Winters-TH, Franza-JR, Radioactivity in Cigarette Smoke, New England Journal of Medicine, 1982; 306(6): 364-365To the Editor: During the 17 years since the Surgeon General's first report on smoking, intense research activity has been focused on the carcinogenic potential of the tar component of cigarette smoke. Only one definite chemical carcinogen -- benzopyrene -- has been found. Conspicuous because of its absence is research into the role of the radioactive component of cigarette smoke.
The alpha emitters polonium-210 and lead-210 are highly concentrated on tobacco trichomes and insoluble particles in cigarette smoke (1). The major source of the polonium is phosphate fertilizer, which is used in growing tobacco. The trichomes of the leaves concentrate the polonium, which persists when tobacco is dried and processed.
Levels of Po-210 were measured in cigarette smoke by Radford and Hunt (2) and in the bronchial epithelium of smokers and nonsmokers by Little et al. (3) After inhalation, ciliary action causes the insoluble radioactive particles to accumulate at the bifurcation of segmental bronchi, a common site of origin of bronchogenic carcinomas.
In a person smoking 1 1/2 packs of cigarettes per day, the radiation dose to the bronchial epithelium in areas of bifurcation is 8000 mrem per year -- the equivalent of the dose to the skin from 300 x-ray films of the chest per year. This figure is comparable to total-body exposure to natural background radiation containing 80 mrem per year in someone living in the Boston area.
It is a common practive to assume that the exposure received from a radiation source is distributed throughout a tissue. In this way, a high level of exposure in a localized region -- e.g. bronchial epithelium -- is averaged out over the entire tissue mass, suggesting a low level of exposure. However, alpha particles have a range of only 40 um in the body. A cell nucleus of 5 to 6 um that is traversed by a single alpha particle receives a dose of 1000 rems. Thus, although the total tissue dose might be considered negligible, cells close to an alpha source receive high doses. The Po-210 alpha activity of cigarette smoke may be a very effective carcinogen if a multiple mutation mechanism is involved.
Radford and Hunt have determined that 75 per cent of the alpha activity of cigarette smoke enters the ambient air and is unabsorbed by the smoker, (2) making it available for deposit in the lungs of others. Little et al. have measured levels of Po-210 in the lungs of nonsmokers that may not be accounted for on the basis of natural exposure to this isotope.
The detrimental effects of tobacco smoke have been considerably underestimated, making it less likely that chemical carcinogens alone are responsible for the observed incidence of tobacco-related carcinoma. Alpha emitters in cigarette smoke result in appreciable radiation exposure to the bronchial epithelium of smokers and probably secondhand smokers. Alpha radiation is a possible etio- logic factor in tobacco-related carcinoma, and it deserves further study.
Thomas H. Winters, M.D.
Joseph R. Di Franza, M.D.
University of Massachusetts Medical Center
Worcester, Ma 01605
To the Editor: In a letter in the Feb 11 issue, Winters and DiFranza (1) correctly point out that alpha radiation from polonium-210 is a possible causal factor in tobacco-related carcinoma, but they incorrectly state that "inhaled" Po210 is a factor and that research on this important possibility has been neglected. I will briefly review recent pertinent research.
Radford and Hunt (2) first suggested that alpha radiation from Po210 in cigarette smoke may be important in the genesis of bronchial cancer. Little et al. (3) found surprisingly high concentrations of Po210 at single bronchial bifurcations in seven of 37 cigarette smokers. Holtzman and others (4 - 6) raised doubts about the validity of these observations because inhaled volatile Po210 is soluble and rapidly cleared. Subsequently, I determined (7) that lead-210 (a beta-emitting precursor of Po210) is highly concentrated in tobacco trichomes and that trichome combustion in burning cigarettes produces insoluble, Pb210-enriched particles in mainstream smoke. Thus, the high concentrations of Po210 observed at segmental bifurcations (4 - 6) can be explained by the persistence of insoluble, Pb210-enriched particles deposited at bifurcations and by the ingrowth of Po210 in these particles. (7,8) Radford and Martell (9) confirmed that the excess Po210 in the bronchial epithelium of smokers is accomplished by a larger excess of Pb210.
