Experimental studies of radiation carcinogenesis in the skin: a review.

INT . J . RADIAT . BIOL .,

1990,

VOL .

57,

NO .

4, 797-808

Int J Radiat Biol Downloaded from informahealthcare.com by Universitaets- und Landesbibliothek Duesseldorf on 06/17/13 For personal use only.

Experimental studies of radiation carcinogenesis in the skin : a review J . E . COGGLE and J . P . WILLIAMS Department of Radiation Biology, Medical College of St Bartholomew's Hospital, Charterhouse Square, London EC 1M6BQ, U .K . Skin has been widely used in radiation carcinogenesis studies because of the accessibility and visibility of its tumours . Both rat and mouse models have proved to be sensitive, reproducible systems to study the dose and time response of cancer induction following different modes and qualities of radiation exposure . This paper discusses the variation in the shape of the low-LET dose responses from purely linear with no threshold to the highly quadratic curves with significant thresholds, although a linear response is more consistently reported following high-LET radiations . Some dose-response curves show no tendency to turnover at high doses, others show a declining incidence of skin cancer at the highest doses . Protraction or fractionation of the dose reduces the carcinogenic effect in rat skin, whilst the reported dose rate studies in mice are equivocal regarding any sparing effect . Mouse skin cancer studies, in particular, have empirically refuted the `hot particle hypothesis' . The extensive studies of Albert and Burns highlight hair follicle damage at 300 ,urn depth as critical in the development of the majority of rat skin tumours . In contrast, mouse studies report a wide variety of cell types as the putative 'cells at risk' in the skin from the spectrum of epidermal and dermal tumours which are induced, and which have been found to be amenable to classification using human pathological categories . Despite these interspecies differences, it is shown that all of the experimental data for radiogenic skin cancer, when expressed per unit area of skin, fall on a relatively narrow and well defined response curve, which is approximately two orders of magnitude more sensitive than the human skin cancer dose response . 1 . Introduction The risk of cancer induction is the most important somatic effect of low doses of ionizing radiation . These risks have recently been reviewed in the UNSCEAR (1988) report, which based its conclusions almost solely on epidemiological studies of the Japanese survivors of the US atomic bombs . The resultant risk assessments, especially at low doses, would be less fragile if we had a better understanding of the mechanisms of radiogenic cancer induction . Such an understanding must rest on the basic theories of radiation and cancer biology, and these theories need to be tested experimentally . It is not, of course, suggested that experimental animal carcinogenesis data can provide quantitative risks applicable to man, since the uncertainties involved in any such extrapolation from animal data to humans surely outweigh the uncertainties of the human data . It is suggested, however, that animal models have provided, and will continue to provide, valuable, semi-quantitative generalizations to such perennial radiological protection problems as the role of dose protraction, radiation quality, partial organ and body exposure and to *Presented at the European Late Effects Project Group (EULEP) at the 22nd Annual Meeting of the European Society for Radiation Biology, 12-16 September 1989, Brussels . 0020-7616/90 $3 .00 © 1990 Taylor & Francis Ltd


798

J . E. Coggle and J . P . Williams

situations where the dose is highly non-uniform as is the case of internal emitters (Coggle 1982, Fry 1981) . Considerations such as these have prompted many of the investigations into the carcinogenic effects of radiation in skin . Skin has been widely used in chemical, viral and physical carcinogenic studies because of its accessibility and the visibility of its tumours . There are a wide variety of cell types in skin that are the putative `cells at risk' for the spectrum of carcinomas and sarcomas that are induced . Both rat and mouse skin have proved to be sensitive and reproducible systems to study the dose response and time response characteristics of cancer induction following a variety of modes and qualities of radiation exposure . 2.

