Documents of the NRPB: Volume 6, No. 2

Synopsis

Summary and conclusions

Recommendations

Effects of Ultraviolet Radiation on Human Health: Report of an Advisory Group on Non-Ionising Radiation

Statement by the National Radiological Protection Board

Scope

1. The National Radiological Protection Board has a statutory responsibility to provide advice and information on standards of protection for exposure to non-ionising radiation. This includes the health effects and hazards associated with exposure to ultraviolet radiation (UVR).
2. In 1990, the Director of the Board set up an Advisory Group on Non-ionising Radiation:

   to review work on the biological effects of non-ionising radiation relevant to human health and to advise on research priorities.

3. In its third report the Advisory Group has reviewed health effects from UVR 1. It has considered both natural and artificial sources of exposure, as well as experimental studies relevant to understanding the effects of UVR on cells and tissues. It has examined information on the clinical effects of UVR and the results of epidemiological studies with the aim of providing advice on the risks of exposure. Some information is also given on the effects of blue light. The Advisory Group has also made recommendations for further research aimed both at improving the basis for assessing exposures to natural and artificial sources as well as furthering knowledge of the effects of UVR on health through experimental and epidemiological studies.
4. This Statement by the Board reviews the main conclusions of the Advisory Group. It also gives advice relevant to the protection of human health. It applies both to members of the public and to those who are occupationally exposed and is intended to provide a framework for reducing UVR exposure and increasing awareness of its effects.

Sources

5. The main source of UVR contributing to personal exposure is the sun and the major risk is associated with personal habits in relation to solar radiation. For some individuals, however, UVR from artificial sources could contribute significantly to their total exposure. Such sources of exposure include: sunbeds used for cosmetic tanning; a few in industry; those used for medical therapy.

Effects

6. It is widely accepted that UVR is carcinogenic and produces other undesirable health effects. The main tissues affected are those in the skin and the eye. There is evidence that UVR can also induce changes in the immune system but the significance of this to human beings is not yet clear.

Skin

7. The most serious health effects, for which exposure to UVR is a recognised risk factor, are the cutaneous malignancies (skin cancers). UVB has been recognised for some time as carcinogenic in experimental animals, and there is increasing evidence that UVA, which penetrates more deeply into the skin, also contributes to the induction of cancer. UVC from the sun is absorbed by the earth's atmosphere, and any arising from artificial sources does not readily penetrate to the sensitive basal layer of the skin.

Short-term effects

8. Short-term effects on the skin may be seen as sunburn, principally consisting of erythema (skin reddening resulting from vasodilation) and oedema (swelling), both of which may be very severe. In some people this sunburn is followed by increased production of melanin and is recognised as a suntan. Racially determined skin pigmentation will provide some protection. A suntan is not an indication of good health and only offers minimal protection against further exposure. It is a sign that damaged skin is attempting to protect itself from further harm 2, 3.

Long-term effects

9. The most serious long-term effect is the induction of cancer. The non-melanoma skin cancers (NMSCs) are mainly basal cell carcinomas and squamous cell carcinomas. They are relatively common in white populations, although they are rarely fatal. The overall incidence is difficult to assess because of under-reporting. Reported NMSCs account for just over 5% of registered malignancies in the UK, but under 0.5% of cancer deaths. These rates are increasing year on year. They occur most frequently on sun-exposed areas of the body such as the face and hands and show an increasing incidence with increasing age. The findings from epidemiological studies indicate that the risk of both of these skin cancers can be related to cumulative UVR exposure, although the evidence is stronger for squamous cell carcinomas. UVR induces NMSCs in experimental animals.
10. Malignant melanoma is the main cause of skin cancer death, particularly in young people, although its incidence is less than NMSC. The risk of developing malignant melanoma has increased substantially in white populations for several decades and the annual incidence in the UK is now approaching 10 new cases per 100,000 population; more than double the rate 15 years ago. Occurring at relatively young ages, it is the cause of 1 in 12 cancers at ages 20-39 years, and 1 in 25 cancer deaths at these ages. A higher incidence is found in people with large numbers of naevi (moles), those with atypical naevi, those with a fair skin, red or blond hair and those with a tendency to freckle, to sunburn and not to tan on sun exposure. Both acute burning episodes of sun exposure and chronic occupational and recreational exposure probably contribute to the risk of malignant melanoma. There are indications that excessive sun exposure within the first two decades of life increases the risk of malignant melanoma later in life, although the precise nature of the relationship is still uncertain.
11. Chronic exposure to solar radiation causes photoageing of the skin which is characterised by a leathery, wrinkled appearance and loss of elasticity. Corroborating evidence for a role of UVR in the aetiology of these responses has been produced from extensive biological studies.
12. A small quantity of UVR is beneficial in terms of vitamin D synthesis in the skin. Whilst this may be important for those of the population on a restricted diet, for the great majority the deleterious effects of UVR in terms of cutaneous damage and increased risk of skin cancer are far more important.

