In early 1995, I received an invitation from Annette Flanagin, R.N., M.A., Associate Senior Editor of the Journal of the American Medical Association (JAMA). The letter invited me to submit a paper for review and possible publication in JAMA's radiation issue of August 1995 --- the 50th anniversary of the Hiroshima-Nagasaki atomic bombings.Part 1. My Suggestion, in Reply to JAMA's Invitation
In response to this invitation, I talked with Ms. Flanagin and Dr. Harriet Meyer by telephone, and followed-up with a letter sent simultaneously to Ms. Flanagin, to JAMA's book editor (Harriet S. Meyer, M.D.), and to the chief editor (George D. Lundberg, M.D.). Text of the letter:
March 14, 1995
As all of you will recall, in answer to your invitation to me to submit an article to be considered for the "Hiroshima" issue of JAMA, we mutually agreed that Dr. Meyer could get an unbiased review of my new book, "Preventing Breast Cancer: The Story of a Major, Proven, Preventable Cause of This Disease," as a contribution to this issue. I not only think this decision is appropriate; I think this is an excellent forum for presenting the issues I have to address.
I am pleased to say that I shall be able to provide finished books, rather than manuscript xeroxes, to you for the Symposium. The printing establishment has informed me that we shall have books before the end of March, and I will forward copies to all of you just the moment we have books in hand. I know that Dr. Meyer needs several copies. These shall be provided, plus any additional copies which would be helpful to any of you.
Meanwhile, I urge you all to examine the several [enclosed] pages from the book, especially the back cover which conveys important messages of the book.
I have a few things to say about medicine, about fairness, and about bias. Medicine and medical research have suffered greatly as a result of charges of conflict-of-interest, and I would hope that medicine will not add fuel to this controversy. My experience, which may not be familiar to you, is very relevant on this issue.
One rarely has an opportunity to do something in medical research that really makes a difference. In 1948-1954, my colleagues and I introduced the discovery of the various classes of low-density lipoproteins and the various classes of high-density lipoproteins, through our work with the ultracentrifuge. And very soon after the initial discoveries, we related the low-density lipoproteins to coronary heart disease. [Insert of January 1996: See Gofman 1996 in the Reference List.]
The resistance to those discoveries was fierce, and there is no doubt that such resistance unnecessarily wasted a decade or more in moving the heart disease field along. Ask Dr. Donald Fredrickson about this as a fact. [Insert of January 1996: See Fredrickson 1993 in the Reference List.]
Today, none of the young scholars even know about that era, or that there was fierce resistance to new ideas. And, after the resistance phase, my work was widely accepted, and I have received several major awards in heart disease research, including the Lyman Duff Lectureship, the Stouffer Prize, and the commendation of the American College of Cardiology. In 1993, the American Heart Association invited me to be the Guest Lecturer at the Annual Arteriosclerosis Section Dinner Meeting in Atlanta. [Insert of January 1996: See Gofman 1993 in the Reference List.]
I am happy that all turned out well, but there was little reason to be happy about the personal attacks during the height of the resistance.
So the issues of bias, resistance, are not novel events in my life. I shall not be surprised that my recent work on the origin of breast cancer will receive a fierce "going over." I welcome honest criticism. There is a monumental difference between honest criticism and exercise of conflict-of-interest. Medicine is the big loser in those instances where eventually it turns out that conflict-of-interest, rather than merit, is at work in an assessment of a piece of work.
Dr. Meyer, you and I have discussed this issue by phone. I would consider choice of a radiologist, or a nuclear medicine specialist, or a radiologic physicist to review my book to be an obvious conflict-of-interest. What one needs is an honest scientist-physician who does not have a conflict-of-interest. I could name many, if asked.
I look to you, Dr. Lundberg, to assess this problem, when the review comes in. I would trust you to decide whether some editorial comment from you might be in order in the Hiroshima JAMA issue. The matter of breast cancer is a monumental issue in our society, and we will all be judged, in our profession, by how we handle the issue of causation. There are some who think the problem will go away if we shut our eyes, ears, and brains to evidence that is as close to a smoking gun as we shall ever see.