Cancer studies in rat skin Although there are a number of early references to experimental skin cancer induction following irradiation, undoubtedly the most extensive and systematic studies in the rat are those of R . E . Albert and F . J . Burns in the 1960s and 1970s, and these workers have summarized and reviewed their own data (Burns and Albert 1986a,b) . They used albino rats, Sprague-Dawley and CD strains, and exposed them to single and fractionated doses of soft X-rays, 91 Y #-rays, monoenergetic electrons and a-particles . The experiments were designed to evaluate the carcinogenic effects of various surface and depth dose distributions and the association with acute and chronic radiation damage of the skin . The spontaneous incidence of skin tumours in the control rats was negligible (Albert et al . 1961) ; in the irradiated rats, tumours began to appear from about 10 weeks post-irradiation, scored when visible at 1 mm diameter, and continued to appear at a fluctuating though accelerating rate throughout the observed period of up to 100 weeks following irradiation . Of the induced tumours, 5 per cent were of connective tissue origin ; the other tumours occurred with the following overall relative frequencies : 35 per cent keratosebaceous tumours, 30 per cent squamous carcinomas, 20 per cent basal cell carcinomas and 10 per cent sebaceous tumours . The growth rate and the time of onset of the keratosebaceous and sebaceous (adnexal) tumours, which are the majority, was shown to be related to the magnitude of the residual hair follicle injury ; the number of atrophic hair follicles remained in a relatively stable ratio to the adhexal tumour yield despite the wide-ranging experimental conditions . There were also periodical peaks of these tumours' appearances at approximately 20, 40 and 60 weeks post-irradiation, which pointed to hair cycle periodicity and to the possible hair follicular origin of many of these rat skin tumours (Albert et al . 1969) . Many of the rats produced multiple tumours and invariably these each had a moat of good skin around them so that the data were consistent with the hypothesis that `existing skin tumours do not affect the formation of new tumours' (Albert et al. 1961) . New tumours continued to occur after irradiation for as long as the animals were observed, although the distribution of appearance times seemed not to depend on the growth rate of the tumours, i .e . there was no difference in the growth kinetics of early- and late-appearing tumours . Similarly the spectrum of histological types was similar in early- and late-appearing tumours, and it did not change significantly with the type of radiation, the geometrical distribution of dose or the temporal pattern of dose delivery . The dose-response curve (figure 1) is from a number of electron experiments and is seen to be highly curvilinear with tumours of all types, increasing abruptly at


Radiation carcinogenesis in the skin

799

Dose (Gy) Figure 1 . The yield of skin tumours as a function of dose at 80 weeks after exposure of rat skin to single doses of electrons (LET=0 . 34keV/µm) ( •) and to argon ions

(LET= 125 keV/µm) (0) . (Reproduced from Burns and Albert 1986a ; courtesy of the authors and Elsevier publishers .) 15 Gy, peaking at 30 Gy and declining at higher doses (Burns and Albert 1986a) . The best fitted line to all the data is given by : fD=AD+BD 2 , where A=O, i .e . it is purely quadratic . The upturn of the tumour dose response corresponds to doses that produce mild-moderate acute skin damage . There was a shift in the distribution of tumour types with dose, with the follicle and sebaceous tumours predominant at intermediate doses, whilst at high, i .e . pilocidal doses (> 40 Gy), the frequency of these adnexal tumours declined and the less frequent squamous carcinomas became more common . The latency time of the tumours was not markedly dose-dependent in the rat . There is a significant dearth of human cancer data following high-LET radiation and so experimental data are a valuable supplementary source of RBE values . Over 20 years ago, Burns et al. (1968) used cyclotron accelerated a-particles to study this problem, and more recently Burns and Albert (1980) used argon ions with an LET of 125 keV/µm . The dose response for argon is shown in figure 1 and is approximately linear up to 9 Gy and the tumour yield declines above 9 Gy where there is appreciable gross tissue damage . The RBE value for argon ions compared with electrons was 2 . 5 at peak tumour yield and increased sharply at lower doses reaching ^__ 50 at a dose of 0. 75 Gy of argon ions . Fractionation of the radiation dose was found to reduce the incidence of skin cancers (Burns and Vanderlaan 1977) . Split doses of Van der Graaf accelerated 0 . 7 MeV electrons were used to study the recovery and repair of radiation carcinogenesis and a repair half-time of 3 h was reported . Further multi-fraction experiments involving lifetime radiation doses between 0. 75 and 1 . 5 Gy per week showed that the split dose repair seen for two doses also operated for up to 60 fractions over a lifetime (Burns and Albert 1986a) .

