National institutes of health summary of the consensus development conference on sunlight, ultraviolet radiation, and the skin

Special article III

I

II

I

I

II

National Institutes of Health Summary of the Consensus Development Conference on Sunlight, Ultraviolet Radiation, and the Skin Bethesda, Maryland, May 8-10, I989 Consensus Development Panel* Bethesda, Maryland

Until the twentieth century, the sun was the main source of human skin exposure to energy within the photobiologic action spectrum. More recently, artificial devices capable of mimicking the emission of some or all of the solar spectrum have been introduced, compounding the opportunities and risks of UV radiation (UVR) exposure. Despite the undeniable importance of exposure to UVR for vitamin D homeostasis, there is little evidence to indicate that there are additional beneficial effects of such exposure. Indeed, overwhelming evidence exists that the skin is damaged in many ways by its direct exposure to natural or artificial UVR. Some exposure is unavoidable and is dramatically dissimilar in different populations depending on climate, geography, occupation, and recreational activities. The consequences of this exposure are also influenced by the degree of melanin pigmentation. The effects of UVR can be divided into two gen*Members of the Consensus Development Panel were David R. Bickers, MD, Case Western Reserve University and University Hospitals; Donald E. DeWitt, MD, East Carolina University School of Medicine; Robert S. Gilgor, MD, University of North Carolina School of Medicine; Pearl E. Grimes, MD, King Drew Medical Center, University of California at Los Angeles; Karen A. Holbrook, PhD, University of Washington School of Medicine; Robert Katz, MD, Georgetown University School of Medicine; Robert W. Makuch, PhD, Yale University School of Medicine; Michael Martin, MD, MPH, MBA, University of California at San Francisco; Lawrence A. Schaehner, MD, University of Miami School of Medicine; Thomas Slaga, PhD, University of Texas M. D. Anderson Cancer Center; Maria L. Chanco Turner, MD, George Washington University Medical Center; Theodore P. Williams,PhD, Florida State University; Vincent A. Ziboh, PhD, University of California at Davis; and Gail M. Zimmerman, National Psoriasis Foundation. No reprints available. 16/1/25517

608

eral types, acute and chronic. Acute effects include sunburn; chronic effects include, among others, the development of certain forms of skin cancer. In addition, UVR is capable of affecting the immune system via its effects on the skin. The skin is also susceptible to degenerative changes evoked by chronic UVR. Expanding knowledge about the hazards of exposure to sunlight and UVR has been accompanied by the development of more effective sunscreen formulations. In addition, pharmacologic agents such as the retinoids may be capable of inhibiting the development of or possibly even reversing certain chronic effects of cutaneous sun exposure. Considerable controversy remains concerning the specific adverse effects caused by various wavelengths of UVR, the magnitude of the adverse effects, and potential strategies for their prevention, treatment, or both. A Consensus Development Conference was undertaken in an effort to define the specific interactions of sunlight, UVR, and the skin as well as to identify methods for preventing and/or treating the adverse effects of UVR. Sponsored by the National Institute of Arthritis and Musculoskeletal and Skin Diseases, the Office of Medical Applications of Research, the National Cancer Institute, and the National Institute of Child Health and Human Development of the National Institutes of Health (NIH), the Food and Drug Administration (FDA), and the Environmental Protection Agency (EPA), the conference brought together physicians, scientists, and other health care professionals, along with representatives of the public, on May 8-10, 1989. After 1 1/z days of presentations and discussions, members of the consensus panel formulated a draft statement in response to several questions:

Volume 24 Number 4 April 1991

9 What are the sources of the UVR, and is the extent of human exposure changing over time? 9 What are the effects of sunlight on the skin? 9 What factors influence susceptibility to UVR? 9 Can UV-induced changes be prevented? If so, how? 9 Are sunlight-induced adverse skin alterations treatable and/or reversible? If so, how? 9 What are the directions for future research? In the application of the recommendations of this consensus conference, it is important to recognize that special circumstances may exist for each patient. These may include unavoidable exposures to U V R or the inability to use certain of the preventive strategies. There are clearly some areas in which final recommendations cannot yet be made because of insufficient data. In these situations, physicians must use their judgment in advising patients. What are the sources of UVR, and is the extent of human exposure changing over time?

