fancy HTML | ASCII text formats
John W. Gofman, M.D., Ph.D.
Professor Emeritus of Molecular and Cell Biology,
University of California, Berkeley.
Ms. Egan O'Connor, Editor, Committee for Nuclear Responsibility.
Table of Contents for These Comments:
* Part 1. FDA's Goal: Reduce X-Ray Exposure to Patients
* This communication, with its three attachments, is offered in complete support of the U.S. Food and Drug Administration's (FDA) proposed performance standards for new x-ray fluoroscopic systems, under the Code of Federal Regulations 21 CFR 1020.30 to 1020.33. Docket 01N-0275.
* Our purpose here is to contribute scientifcally strong evidence that the FDA has greatly underestimated the health benefits of its own proposals (see Parts 4 and 7 below, elaborated in Attachments 1 and 3).
* The goal of the FDA's proposed amendments to the current standards is "to improve the public health by reducing exposure to and the detriment associated with unnecessary ionizing radiation from diagnostic x-ray systems, while assuring the clinical utility of the images ... These proposed amendments will require additional features on newly manufactured x-ray systems that physicians may use to minimize x-ray exposure to patients" (FR, pp.76057-58; please see FR 2002 in our Reference List.)
Main Target: Fluoroscopic X-Ray Systems
* Most of the features apply to fluoroscopic x-ray systems, which use continuous x-ray beams to deliver real-time images to physicians during many common medical/surgical procedures, such as upper GI exams, cardiac catheterization, angiography, angioplasty, urinary/biliary stone removal, various needle-biopsies, placement of catheters, stents, filters, etc.
* Fluoroscopy never irradiates the entire body, but exposes sections of it to x-rays at rates like 1 centi-gray (cGy) up to 20 cGy per minute, with cumulative doses per procedure ranging from less than 1 cGy to over 1,000 cGy (e.g., an estimated 60 cGy per stenosis dilated during angioplasty --- NCRP 1989, p.31). The cGy and rad are identical units of dose.
Some of the Newly Required Features, If Approved
* The required features for new fluoroscopic systems would, if approved, include the capability for last-image-hold ("freeze-frame"), which permits physicians to view and discuss an image during a procedure, without continuing to irradiate the patient. Cost per new system would be about $2,000 (FR, p.76078).
* Very importantly, also required would be display capability to show the operator the dose-rate per minute, the duration of exposure, and the accumulated skin-dose to the patient in "real-time" (during the procedure), and to record such doses. The estimated additional cost per new system is $4,000 (FR, p.76078), whereas addition of such display to a system designed without it is about $10,000 (FR, p.76078).
* Also required would be greater filtration of the x-ray beam (to eliminate many of the x-ray photons which only irradiate the patients but do not contribute to the images), better collimation (to reduce irradiation of tissue not even within the image), and other common-sense features.
* Part 2. The Benefit-Cost Estimate by the FDA
* How great will the health benefits be, if these features are approved, according to the FDA?
FDA's Estimate of Annual Benefits and Costs
* Using the customary rate of replacing older fluoroscopic systems by newer ones in the USA, the FDA expects that, within ten years, all such systems will include the required dose-reducing features (if approved).
* The FDA attempted to estimate the health benefits achieved during the first ten years while the new features are being phased in. It estimated the aggregate expected dose-reduction (based on only a few common procedures) during the ten years. Then it applied a value for the estimated dose which causes one fatal cancer. Then it made adjustments to an annual basis, taking account of gradual phase-in of the new features.
* How many "premature deaths associated with cancer" should be prevented annually by that method of estimation? The FDA's median value is 223 such deaths per year (FR, p.76076) --- a very small number by comparison with over a half-million cancer deaths per year in the USA. The annual benefit associated with 223 prevented deaths is estimated at $320 million per year by the FDA (FR, p.76077).
* The FDA's estimate of annual cost, including one-time costs of re-designing equipment distributed over the ten years, is $40 million per year (FR, p.76079).
* This makes the FDA's benefit-cost ratio for the first ten years 8 to 1 in favor of the benefits ($320 / $40).
* Part 3. Why Is a Federal Mandate Required?
* The FDA asks an important question (FR, p.76072): With such a favorable ratio of benefits over costs, why must a federal mandate be invoked? The FDA's answer: The "market" does not respond to the ratio because the costs accrue to the profession, but the benefits accrue to the patients.