Eye

13. Responses of the human eye to acute UVR exposure include photokeratitis and photoconjunctivitis (inflammation of the cornea and the conjunctiva, respectively). Repeated exposure is also considered to be a major factor in the causation of non-malignant clinical lesions of the cornea and conjunctiva such as climatic droplet degeneration (discrete areas of yellow protein deposits in the cornea), pterygium (an overgrowth of the conjunctiva on to the cornea) and, probably, pinguecula (small yellow growths in the conjunctiva).
14. Epidemiological data on cataract formation in highly exposed people suggest that cumulative UVR exposure is a principal causative factor in the development of, at least, cortical cataracts, although the extent to which this is an important risk factor for cataracts in the general population is unclear. Few appropriate animal studies have been carried out, although acute exposures to UVB at levels above the threshold for photokeratitis have induced anterior cortical opacities.
15. There is good evidence that prolonged gazing at very bright light sources, particularly those emitting shorter wavelength blue light, causes retinal damage resulting in transient or permanent loss of visual acuity. Such an effect would normally be prevented by the natural aversion response invoked by looking at a bright light, but this response can be intentionally suppressed. Similar damage has also been induced in the non-human primate retina following acute exposure; blue light was the most effective. It is not clear to what extent UVA is involved; transmission through the lens is low in adults but is higher in children and in people who have had their lens surgically removed and have not had a suitable UVR-absorbing replacement lens fitted.
16. There is some equivocal evidence that chronic exposure to high levels of solar radiation is a contributory factor in the development of age-related macular degeneration of the retina, a major cause of blindness. There is insufficient evidence of an association between exposure of the eye to UVR and an increase in the risk of ocular melanoma.

Immune responses

17. Biological studies have shown that exposure to UVR can suppress the normal antigen-specific immune response to some skin tumours and to various skin pathogens, although immunity acquired from prior infection is not affected. The significance for human health of UVR-induced immune suppression is not clearly established at present nor is the relationship between exposure to UVR and skin cancer in patients who are immunologically suppressed, e.g. following tissue transplants.

Recommendations

18. The Board welcomes and supports the initiatives that have been taken forward by the government, and the Health Education Authority 4 and other agencies over the past few years in pursuance of The Health of the Nation 5 target to halt the year on year increase in skin cancer by 2005. The actions taken so far are consistent with the recommendations outlined below, which are intended to be applied to the population as a whole. Particular care is needed, however, for children and young people and others at greater risk of UVR-induced skin cancer, such as those with large numbers of naevi.

Public awareness

19. The Board recommends strongly that educational programmes continue to aim at increasing awareness of the health effects of UVR exposure by the general public and by those who may be occupationally exposed. This is particularly important for parents, those working in nurseries, school teachers and others responsible for the day-to-day care of children. The objectives are to improve knowledge, influence attitudes and change behaviour in relation to UVR exposure. The programmes should aim at a reduction in both the cumulative exposure to UVR and, particularly, exposure to high levels resulting in acute damage to the skin or eyes. The overall objective is a reduction in morbidity and mortality due to skin cancer.
20. Information relating to early diagnosis of skin cancer should be readily available to the public. The major signs of suspected malignant melanoma are: an existing mole getting larger, developing an irregular outline, or showing mixed shades of brown and black; or a new mole growing quickly (in months). Minor signs include: a mole becoming bigger than the blunt end of a pencil (around 5 mm), or becoming inflamed or developing a reddish edge, or bleeding, oozing or crusting; or a mole starting to feel different (eg itching or painful).
21. With the aim of increasing public awareness of current levels of UVR exposure throughout the country, the Board supports the regular publication of levels of solar radiation. These measurements will also provide a basis for more realistic exposure assessments.

Clinical diagnosis of malignancy

22. Early detection of malignant melanomas is needed in order to reduce the present high mortality rate. Rapid referral to a medical practitioner is indicated if one or more of the major signs above are observed. Referral to a dermatology specialist should be considered if three or four minor signs are observed, even without the presence of a major sign 6.
23. The Board recommends that those concerned with education and training of medical practitioners and other health-related professionals should consider how to increase skills in the recognition and treatment of skin cancer. This would complement the anticipated improved public awareness and enhance early diagnosis and effective treatment.

Protection from solar radiation and artificially produced UVR

24. Protection of the skin and eyes can be enhanced by wearing hats and clothing, wearing sunglasses which exclude UVR and applying sunscreens. However, sunscreens are used to provide protection against sunburn. The protection against melanoma provided by the use of sunscreens is unclear, possibly because use may lead to longer and more intense exposure. Such protection is less certain than that provided by reducing exposure. The Board recommends that standards organisations and manufacturers develop suitable and scientifically valid protection criteria for these products, which should be agreed internationally. This would enable the public to understand each protection measure and use them effectively.
25. The Board recognises the need for advice limiting exposure to UVR in the workplace. The initiative by the Health and Safety Executive in this area related to outdoor workers is strongly supported 7.
26. The Board recommends that the use of sunbeds and sunlamps for cosmetic tanning should be discouraged. Modern artificial tanning devices use predominantly UVA. However, since the spectra for biological damage extend into the UVA region, exposure to UVR from sunbeds and sunlamps is likely to carry a risk.
27. Some prescribed medicines, drugs, cosmetics and various plant materials can cause sensitisation of the skin and eyes to UVR. The Board recommends that patients and the general public should be warned by health professionals and manufacturers of these interactions with UVR.
28. Approaches to be considered in educational and awareness programmes for protection from solar radiation are given in the table.