800

J . E . Coggle and J . P. Williams


Two of the most interesting areas of study by Albert and Burns concerned the geometrical distribution of dose and the role of hair follicle damage in the production of radiation-induced rat skin cancers . Exposure doses which are of concern in radiation protection are rarely received uniformly, and so it is important to compare effects of non-uniform exposure with those of uniform tissue exposure . Using a variety of grid and sieve patterns, rat skin was irradiated non-uniformly with low-energy X-rays and electrons (Albert et al . 1967a) . The results showed that radiation carcinogenesis in rat skin, when expressed per unit area of skin, is not always proportional to the number of cells irradiated . Also the tumour yields after grid and sieve non-uniform exposure were markedly delayed compared with uniform exposure . However, when the tissue between the heavily irradiated areas was given a low dose, the delayed onset of the tumours was eliminated . This was interpreted to mean that the proximity of unirradiated cells somehow affected the rate of progression of the irradiated cells to cancer . This type of sparing effect of non-uniform exposure has also been investigated in the mouse and will be discussed later . Many of the rat skin experiments were designed to pinpoint the position of the putative cells at risk that eventually produced the characteristic adnexal tumours of the two strains of rat . Doses of electrons having maximum ranges up to 1 . 65 mm in skin showed that if the penetration were less than 180 µm, no tumours at all were observed (Albert et al. 1967b) . Electrons with ranges greater than 180 Jim induced tumours of all histological types and typical results are shown in figure 2 ; there was an initial increase in tumours with dose, followed by a peak and then a decline at high doses . There are obvious differences between the curves in figure 2a where the tumour yields are plotted against surface dose, but since the target cell depth is unknown, the use of surface dose is arbitrary and in figure 2b the data were replotted by Albert and Burns for the doses at a depth of 0 . 27 mm and this reconciles the data to a single curve . This depth of 0 . 3 mm corresponds closely to the bottom of the resting (telogen) hair follicle and fits with their work on the congruity between atrophic hair follicles and tumour yield (Albert et al. 1967b) . 6 -

(a)

Figure 2 . (a) The yield of skin tumours per rat as a function of surface dose maximum penetration depths of electrons (A = 0 . 36 mm; 0 =1 . 40 mm; 0 =1 . 65 mm penetration) . (b) . The tumour incidence as electron dose at 0-27 mm . (Reproduced from Albert and Burns, 1967b; authors and Academic Press .)

for different ∎ = 0 . 75 mm; a function of courtesy of the



801

In later experiments the position of the carcinogenically sensitive cells in the anagen (growing) phase were determined by again varying the energy of the electrons (Burns et al . 1976) . The depth most closely correlated with tumour yield was ^_, 0 . 4 mm, i .e . not significantly different from that for telogen skin of 0 . 3 mm, despite the fact that the depth of the actively growing follicles is ^_-1 .0 mm . Comparable curves for atrophic follicles were best resolved at a depth of 0 .8 mm, i .e . close to the depth of the anagen follicles . It appears therefore that the presumptive target cells (follicle stem cells?) remain at about the same depth throughout the hair cycle . Albert and Burns believe this to be quite plausible, since during the transition (catagen) between growing and resting phases, the entire follicle below the level of the germ cell is resorbed and this presumably destroys any radiation-induced and potential cancer cells at the deeper levels . 3.

