All of the following comments, except those in Parts 1, 2, and 3, were made by people who could not have even seen the book at the time. They express preconceptions. These preconceptions are common, and interesting, and need responses. The preconception examined in Part 7 seems to dominate all others. If a comment comes from a newspaper, we omit the name of the maker because, when newspapers abbreviate, the residual quotations may not be exactly what the maker really said.
Part 1. There Was Breast Cancer before the Discovery of X-Rays
In the London Herald of August 3, 1995, one of the authorities at the National Radiological Protection Board is quoted as follows:
"He [Gofman] is holding out the prospect to women that if they never have an X-ray in their life, they will not get breast cancer, and that's absolutely not true. Breast cancer did not start when X-rays began to be used."
"Absolutely not true" is the claim that we have ever held out the prospect of making breast cancer disappear, if women would refuse all x-rays for themselves and their female children. Again and again, we discuss the role of additional causes of breast cancer (pp. 6, 280, 281, 282, 284, 286 --- always the same in both editions, through page 300 (end of Chapter 42)).
The occurrence of breast cancer prior to the discovery of x-rays is totally consistent with our thesis, that medical x-rays have become the dominant cause of breast cancer. Moreover, some of the pre-x-ray cases were radiation-induced, because natural background radiation irradiates every female in-utero and, thereafter, every year of her life. Surely the NRPB's expert knew that.
It is important to note, too, that no one should glibly dismiss the role of natural background radiation in the inherited predisposition to breast cancer. Our hypothesis (Gofman 1994) is that humanity's aggregate burden of inherited afflictions is very largely the consequence of exposure to natural background radiation over the centuries and millenia --- and the consequent build-up of radiation-induced genetic lesions in the population.
Part 2. Complaint: We Did Not Discuss Nuclear Power Plants
Drs. Ernest Sternglass and Jay Gould have suggested that radioactive releases from civilian nuclear power plants may explain above-average rates of breast cancer in various counties of the USA. We have been scolded for not dealing, one way or the other, with their suggestion.
The main reason for not dealing in this book with the Sternglass-Gould hypothesis is that our Master Table in Chapter 39 covers doses received in the 1920-1960 period. Civilian nuclear power plants in the USA got underway later, during the 1960s. Regarding the post-1960 period, we are unaware of any dose-estimates associated with the Sternglass-Gould hypothesis. The number of curies of various radionuclides released, per plant, is not an estimate of dose at all.
Our guiding principle in this work is to direct attention first to the dominant sources of radiation, and we believe that the dominant source of controllable radiation exposure in this country is still medical irradiation (see Chapter 48). On the other hand, nuclear pollution could become the dominant source one day, if citizens don't watch out.
Part 3. Complaint: We Omitted Breast-Dose from Past Dental X-Rays
One of our most intelligent critics over many years has been Miriam Goodman, of Long Island. We learn a lot from her. And she raised an interesting question about the First Edition, when she asked: Why didn't you discuss breast-dose from past dental x-rays received during the 1920-1960 period?
We should have mentioned dental x-rays in Chapter 38, as a potential source of underestimation of breast-dose. Mrs. Goodman suggests that the x-ray beam was probably not well focused on just part of the jaw. "Even today, the technicians shield themselves behind a wall when they take dental x-rays. What was going on in the 1920-1960 period? Did they ever shield the patient's chest with a leaded apron?" We do not know how to answer such questions.
Part 4. "There Are No Data to Show that BC is Caused by Mammograms"
"As far as we're concerned, there are no data to show that breast cancer is caused by mammograms," commented the Medical Director of a major Comprehensive Breast Center in California, when asked by a newspaper to comment (May 5, 1995) on the 75-percent estimate of this book. The response reflects two common and mistaken assumptions, occurring in high places.
The First Mistake
The safe-dose fallacy is showing in this statement. Unless the Medical Director assumes a safe threshold-dose, the Medical Director has overwhelming evidence that the x-rays from mammograms are capable of inducing breast cancer. The evidence for radiation-induction of breast-cancer is not in dispute (see flagged # entries in our Reference list). The evidence against any threshold dose is also overwhelming (Chapter 45). Nonetheless, this "authority" --- Medical Director of a Comprehensive Breast Center --- is assuring women that there is no risk of mammogram-induced breast cancer "as far as we're concerned."