I have never taken the position of being "anti-medical x-rays." The enclosure here of the back cover of the new book proves beyond doubt that I do not fall in that category. The text of my book is laden with compliments to the contributions of famous roentgenoscopists in the era of pneumo-thorax management of tuberculosis. They gave many women decades of good life after recovery from pulmonary tuberculosis. Yes, some [women] later developed breast cancer as a result. I applaud those roentgenoscopists for making it possible for those patients to live many decades, even if some paid a price in breast cancer for that gift of decades of good life, after recovery from tuberculosis.
My suggestions will lead to better roentgenology, not to less roentgenology. Just look at the evidence. Mammography was correctly criticized in the early 1970s for being a dose-related health hazard. Mammography today is performed with vastly lower doses and with much wider application. Criticism had the opposite effect from that of destroying mammography.
So, if medicine cannot look back on the first hundred years of Roentgen's discovery, and assess honestly the good and the bad, that will be a very sad day for medicine. I believe I have made a constructive effort in this book to see that the second hundred years of application of Roentgen's discovery is an era where we achieve the benefits of his discovery, with a drastic reduction in the health costs. That is possible and desirable. That will earn the appreciation of the public. Any conflict-of-interest denial of the undeniable will only earn well-deserved scorn for our profession.
I noted with great interest that you, Annette Flanagin, and you, George Lundberg, specifically mention receiving [inviting] reports of original research. The methods I have used for this research are indeed original, and they were not in existence when I started. I feel that this is my best research since my work 47 years ago on the discovery and significance of the low-density lipoprotein classes. And these  research methods can undoubtedly be applied to other cancer research efforts.
The issue is not minor. I look forward to working with all of you toward a constructive approach. This is an opportunity, not a burden for any of us. Think of it --- the findings mean that many, possibly most, breast cancers are avoidable.
John W. Gofman, M.D.
P.S. If you, Dr. Lundberg, do not feel an editorial by you is warranted, you might wish to consider publishing this letter, as a Letter to the Editor, in the Hiroshima issue.Part 2. The Subsequent Sequence of Events at JAMA
In very early April of 1995, we sent books to JAMA, as promised (above): Copies to Annette Flanagin, George Lundberg, and three to Harriet Meyer.
Who was chosen by JAMA to review the book? Clark W. Heath, Jr., M.D., vice president for epidemiology and surveillance research at the American Cancer Society (ACS). The ACS is a strong advocate of irradiating people, both for diagnostic and therapeutic purposes with respect to cancer. Indeed, the ACS is a leading advocate of mammography --- and was advocating widespread mammography even when the radiation doses were 10 times higher per exam than they are now (Chapter 28 of this book, Part 5).
Dr. Heath's review of the book was published in JAMA on August 23, 1995 (Heath 1995). Meanwhile, JAMA's associate editor for the Medical News and Perspectives section, Andrew Skolnick, wrote two pages about the book in the JAMA "Hiroshima" issue of August 2, 1995 (Skolnick 1995). So JAMA really did its job in contributing to peer-review, although almost all of the substance comes from a single person, Dr. Heath.
We responded to both Skolnick 1995 and Heath 1995 (see Parts 3 and 4, below).
Our response to Skolnick 1995 focused largely on Heath's claim (used by Skolnick) that our calculations are "based on two serious errors." Our response to Heath 1995 focused on his assertion (for which Heath cites Evans 1986) that medical uses of radiation probably account for "less than one percent" of breast cancer in the USA.
Following the instruction of Annette Flanagin, we submitted two 500-word Letters to the Editor, for consideration. The response to Skolnick 1995 was sent by us on August 21, and the response to Heath 1995 was sent by us on September 19.
On October 10, 1995, JAMA notified us that JAMA would publish our response to Skolnick 1995. This occurred in the December 13, 1995 issue of JAMA. Our published letter is followed by the notation: "This letter was shown to Dr. Heath, who declined to reply. -- ED."
On October 11, 1995, JAMA notified us that JAMA would not publish our response to Heath 1995: "After considering the opinions of our editorial staff, we determined that your letter did not receive a high enough priority rating for publication in JAMA ... However, we are forwarding a copy of your letter to the author of the article [Heath]. The author may or may not reply to you personally." The letter says: "cc: Clark W. Heath, Jr., MD" at the bottom. We have not heard from Dr. Heath yet (January 1996).