* We believe that the above dynamic does operate, but that it operates only because the medical profession has been taught for decades that the cancer hazard from medical x-rays is negligible. That same message continues to be repeated today, not only by the FDA (Part 2, above), but elsewhere with greater vigor (Part 6, below).
* Part 4. Gofman Evidence on Benefits from Dose-Reduction
* In great contrast to claims of very low hazards from customary medical x-ray practices (and thus, negligible health benefits from dose-reduction), we have uncovered powerful evidence that customary x-ray practices became and remain one of the necessary causal co-actors in over half of the fatal cases of cancer and over half of the fatal cases of ischemic heart disease (coronary artery disease) in the USA (Gofman 1999). The study's method and findings are most succinctly summarized in Attachment-1 (i.e., Parts 4 and 5 of Gofman 2002).
How Do X-rays Cause Coronary Artery Disease?
* Our Unified Model of Atherogenesis and Acute IHD Death (Gofman 1999, Chapters 44, 45, 46) proposes that a lipid-containing arterial plaque arises where atherogenic mutations (acquired after conception) produce a clone of dysfunctional cells (mini-tumors) which do an incomplete job of clearing the lipids out of that patch of dysfunctional arterial tissue and of protecting the arterial lumen from the accumulated thrombogenic lipids within the plaque.
* This model is consistent not only with previously established causal co-actors for IHD, but it also explains why arterial plaques occur only in discrete patches, surrounded by normal tissue. Some supplemental evidence for acquired atherogenic mutations is summarized in Note 1 of Attachment-1 (Gofman 2002).
Peer-Review of the 1999 Gofman Medical Monograph
* The new evidence presented in Gofman 1999 deserves serious consideration, according to Arthur C. Upton, M.D., former Director of the National Cancer Institute, and very active member of all the quasi-official radiation committees (BEIR, UNSCEAR, ICRP, NCRP). Dr. Upton's comment on the Gofman 1999 monograph, verbatim in its entirety, follows (Upton 1999, which is Attachment-2):
* "Thank you for kindly sending me a copy of your recent book entitled `Radiation from Medical Procedures in the Pathogenesis of Cancer and Ischemic Heart Disease.' Your observations are impressive and are consistent with the linear-nonthreshold dose-response hypothesis for the genetic and carcinogenic effects of ionizing radiation, and they support the wisdom of the ALARA [as low as reasonably achievable] principle in radiation protection." And:
* "At the same time, however, the associations you have so skillfully demonstrated cannot be taken as proof of causal relationships, owing to the possible influence of confounding variables. Just as the inverse relationship between lung cancer rates and county residential radon levels, as reported by Bernard Cohen, does not suffice to prove that low-level exposure to radon protects against lung cancer, neither do your observations suffice to establish medical radiation as a causal factor in the associations you have identified." And:
* "Nevertheless, I find your observations intriguing, and your interpretation of them to be thoughtful and constructively hypothesis-generating. I hope that your book stimulates the productive follow-up research that your findings clearly call for. Many thanks, again, for sharing your findings with me, and best wishes for continuing productivity in the new millenium. Arthur C. Upton."
* The Executive Summary of that 699-page medical monograph (Gofman 1999) and a discussion of its extensive peer-review (Gofman 2002) are provided as an integral part of this submission, as Attachments 1 + 3. So far, the main critiques of the 1999 monograph have produced no reason to reject, modify, or ignore its conclusions.
Gofman's Estimated Annual Benefit from Dose-Reduction
* The correlations uncovered in Gofman 1999 strongly indicate that accumulated exposure to medical x-rays is a necessary causal co-actor in over half of the fatal cases of cancer and IHD in the United States. Those two diseases account for over one million deaths per year in the USA.
* The concept, "necessary causal co-actor," is very commonly accepted with respect to cancer and to ischemic heart disease (IHD). It means that every case has more than one cause. Thus, in the absence of any one of the case's necessary causal co-actors, the case could not happen as it did.
* From this, plus the observations in Gofman 1999, it follows that if there were no exposure to medical x-rays, and if all other causal co-actors were held constant, the mortality rates from cancer and IHD would gradually fall in half or more.
* Also it follows, if all other things were held constant, that half of that half (i.e., about 250,000 premature cancer and IHD deaths per year) would be prevented over time, if customary x-ray doses were cut in half instead of completely eliminated. Part 5 discusses the feasibility.