Cancer studies in mouse skin These findings by Albert and Burns, that damage to a 'macrotissue unit' (hair follicles?) is involved in skin cancer induction, was taken up by Geesaman (1968) and Tamplin and Cochran (1974), in the so-called `hot particle hypothesis' . This implied that a highly non-uniform particulate exposure pattern in, for example, skin or lung might be five orders of magnitude more harmful than uniform exposure . Many national and international bodies stated their objection to the theory and opined on theoretical grounds that non-uniform exposure was more likely to be less carcinogenic than uniform exposure . Whilst work in this laboratory in mouse lung models empirically refuted the theory (Lambert et al. 1982, Coggle et al . 1985), few data were available to refute the theory for skin . Since particulate exposure for a- and fl-particles is a serious practical problem in the nuclear power and other radiological industries (Charles 1986) and since ICRP's advice regarding highly non-uniform skin exposure is inadequate, a skin tumour experiment was carried out to compare tumour incidence after large-area uniform #-particle irradiation with exposure to the same mean dose delivered via arrays of small sources . It was felt that mouse skin might prove more appropriate than rat skin since the tumour profile of the latter showed such a strong association with hair follicle damage, yet such an association is not recognized in human skin cancer ; also mice have a much longer latent period for radiogenic skin cancer than rats and are less susceptible to multiple tumour formation, which is also closer to the human situation . The immediate antecedent to this `hot particle' experiment was the work of Hulse and his colleagues (Hulse 1967, Hulse et al . 1983, Papworth and Hulse 1983) who irradiated CBA/H female mice with thallium-204 #-particles using 12 different surface doses from 5 . 4 to 260 Gy, given at four different dose rates from 200 to 1 . 7 cGy per minute . The average latent period for tumour formation was 7 months and over 70 per cent of the tumours were of dermal origin, 30 per cent epidermal and over 60 per cent were malignant . Hulse (1967) reported that tumour yield was proportional to the area of skin irradiated and so incidences in figure 3 are given per unit area of skin . The original data of Hulse (1967) for dermal tumours fitted a D 2e -D/ D ° response function, a model which predicted a tumour incidence that rose to a peak at a surface dose of 200 Gy then declined to negligible values at ^_- 500 Gy . The more recent data by Hulse et al . (1983) and Papworth and Hulse (1983), and shown in

J . E . Coggle and J . P . Williams


80 2

t$.

i 0

i

i 100

r 200

1 300

Dose (Gy)

0

100

200

300

Dose (Gy) Figure 3 . (a) . Incidence of dermal tumours per 1000 cm 2 of thallium-204 #-particleirradiated skin of CBA mice . (b) Incidence of epidermal tumours per 1000cm 2 of thallium-204 beta particle-irradiated skin of CBA mice . Dose rate (per cGy) : p =1 . 7-2 . 4 ; V=5-8 ; 0 =11-21 ; 0=110-200 . (Reproduced from Hulse et al . 1983 ; courtesy of the authors and Taylor & Francis Ltd .)

figure 3, no longer support that view . The data now indicate thresholds of 10 . 6 Gy and 16 . 3 Gy for dermal and epidermal tumours, respectively, followed by a linear dose induction curve, a turnover point due to a cell sterilization component with D o values between 200 and 300 Gy . Hulse et al. (1983) suggested that the thresholds were more apparent than real because `once they were exceeded the dose-response



803

curves rise very steeply' and, like Albert and Burns, they proposed some 'restraining factor' rather than intrinsic radioresistance to account for the threshold doses . Both Hulse et al . (1983) and Cole (1986) have speculated on the role of the skin's immune system in restraining the appearance of radiation-induced skin cancers in their CBA/H mice, the latter looking solely at the possible role of the Langerhans cell (LC) . However, the Cole data (1986) involved the use of only a single 20 Gy local dose to the mouse footpad, which produced no significant fall in the mean Langerhans cell density until 15 months post-irradiation . This long-delayed cellular immune response must be compared with a skin cancer latency in CBA/H mice of "7 months (Hulse et al. 1983) ; it seems therefore that any LC-mediated immune deficiency occurs only after the development of many of the skin cancers, and its role in restraining their appearance must be regarded as merely speculative . With regard to dose rate dependency, Hulse et al . (1983) reported that there was little dependence on dose rate over a range 0 . 1-2 Gy min -1 , but a further 10-fold reduction in dose rate reduced the incidence of skin tumours in CBA/H mice at the highest dose of 32 Gy . These mouse skin data by Hulse were the basis for an experiment in our laboratory designed to test the `hot particle' hypothesis . The skin of 1200 male SAS/4 mice were exposed to a range of uniform and non-uniform sources of thulium-170 #-particles . Non-uniform exposures were produced using arrays of 32 or eight, 2 mm diameter sources distributed over the same rectangular 8 cm 2 area as the uniform control . The details of the experiments have been published (Williams et al. 1986) and among the 1200 irradiated animals, 396 skin tumours were histologically confirmed and classified, with some 96 per cent being of dermal origin ; the range of doses given was 2-1000 Gy and the dosimetry was carried out at 16 ym into the skin, i .e . the base of the epidermis, whilst the dose to the base of the dermis was 50 per cent of the `surface' dose . No tumours developed during the lifespan of 152 control mice . Figure 4 shows the comparison of the 120-129-week dose response for uniform and non-uniform thulium-170 sources and indicates no significant reduction in the incidence above 200 Gy and so the data could not be fitted by any acceptable model combining induction and cell sterilization terms . However, the combined data of two experiments using a uniform thulium-170 source (figure 5) at doses up to 200 Gy is best fitted by a linear induction term plus an exponential cell killing exponent : I = aDe bD The linear regression of the induction data up to 100 Gy when constrained to go through the origin, fitted the data and explained 97% of its variability . The results of the eight-array non-uniform exposures show a significant reduction in tumour yield at all mean doses in comparison to uniform exposure, and provide no evidence for a `hot particle' effect . The results also generally provide evidence for a sparing effect for non-uniform exposures, perhaps resulting from repair, mediated by the relatively undamaged cells surrounding the foci of highly irradiated, damaged tissue . These data are in agreement with the sieve and grid pattern exposures in rats described above, in that at high doses the tumour yield after non-uniform exposures tended to approximate those of the uniform irradiation . Albert and Burns with rats (figure 1), Hulse with mice (figure 3) and reports by Ootsuyama and Tanooka (1988) have reported significant thresholds for their skin