The sun emits a wide variety of electromagnetic radiation: infrared, visible, and U V A (320 to 400 nm), UVB wavelengths that reach the earth's surface are UVA and UVB. U V A radiation is 1000-fold less effective than UVB in producing skin redness. However, its predominance in the solar energy reaching the earth's surface (10-fold to 100-fold more than UVB) permits U V A to play a far more important role in contributing to the harmful effects of sun exposure than previously suspected. Sunlight is the greatest source of human U V R exposure, but the extent of a person's exposure varies widely. There is greater U V R exposure with decreasing latitude. Residing at higher altitude results in a greater U V R exposure (e.g., for every 1000 feet above sea level, there is a compounded 4% increase in U V R exposure). U V R exposure increases with decreased stratospheric ozone. Other factors that influence exposure to U V R include heat, wind, humidity, pollutants, cloud cover, snow, season, and time of day. Solar flares (sunspots) also alter the amount of U V R reaching the earth. Solar flares increase ozone concentration in the stratosphere (above 50 kin), thereby reducing the amount of surface UVB. This 11-year cycle of solar flares causes as much as a 400% variation in UVB at 300 nm reaching the earth. When solar flares are inactive, there is a decrease in the ozone concentration, allowing in-

Sunlight, UVR, and the skin 6 0 9 creased UVB to penetrate to the earth's surface. There is also serious concern about depletion of stratospheric ozone by man-made chlorofluorocarbons (CFCs). In a recent risk assessment d o c u m e n t , the E P A predicted that without controls on C F C production, a 40% depletion of ozone would occur by the year 2075. The EPA further concluded that for every 1% decrease in ozone, there will be a compounded 2% increase in the more damaging shorter UVB wavelengths reaching the earth's surface. This is predicted to result in an additional 1% to 3% increase per year in nonmelanoma skin cancer (NMSC). Recent measurements indicate a worldwide decrease in stratospheric ozone.during the last decade and a seasonal hole in the ozone layer over the A n t arctic secondary to its destruction by CFCs. Although increased surface U V B has been m e a s u r e d in the Antarctic, there has not been a measurable change in UVB in the stratosphere over the U n i t e d States. The most common sources of artificial U V R exposure are various kinds of lamps used primarily for recreational tanning and phototherapy for skin diseases. U V R lamps can emit UVA, UVB, a n d / o r UVC. Some U V A lamps generate more than five times more U V A per unit of time than does solar U V A radiation reaching the earth's surface at the equator. At these doses, "pure U V A " is likely to have adverse biologic effects. However, U V B remains a potential problem with most of these sources. Even 1% UVB emission from a U V A source can cause a significant increase in the potential for skin cancer. The tanning industry is rapidly growing in t h e United States. Currently, more than 1 million Americans use commercial tanning facilities every day. The biggest categories of users are adolescents and young adults, especially women. The use of artificial U V sources in the phototherapy for cutaneous diseases has increased substantially in recent years and has exposed these patients to markedly increased doses o f U V R . Epidemiologic studies have shown an unequivocal dose-dependent increase in the incidence of N M S C , especially squamous cell carcinoma (SCC). Another potential source of artificial U V R is unshielded fluorescent bulbs. An unresolved issue is t h e amount of U V A they emit and the long-term effects of this exposure.

610

Consensus Development Panel

What are the effects of sunlight on the skin?

Marked morphologic changes occur as consequences of exposure to UVR. It is unclear how much exposure and how much time is required to produce these changes, although clinically normal-appearing skin can show pathologic signs of sun damage. Persons with fair complexions are more susceptible to this damage. Sun-exposed epidermis becomes thickened as much as twofold and is disorganized, showing evidence of hyperkeratosis, parakeratosis, and acanthosis. Keratinocytes lose their typical alignment and progressive flattening, show inclusions in the nucleus, and accumulate excessive amounts of melanosome complexes above the nucleus (capping). At the ultrastructural level, clumped keratin filaments and alterations in electron density of some basal cells are characteristic. Keratinocytes of the more differentiated epidermal layers (upper spinous, granular, and cornified) show few, ff any, cytologic changes. There are no data indicating altered keratinocyte differentiation as a result of sun exposure. Furthermore, it is not known how UVR interacts with lightabsorbing molecules within the keratinocytes. The melanocyte is the primary cell involved in photoprotection of the skin. In sun-damaged epidermis, these cells enlarge, increase in number, and migrate to higher levels of the epidermis. UVR also affects Langerhans cells in both animal and human skin by altering their immunologic function. Even low doses of UVB can reduce their antigen-presenting capability, block the normal effector pathway, and evoke an inappropriate response by activating T suppressor networks. It is unclear whether UVR affects Langerhans cells both directly and indirectly through soluble factors released by damaged keratinocytes. The dermoepidermaljunction loses its rete ridges, thereby producing a flattened interface. This abutment is more susceptible to shearing forces. At the ultrastructural level, regions of reduplicated lamina densa are evident. This change is not unique to photodamage but is characteristic of trauma to the epidermis. UVR causes dermal damage such as alterations in architecture, matrix composition, vascular structure and function, and cellular activities. The connective tissue immediately beneath the epidermis contains large bundles of densely packed, normalappearing collagen fibrils. Beneath this region, a broad zone of electron-dense elastotic material is