Fleischer and Parungo (10) provided experimental evidence indicating that radon decay products and Pb210 are concentrated on trichome tips. Mechanisms of accumulation of Pb210 on tobacco trichomes are discussed by Martell and Poet. (11)
Two recent studies (12,13) indicate that alpha radiation from inhaled indoor radon progeny may explain the incidence of lung cancer in nonsmokers. Martell and Sweder (14) report that indoor radon decay products that pass from the room air through burning cigarettes into mainstream smoke are present in large, insoluble smoke particles that are selectively deposited at bifurcations. Thus, the smoker receives alpha radiation at bronchial bifurcations from these three sources: from indoor radon progeny inhaled between cigarettes, from Po214 in mainstream smoke particles, and from Po210 that grows into Pb210 enriched particles that persist at bifurcations. I estimate that the cumulative alpha dose at the bifurcations of smokers who die of lung cancer is about 80rad (1600rem) -- a dose sufficient to induce malignant transformations by alpha interactions with basal cells.
Edward A Martell, Ph.D.
National Center for Atmospheric Research
Boulder, CO 80307
To the Editor: The presence of Po210 and Pb210 in cigarette smoke may help to explain a paradox found in smokers of low-tar, low-nicotine cigarettes.
Hammond et al. (1) noted that the number of deaths from lung cancer was greater in subjects who smoked 20 to 39 low-tar, low-nicotine cigarettes a day than in those who smoked one to 19 high-tar, high-nicotine cigarettes a day. Thus, the number of cigarettes smoked may be more important than their tar and nicotine content.
Two features of low-tar low-nicotine cigarettes that help to reduce the amounts of tar in inhaled smoke may have little effect or adverse effects on the amounts of Po210 and Pb210 in inhaled smoke. In the first place, the use of higher porosity paper and perforated filters may enhance the completeness of combustion. Although this may decrease the tar and nicotine content in inhaled smoke, it may increase the pyrolysis of trichomes, resulting in smoke particles with higher specific activities of Pb210. Secondly, cigarette filters have been shown to have no noticeable protective effect against Po210 inhalation. (2) If Po210 and Pb210 contribute to tobacco related cancer, then the number of cigarettes smoked may be more important than the tar or nicotine content.
Although intensive effort has been successful in producing low-tar, low-nicotine cigarettes, perhaps future research should be aimed toward the development of low Po210, low Pb210 cigarettes.
Jeffrey I. Cohen M.D.
Duke University Medical Center
Durham, NC 27710
To the Editor: Contrary to the contention of Winters and DiFranza that research into the carcinogenic potential of the radioactive component of cigarette smoke is conspicuous by its absence, we and others have studied and reported on this risk since the theory was first proposed by Radford and Hunt in 1964. (1) Within five years of the initial report that the radioactive alpha emitter Po210 was present in mainstream smoke and in samples of bronchial epithelium from cigarette smokers, results from over two dozen related studies were published. The source of the Po210 and Pb210 (The beta emitter Pb210 is the long lived precursor that supports the Po210) was investigated, (2) the contents of these nuclides in various tobaccos documented, (3) the fraction transferred to the mainstream or sidestream smoke (or both) determined, (4) and the concentration in the whole lungs of smokers and nonsmokers measured. (5)
Measurements made with cigarette smoke condensate demonstrate that although radium and thorium are also present in cigarette smoke, 99% of the alpha activity is from Po210. (6) Measurements of the whole lungs of smokers and exsmokers show that the inhaled Po210 is retained in the lower lung. (7)
A relatively new detection technique using nuclear-track-etch film has allowed us to determine the amount and microdistribution of alpha activity on the bronchial mucosa in fresh autopsy specemins. (8) We examined about one-fourth of the upper respiratory tract in each of seven persons (Three smokers, two exsmokers, and two nonsmokers). A few areas of slightly elevated alpha activity were found in each of the bronchial trees examined except that of one young smoker, in which efficient bronchial clearance would be expected. The average dose rate to the basal cells of the bronchial epithelium from alpha activity in these seven persons ranged from 2.0 to 40mrem per year. For comparison, the natural background dose from inhaled radon-daughter alpha activity is about 2000mrem per year. One area of a few square millimeters, containing markedly elevated activity, was found in the bronchii of an older smoker. This area could deliver an annual dose of about 20,000mrem, comparable to the results originally reported by Bradford and Hunt. This activity can lead to a lifetime dose similar to the alpha dose that appears to yield an elevated risk of lung cancer in underground miners. However, the total dose cannot be calculated, since the residence time of such an alpha emitting spot on the bronchial tree is not known.