The Second Mistake
The second mistake is the assumption that this book blames a lot of breast cancer on mammography. Reality: The 75-percent estimate of this book does not include a single case caused by mammography. Mammography is not included at all in our Master Table, which covers 1920-1960. But is this book hostile to mammography nonetheless?
We have taken no position on mammography. On page 173, we draw the distinction between wanting to estimate the cancer-consequences of modern mammography, versus wanting to obstruct use of mammography. On pages 180-181, where we show women how to estimate their personal risks from mammography, we emphasize three separate times that the chance of not developing mammogram-induced breast cancer far exceeds the chance that a mammogram-induced cancer will ever develop.
Hostile? The First Edition pointed to mammography as a shining model of how much has been achieved in reducing unnecessarily high radiation dosage, whenever the medical world got serious about doing so (p.293). And we emphasized the point again (bottom, p.298).
Part 5. "Not Plausible" Because X-Ray Dose Falling without BC Decline
On May 7, 1995, the London Sunday Telegraph carried "X-ray link to breast cancer," a front-page news story about this book. The reporters (Robert Matthews and Victoria Macdonald) asked for a comment from a cancer specialist at Hammersmith Hospital in London. The professor said:
"There has been increasing awareness of the use of X-rays, and over the past 30 years, there has been a reduction in radiation doses, so one would have expected to see a decline in breast cancers if these claims were correct. Instead, breast cancer cases have dramatically increased across the world in 50 years." Several additional experts have voiced the same assumptions and the same mistaken conclusion: If x-rays were important in breast-cancer causation, then the breast-cancer problem would be declining today.
Not so.
Such critiques fail to take into account the long and variable latent period which follows breast-irradiation (Chapter 2), and the radiation-sensitivity of the women whose x-ray-induced breast-cancers are delivered after 1965. Because of these factors, it is easily possible for x-ray-induced breast-cancer to be flat, or even to increase, during a period when breast-irradiation is gradually declining.
5a. The "Breast-Cancer Years"
Except for the early-onset cases (Chapter 3), the x-ray-induced cases wait until the breast-cancer years --- ages 35 to 85, and beyond. What we mean by "the breast-cancer years" can be illustrated by the rates of breast-cancer deaths, per 100,000 white women in the USA, for 1980 (from USDHHS 1993, Table 39, p.71). For incidence, the climb has to begin at younger ages than for deaths.
Ages 25-34 3.0 Ages 35-44 17.3 Ages 45-54 48.1 Ages 55-64 81.3 Ages 65-74 103.7 Ages 75-84 128.4 Ages > 85 171.7What Happens between 1945 and 1980?
We can use the year 1980 also to illustrate the role of latency in correct expectations for x-ray-induced breast-cancers. We will use the year 1945 as a typical year of irradiation from our analysis. The interval = 35 years, from 1945 to 1980.
In the year 1945, some infants were receiving thymus irradiation and "well-baby" fluoroscopic check-ups. Some children were receiving pre-tonsillectomy irradiation. Some women in their 20s were receiving artificial pneumothorax therapy for tuberculosis. In diagnostic radiology, the x-ray beam was poorly collimated, and breasts were in the beam when they did not need to be. Films were slow, and exposures higher than today. Fluoroscopy was popular. We need not review the entire book here.
Now the year 1980 arrives 35 years later. Delivery of the x-ray-induced breast-cancers, produced during 1945, is not yet finished. Far from it. Because most diagnosis occurs during "the breast-cancer (mortality) years," some delivery from 1945 is just beginning in 1980. The infants irradiated in 1945 are age 35, and are just barely arriving at their passage through the breast-cancer years. The age-10 children irradiated in 1945 are only age 45 in 1980 --- with most of their breast-cancer years still ahead of them. The age-25 women irradiated in 1945 are age 60 in 1980, with many breast-cancer years still ahead. The age-40 women irradiated in 1945 are age 75 in 1980, and even at that age, discovery of new cases continues.
5b. Bottom Line: Not Only Possible, but Also Plausible
And our analysis did not stop in 1945. It considers x-ray practices out to 1960. So in 1980, the delivery of x-ray-induced cases produced in 1960 is even far less complete than the illustrations above from exposure in 1945. The illustrations above used a 35-year interval between irradiation and observation. For irradiation in 1960, a comparable interval would bring the observer to 1995.