Approaches to minimising sun-induced skin and eye damage

• Remember that there is no such thing as a safe or healthy tan.
• Take sensible precautions to avoid sunburn, particularly in children.
• Limit unprotected personal exposure to solar radiation, particularly during the four hours around noon, even in the UK.
• Seek shade wherever possible, but remember sunburn can occur even when in partial shade or when cloudy.
• Remember that sunburn can occur while swimming and is more likely when there is a high level of reflected UVR, such as from snow and sand.
• Wear suitable head wear, such as a wide-brimmed hat, to reduce exposure to the face, head and neck.
• Cover exposed skin with clothing giving good protection - examples are long-sleeved shirts and loose clothing with a close weave.
• Sunglasses should be designed to exclude both direct and peripheral exposure of the eye.
• Apply sunblocks, or broadband sunscreens with high sun protection factors ( SPF around 15) to exposed or uncovered skin. Apply generously and reapply frequently.
• Remember that certain prescribed drugs, medicines, cosmetics and plant materials can make people more sensitive to sunlight.

The sun protection factor (SPF) is the ratio of the UVR exposure to produce minimal erythema on a skin site protected by sunscreen to the UVR exposure to produce a comparable erythema on unprotected skin 8.

Research

29. The Board supports the recommendations for further research made by the Advisory Group. The Board recommends that the current research programmes to establish the biological mechanisms and the risks of cancer induction from UVR exposure should be continued. A significant research challenge exists to improve the effectiveness of screening for malignant melanomas.

Development of formal guidance on UVR exposure

30. The Board will develop formal guidance on UVR exposure appropriate to the UK population. This guidance will not apply to patients receiving UVR as part of medical treatment.

UVR is divided by wavelength into UVA 315-400 nm, UVB 280-315 nm and UVC 100-280 nm. Blue light lies in the range of about 400-500 nm.

1 NRPB. Health effects from ultraviolet radiation: Report of an Advisory Group on Non-Ionising Radiation. Doc. NRPB, 6, No. 2, 7-190 (1995).

2 CRC. Malignant melanoma. London, Cancer Research Campaign, Fact Sheet 4.1 (1995).

3 UK Skin Cancer Prevention Working Party. Consensus statement. London, British Association of Dermatologists (1994).

4 HEA. 'Sun Know How' campaign. London, Health Education Authority (1994).

5 DH (Department of Health). The Health of the Nation: A strategy for health in England. London, HMSO, Cmd. 1986 (1992).

6 CRC. Malignant melanoma: A guide to early detection. London, Cancer Research Campaign (1989).

7 HSE. 'Keep Your Top On'. London, Health and Safety Executive, (IND CG) 147L (1995).

8 CIE. Sunscreen testing (UVB). Vienna, International Commission on Illumination, Technical Report CIE 90 (1991).

Summary and conclusions

Introduction

1. This report by the Advisory Group on Non-Ionising Radiation examines the effects of ultraviolet radiation (UVR) on human health. In preparing the document both natural and artificial sources of exposure have been considered, as well as experimental studies relevant to understanding the effects of UVR on cells and tissues. The clinical consequences of exposure and the results of epidemiological studies have been reviewed with the aim of providing, as far as is possible, information on the risks of exposure. Effects on plants and animals in the environment are not addressed as this is the concern of other bodies.
2. Ultraviolet radiation lies within the range of wavelengths from 100 to 400 nm. In terms of direct hazards, 180 nm is often considered as the practical lower wavelength limit as radiation of shorter wavelengths is strongly attenuated by air. As a result of its limited penetration into biological tissues, the direct hazards of exposure to UVR are mostly confined to the skin and the eyes. Widely used subdivisions of the UVR region of the spectrum are based on their relative efficiency in producing the most important biological effects in human beings. These are: UVC 100-280 nm, UVB 280-315 nm, and UVA 315-400 nm.