8 04

J. E. Goggle and J . P . Williams

0

50

100

200

300

500

1000

Dose (Gy)

Figure 4 . Cumulative skin tumour incidence as a function of mean surface dose for uniform, 32- and 8-point source non-uniform arrays of thulium- 170 #-particles .

tumour induction, which is at variance with our own work in figure 5 . Albert and Burns's dose fractionation studies in rats, and Hulse et al .'s dose rate work in mice, showed positive sparing effect for tumour induction whilst we did not observe any dose rate effectiveness factor over a 1000-fold range of thulium /3-particle doses down to rates as low as 0 . 01 Gymin -1 (Williams 1988) .

100-

80-

oO

Surface Dose (Gy)

Figure 5 . The dose-response curve for skin tumours in SAS/4 mice for a uniform thulium170 source.


805


We know of no reasonable biological explanation to account for these reported species and strain differences in radiosensitivity for skin cancer induction . Parallels with human racial variations in skin cancer proneness are superfically attractive (Shore et al. 1984, Albert et al . 1986) . Such parallels are likely to be misleading, however, since the human skin cancer data involve a significant component of tumours induced by the synergism between ultraviolet light (UV) and ionizing radiation . UV is unlikely to play any role in skin cancer induction in laboratory rodents (other than nude species) because so little UV reaches the epidermis through the overlying fur . 4.

Interspecies differences in radiation-induced skin carcinogenesis Despite these variations, useful inferences about temporal and dose responses can be drawn from comparing the incidences of cancer induced in experimental animals with those reported for man. Albert et al . (1978) noted the remarkably close similarity in the time sequence of the appearance of skin tumours in rats and humans provided adjustments were made for differences in lifespan . Similarly the time response function in irradiated mice and man is proportional to their median lifespan (Williams 1988) . Figure 6 shows the comparison between the number of tumours per cm2 induced in mice and rats together with the human data . Allowing for the log scale, it is apparent that there is generally close agreement between the rodent data .

Figure 6 . Radiation-induced skin cancer in rodents and man . Mouse data derived from Papworth and Hulse 1983 (∎); Williams et al . 1986 (0) ; Albert et al . 1972 (0) ; Leith et al. 1973 ( •) . Rat data derived from Burns and Albert 1986a (A) . Human data points (A) are from the literature (see Shore 1989) . The fitted lines for whole body skin, - - for UV exposed skin and - - - for UV-shielded skin are based on either an absolute (AR) or relative (RR) risk model (Charles, personal communication) .