* Part 5. How Much Can Fluoroscopic Doses Be Reduced?
* Fluoroscopy and CT scans account today for the bulk of the doses received by patients from medical x-rays. Because doses still are seldom measured or recorded, no one can presently know what share of the population's past or current average accumulated x-ray dose comes from fluoroscopy --- but the share must be a large one.
* Dr. Orhan H. Suleiman, health physicist with the FDA, estimates that "Doses from fluoroscopy can easily be reduced by orders of magnitude [well over 10-fold] when one uses currently available technologies such as a cumulative timer, pulsed fluoroscopy, more filtration, better collimation, last frame hold, dose and dose-rate display" (Suleiman 2001, p.8). Dr. Suleiman stands by that opinion, as ascertained by telephone on Feb. 27, 2003.
* Dr. Joel E. Gray, recently retired professor and medical physicist at the Mayo Clinic, warns that "Many users of hospital fluoroscopy equipment do not understand the basic principles of radiation protection and proper use of fluoroscopic equipment. [Many types of physicians besides radiologists use fluoroscopy.] It is necessary that physicians receive basic education about radiation and fluoroscopy before using the equipment. The [hospital's] Radiation Safety Committee can credential individuals who have obtained the appropriate education, and the hospital can grant privileges based on credentials. Continuing education should be required to maintain physicians' credentials and privileges" (Gray 1998, p.70).
* The Editor-in-Chief of AJR (American Journal of Roentgenology), Lee F. Rogers, M.D., has exhorted his readership --- mainly radiologists --- in an editorial (Rogers 2001, p.1): "The [fluoroscope] operator must be aware of the amount of radiation used during a procedure. Cavalier attitudes about fluoro time, failure to record the time, and conscious avoidance of steps that minimize exposure are to be condemned and discouraged. Using intermittent fluoro, pulsed fluoro, simply taking your foot off the fluoro pedal, reviewing the previous run on video rather than repeating it --- there are many ways to reduce exposure without compromising a procedure, but you must recognize the need to do so. Some physicians, unfortunately, don't. As a result, some operators and their patients are needlessly overexposed. You can learn to do with less. It is all a matter of making a conscious effort to do so. You `gotta wanta'!"
* Part 6. Claims of Threshold Doses and Hormesis (Protection against Cancer by Low-Dose Radiation): Not a Challenge to Gofman's 1999 Findings
* Will radiologists and other users of fluoroscopy want to reduce x-ray doses to their patients?
* A prime example of how radiologists receive assurance, that the harm from pre-cancer medical x-rays is negligible or non-existent, is the recent review article in the AJR by Dr. Bernard Cohen (Cohen 2002), in which Gofman 1999 and many other pertinent studies were not mentioned. Dr. Cohen concludes his review as follows (Cohen 2002, p.1141):
* "The evidence presented in this review leads to the conclusion that the linear no-threshold [LNT] theory fails badly in the low-dose region because it grossly overestimates the risk from low-level radiation. This means, for example, that the cancer risk from diagnostic radiography is much lower than is given by usual estimates, and may well be zero."
* Cohen 2002 summarizes nonhuman and human evidence suggesting that exposure to low doses of ionizing radiation (e.g., 0.2 cGy to 10 cGy) may temporarily stimulate production of DNA repair-enzymes, stimulate the apoptosis process (removal of damaged cells by cell-suicide), or stimulate the immune system. These effects, according to Dr. Cohen and some others, may create a non-linear dose-response, a threshold dose-level below which there is no cancer-risk, or a low-dose interval in which exposure is protective against cancer (the hormetic or J-shaped dose-response).
Why Cohen's Article Is No Challenge to Gofman's Findings
* The 2002 Cohen article does not represent a challenge to the 1999 Gofman findings. Why not? Because Gofman 1999 observes and reports the harms which resulted from accumulated medical x-rays anyway, despite any potential risk-reductions which might have occurred due to briefly stimulated DNA-repair, stimulated apoptosis, stimulated immune system, or hormetic "protection."
Shape of the Dose-Response
* The method of the 1999 Gofman study, described in both Attachments 1 + 3, means that the validity of its findings are completely compatible with any shape of dose-response: Linear, supra-linear, concave upward, S-shaped, or J-shaped. Whatever the true shape, the shape reflects a human biological response to ionizing radiation --- a response which will be alike in all nine separate populations in the Nine Census Divisions into which the USA is divided.