Solar radiation

3. The human race has evolved in an environment in which intense, broad spectrum UVR from the sun has always been present. Short exposure to UVR may be beneficial for health (around 15-20 minutes per day) as the exposed skin can generate vitamin D. For most people exposure to solar radiation is mainly incidental, in the course of their work or otherwise in normal daily life; for others, much of their exposure is by choice as from sunbathing. If solar radiation reached the surface of the earth unscreened it would be lethal to exposed living organisms; in practice the shielding provided by the atmosphere results in substantial attenuation, with ozone generated by the interaction of short wavelength UVR (less than approximately 240 nm) and oxygen in the atmosphere providing a major contribution. There has been concern in recent years that atmospheric pollution, which depletes stratospheric ozone, is resulting in higher intensities of UVR at ground level. Ground-level measurements of solar UVR have been made worldwide for some time and such measurements have become more common in recent years. Generally, however, these measurements have been sporadic and not coordinated internationally so that they provide only a limited database for assessing personal exposure to solar UVR at ground level and any temporal trends.
4. Within the UK, simultaneous measurements of visible radiation, UVA and erythemally weighted UVR have been made at Chilton (approximate latitude 52° N), Leeds (54° N) and Glasgow (56° N) since 1988. This programme of measurements is intended to provide information regarding the range and variation of solar radiation at different latitudes in the UK with the time of year, as well as to establish baseline levels for natural UVR which can be used to monitor trends as well as to compare with measurements from artificial sources. An analysis of the annual integrated measurements at Chilton between 1988 and 1994 suggests a 5% decrease in UVR over the period, but this downward trend is not statistically significant and is also consistent with there being no change at all.
5. There is no convincing, consistent evidence from published measurement data worldwide for populated areas to indicate a global trend of changing solar UVR at the earth's surface with time or with stratospheric conditions. This lack of a discernible effect is likely to be due to dominant tropospheric variables, such as atmospheric absorption and scattering processes, localised atmospheric pollution and cloud cover, all of which affect levels of UVR at ground level.

Artificial sources

6. In addition to solar radiation, there are many artificial sources to which exposure may be deliberate, such as those used for medical therapy or the production of a tan, or adventitious, as from some tungsten halogen lamps, sources used in industrial processes and welding arcs. While there are few artificial sources that result in human exposure greater than from solar radiation, there are some people for whom UVR from artificial sources may contribute significantly to their total exposure.
7. Incandescent sources such as tungsten filament bulbs intended for general lighting emit levels of UVR insignificant to human health; however, some unshielded tungsten halogen lamps can emit amounts of UVR sufficient to cause erythema. Fluorescent lamps intended for general lighting purposes are specifically designed to emit visible radiation and only give out very small amounts of UVR at the levels of illumination normally encountered in the home and in the workplace. However, fluorescent lamps for special applications, such as those used for cosmetic tanning, emit levels of UVR sufficient to cause injury to the skin and eye. The most potent artificial sources of UVR, and particularly of UVB and UVC, are those characterised as high intensity discharge (HID) lamps. These include high pressure mercury, mercury metal halide and xenon lamps. The spectral emission of xenon lamps closely matches that of a 'black-body' radiator at about 6000 K enabling them to be used as solar radiation simulators in experimental studies. HID lamps used for lighting purposes are normally double enveloped with an outer envelope which attenuates the UVR emitted and, when used within properly designed luminaires, do not present a hazard. High pressure mercury vapour lamps with little or no filtration were once commonly used for cosmetic tanning. These lamps are now little used, having been largely replaced by UVA fluorescent and filtered HID UVA cosmetic tanning systems.
8. Gas welding, brazing and cutting processes operate at temperatures insufficiently high to cause the emission of intense UVR, but advised limits for protection may be approached if exposures at short distances are very prolonged. Arc welding processes are particularly potent sources of UVR and even very short exposures may be hazardous to the eyes and skin. Both gas and arc welding also emit visible and infrared radiations which may be hazardous to the retina. Appropriate personal protection always needs to be worn during welding.

Guidelines

9. There are at present no regulations specifically controlling exposure to artificial UVR. Where recommendations specifically advise limiting exposure to optical radiation (ultraviolet, visible and infrared), they do so in the form of voluntary standards and not as regulations.