The importance of proper assessment of the risk to cigarette smokers from radionuclides in the smoke cannot be overstated. In view of the present knowledge, it is improbable that a single area of a few square millimeters of high alpha activity in the bronchial tree is important. Nonetheless, Po210 is the only component in cigarette smoke tar that has produced cancers by itself in laboratory animals as a result of inhalation exposure. (9)
We firmly believe that the role of alpha radiation in tobacco related carcinogenesis deserves further study. The techniques to define its role in this disease are now available.
Beverly S. Cohen, Ph.D.
Naomi H. Harley, Ph.D.
New York University School of Medicine
New York, NY 10016
To the Editor: The letter of Winters and DiFranza has renewed the earlier suggestion that the radioisotope Po210 may have an important role in the induction of lung cancer in smokers. In particular, it is claimed that the radionuclide may be deposited very inhomogeneously in the bronchial epithelium, in the form of a limited number of relatively "hot" particles, and that such hot particles may be much more effective carcinogenically than the same amount of radioactivity would be if it were more uniformly distributed. The basis of both these claims must be questioned.
Evidence on the question of the carcinogenicity of hot particles has been reviewed by the International Commission on Radiological Protection, (1) which found the actual situation to be just the reverse of that suggested by the correspondents. The evidence cited for the actual formation of hot particles (2) comes from a study of the Po210 in a series of several very small samples of bronchial epithelium (usually less than 25mg) collected from smokers' lungs. In these measurements, the activities in individual samples were so low that for a proportion at least, only about 20 counts were recorded in a counting period of three to seven days against a background of 40 counts. Proper analysis of the statistical validity of these observations was not given by the original authors and is not possible from their reported data. Contrary evidence, not cited by the correspondents, is provided by a somewhat earlier paper (3) that reported the results of auto radiographic examination of excised segments of bronchial epithelium; this study found no evidence of surface concentrations of alpha activity of more than 0.01pCi per square centimeter, corresponding to a mean dose rate of about 10mrem per year. Finally, the correspondents' suggestion that the "major source of the polonium is phosphate fertilizer" is not substantiated and is at variance with published data (3,4) indicating that it originates as atmospheric fallout of the decay products of natural radon-222.
C.R. Hill, M.D.
Institute of Cancer Research
Royal Marsden Hospital
Sutton, Surrey SM2 5PX,
To the Editor: The Surgeon General's recent denunciation of tobacco smoking and the American Cancer Society's pessimistic prognosis that lung cancer will be the number one cause of death from cancer in women by 1985 (1) provide timely emphasis on the recent NEJM letter on radioactive alpha emitters in tobacco smoke. Some of the further study encouraged by Winters and DiFranza has in fact been performed, yielding results far more foreboding than expected.
In two separate studies, Little et al. (2,3) have induced respiratory tumors in hamsters by intratracheal instillation of Po210 in various amounts down to less than one-fifth that inhaled by a heavy cigarette smoker (one who consumes two packs a day) during 25 years. The incidence of tumors at the lowest dose was 13%, including borderline carcinomas, and was 11% for frankly malignant tumors.
