Dr. Cupido - Independent Medical Opinion

Thank you for allowing me to comment on this controversial case. I will restrict my comments to the facts as dictated by my expertise in Diagnostic Medical Physics and Radiation Safety. In brief, I am currently the Head of Diagnostic Medical Physics at the Capital District Health Authority, and Professor of Radiology at Dalhousie University, Halifax, Nova Scotia. I also hold academic appointments in Radiology at the University of Ottawa, and Memorial University of Newfoundland. I am certified by the Canadian College of Physicists in Medicine for competence in physics as applied to medicine.

In preparing to submit this report, I have read all of the documentation provided to me by the Veteran's Review and Appeal Board (the Board), drawn upon my over 25 years of experience in Radiation Safety and Diagnostic Medical Physics, and read over 500 published articles relating to this matter.

Specifically, I have been asked by the Board to express my opinion on 3 specific points of clarification.

POINT A

In my opinion, is the Board's calculation of the radiation exposure received by the Appellant accurate?

RESPONSE A

The two issues which relates to this question are

• the source of the data, and
• the accuracy of the recorded data

On the first issue, the documentation provided is very clear on the recorded measurements of 940 mrem obtained during the Nevada test site exposures. I have no issue relating to the source of this information. In the early days of this appeal there was a question of the physical units associated with the physical number, but I'm satisfied that the proper unit of "mrem" has been settled on. The modern unit is mSv with 1 mrem = 0.01 Smv, hence 940 mrem = 9.4 mSv.

I am less clear on the source of the documentation which was provided relating to the Maralinga radiological surveys. The only documentation at my disposal was 2 sheets with names and occupational radiation exposure values, with no indication as to the source of this information. The letter head on the sheets were blinded to me. If the Board has more information that specifically states that this data comes from the RAF Radiological Survey Lab, then I can concur that the Appellant's exposure of 460 mrem (or 4.6 mSv) from the Maralinga experiments is credible. If this is indeed the case, then I want to point out that of the 16 Canadian soldiers who participated in these tests, 3 had recordings exceeding 2,000 mrem (or 200 mSv) which is more than 4 times the levels received by The Appellant.

The second major issue relating to this question is if the recorded data accurately reflect the amount of radiation which the Appellant received. Given the spatial variations in radiation incident on the body, and the fact that typically only one part of the body is monitored by radiation monitors, occupational radiation exposure measurements at the very best of times is merely an indication of the amount of radiation which a person received. Also, in the 1950's film, with its many inherent physical shortcomings as radiation monitors, was used for personnel exposure monitoring. Given these facts, it is not unreasonable to assume that the actual exposure that the Appellant could have received varied by as much as ± 50%, meaning that his actual total recorded occupational exposure of 14 mSv, (Nevada = 9.4 mSv, and Maralinga = 4.6 mSv) could possibly have been as high as 21 mSv. Traditionally, radiation safety calculations are made based on the higher estimate to err on the conservative side. As such I have chosen the amount of 21 mSv as the basis for discussing the other issues which I've been asked to address. However, to put this number into immediate perspective, it is similar to the annual maximum permissible dose (20 mSv) for radiation workers in Canada.

In summary, based on the exposure records provided, the Board's estimate of the amount of radiation received is probably correct, but it most likely does not represent the actual amount of radiation which the Appellant received. For purposes of further discussion, I have made allowances that it could possibly have been up to 50% higher than what the records indicate.

Top of Page

POINTB

Does the attached chapter from Textbook of Cancer Epidemiology represent the consensus with regard to radiation exposure and cancer of the oral cavity?

RESPONSE B

There is no direct answer to this question. I could not identify any published literature which answers this question as bluntly as the authors of the textbook in question. I also did not identify any publication which categorically states that external exposure to external radiation does play a role in causing oropharyngeal cancers.

To address this issue I researched the cancer incidence and mortality rates of cancers of the oral cavity in populations who routinely receive, or who have received radiation levels similar to my elevated estimate of the Appellant's exposure. These populations include:

(1) Patients who undergo radiological Computerized Tomography (CT) of the ENT (ear-nose-throat) area. CT is an x-ray imaging modality which was first introduced in 1972. The typical radiation absorbed dose from a head and neck examination ranges from 20-50 mSv, depending on the technical factors used to acquire the images. My elevated estimate of the Appellant's total exposure is therefore similar to the lower limit of routine head and neck CT examinations, which are typically delivered over a period of seconds.

In my medical district which serves as the referral center in Atlantic Canada for ENT radiology, approximately 50 ENT CT examinations are performed weekly. I estimate that more than 10,000 of these examinations are performed in Canada annually. Given the fact that CT scanning was introduced as an x-ray imaging modality more than 30 years ago, which is more than the latent period for most radiation induced cancers, an increase in ENT cancers should have been observed over the last 10 - 30 years if these levels of radiation indeed caused cancers of the oral cavity. The cancer statistics of the National Cancer Institute of Canada for 1991-1998 show age-adjusted incidence and age-adjusted mortality rates for oral and laryngeal cancers in males to have decreased by about 3.5% over the similar previous period. This leads me to conclude that there is not a causal-effect relationship between radiation in the 20 - 50 mSv range and oropharyngeal cancers.