Journal of the American Acadcrny of Dermatology

evident. There are no data that demonstrate how newly synthesized or degraded, previously existing elastic fibers contribute to this material. Abnormal collagen fibrils can be admixed with the elastotic substance. Other studies show changes in the type III/I collagen ratio and an increase in glycosaminoglycans. Fibroblasts appear to be metabolically active. It is not clear whether this is a transient response to the UVR or whether there is a change in cell phenotype that can be retained in vitro. Dermal vessels become dilated and leaky and accumulate excessive basement membrane-like material. Inflammatory cells collect around the vessels; mast cells are increased and may show evidence of degranulation and apparent physical associations with fibroblasts. Generation of the prostaglandins associated with UVB erythema produced within the first 6 to 12 hours can be blocked by topical nonsteroidal antiinflammatory agents such as indomethacin. These anit-inflammatory agents, however, cannot inhibit the delayed, post-24-hour erythema that is modulated by lipoxygenase products. The time-dependent release of varying mediators during the UV-induced inflammatory process underscores the need for further exploration into selective inhibitors of both the cyclooxygenase and lipoxygenase pathways in the prevention and treatment of sunburn erythema. Also associated with UVR is the appearance of dyskeratotic keratinocytes (sunburn cells) in the superficial layers of the epidermis. The mechanisms of the development of these cells are still unclear. Tanning is mediated by a combination of immediate pigment darkening (IPD) and delayed pigment darkening (DPD). IPD is caused by UVA and is due to photooxidation of preformed melanin. It is not protective against UVB erythema. DPD occurs approximately 72 hours after UVR exposure and does not afford much protection against UVB erythema and pyrimidine dimer formation. It is accompanied by an increase in the number of dopapositive melanocytes, an increase in the number and melanization of melanosomes, and an increase in dendricity of melanocytes. The degree of protection afforded by melanin is unclear. Persons with dark complexions are still susceptible to UVR-induced photodamage. UVR also increases the transfer of melanosomes from melanocytes to keratinocytes. After UVR exposure melanosomes diffusely distributed within keratinocytes collect above the nucleus, forming a "cap." DPD occurs with either UVB or UVA. DPD induced by UVB is more protective

Volume 24 N umber 4 April 1991

against UVB erythema than is DPD induced by UVA. Both UVB- and UVA-induced DPD protect equally well against UVB dimer formation. U V R modifies local and systemic immune responses, functionally alters Langerhans cells, and activates the suppressor T cell pathway. Soluble factors released from UV-irradiated epidermal cells also m a y be responsible for this altered immune response. In certain experimental systems, UVRinduced tumors transplanted into genetically identical animals are normally rejected. If these host animals are UV irradiated before transplantation, the t u m o r will be accepted. The role of U V R in the immunobiology of human skin cancer and particularly in susceptibility against ceratin cutaneous infectious diseases is unclear. Epidemiologic evidence is extensive to support the direct role sunlight plays in human skin cancer. Basal cell carcinomas (BCCs) are found primarily on sun-exposed areas. Furthermore, persons with a light complexion who sunburn have a higher incidence of tumors. There is even stronger evidence that sunlight contributes to development of SCCs. Although both BCCs and SCCs are more prevalent in geographic areas of high sun exposure, there is a m u c h greater increase in SCC. A reasonable correlation exists between sunlight exposure and melanoma, but the relation is not as clear as with NMSC. The incidence of N M S C and melanomas has been steadily increasing. Unlike NMSC, melanomas occur most frequently on the upper part of the back in men and lower extremities in women. Melanoma incidence does not follow a pattern of increased risk with cumulative U V R exposure whereas the incidence of N M S C does. In mice and guinea pigs, UVR induces mainly SCC, whereas both SCCs and BCCs are produced in rats. In general, U V R induces SCCs somewhat more effectively in young mice than in older ones. The cancer is preceded by photodamage to the epidermal DNA, inflammation, epidermal hyperplasia, and dysplasia. The induction of melanomas by U V R has been difficult, but the opossum may be a reasonable model. Experiments in animals indicate that UVB is m u c h more effective than UVA (320 to 400 nm) in causing NMSC. Nevertheless, U V A can induce D N A damage, erythema, and SCC in both pigmented and albino mice and in guinea pigs. The longer UVA wavelengths (UVA I: 340 to 400 nm) of the UVA spectrum may be less damaging than