During the past 30 years, delivery of radiation-induced breast-cancers just began to include cases from the most sensitive age-groups in our analysis (the younger ones in the 1920-1960 period). Their radiation sensitivity is greater than that of women who were older in 1920-1960, and who were already passing through the "breast-cancer years" before 1965. The greater sensitivity to x-rays, of the groups starting to "deliver" after 1965, could easily offset a gradual decline in dosage.
So it is not only possible, but also plausible, that a radiation-induced cancer problem can be flat or increase while, simultaneously, the x-ray dosage is gradually declining. The long and variable "delivery" period, for cases produced by earlier x-rays, must be kept in mind.
Breast-Cancer Rates per 100,000 Women
It should be noted that the estimates of this book are based on absolute numbers of cases per year: 114,336 and 182,000 (p.285). Our method (a) estimated that 114,336 was the absolute number of radiation-induced breast-cancers "put on the shelf" (produced) each year during the 1920-1960 period, and (b) approximated that the absolute number of cases "delivered" later, per year, would approach equality with the absolute number produced earlier, per year (Chapters 4 and 5). One of the stated approximations was stability of the conversion-factors over decades (p.286).
Our absolute number for (a) and (b) above was 114,336 cases per year. It was compared with 182,000 total cases in 1994, not with trends in breast-cancer rates per 100,000 women. The interesting topic of rates is examined in the forthcoming HEIR-3 Report from this Committee (1996, ISBN 0-932682-97-9).
Part 6. "Can't Be 75%" When Only 5-10% of My Patients Had Radiation
The Oakland (California) Tribune published a story (May 3, 1995, p.A-1) about this book, in which the reporter quoted an assistant clinical professor at UCSF Medical School responding to the 75-percent estimate: "I think it seems implausible." Reason: He said only 5 percent to 10 percent of his patients have a history of radiation exposure. Presumably he meant his patients with breast cancer.
Response:
The comment indicates that the professor asks his breast-cancer patients about radiation exposures in the 1920-1960 period --- which is the source of our 75-percent estimate. We have yet to meet a breast-cancer patient who was ever asked. But we are sure there must be some physicians asking.
For the physicians who do ask about the 1920-1960 period, how can the percentage in the response be meaningful? It is not realistic, for example, to think that many women --- say, age 60 at diagnosis in 1990 --- will know that they received irradiation for "enlarged thymus" as a newborn in 1930, or monthly fluoroscopy from birth to age 2, or pre-tonsillectomy thymus irradiation, etc.
How many parents would have thought to tell their children, when such practices were just routine? By contrast, women who were found in childhood to have a congenital problem (say, a heart murmur) would be more likely to have learned something about their childhood radiation history. But why would parents ever mention diagnostic x-rays and fluoroscopy to the even greater number of children whose exams ruled out such a diagnosis?
Even though positive response is bound to be falsely low, to questions about a patient's radiation history in the 1920-1960 period, we strongly encourage physicians to ask. If a physician realizes that the response is falsely low, then a positive response of just 5 percent or 10 percent ought to put a crack in the existing "mental concrete" surrounding the topic of x-ray-induced breast cancer.
Part 7. "Can't Be 75%" Because Leaves Too Little Room for other Causes
During the past eight months, we have learned that the dominant preconception about this work is that the 75-percent estimate "can't be right, because it leaves too little room for other causes." This prejudice is so powerful, that people with this belief never read the work! If they did, they would learn how the 75-percent estimate is based on synergism and how it already incorporates other causes.
7a. Synergistic and Additive Co-Action with Radiation
The 75-percent estimate arises out of conversion factors (number of breast-cancer cases per dose-unit of radiation). These conversion factors start with the A-Bomb Study, which already includes radiation's co-action with the other causes. This is no different from studies which may find a dose-response between, say, pesticide-exposure and breast cancer. The response to pesticides would include the co-action with medical x-rays and with the other causes.
What is measured in such studies is the difference in breast-cancer rate when one co-actor is increased and other co-actors are presumed to be constant in all the dose-groups. Then it is correct to say that the increased agent accounted for the difference in breast-cancer response. But also, the observed breast-cancer rate in every dose-group includes the contribution from all the co-actors.
Synergistic co-action and additive co-action are different. Synergism, defined, is "the simultaneous action of separate agencies which, together, have greater total effect than the sum of their individual effects."