806

J . E . Coggle and J . P . Williams

Individual authors claim striking differences between rats and mice, e .g . the yield of tumours in the rat is reported as being 40 times greater than in the mouse, although this difference reduced to 8-fold when comparison was made in terms of tumours per unit area (Burns and Albert 1986a) . The incidence data in Caucasian man in figure 6 are taken from the epidemiological literature . It is not surprising that they show significant scatter since they are derived from people of widely differing ages exposed under very different conditions to ionizing radiation to various areas of the skin which may or may not have been exposed to UV (Shore, this issue pp . 809-827) . There is no clear dose-response relationship for these human data, but there is nevertheless some discernible trend that is not inconsistent with that of the rodent data . The tentative lines drawn through these human data by M . W . Charles (personal communication) are predictions based on a linear response for absolute (AR) and relative (RR) risk models, and the animal data should be compared with the lowest line, since rodent skin must be considered to be non-UV-exposed . The human data undoubtedly support a significantly lower skin cancer incidence, at least two orders of magnitude less than in the rodents ; some of this discrepancy may be because the human tumours are almost wholly epidermal (mostly basal cell carcinomas), whereas such cancers make up only a small percentage of the radiation-induced cancers in the experimental animals, e .g . about 1 % in SAS/4 mice (Williams 1988) . Finally, it must also be remembered that human skin tumours are widely under-reported . The human skin cancer induction rate for figure 6 is approximately 10 -4-10 -5 cm -2 Gy -1 , and if this were applicable to the whole body skin area (2 x 104 cm') then the skin cancer risk incidence for total body skin exposure would be some 10-100 per cent Gy -1 . This analysis of skin cancer risks in man is currently under discussion by the ICRP Task Group on Skin (Charles, personal communication, Shore, this issue pp . 809-827) .

5.

Interactions between radiation and co-carcinogens in skin cancer induction

Experimentally, radiation has been found to be a complete carcinogen, i .e . initiator and promotor, in rodent skin cancer after either single, fractionated or chronic doses . In man, a large proportion of radiogenic skin cancer is probably associated with UV and there are data stressing the synergistic effect of UV and ionizing radiation (Shore et al . 1984) . Despite the importance of these interactions, there has been relatively little work on the interaction of X-rays and other cocarcinogens . Also, the work is outside the remit of this article and the reader is referred to papers by Fry et al. (1986) and Jaffe and Bowden (1986) . 6.

Conclusions

The following conclusions may be drawn from the literature on experimental skin cancer induction . (1) The dose-response curves for skin tumour induction by low-LET radiation show a wide variation in shape from the purely linear with no threshold to quadratic with significant threshold . A linear response is found for highLET radiation . (2) Some dose-response relationships show a tendency for a declining incidence of skin tumours at high radiation doses ; others do not .

Radiation carcinogenesis in the skin (3) In rats and mice, and also

in

807

man, the times between irradiation and the

appearance of tumours, as fractions of life span of the species, are similar . (4) Protraction of the radiation dose produces a reduction in its carcinogenicity in rats, whilst dose rate studies in mice are equivocal regarding a sparing effect .


(5) Mouse skin cancer studies have empirically refuted the `hot particle hypothesis' . (6) The extensive rat studies of Albert and Burns highlight hair follicle damage, specifically at 300µm, as critical in the development of the majority of their induced skin tumours ; mouse studies report a very wide spectrum of cancer types of dermal and epidermal origin which are amenable to classification using human pathological categories . (7) Despite all the interspecies differences, the experimental data for radiogenic skin tumours when expressed per unit area of skin fall on a relatively well defined dose-response curve and are approximately two orders of magnitude more sensitive than the human radiogenic skin cancer data (figure 6) . (8) There is a surprising dearth of information on the interaction between radiogenic skin cancer and other co-carcinogens .

References ALBERT, R. E ., BURNS, F. J ., and HEIMBACH, R . D ., 1967a, Skin damage and tumor formation