Cellular and molecular structures

10. Biological evidence from studies at the molecular and cellular level suggests that UVR is a complete carcinogen. It can bring about initiation, which results from stable changes in the genetic information carried in DNA, and promotion, which involves cellular proliferation and progression to the fully malignant state.
11. The principal type of DNA damage produced by direct absorption of UVR at shorter wavelengths is the formation of pyrimidine dimers. For cultured cells damage is maximal in the UVC region but in animals and human beings the effectiveness of radiation at these shorter wavelengths is considerably attenuated by absorption in the outer layers of the epidermis, resulting in maximum damage for radiation in the UVB region. The formation of pyrimidine dimers is strongly implicated as a pre-mutagenic lesion, whereas the importance of other types of damage is less clear. As the wavelength increases, direct absorption by DNA reduces and damage increasingly occurs indirectly through the action of photosensitisers which absorb UVR. The types of induced damage change with increasing wavelength, with single strand breaks, double strand breaks and protein-DNA crosslinks becoming dominant. The sensitisers involved in these reactions remain to be definitively identified, although riboflavin, bilirubin, porphyrins, and the photoprotective pigment melanin, in melanocytes, are all possible candidates. In addition, a number of exogenous compounds to which people may be exposed could act as sensitisers. Examples of such compounds are to be found amongst dyes, some drugs, aromatic hydrocarbons, cosmetics, sunscreens and natural products such as plant extracts.
12. Although photodamage to DNA has the potential to contribute to the development of cancer, this will occur only if the damage leads to a stable change in critical genes. In the majority of cases, photodamaged DNA is rapidly and accurately repaired prior to replication. However, if the cell is already committed to replication, or if damage is excessive, then inaccurate DNA repair is more likely and may lead to the induction of mutations and subsequent cellular development into cancer. The most important error-free pathway appears to be excision repair, which is relatively well understood; post-replication repair pathways, whether error free or error prone, are much less well understood. It is clear, however, that the cellular response to damage is a major factor in determining whether DNA damage results in tumour development.
13. Several frequently mutated genes have been identified in skin tumours, although inherent bias in the assays used for their detection may overemphasise their importance. Thus, analysis of DNA from UVR-induced mouse skin tumours and clinical samples of human skin tumours has indicated that a small number of specific mutations at sites within the ras gene occur with high frequency in non-melanoma skin cancers. The same mutations appear to be associated with some malignant melanomas, and may be sufficient to commit melanocytes in culture to the process of malignant transformation. However, the assays are selectively biased towards detecting ras gene mutations. Inactivating mutations in genes encoding growth regulators, such as the p53 tumour suppressor gene, have also been identified in human skin tumours. There is, however, no conclusive evidence that any of these genes are the principal targets for the initiating events which result in skin cancer.
14. Cells appear to mount a coordinated response to the effect of UVR which is probably important in determining the final outcome of exposure. This occurs through the synthesis and secretion of intercellular signalling molecules such as cytokines and growth factors which elicit a range of short-term and long-term effects, including increased proliferation of keratinocytes and melanocytes. The altered expression of regulatory genes is probably important in this response, but may also be the means through which UVR exerts a tumour promoting effect through the activation of growth stimulatory signalling pathways involved in growth stimulation. The responses of the cellular proto-oncogenes fos, jun and myc to UVR appear to be similar to their responses to chemical tumour promoters such as phorbol esters. Increased cellular proliferation is a late effect of exposure to UVR, and there is evidence that the changes in gene expression include the activation of secreted proteases required for local remodelling of tissue which would be associated with this process. Another consequence of the response of cells to UVR is the activation of quiescent viral sequences in exposed cells.
15. Several genetic factors exist that render some people more sensitive to UVR than the majority. The most important factor that is specifically associated with an altered risk of UVR-induced skin cancer is skin type. Other factors include DNA repair defects which result in clinical conditions such as xeroderma pigmentosum. Whilst the homozygous defects which underlie such conditions result in greatly elevated risks of developing skin neoplasms, they are relatively uncommon. In addition, other heterozygous mutations, of tumour suppressor genes, for example, could also result in a predisposition to cancer.

Effects on tissues

16. Information on the effects of UVR on people is available from clinical observations and from volunteer studies. Animal studies provide insights into the mechanisms and biological effects of exposure to UVR, although quantitative extrapolation to human beings is difficult. Acute tissue damage results from excessive exposure to UVR while various irreversible degenerative changes are associated with chronic exposure including photoageing of the skin, the development of skin cancer and, probably, cataracts.

Acute effects

17. Exposure to UVR elicits acute effects in the skin, the severity of which increases with increasing exposure. Transient responses include adaptive changes such as immediate pigment darkening, melanogenesis and epidermal hyperplasia, and inflammatory responses such as erythema (skin reddening resulting from vasodilation) and oedema (swelling). The sensitivity of human beings to many of these responses varies within six broad categories of skin type (I-VI), identified on the basis of qualitative assessments of skin pigmentation, tanning ability and erythemal sensitivity. People with skin types I and II show greater sensitivity to UVR and are less able to tan than those with skin types III to VI.
18. In the eye, photokeratitis (inflammation of the cornea) and conjunctivitis (inflammation of the conjunctiva which overlies the visible white 'scleral' region of the eye and the inner surface of the eyelid) also occur as a result of acute exposure to UVR. Acute exposure to ambient UVR from highly reflective surfaces, such as snow, may damage the surface of the eye, causing photokeratitis. Acute exposure to UVB, at levels exceeding photokeratitis thresholds, has been shown to induce anterior lens opacities in the eyes of rabbits; sufficient damage may eventually induce posterior subcapsular cataracts due to the backward migration of cellular debris. The retina in the eye is normally protected from acute damage caused by looking at bright sources of light, such as the sun, by involuntary aversion reflexes.
19. Most of the acute effects observed show maximal sensitivity in the UVB region, and are much less sensitive to UVA radiation. However, radiation in the UVA region is more effective than UVB in inducing immediate pigment darkening in human beings.

