being proportional to the length of gestation before induced abortion or the number of abortions a woman has.

Another important area of study would be the incidence of breast cancer for women whose first pregnancy ended in an abortion and their breast cancer risk if that is followed by either more abortions or no pregnancy for the next eight years (not unlikely scenarios for many teenage girls who get an abortion at 16 but do not bring a baby to term until they are 25).

Other hypotheses could be investigated to elucidate those circumstances that most impact breast cancer risk for a woman who chooses abortion. The creation of these data bases for study would eliminate another round of macro “yes there is a link” vs. “no, there is no link” studies by parsing out the circumstances that are more cancer-risk increasing, such as whether the abortion occurred before or after a full term pregnancy or at a very early or much later age.

VI. Conclusions:

Scientific investigation in any realm must be free of political or ideological agendas no matter how worthy those agendas may be judged to be by the scientists conducting those investigations.

Science builds especially by being open to data that is contrary to treasured hypotheses. This is the very means of the constant revolution in science and is one of the main pathways to new insights in all realms of science. Science cannot move forward without that willingness to abandon long held theories once the data make themselves repeatedly disturbing of the status quo. This is the duty of every scientist and the hallmark of greatness in scientists especially when there is a personal cost, whether in funding, peer esteem or even self esteem. New paradigms discovered by scientists throughout history are difficult to accept even when they are not as politically and ideologically charged as is the subject of this paper. But the health of women is at stake and the data covered above indicates that abortion is very likely a culprit in the leading cause of cancer death among young women in their prime, breast cancer.

Yet it is clear that the above well-documented scientific data which supports a causal relationship between abortion and breast cancer has been both ignored and suppressed by many of the leading scientists in the field, and especially by those connected to the National Cancer Institute. Rather than practicing good science these academics may well be open to a charge of willing distortion of science at grave cost to many women.

Furthermore, if the data and interpretations above are correct, the data will eventually out. And the opposite is also true: if the above interpretations are incorrect there is nothing to fear in being open in the discourse. They will eventually be proven wrong: Nature’s biological processes are not changed by political or ideological agendas, neither by pro-abortion agendas nor by pro-life agendas.

This debate can move forward to good science and good medicine by the establishment of open, easily accessed, solid, uniform and multiple data bases, as described above, generating gold-standard scientific studies.

May an open scientific inquiry and discourse begin.

-----------------------

[i] Lambe, M “Chapter Six Reproductive Factors” in Breast Cancer Epidemiology ed Li,C (New York Springer2010) 11-129.

[ii] Clavel-Chapelon F, Gerber M Reproductive factors and breast Cancer Risk Breast Cancer Res Treat 2002 72(2):107-115

[iii] Decarli A,et al. Age at any birth and breast cancer in Italy. Int J Cancer Jul 1996 17; 67(2):187-189

[iv] J.R. Daling et al., “Risk of Breast Cancer among Young Women: Relationship

to Induced Abortions,” Journal of the National Cancer Institute 86 (1994):

1584–1592;

[v] H. Lambe et al., “Parity, Age at First and Last Birth, and Risk of Breast Cancer: A Population Study in Sweden,” Breast Cancer Research and Treatment 38 (1996):305–311

[vi] Shah P, Zao J on behalf of Knowledge Synthesis Group of Determinants of Preterm/LBW Births. Induced termination of pregnancy and low birthweight and preterm birth: a systematic review and meta-analyses. BJOG 2009; 116:1425–1442. H. Swingle et al., “Abortion and Risk of Subsequent Preterm Birth: A Systematic Review and Meta-Analyses,” Journal of Reproductive Medicine 54 (2009): 95–108;

[vii] B. Rooney and B. Calhoun, “Induced Abortion and Risk of Later Preterm Births,”

Journal of the American Physicians and Surgeons 8 (2003): 46–49.

[viii] Richard E. Behrman, Adrienne Stith Butler, Editors. Preterm birth: Causes, Consequences and Prevention. Committee on Understanding Premature Birth and Assuring Healthy Outcomes. Institute of Medicine. Appendix B, Table 5, p. 519.  Available at:

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download