Contrary to the expected results of most radiobiologists, dose reduction did not result in either a constant dose-response ratio (the linear response hypothesis) or a larger dose-response ratio (The threshold or sigmoid hypothesis) but instead produced a marked decrease in the dose-response ratio. In one study, a reduction in activity from 0.700microCi of Po210 instilled to 0.00375microCi of Po210 instilled -- about a two hundred-fold decrease -- resulted in a decrease in the incidence of tumors from 61% to 13% (including borderline cases) -- only a fourfold decrease.
This decrease in the dose-response ratio with decreasing dose has also been observed in other studies of the effects of low dose alpha radiation and other radiation particles with high linear energy transfer (LET). In a study of osteosarcoma induction by alpha radiation, Muller et al. (4) had over a 100-fold decrease in the dose-response ratio from their highest dose (1500rad) to their lowest dose (15rad). For neutron radiation, Rossi et al. (5) found similar results, with leukemia induction having the smallest dose-response ratio in the lowest dose in survivors of the atomic bomb. Similarly, Hall et al. (6) found that both dose protraction and dose reduction for neutron radiation increased the cell-lethality-dose ratio of hamster cells in vitro.
The importance of these results with low dose irradiation by high LET particles should not be overlooked. Doses in the range of several thousand to 10^5 rad have generally been necessary for the experimental induction of lung cancer by beta or gamma radiation (with low LET), (7,8) as compared with the studies by Little et al., in which the lowest dose of 15rad (0.00375microCi in the lung volume for the lifetime of the hamsters) induced cancer at an incidence of about 13%.
Presumably, the high density of ionization along the track of alpha radiation (about one ion pair for every 2 Angstrom traveled) and other high-LET radiation is the prime factor causing Po210 to be an extremely efficient carcinogen.
Clearly, further work is warranted in this area, but we should not hesitate to disseminate the information already at hand -- that the alpha-radiation exposure to the lungs of tobacco smokers is extremely important.
Walter L. Wagner, B.A.
San Francisco, CA 94121
To the editor: The letter by Winters and DiFranza rivets much needed attention on the earlier finding of Radford and Hunt, (1) which is crucial to an understanding of the pathogenesis of smoking diseases. (2,3)
Although Winters and DiFranza tellingly describe the mechanisms by which Po210 on insoluble particles in cigarette smoke causes lung cancer, they neglect the even more important matter of how Po210 and other mutagens from tobacco smoke cause malignant neoplasms, degenerative cardiovascular diseases, and other diseases throughout the body of smokers (Table 1).
|Effects of Smoking on Tissues Directly and Indirectly Exposed to Radiation in Current Cigarette Smokers*|
|Cause of Death||Number of Deaths||Observed/Expected (ratio)|
|Of directly exposed tissue||3,061||296||10.34|
|-||Of buccal cavity||110||26||4.23|
|-||Of lung and bronchus||2,609||231||11.29|
|Of indirectly exposed tissue||4,547||3,292||1.38|
|-||Of liver and biliary passages||176||75||2.35|
|-||All other cancers||597||407||1.47|
|All cardiovascular diseases||21,413||13,572||1.58|
|-||Coronary heart disease||13,845||8,787||1.58|
|-||All other cardiovascular||6,624||4,605||1.44|
|Ulcer of stomach, duodenum or jejenum||289||93||3.10|
|Cirrhosis of liver||404||150||2.69|
|*Data adapted from Rogot and Murray. (4)|
The proof of circulating mutagens from smoking is that Po210 and other mutagens can be recovered not only from tobacco smoke and bronchial mucosa, but also from the blood and urine of smokers. (1,7)
R.T. Ravenholt M.D., M.P.H.