from grid and sieve patterns of electron and beta radiation in the rat . Radiation Research, 30, 525-540 . ALBERT, R . E ., BURNS, F . J ., and HEIMBACH, R . D ., 1967b, The effect of penetration depth of electron radiation on skin tumor formation in the rat . Radiation Research, 30, 515-524 . ALBERT, R . E ., BURNS, F . J ., and SHORE, R . E ., 1978, Comparison of the incidence and time patterns of radiation-induced skin cancers in humans and rats . In Late Biological Effects of Ionizing Radiation : Proceedings of the Symposium, Vol . 2 (International Atomic Energy Agency, Vienna), pp . 499-505 . ALBERT, R . E ., NEWMAN, W ., and ALTSCHULER, B ., 1961, The dose-response relationships of beta-ray-induced skin tumors in the rat . Radiation Research, 15, 410-430. ALBERT, R . E ., PHILLIPS, M . E ., BENNETT, P ., BURNS, F ., and HEIMBACH, R ., 1969, The morphology and growth characteristics of radiation-induced epithelial skin tumors in the rat . Cancer Research, 29, 658-668 . ALBERT, R. E ., SHORE, R . E ., HARLEY, N ., and OMRAN, A ., 1986, Follow-up studies of patients treated by X-ray epilation for tinea capitis . In Radiation Carcinogenesis and DNA Alterations, e d . by F . J . Burns, A . C . Upton and G . Silini (Plenum Press, New York), pp . 1-25 . BURNS, F . J ., and ALBERT, R. E ., 1980, Dose-response for rat skin tumors induced by single and split doses of argon ions . In Biological and Medical Research with Accelerated Heavy Ions at the Bevelac (University of California, Berkeley), UCRL 11220, pp . 233-235 . BURNS, F. J ., and ALBERT, R . E ., 1986a, Radiation carcinogenesis in rat skin . In Radiation Carcinogenesis, e d. b y A . C . Upton, R . E . Albert, F . J . Burns and R . E . Shore (Elsevier, New York), pp . 199-214 . BURNS, F . J ., and ALBERT, R . E ., 1986b, Dose-response for radiation-induced cancer in rat skin . In Radiation Carcinogenesis and DNA Alterations, ed . by F . J . Burns, A . C . Upton and G . Silini (Plenum Press, New York), pp . 51-70 . BURNS, F . J ., ALBERT, R . E ., and HEIMBACH, R ., 1968, The RBE for skin tumors and hair follicle damage in the rat following irradiation with alpha particles and electrons . Radiation Research, 36, 225-241 . BURNS, F . J ., SINCLAIR, I . P ., ALBERT, R. E ., and VANDERLAAN, M ., 1976, Tumor induction and hair follicle damage for different electron penetrations in the rat skin . Radiation Research, 67, 474-481 .

808


J ., and VANDERLAAN, M ., 1977, Split-dose recovery for radiation-induced tumours in rat skin . International Journal of Radiation Biology, 32, 135-144 . CHARLES, M . W., 1986, Recent developments in the dosimetry of superficial tissues . In Radiation Damage to Skin, British Journal of Radiology, Suppl . 19 (British Institute of Radiology, London), pp . 1-7 . COGGLE, J . E ., 1982, Dose effect relationships for radiation induced cancer : relevance of animal evidence. Journal of the Society for Radiological Protection, 2, 15-21 . COGGLE, J . E ., PEEL, D . M ., and TARLING, J . D ., 1985, Lung tumour induction in mice after uniform and non-uniform external thoracic irradiation . International Journal of Radiation Biology, 48, 95-106 . COLE, S ., 1986, Long-term effects of local ionizing radiation treatment on Langerhans cells in mouse footpad epidermis . Journal of Investigative Dermatology, 87, 608-612 . FRY, R . J . M ., 1981, Experimental radiation carcinogenesis : What have we learned? Radiation Research, 87, 224-239 . FRY, R . J . M ., STORER, J . B ., and BURNS, F . J ., 1986, Radiation induction of cancer of the skin. In Radiation Damage to Skin, British Journal of Radiology, Suppl . 19 (BIR, London), pp . 58-60 . GEESAMAN, D . P., 1968, An analysis of the carcinogenic risk from an insoluble alpha emitting aerosol deposited in deep respiratory tissue (University of California, Berkeley), UCRL 50387 and addendum). HULSE, E . V ., 1967, Incidence and pathogenesis of skin tumours in mice irradiated with single external doses of low energy beta particles . British Journal of Cancer, 21, 531-547 . HULSE, E . V ., LEWKOWICZ, S . J ., BATCHELOR, A. L ., and PAPWORTH, D . G ., 1983, Incidence of radiation-induced skin tumours in mice and variations with dose rate . International Journal of Radiation Biology, 44,197-206 . JAFFE, D . R ., and BOWDEN, G . T ., 1986, Ionizing radiation as an initiator in the mouse twostage model of skin tumour formation . Radiation Research, 106,156-165 . LAMBERT, B . E ., PHIPPS, M . L ., LINDOP, P. J ., BLACK, A ., and MOORES, S . R ., 1982, Induction of lung tumours in mice following the inhalation of 239 PU0 2 . Proceedings of