To illustrate the distinction between additive co-action and synergistic co-action, we can imagine that there are only two types of agents which cause breast cancer: Radiation and non-radiation. And we can suppose that these two agents co-act to produce 180,000 new cases per year in the USA. We might suppose that the co-action is exclusively additive --- say 70,000 cases caused by radiation acting independently of non-radiation agents, and say 110,000 cases caused by non-radiation agents acting independently of radiation.
Or, we might suppose that the 180,000 cases are caused exclusively by synergistic co-action between the two types of causes. We have ruled out independent action by either agent in this illustration. So radiation alone, in the absence of non-radiation agents, causes no breast cancer ... and non-radiation agents alone, in the absence of radiation, cause no breast cancer. Without the other, each type of agent is impotent, and "the sum of their individual effects" is zero. But together, their total effect would become very much larger than zero --- some 180,000 new cases per year. That would be synergistic action, with each agent amplifying the effect of the other.
Although we illustrated either one type of co-action exclusively, or the other type exclusively, some carcinogens may act in both additive and synergistic (multiplicative) ways --- and thus have some independent power as well as synergistic power. The issue of co-action is also discussed at pages 280-282, 286, and Chapter 44, Part 3.
7b. "Equal Opportunity" for Non-Radiation Factors
In human experience, radiation as a co-actor is never fully absent, due to natural background sources. Thus, measurements of every other cause of cancer always include radiation co-action (either synergistic, or additive). And we know of no radiation study where all non-radiation co-actors were absent.
We who study radiation are well aware of the uncertainties about co-action (see pages 280-282, 286, and Chapter 44, Part 3). We must consider endogenous and exogenous non-radiation carcinogens (including promoters). Does their presence make the carcinogenicity of ionizing radiation much worse than it is in their absence? If so, how much worse? Indeed, does ionizing radiation alone, in the absence of other carcinogens, produce any cancer at all?
With these questions unanswered today, it is clearly a mistake --- a correctable one --- for anyone to assume that our 75-percent estimate "leaves too little room for other causes!"
o - We have said explicitly (pp.280-282) that the 75-percent estimate incorporates the assumption of synergism between radiation and co-factors (other causes). So other causes are already present in the 75-percent estimate. By definition, synergism means that radiation is multiplying the power of the other causes (co-factors), and the other causes are multiplying the power of the radiation. Does this sound as if there is "too little room for other causes?"
o - We have said that without co-factors, the radiation-effect might even be zero (p.286). And we assume that analysts who explore the carcinogenicity of pesticides, estrogens and xeno-estrogens, nutrients, or other exogenous and endogenous substances, sometimes ask themselves:
(a) Does the carcinogenic potency of these substances depend upon the presence of radiation-induced genetic lesions, either acquired or inherited? (See Chapter 1, Part 5). (b) Do radiation-induced genetic lesions make the carcinogenicity of the substances many times worse than they are in the absence of such lesions? (c) Do these substances cause any cancer at all in the absence of radiation-induced lesions?
o - In terms of preventing breast-cancer, we have said (p.5) that "a very large part of the cancer problem can be eliminated, if people correctly identify and eliminate the non-inherited forces which act alone or act in concert with inherited genetic lesions, in producing malignancy." Ionizing radiation is one of the proven causes which can be reduced very easily, and if synergism occurs between cancer-causes, reduction of radiation will also reduce the carcinogenic potency of the co-factors. This would be very good news.
o - "There is no inherent conflict or competition between carcinogens. The multi-step genetic model of cancer development `permits' contributions even to a single case of cancer, from heredity, ionizing radiation, viruses, and chemicals (for example). It is correct to say that each contributor caused the cancer, if the case would not have occurred when it did without that contributor" (p.6). Does this sound like "too little room for other causes?"
There are a lot of reasons to "bet" that synergism occurs among cancer-causes. There is certainly no scientific basis to rule out synergism at this time. And yet that is what people do, whether or not they realize it, when they shut their minds to our 75-percent estimate "because it leaves too little room for other causes." Our 75-percent estimate leaves all the room in the world for other causes to be full partners, as extremely important or probably essential co-factors.
"Every truth passes through three stages before it is
recognized. In the first, it is ridiculed. In the
second, it is opposed. In the third, it is regarded as
self-evident."o - Arthur Schopenhauer, 1788-1860.
German philosopher.
# # # # #