Sunlight, UVR, and the skin 611 the shorter UVA wavelengths (UVA II: 320 to 340 am). What factors influence susceptibility to UVR? Genetic disorders such as keratinization and pigmentation can increase the susceptibility to U V R damage. In addition, numerous acquired diseases manifest increased susceptibility to UVR. Other significant factors that influence susceptibility to U V R damage include race, ethnicity, eye and hair color, and the tendency toward formation of freckles and nevi. Numerous systemic medications may also augment U V R susceptibility. Can UV-induced changes be prevented? If so, how? In 1978, there were more than 500,000 new cases of skin cancer. This is probably a substantial underestimate for 1989. The following measures can be taken to diminish the risk of U V R exposure: 9 The wearing of proper clothing 9 The proper use of physical and chemical sunscreenhag products 9 Enhancement of behavior that limits sun exposure. Fifty percent of a person's total lifetime UVR exposure occurs by 18 years of age. Therefore parental education is important. Time of day and time of year have a major impact on the extent of UVR exposure. For example, on a sunny day in June between 10:00 AM and 3:00 PM, 60% of the daily UVB radiation reaching the earth's surface arrives. If exposure during this time were minimized, a significant reduction in the number of NMSCs would almost certainly Occur.

9Awareness of photosensitizing medications and chemical o Awareness of the adverse effects of intentional UVR exposure Are sunlight-induced adverse skin alterations treatable and/or reversible? If so, how? To improve the features of chronic photodamage various therapies (e.g., chemical peels, topical 5-fluorouracil, a-hydroxy acids, all-trans-retinoic acid) have been tried. Although the beneficial cosmetic effects of some of these treatments have received wide publicity, data are insufficient to demonstrate sustained improvement, reversibility of tissue pathology, or perservation of normal skin function. No information is available on long-term toxic effects of these agents. Data demonstrate both prevention and

612

Consensus D e v e l o p m e n t Panel

Journal of the American Academy of Dermatology

potentiation effects of topical retinoids in the develo p m e n t o f U V R - i n d u c e d skin tumors in animals. L o n g - t e r m , large-scale studies o f n o r m a l persons in the general population are in progress.

Free, single copies of the complete NIH Consensus Statement on Sunlight, Ultraviolet Radiation and the Skin may be obtained from the Officeof Medical Applications of Research, National Institutes of Health, Building 1, Room 260, 9000 RockviUe Pike, Bethesda, MD 20892; phone: 301-496-1143.

W h a t are the d i r e c t i o n s f o r f u t u r e r e s e a r c h ?

Presenters at the conferencewere Paul R. Bergstresser,David L. Correll, Peter M. Farr, Irene E. Kochevar, Robert M. Lavker, P. Donald Forbes, Michael F. Holick, Margaret L. Kripke, John A. H. Lee, Kenneth H. Kraemer, Arthur J. Sober, Robert S. Stern, Edward A. Emmett, Joseph Scotto, Janice Longstreth, Robert T. Handren, Jr., Leonard C. Harber, Maxim F. Mutzhas, Richard W. Gange, Nicholas J. Lowe, Joseph S. Rossi, John J. Voorhees, Eugene J. Van Scott, and John H. Epstein.

T h e following are not listed in any particular priority. 9 Updated epidemiologic data on N M S C in the United States 9 Definition of the action spectra of sunlight and UVR in the pathogenesis of melanoma and the development of more suitable animal models 9 More precise definition of "photoaging" of the skin and a comparison of this with chronologic aging insofar as pathophysiologic mechanisms are concerned . Clearer definition of the immunologic effects of U V R exposure and its potential role in the development of skin cancer and cutaneous infections 9 Further study of the biologic effects of UVA radiation 9 Further study of the importance of UVB radiation in vitamin D homeostasis ~ Development of new approaches for behavior modification and education to reduce skin exposure to UVR during childhood and adolescence 9 Development of more effective sunscreens and anticarcinogenic agents 9 Monitoring of the risk/benefit ratio of long-term sunscreen use 9 Research on the age-related optical properties of skin and acute and chronic cutaneous responses to UVR 9 Quantitative assessment of U V R exposure in normal populations

Members of the planning committee were Alan Moshell, MD, National Institute of Arthritis and Musculoskeletal and Skin Diseases; David R. Bickers, MD, Case Western Reserve University and University Hospitals; Linda W. Blankenbaker, NIH; Patricia Blessing, National Institute of Arthritis and Muscoloskeletal and Skin Diseases; Delbert Dayton, MD, National Institute of Child Health and Human Development;John Epstein, MD, University of California at San Francisco; Carnot Evans, MD, FDA; Irwin Freedberg, MD, New York University Medical Center; Lowell Goldsmith, MD, University of Rochester School of Medicine; William H. Hail, NIH; Stephen I. Katz, MD, PhD, National Cancer Institute; Kenneth H. Kraemer, MD, National Cancer Institute; Janice Longstreth, PhD, ICF-Clement Associates, Inc.; Andrea Manning, Prospect Associates; John Parrish, MD, Massachusetts General Hospital; and Steven Sykes, FDA.