Centers For Disease Control
Rockville, MD 20857
To the editor: We concur with Drs. Winters and DiFranza that the scientific and medical community as well as public health officials should be more concerned with the detrimental effects of cigarette smoking. Reviews on the carcinogenic effect of cigarette smoke are made available to United States physicians at regular intervals through the Surgeon General's reports entitled Smoking and Health. (1) From these reports it is clear that benzo(a)pyrene is by far not the only carcinogen identified in cigarette smoke. Benzo(a)pyrene serves merely as an indicator for the wide spectrum of carcinogenic polycyclic hydrocarbons, all of which are pyro synthesized by the same mechanism during smoking. Aside from these hydrocarbons, cigarette smoke contains other carcinogens such as aza-arenes, aromatic amines (including beta-napthylamine), nickel, volatile nitrosamines, and especially tobacco-specific N-nitrosamines. (1-3) The N-nitrsamine compounds are formed by nitrosation of nicotine and other alkaloids; their concentrations in tobacco and smoke exceed those of nitrosamines found in other consumer products by at least several hundred fold. These nitrosamines are probably formed from nicotine in vivo. (2,3) Above all, one needs to consider that the carcinogenic potential of tobacco is a composite effect of tumor initiators, tumor promoters, or co-carcinogens, and organ-specific carcinogens. (1,2)
Dietrich Hoffmann, Ph.D.
Ernst L. Wynder, M.D.
American Health Foundation
New York, NY 10017
To the editor: We thank Dr. Martell and Drs. Cohen and Harley for their reviews of the literature. Judging by the response to our original letter, research into the radioactive component has been in progress since the early 1960's, but the existence of this research is largely unknown outside a small segment of the scientific community. We were gratified to receive hundreds of phone calls from smokers who quit on learning about the alpha radiation in cigarette smoke.
Hill examined the lungs of only two smokers old enough to have metaplastic lesions. In addition, he analyzed whole bronchial specemins weighing 5g to 15g, of which only 2% by weight was epithelium. His result of 0.007 pCi per gram of tissue is in reasonable agreement with Little's result of 0.012pCi per gram of whole bronchus and thus does not disprove the existence of hot spots. In addition, the accumulation of Pb210 on tobacco leaves is from natural and unnatural radon-222 decay products and from phosphate fertilizers.
We thank Dr. Wagner for pointing out that alpha radiation now appears to be 1000 times more carcinogenic than gamma radiation. Standard practice reguards alpha radiation as only 10 to 20 times as carcinogenic as gamma radiation.
The growing list of malignant diseases associated with smoking, presented by Dr. Ravenholt, begs for causal explanation. Smokers have higher levels of Po210 in the lungs, bone blood and urine. (1-3) Higher levels of Po210 have been consistently found in smokers in the liver, kidney, spleen, pancreas, and gonads. (4,5) A study with an adequate number of subjects would probably demonstrate a significant difference. The Po210 must be strongly considered as a cause of these cancers.
Drs. Cohen and Harley report finding one "hot spot" on studying the alpha activity of alpha Po210 in tracheal autopsy specemins of seven people, three of whom were smokers. (6) This supports Little and his colleagues' previous findings of :hot spots" in 7 out of 37 smokers.
We thank Drs. Hoffmann and Wynder for correcting us about the variety of chemical carcinogens present in cigarette smoke. It is possible that chemicals and Po210 act as cocarcinogens in the following manner. Chemical and possibly physical agents create metaplastic nonciliated epithilial lesions. Auerbach demonstrated such lesions in 100% of heavy smokers. (7) The Po210 present on insoluble particles gains entrance to epithelial cells in such non-ciliated areas of mucous stagnation. Ingrowth of Po210 from the decay of Pb210 results in high doses of alpha radiation to already metaplastic cells. (8) Continued smoking ensures a steady delivery of Pb210 to these stagnant sites. Little and his co-workers have demonstrated synergism between benzo(a)pyrene and Po210 in an animal model. (9)
In view of the potential role of alpha radiation in a variety of tobacco related neoplasias, we believe that this area deserves more intense research. We find it surprising that the National Cancer Institute, with an annual budget of $500 million, has no active grants on alpha radiation as a cause of lung cancer (National Cancer Institute: personal communication).
We have found when educating smokers that more are encouraged to quit as they learn of the presence of radiation in cigarette smoke.
Joseph R. DiFranza, M.D.
Thomas H. Winters, M.D.
University of Massachusetts Medical Center
Worcester, MA 01605