Both of our responses, to the critiques in JAMA, are presented below. New references cited in the letters have been added to our Reference List.Part 3. Our Response to Skolnick 1995
Our response to Skolnick 1995 was entitled (by us) "X-Rays Not Exonerated As the Major Cause of Current Breast Cancer." The title's beginning and end were still present when it appeared in the December 13, 1995 issue of JAMA (Gofman 1995-b):
"X-Rays and Breast Cancer."
To the Editor. --- My estimate, that at least 75% of current breast-cancer in the USA is due to earlier medical irradiation of breasts (Gofman 1995), was called a 12-fold overestimate in the article by Skolnick 1995. He is quoting Clark Heath of the American Cancer Society, who asserts that I made "two serious errors."
First, an alleged 2-fold overestimate comes from my "assumption" that dose-response is supralinear. This is no assumption. The human evidence from the atomic-bomb survivors for all cancers combined shows supralinearity fitting the observations provably better than linearity (Gofman 1990). Supralinearity specifically for breast-cancer is visible to anyone who inspects the figure at page S26 of Thompson 1994, although Land's analysis raises questions (Land 1995).
Second, an alleged 6-fold overestimate comes from transport of Japanese evidence to the USA. Why did I not use North American data? Because (a) North American data on risk are inconsistent by a factor of six from one study to another, and (b) such studies exclude infants and children. For these and other reasons, the A-Bomb Study is more informative.
How to transport observations internationally is an unsettled issue, with the uncertainties fully discussed in my book (Gofman 1995). Although Heath asserts he knows which way of transport is a "serious error" and which way is correct, no one today can possibly be sure.
Nonetheless, Heath claims the correct way would be to make no adjustment of the Japanese observations for application here. His unstated assumption is that radiation and other causes of breast-cancer each act alone, without co-action, and therefore, a unit of radiation produces the same number of breast-cancers in all countries. My clearly stated assumption is that radiation and other causes co-act, and therefore, adjustment is required for transport of evidence from one milieu of co-actors to another.
Land 1995 presents evidence on both sides of this wide-open issue. Although he takes a position like Heath's at first, Land quickly allows for radiation co-action with breast-cancer risk-factors like benign breast disease, reproductive history, and inherited lesions.
Skolnick's report (Skolnick 1995) raises the safe-dose issue twice, by quoting Dr. Feig and alluding to "many radiation physicists" who believe DNA damage is fully repaired at doses which do not "overwhelm" the repair-system. Their speculation is refuted by mainstream human evidence from serial breast-irradiation at doses below 0.1 Sv --- in fact, at doses close to the lowest conceivable radiation dose and dose-rate, namely one ionization track per cell-nucleus per exposure (Gofman 1990). [Additional information in Chapter 45.]
Thanks to rejection of the safe-dose fallacy, radiation dosage from mammography is now 30-fold lower than in 1970. Don't we owe it to women also to achieve much lower doses for millions of other useful exams which irradiate the breasts, including upper spine, upper GI tract, and heart procedures, plus neonatal x-rays? In the medically relevant dose-range of 0.1 to 0.5 Sv of accumulated dose, Land's table for Japan (Land 1995) estimates that 10.5 percent to 38.0 percent of all breast-cancers were radiation-induced.
John W. Gofman, M.D., Ph.D.
University of California, Molecular & Cell Biology
Berkeley, California 94720Part 4. Our Response to Heath 1995
Our response to Heath 1995 was entitled "Relevant and Irrelevant Approaches to Explaining the Breast Cancer Problem." The response was not published at all by JAMA, as related in Part 2.
To the Editor. --- An interesting error merits correction. Clark Heath (Heath 1995) features a 1986 estimate by Evans et al that less than one percent of all breast cancer in the USA results "from diagnostic radiography" (Evans 1986). Then Heath asserts my new book (Gofman 1995) "disputes these facts" because I estimate that 75 percent of cases are due to earlier exposure to medical irradiation (including therapeutic radiation for numerous non-malignant conditions).
Evans' estimate and Heath's use of it fail to take account of the long latency periods of radiation-induced breast cancer. The A-Bomb Study now shows that excess breast cancers are gradually "delivered" over at least 45 years, when a female population (all ages) is irradiated on the same day. Prudent analysts must assume radiation-induced cases will continue for 60-75 years among females irradiated in infancy and childhood.