* These nine populations demonstrated very strong positive and linear nine-point dose-responses between physician-density and age-adjusted cancer and ischemic heart disease mortality-rates, by Census Divisions. But this observed linearity does not necessarily reflect the biological dose-response; it is also compatible with non-linear biological dose-responses. How? The observed linearity was discussed in Gofman 1999 (pp.93-94) as follows:
* "Physician-density is proportional to average accumulated per capita population dose from medical radiation because the more physicians there are per 100,000 population, the more radiation procedures are done per 100,000 persons. [Supporting evidence in Mettler 1987, p.134.] The increase in procedures occurs chiefly because more persons per 100,000 receive such attention --- not because the same persons get irradiated more often." [It would be unreasonable to think that physicians who realize they work in a Census Division having a high physician-density therefore decide to give each patient extra x-rays.] And:
* "In other words, the average per-patient dose is about the same in the Census Divisions with low physician-density values as in Divisions with high physician-density values, but the average per-capita dose is higher in high physician-density Census Divisions than in low physician-density Divisions because there are more patients per 100,000 population in high physician-density Divisions." And:
* "At the cellular level where x-ray-induced mutations occur, the average per-patient dose-level is likely to be very similar in all Nine Census Divisions. Therefore, the observed absence of curvature (e.g., the absence of supra-linearity) matches expectation, in dose-responses between physician-density and [age-adjusted] mortality rates."
Threshold and Hormesis: Some Evidence Omitted by Cohen
* The analysis in Gofman 1999 does not depend upon the no-threshold premise --- even though we and others have presented very strong evidence that no threshold-dose exists for low-LET ionizing radiation (for example, Gofman 1990, NRPB 1995, and other studies below --- none of which were mentioned in Cohen 2002).
* If either a threshold-dose or an hormetic dose-interval were actually to exist, it would mean that all the harms revealed by the observations in Gofman 1999 come from accumulated medical exposures each of which was higher in dose than the very low dose-range where there is any conceivable room left for either a threshold-dose or for a protective hormetic effect. A few examples suggest the upper limits of that potential dose-range:
* 7.5 cGy per exposure: The Nova Scotia fluoroscopy study (Boice 1978; not mentioned by Cohen) produced a host of excess breast cancer from serial absorbed doses to the breasts estimated at only 7.5 cGy each (BEIR 1980, p.276; Gofman 1990, Chapter 21).
* 2 cGy per exposure: The Lloyd Study (Lloyd 1988; not mentioned by Cohen), of human chromosome aberrations induced in vitro by low-let radiation, showed a linear dose response right down to 2 cGy. Additionally, the Sutherland Study (Sutherland 2000, p.106; not mentioned by Cohen) yields evidence confirming that a single track of low-LET ionizing radiation is capable of causing clustered DNA damage and double-strand breaks.
* Less than 0.6 cGy per exposure: The Scoliosis Study (Doody 2000, pp.2057-2058; not mentioned by Cohen) revealed a clear excess of breast cancer in women who received, during childhood, serial x-ray doses to their breasts estimated at less than 0.6 cGy per exposure. For example, when the average number of radiographic exams was 76.2, the average cumulative dose to the breasts was estimated at 32.6 cGy (Doody, Tab 6).
* The dose-range of 0.6 cGy to 7.5 cGy per exposure is very commonly exceeded during medical x-ray imaging. Therefore, it is conceivable (but we do not think it is the reality) that the x-ray-induced cancer and IHD revealed by the 1999 Gofman analysis results exclusively from accumulated x-ray doses each of which was higher than (say) 7.5 cGy per exposure.
* Part 7. A Scientifically Powerful Reality-Check on Risks
* Dr. Cohen's article (2002) challenges operation of the linear no-threshold (LNT) dose-response in radiation's "low-dose region" (outset of Part 6 above). As explained in Part 6 above, the harms estimated in Gofman 1999 are not tied to assuming operation of the LNT in the low-dose region. Therefore, Dr. Cohen's challenge to the LNT does not conflict with any of the observations and conclusions of Gofman 1999 (Part 4 above).
* Nonetheless, Dr. Cohen assumes (at p.1141) that "failure" of the LNT in the low-dose region (if true) would mean that "the cancer risk from diagnostic radiography is much lower than is given by usual estimates, and may well be zero."
