Photodermatology: progress, problems and prospects

ARTICLE

Major progress was made in the 1990s as there was a fundamental understanding of some key events in nonmelanoma skin cancer ­ p53 in SCC and ptc in BCC ­ specifically in terms of their involvement in apoptosis and their roles in skin homeostasis [1, 2]. During the 1990s we also became more fully aware of the hazards of sun exposure and the lack of universal protection by sunscreens [3]. In this period, sunscreens moved from being touted as the sole form of protection of sun exposed skin to one of three approaches (in addition to avoiding sun and covering up). In fact this retreat is further highlighted by the refrain "do not use sunscreens to extend your time in the sun" as a concern has arisen that even the suberythemal damage that occurs through sunscreen protected skin could have long term deleterious effects [4]. The recognition in the 80s and 90s of the effects of solar UVR on skin immune function led to a desire to demonstrate that sunscreens can also protect immune function. Several studies have been published [5]. All seem to indicate some level of protection but not at a level approaching the SPF number on the bottle [6]. This kind of data suggests that the wavelengths that cause erythema (what SPF is a measure against) and those that cause immune suppression are different (to some degree). Yet there has been listless development of new sunscreens.

El-Ghorr and Norval [7] have shown that the UV wavelength dependence for the suppression of contact hypersensitivity, delayed type hypersensitivity and cis-urocanic acid isomerization in mice skin differ from each other. Until similar studies are performed in human skin, it will be difficult to select a proper endpoint for the determination of the ability of a sunscreen to protect immune function. In this vein, Young and Walker recently noted that sunscreens are widely advocated to reduce the long-term effects of sun exposure [8]. They point out that these recommendations are based on an extrapolation from animal studies (as in the study by El-Ghorr and Norval) rather than an actual analysis of protective effects in human skin. On the other hand, Halliday and co-workers have used a nickel contact hypersensitivity model in human skin to evaluate the immune-protecting effects of sunscreens [9]. A recent study comparing nickel-induced reactions in irradiated and non-irradiated skin showed a linear dose dependence in both unprotected and sunscreen-protected skin. Protection factors were determined by dividing the immune function score for protected skin by that for unprotected skin.

In the coming years, we can look forward to additional advances in our understanding of the effects of UVR on human skin immune function. However, the latter is fraught with a unique set of problems. Specifically, there are many immune endpoints that can be measured. At some point it needs to be demonstrated which of these endpoints is the best match for the immune function in human skin that keeps pre-cancerous cell populations in check.

A burning issue for which there is little molecular information concerns sunscreen use and melanoma, obviously a very important public health issue. In an excellent review Weinstock has surveyed the findings of several epidemiology studies on sunscreen use and skin cancer [10]. Late in 1998, Autier et al. showed that more nevi developed in children who used sunscreens [11]. The kind of criticisms this study was met with has recently been summarized by Bigby in the Archives of Dermatology [12]. In December 1999 the Archives of Dermatology initiated a new section called Evidence-based Dermatology. The purpose of the new section is "to alert clinicians to important advances in dermatology by selecting from the biomedical literature those original and review articles whose results are most likely to be true and useful. These articles are summarized in structured, value-added abstracts and commented on by knowledgeable clinicians". One of the first contributions summarized Autier's findings and cautioned that the uproar and knee-jerk reaction should be reined in [13]. This is a major advance towards developing a dialog between hypothesis driven scientists and the advocates of a blanket recommendation for skin protection without a full awareness of the scientific facts as a backing.

Regarding sunscreen use, some would say there has been more heat than light shed on this topic by the spate of media reports in the late 1990s. Photobiology and epidemiology studies have not confirmed that covering skin with sunscreens will necessarily prevent skin cancer [14] although there is some evidence for the prevention and/or remission of actinic keratoses¹ [15]. This multifaceted issue is related to three phenomenon. First we do not know the action spectra for all of the effects that sunlight can cause in human skin. Therefore we do not know which wavelengths need to be filtered and to what extent. Second, laboratory studies have not used biologically-relevant light sources to mimic the effects of sunlight (see below). Third, the relationship between DNA damage and immune suppression needs to be better understood.

On the therapeutic front, photobiology has been and will continue to be an active part of dermatological research. As evidence for this it is noteworthy that the number of photobiology based papers in the Journal of Investigative Dermatology increased during the latter part of the 1990s.

PUVA use has been receding ­ partly due to alarmist reporting about skin cancer in PUVA treated psoriasis patients (especially in terms of melanoma incidence) [18]. Molecular epidemiology studies indicate that the cause and effect between PUVA and skin cancer might not be as straightforward as anticipated [19]. There has been a concomitant interest in developing other therapies for psoriasis. In as much as psoriasis is an immunological disorder, a successful therapy will impact the immune system and hence replacement therapies may be fraught with the same long term consequences.