Because of long latencies, anyone who wants to uncover the radiation-contribution to our contemporaneous incidence of breast cancer must evaluate average breast doses received during many decades beforehand, by females of all ages. My study did it for 1920-1960, whereas Evans et al considered the single year of 1977.
How much does the choice matter? For children up to age ten, the Evans' per capita estimate of average annual breast-dose in 1977 (from diagnostic x-rays) is 35 times lower than my estimate per year during 1920-1960 (from diagnosis plus noncancer therapies). The difference is really bigger, because my dose estimate excluded any contribution by fluoroscopy, despite the documented practice of doing fluoroscopic exams as part of routine well-baby and pediatric check-ups. So proper attention to latency introduces more than a 35-fold disparity between Evans and me in the relevant per capita dose-estimate for children --- the most vulnerable group.
For everyone above age ten, the Evans' estimate is about 10-fold lower for 1977 than my estimate for the earlier time-period. Again, the disparity is really bigger, because my analysis excluded all uses of radiation in dermatology, all radiation therapies for other inflammatory conditions (from tuberculosis of the lymph glands to peptic ulcer), and almost all uses of fluoroscopy.
An undeniable, major flaw in the Evans' analysis is its exclusion of doses from the relevant period. When Heath presents its conclusion as a "fact" which contradicts my conclusion, he is just wrong.
My 75-percent estimate is the product of two factors: (Dose in 1920-1960) times (Cancers per Unit Dose). For the second factor, all analysts including Evans have to incorporate important assumptions, pending future evidence. If some JAMA readers accept Heath's opinion (Heath 1995) that my second factor is too high, they must also recognize the first factor: Past dosage. My book shows repeatedly why my dose estimates are probably many-fold too low --- a point not mentioned by Heath.
Thus the bottom line is that my critics and I must already be much closer than Heath indicates, in recognizing that past medical irradiation explains a large fraction of the current breast cancer problem.
John W. Gofman, M.D., Ph.D.
University of California, Molecular & Cell Biology
Berkeley, California 94720Part 5. A Headline Which Merits Dispute
JAMA is one of the leading medical journals of the world. Such journals are meant to be an important vehicle for exchange of well-informed opinion --- rational dialog --- about the causes, prevention, and treatment of illness. Both Skolnick and Heath acknowledge our standing for such a dialog. Skolnick characterizes Gofman as "a respected authority on the biological effects of ionizing radiation" (Skolnick 1995, p.367), and he quotes Heath as saying that Gofman is "a very good radiation physicist who has had a remarkable career" (p.368).
One of the major diseases of our time is breast cancer. Countless papers are published about it. Along comes an experienced researcher who produces evidence that the main cause of breast cancer is a preventable one: Irradiation of the breasts. In response, someone at JAMA creates a three-column headline:
"Claim That Medical X-rays Caused Most US Breast Cancers Found Incredible"(Skolnick 1995, p.367).
"Found incredible?" Who at JAMA selected and permitted two words which indicate that the work had flunked peer-review, when peer-review was just in its very first step? It had to be people who were not even interested in whether or not our response would invalidate the criticism. This is not the way a leading medical journal serves as a neutral vehicle in a debate of singular importance for women and their families. But every institution has failures from time to time! [And it's easy for editors to miss the boat. E. O'Connor, Editor.]
By contrast, Skolnick's text --- for any readers remaining after its headline --- was an excellent introduction to the issue. Skolnick even included some of our position on peer-review: We want it to occur widely and in the open, not "behind closed doors."
Both JAMA items (Skolnick and Heath) relied almost entirely on Heath for peer-review --- one peer. And one peer can make numerous claims in a short space. Everyone knows that it takes more space to refute a claim than just to state one. Although JAMA allowed us a response of only 500 words (with a limit of five references), it is likely that JAMA will return to our work in the future.
JAMA-readers have a right to expect a series of in-depth exchanges until there is some resolution of an issue which actually addresses prevention of future breast cancer. Meanwhile, JAMA's headline is sure to stick in many memories, but our 75-percent thesis has certainly not been shown by anyone to be "incredible" --- as readers of this book's Section 5 will see for themselves. Above, we have shown why "incredible" is the appropriate adjective for Heath's one-percent claim. JAMA got it wrong this time.
# # # # #