Chronic effects

20. In chronically exposed people, photoaged skin is characterised by wrinkling, the appearance of thickened elastic fibres and a loss of collagen. Prolonged exposure may also adversely affect the microcirculation in the skin with blood vessels becoming dilated and tortuous, and eventually very sparse. Alterations in skin structure that have been induced in animals appear to be similar to changes reported in human skin. In mice, these changes include aggregates of thickened elastic fibres, skin thickening, wrinkling and sagging and a proliferation of fibroblasts and dermal cysts. Action spectra for most of these effects are highest in the UVB region; in contrast action spectra for skin sagging may be maximal in the VA region.
21. UVR exposure is a major factor in the causation of corneal and conjunctival disorders such as climatic droplet degeneration, pterygium and, probably, pinguecula. Information from human studies suggests that UVR exposure is a cause of cortical, and possibly some other, cataracts. The UVB component of UVR is thought to be the main causative factor, but longer wavelength UVA is also likely to play a part. UVR is absorbed by a yellow chromophore which accumulates in the lens with age. Thus, little UVR reaches the retina in adult human beings with intact lenses. There is, however, strong evidence that environmental exposure to very high levels of solar radiation can damage photoreceptors and the retinal pigment epithelium, particularly in the macular and paramacular regions. These effects may be associated with solar retinitis (inflammation of the retina) and blindness caused by prolonged staring at the sun, such as during an eclipse. There is ome evidence to suggest that exposure to sunlight may contribute to the production of age-related macular degeneration.
22. Exposure to natural sunlight increases the risk of developing squamous cell carcinoma, basal cell carcinoma and malignant melanoma in human beings. Animal studies have also shown that UVR can act either as a complete carcinogen, capable of inducing tumours when applied by itself, or as a co-carcinogen in combination with tumour initiators and promoters. Squamous cell carcinomas can be induced in animals by exposure to UVA, UVB or UVC alone. However, the commonly used animal models, such as hairless mice, may be more sensitive to UVR than human beings. Basal cell carcinomas, the most common human malignant skin tumours, have only rarely been observed in animals. Malignant melanomas have not been induced in animals by exposure to UVR unless this has been given in conjunction with a chemical carcinogen. Transgenic mouse models of malignant melanoma have, however, been developed in which the series of changes and types of lesion parallel in many respects those in the human disease, including the tendency to metastasise. These models may offer useful experimental models for melanoma induction in human beings.

Immune response

23. In animals, exposure to UVR suppresses specific aspects of the immune response, although it does not induce general immunosuppression. The immune response of irradiated animals to newly encountered antigens is suppressed, although existing responses acquired from prior immunisation are not affected. In particular, thecytotoxic T-cell response is depressed in animals exposed to UVR. This is a key element in the immune response to viral infections. UVR-induced suppression in response to viruses, including herpes simplex, bacteria, and parasitic infections, has also been demonstrated. It has been suggested that such antigen-specific suppression may have evolved in order to prevent an autoimmune response to UVR-induced antigens expressed on exposed cells in the skin.
24. Few studies of the effects of UVR exposure on human immune response have been carried out; the results of studies on the suppression of contact hypersensitivity have been variable, but at least some UVR-exposed individuals became immunologically tolerant to the sensitiser, indicating that human beings may exhibit a similar response to animals. It has been suggested that the variable response observed in volunteers may be a reflection of a genetically determined sensitivity similar to that observed in mice.

Epidemiology

25. Non-melanoma skin cancer is relatively common in white populations. These tumours are seldom fatal. The evidence overall strongly supports the hypothesis that the risk of non-melanoma skin cancer is related to cumulative solar radiation exposure. No competing hypothesis has been put forward which would satisfactorily explain the anatomical distribution of these tumours, mainly on the most frequently exposed body sites, the increasing incidence with age, and the findings from epidemiological studies which suggest that both squamous and basal cell carcinomas are related to UVR exposure, with squamous cell carcinoma being the more closely related. UVB has been recognised for some time as potentially carcinogenic and there is increasing evidence that UVA may also contribute to the production of cancer as it penetrates deeper into the skin and has the capacity to damage DNA indirectly and to induce skin tumours in animals. Data from patients treated with PUVA (psoralen plus UVA irradiation) show a strong dose-response relationship of the treatment to the risk of squamous cell carcinoma and a much weaker relationship to the risk of basal cell carcinoma, although as yet the separate contributions of psoralen administration and UVA have not been defined.
26. The risk of cutaneous melanoma has increased substantially in white populations for several decades and the major risk factor appears to be exposure to solar radiation. A high risk of melanoma is seen in people with a light skin (melanin absorbs UVR), red or blond hair, blue eyes, a tendency to burn and not tan on sun exposure, a tendency to freckle, large numbers of naevi and atypical naevi. Cumulative exposure is probably the main cause of melanomas of the head and neck. There are insufficient data to assess the relationship of melanoma in sites normally unexposed to solar radiation. Intense short-term exposures of untanned skin appear likely to be the main cause of melanoma of intermittently exposed skin sites. There is equivocal evidence that the use of sunlamps and sunbeds may cause melanoma. There is also evidence, particularly from data on migrants and frequency of sunburn, that exposures in childhood and adolescence may be of special importance to the risk of melanoma. Several of the studies on this issue are, however, potentially biased, and the precise relationship is still uncertain.
27. The evidence on the relationship of ocular melanoma risk to solar UVR exposure is limited, inconsistent and inconclusive. Three studies have found raised risks of ocular melanoma in relation to exposure from lamps emitting UVR, but there is insufficient evidence to be sure that it is a real relationship.
28. Some recent evidence suggests that UVR may increase the risk of non-Hodgkin's lymphoma and chronic lymphatic leukaemia. This is based in part on studies of second cancers which suggest that these two conditions share a common aetiology with squamous carcinoma of the skin and malignant melanoma.
29. Although the epidemiology of lip cancer generally accords with causation by cumulative solar UVR exposure, pipe smoking is also a risk factor, and further factors may possibly be involved. The worldwide distribution of lip cancer does not parallel closely the levels of ambient UVR; there is, for instance, a particularly high incidence in Newfoundland. This suggests that there are other aetiological factors beyond a simple relationship with sun exposure: as well as pipe smoking there might be further influences.
30. Overall, the epidemiological data for cataract suggest that cumulative UVR exposure is an important cause at least of cortical cataracts. Since the most convincing evidence has come from highly exposed men, and since the gradient of risk with latitude in the general population in the USA appears to be relatively small, it is unclear to what extent UVR radiation is an important risk factor for cataracts in the general population in western countries.