(2) The second population which I investigated is workers who are occupationally exposed to radiation. Until very recently, the annual maximum permissible dose for radiation workers was 50 mSv. This has subsequently been reduced to 20 mSv in Canada, following a re-evaluation of the data from the Japanese atomic bomb survivors. There is no question that there were higher incidences of certain cancers in early radiation workers. In a very recent study (September, 2004) results from a meta-analysis found an increase rate of leukemia, breast cancer and melanoma in workers before 1950, when the maximum permissible doses were 30,000 mSv, then reduced to 300 mSv. Another study demonstrated an increase of 5% in all cancers per 10 mSv cumulative dose for workers at the Oak Ridge National nuclear laboratory who were 45 years or older at the time when they were exposed to the radiation. Of over 20 studies which I consulted on this topic alone there was no study which demonstrated any increased cancer rates since the maximum permissible doses have been reduced to 50 mSv. Not one study (even in the early, pre-1950 days) reported a link between increased oropharyngeal cancers in radiation workers exposed to external radiation.

Another group of workers who is not classified as radiation workers, but who receive more radiation while performing their routine duties than most radiation workers, is airline cockpit crews. In a study of German cockpit crew (1960-1997) it was determined that some crew members receive as much as 80 mSv effective dose, mainly from cosmic radiation. The study reported no adverse health effects in German airline cockpit crews.

(3) The third population which I investigated was the atomic survivors from the Japanese bombings. In particular, I focused on The Life Span study by the Radiation Effects Research Foundation, which studies survivors who received an estimated effective dose higher than 5 mSv (the average effective dose of the survivors was estimated to be 200 mSv). Of 4,687 cancer deaths which occurred from cancers other than leukemia, it is estimated that 7% (339) of these deaths was caused by radiation. These cancers were mainly lung, stomach, breast, colorectal, and liver cancers. There is no mention of excess cancers of the oropharynx in this population who, on average, received an estimated 200 mSv effective radiation dose, i.e. 10 times my elevated estimate of the Appellant's exposure.

(4) The fourth population which I investigated is people who live in very high natural background radiation areas. The average global average level of background radiation is 2.4 mSv (typically 2.5 to 3.0 mSv in North America) per year. However, there are several abnormally high natural background radiation areas in the world, such as the Yanjiang region of China where the average level is 6.4 mSv per year. In the northern Iranian city Ramsar, the annual level is up to 260 mSv, I00 times the global average. Study after study looking for increased incidences of cancer, chromosomal abnormalities and hematological alterations in the inhabitants have not demonstrated any such effects.

In summary, there is no evidence to suggest that exposure to low levels of external radiation (<50 mSv) cause oropharyngeal cancers, and in that sense the statement by the authors does represent a concensus. The authors are the only ones who state this fact so bluntly, but there is no evidence to suggest the contrary. However, for completion sake, one should include radiation exposure from internal sources caused by radioactive elements. I identified one study which reported a positive radiation dose-effect relationship for cancers of the upper aerodigestive tract in nuclear workers who were chronically exposed to internal radiation from uranium and mixed-fission products.

Top of Page

POINTC

If a CT scan of the abdomen (for example) with a dose of 10 mSv were to represent a 1 in 2,000 risk of cancer, what would be the corresponding risk of cancer of the oral cavity?

RESPONSE C

To estimate cancer risks at relatively low doses of radiation is in itself a risky business. Cancer risks at low radiation doses is extrapolated from known effects at high doses, and the conservative model by which this is often done, is based on the linear, no-threshold dose response model. Many scientists do not agree with this model. An oft quoted percentage, based on this model, is 5% theoretical risk of cancer per Sv, the basis for the 1 in 2,000 risk of cancer from 1 abdominal CT examination. If this was indeed a true risk, we should be observing a tremendous increase in abdominal cancers in Canada, because more than a million abdominal CT scans are performed in Canada annually. Be that as it may, the orophaynx is rather insensitive to radiation, and hence my best possible answer is that the risk of developing cancer at this site from being exposed to 21 mSv (which is similar to the radiation dose for an abdominal CT) to the whole body relates to a much smaller risk of cancer induction, which is impossible to estimate.

In summary, the risk from developing oropharyngeal cancer from a relatively low (< 50 mSv) amount of radiation is negligible.

Top of Page

EXECUTIVE SUMMARY

Based on the scientific facts at my disposal I conclude that there is no proven causal-effect relationship risk between 14 mSv (1,400 mrem) of external radiation and oropharyngeal cancer. The risk of developing such cancers pales in comparison to the known causal-effect relationships with other factors such as diet, alcohol, and smoking.