* By contrast, the evidence in Gofman 1999 provides the first scientifically powerful reality-check on all those "usual estimates" --- and shows that they have been gross underestimates of the health risks from medical x-ray imaging procedures.
Why the "Usual Estimates" Have Been So Wrong
* Among the many reasons that the "usual estimates" have been so wrong is that there have been no reliable data for past or present decades on x-ray doses per procedure, the annual numbers of procedures, the collective accumulated doses by age and organ, and the corresponding risks per absorbed cGy of xray-dose (Gofman 2002, Part 3). Small studies of specific x-ray-exposed participants --- whose lifelong x-ray histories cannot be quantified --- have produced a wide range of risk-per-cGy values.
* Efforts to evaluate x-ray risk by substituting risk-per-cGy values from gamma-exposed populations (e.g., A-bomb survivors, nuclear workers) no doubt contribute to the underestimation of x-ray risk. Why? Not only are the participants' gamma doses poorly known and their x-ray histories rarely even considered, but also per cGy, 90 kVp x-rays are probably two-fold to four-fold more mutagenic than such gamma rays (ICRU 1986; Gofman 1990, Chapter 13, Part 4; BEIR 1990, p.218; Gofman 1996, p.275, Box 5; Gofman 1999, pp.46-48).
The Gofman Study: Not Based on "Thin-Air" Input
* One of the main scientific strengths of the 1999 Gofman Study is that it does not repeat the "usual" methods of estimating x-ray risk. Instead, it demonstrates a way to estimate the health impacts of accumulated exposure to medical x-rays without using the usual "thin air" dose estimates or the consequently unreliable estimates of risk-per-xray-cGy (Gofman 2002, Part 3).
* Part 8. "If You Care, You Measure": A Moral Imperative
* We acknowledge that no one, ourselves included, knows the average risk-per-cGy from medical x-rays. It is unknowable, thanks to failure to measure and record x-ray doses accumulated by patients over their lifespans. Is there a single patient anywhere who could find out his/her lifetime accumulated x-ray dose to any organ (breasts, lungs, heart, testes)?
* But no doubt exists that low-LET ionizing radiations, including x-rays, are a proven mutagen and carcinogen. This has been demonstrated for decades. Within irradiated cells, low-LET radiations instantly deliver biologically unnatural amounts of energy (e.g., 60 ev) within a volume estimated at only 4 nanometers in diameter (Ward 1988, p.103). This unique physical property makes them capable of producing especially complex, unrepairable, and consequential mutations (discussion in Gofman 1999, Appendix-C).
A Moral Imperative, When Using a Potentially Lethal Agent
* Certainly not all x-ray-induced mutations are consequential, but when they are carcinogenic or atherogenic, the consequence can be fatal. Therefore, a moral medical imperative exists to reduce doses from medical imaging procedures to the lowest effective levels --- not just to the lowest convenient levels. And this imperative would exist even if our estimate of benefit from dose-reduction ever turns out to be too high.
* "If you care, you measure." This axiom reflects the well-known fact in business and education that, if you are serious about achieving a goal, you establish a system to measure progress or its absence. "What you measure improves." Without seeing the improvement, or knowing of its absence, people lack guidance and motivation, and are robbed of their pride in achievement.
* It is impossible to believe that doses during fluoroscopy will be cut in half (and much more) unless the measurement of fluoroscopic x-ray dose becomes easy and automatic.
Medical Ethics and FDA's Proposed Performance Standards
* After more than ten years of study, the FDA is finally proposing performance standards which guarantee dose measuring devices on new fluoroscopic equipment, plus some additional important dose-reducing features. In our opinion, further delay in acceptance of these performance standards would be medically unethical, for each additional year of delay causes irreversible, unnecessary, and deadly harm.
# # # # #
- BEIR 1980. Committee on the Biological Effects of Ionizing Radiation (appointed by the Natl. Research Council). The Effects on Populations of Exposure to Low Levels of Ionizing Radiation (BEIR-3 Report). National Academy Press, Washington DC.
- BEIR 1990. Committee on the Biological Effects of Ionizing Radiation (Natl. Research Council). Health Effects of Exposure to Low Levels of Ionizing Radiation. ISBN 0309039959.