Dose-response relationships for skin cancer

31. A model for estimating the risks of basal cell carcinoma and squamous cell carcinoma from exposure of a population to UVR has been described. The model gives close agreement between observed and calculated age-standardised incidence rates (both by latitude and country of birth) in a recent national survey in Australia, the country with the highest recorded incidence of non-melanoma skin cancer. The model was used to estimate the risks (together with 95% confidence intervals) of basal cell carcinoma and squamous cell carcinoma associated with elective UVR exposure (use of UVA sunbeds for whole body tanning) and adventitious UVR exposure (occupational exposure in a hospital phototherapy department) and the possible consequences of stratospheric ozone depletion.
32. There is no convincing evidence to date of global trends in levels of UVR at the earth's surface in populated areas. The increased risks of adverse health effects on the skin of even the most pessimistic predicted reductions in stratospheric ozone levels and consequential increases in solar UVR at the earth's surface, should they occur, are likely to be small in comparison with reductions in risk that can be achieved by changes in personal behaviour. For British adults alive today, stratospheric ozone depletion continuing indefinitely at current rates is predicted to result in a relatively small additional lifetime risk (<5%) non-melanoma="non-melanoma" skin="skin" cancer,="cancer," assumption="assumption" no="no" climate,="climate," time="time" spent="spent" out="out" doors,="doors," behaviour,="behaviour," clothing="clothing" habits="habits" or="or" ground-level="ground-level" pollution.="pollution." case="case" children,="children," predicted="predicted" rate="rate" is="is" 10-16%="10-16%" higher="higher" than="than" expected="expected" without="without" changes="changes" stratospheric="stratospheric" ozone.="ozone." Although="Although" current="current" knowledge="knowledge" does="does" not="not" allow="allow" quantification="quantification" risk="risk" estimates="estimates" to="to" made="made" for="for" malignant="malignant" melanoma,="melanoma," it="it" must="must" be="be" assumed="assumed" that="that" an="an" increase="increase" in="in" UVR="UVR" could="could" incidence="incidence" of="of" this="this" disease,="disease," and="and" have="have" some="some" effect="effect" on="on" the="the" eyes.</li="eyes.</li" />
33. Personal behaviour in relation to sun exposure is an important factor in the risk of adverse health effects resulting from exposure to UVR and particularly so in respect of the risk of skin cancer. Increases in the rates of skin cancer over the past 50 years may be attributed to changes in behaviour, such as increased leisure time, increased recreation in the sun, lighter clothes and exposure of larger areas of the skin.

Unless otherwise indicated, the animal studies referred to in this chapter were carried out using mammals, mostly rodents.

Recommendations for research

Introduction

1. A number of specific recommendations are made concerning the need for further research relating to health effects resulting from exposure to ultraviolet radiation (UVR). First, there is a need to understand better the aetiology of UVR-induced skin cancers, particularly malignant melanoma, and to relate this to individual UVR exposure. In this context, coordinated measurement programmes of trends in solar UVR and personal exposure are important, as is the identification of sensitive subgroups within the population.
2. There is a need for further study of the potential role of UVR in the induction of cataracts and long-term damage to the exposed skin, the ocular surface (pterygium and climatic droplet keratopathy) and the retina (photopic retinopathy). In addition, the nature and relevance to health of the UVR-induced changes in the response of the immune system need to be determined, particularly in relation to the development of skin cancers.
3. In all these areas of study, the derivation of relevant action spectra and dose-response relationships are essential in order to provide a basis for estimating risks; the validation of risk models for carcinogenesis is considered to be of high priority. In this context, determining the risks from exposure to UVA, which forms the major part of solar UVR and almost all of the UVR output of 'tanning' solaria, is paramount.

Solar and artificial radiation sources

4. Measurement programmes are necessary for obtaining both broadband and spectral data. These should be conducted with the objective of providing a reliable, quality assured database for assessing solar radiation exposure levels throughout the world, for analysing temporal trends and for assessing biological responses.
5. Characterisation of personal exposures from solar UVR and in particular its distribution across the body for a variety of postures and habits are needed to relate environmental measurements to individual exposure. Compact and biologically relevant personal exposure meters are required to provide a more precise measurement of solar UVR radiation exposure over long periods.
6. The major source of UVR exposure is the sun, although many people are exposed to artificial sources. In particular, the effects of exposure to sources with a high UVA emission, such as sunbeds, need further study as the role of UVA in biological interaction processes in human beings is as yet uncertain.
7. Direct viewing of the sun is unlikely to represent a hazard to the retina due to the normal aversion responses of the eye. Further studies need to be carried out to identify the role of artificial sources of optical radiation in causing ocular damage, with emphasis on the relative contributions of UVR and blue light (400-500 nm) to photopic retinopathy and, chronically, to age-related macular degeneration.