BURNS, F.

the Third International Symposium of the Society for Radiological Protection, vol . 1

(SRP Publications, Reading), pp . 370-375 . J . T ., WELCH, G . P., SCHILLING, W. A ., and TOBIAS, C . A ., 1973, Life span measurements and skin tumorigenesis in mice . Radionuclide Carcinogenesis (Department of Energy, Oak Ridge, TN), pp . 90-105 . OOTSUYAMA, A ., and TANOOKA, H . A ., 1988, 100% tumour induction in mouse skin after repeated beta irradiation in a limited dose range . Radiation Research, 115, 488-494. PAPWORTH, D . G ., and HULSE, E . V ., 1983, Dose-response models for the radiationinduction of skin tumours in mice . International Journal of Radiation Biology, 44, 423-431 . SHORE, R . E ., ALBERT, R ., REED, M ., HARLEY, N ., and PASTERNACK, B ., 1984, Skin cancer incidence among children irradiated for ringworm of the scalp . Radiation Research, 100,192-204 . TAMPLIN, A . R., and COCHRAN, T. B ., 1974, A report on the inadequacy of existing radiation standards related to internal exposure of man to insoluble particles of plutonium and other alpha emitting hot particles . Radiation Standards for Hot Particles (Natural Resources Defence Council, Washington, DC) . UNSCEAR (UNITED NATIONS SCIENTIFIC COMMITTEE ON THE EFFECTS OF ATOMIC RADIATION), 1988, Report to the General Assembly (United Nations, New York) . WILLIAMS, J . P ., 1988, Skin carcinogenesis in the mouse following uniform and non-uniform beta irradiation . PhD thesis (University of London) . WILLIAMS, J . P ., COGGLE, J . E ., CHARLES, M . W ., and WELLS, J ., 1986, Skin carcinogenesis in the mouse following uniform and non-uniform beta irradiation . In Radiation Damage to Skin, British Journal of Radiology, Suppl . 19 (BIR, London), pp . 61-64 . LEITH,

Progression and multiple events in radiation carcinogenesis of rat skin.

Skin cancer: role of ultraviolet radiation in carcinogenesis.

Melatonin's protective effect against UV radiation: a systematic review of clinical and experimental studies.

Milestones in skin carcinogenesis: the biology of multistage carcinogenesis.

Visceral pain: a review of experimental studies.

Protecting the radiation-damaged skin from friction: a mini review.

Skin Carcinogenesis Studies Using Mouse Models with Altered Polyamines.

Cdc6: Skin in the carcinogenesis game.

Peroxiredoxin 6 in skin carcinogenesis.

Multistage carcinogenesis in mouse skin.

Concerning radiation carcinogenesis.

Studies of the carcinogenesis and tumorigenesis of skin applications of dodecylbenzene on hairless mice.

Selective promoting activity of phorbol myristate acetate in experimental skin carcinogenesis.

Correction: Protective role of p53 in skin cancer: Carcinogenesis studies in mice lacking epidermal p53.

Modification of multistage skin carcinogenesis in mice.

The use of PIXE in experimental studies of the physiology of human skin epidermis.

The pig as an experimental model for skin flap behaviour: a reappraisal of previous studies.

Some mechanisms operative in carcinogenesis a review.

The contributions of 131I to the understanding of radiation carcinogenesis.

S100A9 has a protective role in inflammation-induced skin carcinogenesis.

The role of androgens in experimental rodent mammary carcinogenesis.

Molecular mechanisms of ultraviolet radiation carcinogenesis.

Radiation signatures in childhood thyroid cancers after the Chernobyl accident: possible roles of radiation in carcinogenesis.

Experimental animal models of pancreatic carcinogenesis for prevention studies and their relevance to human disease.