- Boice 1978. John D. Boice et al, "Estimation of Breast Doses and Breast Cancer Risk Associated with Repeated Fluoroscopic Chest Examinations of Women with Tuberculosis," Radiation Research 73: 373-390.
- Cohen 2002. Bernard L. Cohen, "Cancer Risk from Low-Level Radiation; Review" AJR (Amer. J. Roentgenology) 2002; 179: 1137-1143. November.
- Doody 2000. Michele Morin Doody et al, "Breast Cancer Mortality after Diagnostic Radiography: Findings from the U.S. Scoliosis Cohort Study," Spine 2000; 25: 2052-2063.
- FR 2002 = Federal Register 2002; Vol.67, No.237: 76055-76094. Dec. 10, 2002. Available online at: http://www.fda.gov/OHRMS/DOCKETS/98fr/02-30550.htm
- Gofman 1990. John W. Gofman. Radiation-Induced Cancer from Low-Dose Exposure: An Independent Analysis. ISBN 0932682898.
- Gofman 1996. John W. Gofman. Preventing Breast Cancer: The Story of a Major, Proven, Preventable Cause of This Disease. ISBN 0932682960.
- Gofman 1999. John W. Gofman. Executive Summary of the next entry, a 699-page medical monograph. The Executive Summary is 32 pages.
- Gofman 1999. John W. Gofman. Radiation from Medical Procedures in the Pathogenesis of Cancer and Ischemic Heart Disease: Dose-Response Studies with Physicians per 100,000 Population. ISBN 0932682979 (hardcover). ISBN 0932682987 (softcover). November 1999.
- Gofman 2002. John W. Gofman, "What Are the Main Critiques of the 1999 Study by Gofman, after Three Years of Peer-Review? Six Critiques of `Radiation from Medical Procedures in the Causation of Cancer and Ischemic Heart Disease (IHD)'." 5-page document. Committee for Nuclear Responsibility, San Francisco. November 2002.
- Gray 1998. Joel E. Gray, "Optimize X-Ray Systems to Minimize Radiation Dose," Diagnostic Imaging 1988; Vol.20, No.10 (Oct): 62-70.
- ICRU 1986. International Commission on Radiation Units & Measurements. The Quality Factor in Radiation Protection. ICRU Report 40. Bethesda, MD (see NCRP below).
- Lloyd 1988. David C. Lloyd et al, "Frequencies of Chromosomal Aberrations Induced in Human Blood Lymphocytes by Low Doses of Xrays," International Journal of Radiation Biology 1988; Vol.53, No. 1: 49-55.
- Mettler 1987. Fred A. Mettler et al, "Analytical Modeling of Worldwide Medical Radiation Use," Health Physics 1987; 52: 133-141.
- NCRP 1989. Natl. Council on Radiation Protection & Measurements. Exposure of the U.S. Population from Diagnostic Medical Radiation. Report 100. Bethesda, Maryland.
- NRPB 1995. Natl. Radiological Protection Board (Britain). Risk of Radiation-Induced Cancer at Low Doses and Low Dose Rates for Radiation Protection Purposes. ISBN 0859513866.
- Rogers 2001. Lee F. Rogers, "From the Editor's Notebook: Serious Business: Radiation Safety and Radiation Protection," AJR (Amer. J. Roentgenology) 2002; 177: 1. July.
- Suleiman 2001. Orhan H. Suleiman, quoted by Robin Anderson in "Radiation Dose: The Elephant in the Radiology Department," ASRT Scanner 2001; Vol.33, No.12 (Sept): 6-8. ASRT = Amer. Society of Radiologic Technologists.
- Sutherland 2000. Betsy M. Sutherland et al, "Clustered DNA Damages Induced in Isolated DNA and in Human Cells by Low Doses of Ionizing Radiation," PNAS (Proc. Natl. Academy of Sciences) 2000; Vol.97, No.1 (Jan. 4); 103-108.
- Upton 1999. Arthur C. Upton. Personal communication to John W. Gofman via E-mail, Nov. 29, 1999.
- Ward 1988. John F. Ward, "DNA Damage Produced by Ionizing Radiation in Mammalian Cells: Identities, Mechanisms of Formation, and Reparability," Progress in Nucleic Acid Research & Molecular Biology 1988; Vol.35.
and can always be found at http://www.ratical.org/radiation/CNR/JWGtoFDAP.html