Experimental studies

8. The adequacy of using six different categories of skin type as predictors of the response to acute UVR exposure should be reviewed. Other phenotypic characteristics may more accurately predict the risk of UVR damage and skin cancer.
9. With regard to chronic changes, the mechanisms of photoageing in the skin, particularly the immediate changes in cytokine production, should continue to be investigated at the cellular, animal and volunteer level.
10. The mechanisms of damage to DNA occurring at longer UVR wavelengths, particularly the more penetrating UVA, needs further study. In this context, the role of both potential endogenous sensitisers, such as melanin, and potential exogenous sensitisers,such as some medications, should be examined. Action spectra of DNA damage in the different cell types in the human skin following volunteer exposures, would be of value. In addition, studies of DNA repair bias, mechanisms of error-prone repair, and the molecular cloning and characterisation of the relevant genes and their implications for human health should receive attention.
11. The identification of target genes has been valuable in understanding the development of a range of cancer types. The screening of oncogene and tumour suppressor gene mutations already associated with skin cancers should be continued and efforts made to identify other genes which may be involved.
12. Experimental carcinogenesis studies should principally be directed towards the development of an adequate model of human malignant melanoma and an understanding of the influence of human genetic predisposition, particularly in the case of familial melanoma.
13. The mechanisms by which UVR promotes skin carcinogenesis have not been investigated in any detail, but probably involve activation of cellular signal transduction pathways leading to clonal expansion of initiated cells. The identification of intracellular and extracellular signalling events involved in this process, and the determination of the spectral dependence for their activation, need further study. Another consequence of these signalling events that requires investigation is the reported increase in genetic instability, a process that could be involved in the accumulation of genetic changes required for progression to the malignant phenotype. Moreover, UVR-induced epigenetic changes, such as alterations in DNA methylation patterns, may also be relevant to phenotypic changes and should be pursued.
14. There is scope for studying the mechanism by which UVR causes irreversible changes in proteins at the surface of the eye in climatic droplet keratopathy and the manner in which UVR energy absorption within the lens results in different forms of cataract. A study of the relative importance to UVR-induced cataract of DNA damage or damage to lens membranes or proteins would also be of value. Further work is required on the action of photosensitisers in enhancing UVR damage to ocular tissues.
15. Further cellular, animal and volunteer studies are necessary to clarify the nature of the changes induced in immune response by UVR exposure and their role in skin tumorigenesis, photosensitivity and disease, and whether there are groups that are genetically predisposed to UVR-induced immunosuppression.

Epidemiology

16. One priority for research must be to delineate the relationship between UVR in solar radiation and the risk of cutaneous melanoma since malignant melanoma is the main proven fatal consequence of exposure to sunlight in Britain. Another is to explore further the postulated relationship of UVR with non-Hodgkin's lymphoma and chronic lymphatic leukaemia.
17. Issues requiring clarification are: the types of exposure of aetiological importance; whether tanning, other skin changes consequent on exposure to solar radiation and sunscreen preparations can alter the risk; the extent to which exposures at specific ages, e.g. childhood, are of particular importance.
18. Monitoring of trends in incidence of and mortality from melanoma and non-melanoma skin cancers needs to be continued. In particular, improved methods to gain reliable population-based data for non-melanoma skin cancer are required.
19. Since numbers and unusual appearance of melanocytic naevi are the principal risk factors known for melanoma, research is also needed on the relationship of UVR exposure, especially in childhood, to their development.
20. Epidemiological research is needed into the relationship of exposure to artificial UVR sources, such as lamps and sunbeds, especially for tanning, to risks of cutaneous melanoma and non-melanoma skin cancer, ocular melanoma, cataract and retinopathy, including age-related macular degeneration.
21. The risks if any, to the eye, of intense sources of artificial illumination should also be studied. Such investigations are required, firstly, because the risk, unlike that from solar UVR, is in relation to an artificial source about which clear, preventive action could be taken if a risk were demonstrated, and, secondly, because, unlike solar UVR, the relation to artificial sources with different spectra gives the best prospect of discovering in human beings the carcinogenicity of different parts of the UVR spectrum.
22. Although cataract is a far less serious disease for the individual than melanoma, the high incidence of cataract at older ages makes it important to continue research on the relationship of cataract risk to UVR exposure. There is a particular need to determine whether solar UVR exposure in temperate latitudes represents a risk for cataract. Large-scale studies are required to determine the contribution of chronic UVR and other optical radiation to age-related macular degeneration.

Risk estimates

23. The creation of quantitative guidelines limiting exposure to UVR must depend on sound information on dose-response relationships. Further models are therefore needed for assessing the risk to human beings of the various adverse health effects following exposure to UVR which incorporate information on action spectra. In particular, the use of erythemal action spectra as a surrogate estimate of the action spectra for other biological responses, particularly skin cancers, needs to be validated.

Additional information

 

Last reviewed: 6 October 2009