In the 1990s while one form of psoralen photochemotherapy faded sightly, another was coming to the forefront in a limited way. Photopheresis, an extracorporeal form of psoralen photochemotherapy [20], has become the treatment of choice for certain stages of cutaneous T cell lymphoma [21]. In addition, there have been suggestions that it may be effective for scleroderma [22], graft versus host disease [23] as well as the prevention of transplanted organ rejection [24].

In the latter part of the 1990s, after years of development, another phototherapy employing a range of different photosensitizers has reached the clinic. Photodynamic therapy (PDT) employs organic chemicals which absorb visible radiation [25]. Among those receiving much study are photofrin, aminolevulinic acid and the texaphyrins. Recently it was announced that the FDA had approved the first drug-device combination for the treatment of actinic keratosis [26]. The treatment combination consists of a topically applied solution of aminolevulinic acid (ALA) to be used in conjunction with an activating light source (product name ­ Levulan Kerastick). In a multi-center study it has been shown that 77% of patients treated with this photodynamic therapy had at least a 75% reduction in lesions. 66% had complete resolution of treated lesions after 8 weeks compared to ~ 12% in vehicle-treated controls. The topically applied ALA is rapidly metabolized by fast-growing AK cells and converted in situ to a photosensitizing form of porphyrin (protoporphyrin IX). While ALA and photofrin are derived from natural products, another approach has been to synthesize highly specific metal binding compounds to improve upon these. Expanded porphyrins ("Texas-size" texaphyrins) have been designed and synthesized so that they could accommodate large metal ions [27].

In another photobiology approach, a portion of the UVA spectrum (340-400 nm, UVA-1) has been shown to be capable of ameliorating certain skin diseases [28]. High doses of UVA-1 which can penetrate to the basal layer and beyond have been used successfully for scleroderma [29] and CTCL [30] ­ but long term sequellae are unknown at this time. High dose UVA-1 therapy may achieve therapeutic responses by penetrating deep into the dermis without the usual side effects caused by less penetrative UVB-like wavelengths in the UVA-2 region (320-340 nm).

For additional meaningful progress to be made in skin photobiology, there is an essential issue that must be addressed. While laboratory studies on skin photobiology have advanced by implementing modern molecular biology techniques, there have not been similar advances in the kind of UVR sources that are employed in these studies. There is a profound need for a cost-effective artificial source of solar radiation. Photobiological studies depend on the availability of light sources to provide radiation of appropriate wavelengths and of sufficient energy to conduct experiments in a facile and reproducible manner. Pioneering experiments utilized mercury arc lamps and carbon arc lamps but these were bulky and expensive and were not commonly available. Eventually solar simulators became available and although they provided an excellent representation of the solar spectrum, due to their cost they too were not accessible for routine research studies. The development of fluorescent sunlamps (FS²) greatly improved the ability to perform UV related studies [31]. Due to their low cost, easy set up and replacement, sunlamps have become the most commonly used source of UV radiation for photobiology research. Although they are convenient to use, a drawback is that their output is a poor representation of the solar spectrum. These lamps contain significant amounts of UVC which is not found in sunlight at the surface of the earth. Approximately 52% of the total emission spectrum from these lamps is in the UVB (285-290 to 315-320 nm) range; 45% in the UVA (315-320 to 400 nm) range; and 3% in the UVC (250 to 285-290 nm) range. Often the contributions of UVC and UVA components are discounted based on the small percentage of the total output for the former and the supposed relative ineffectiveness for the induction of biological effects for the latter. An example of the potent impact of a small percentage of higher energy photons is illustrated in a study which showed that the 0.8% ultraviolet B content of an ultraviolet A sunlamp induced 75% of cyclobutane pyrimidine dimers in human keratinocytes in vitro [32].

While it may be tempting to say that "a photon is a photon is a photon", it must be remembered that the effects of photons on skin biology are wavelength dependent. Typically different effects in skin have different wavelength dependencies. For this reason it is important in any study of skin photobiology to use a source that closely mimics the spectrum of the sun, otherwise the unnatural source may over-emphasize one molecular phenomenon in favor of another.

Some have argued that there is no representative solar spectrum to mimic. Solar irradiance varies by geographical location, season of the year and time of the day. Yet even when such extremes are considered, the terrestrial solar spectrum does not resemble anything like that of a sunlamp. Figure 1 shows the spectral irradiances for the sun (panel A) and for an FS-40 sunlamp³ (panel B) over the wavelength range 250-375 nm. Also shown are the demarcations for the regions of the UV spectrum (UVC boundary at 290 nm, UVB at 320 and UVA at 400 nm⁴). It is not uncommon for data obtained using unfiltered sunlamps to be published. Because UV induced DNA photochemistry is strongly wavelength dependent, some of these published studies may not be relevant to human photobiology caused by solar exposure [34]. Recently it was shown that some discrepancies in published data could be resolved by removing the non-solar type UVC (275-290 nm) portion of the spectrum of these lamps using an acetate filter [33]. It has been shown that these wavelengths make a significant contribution to the experimental effects. Figure 1 also compares the solar spectral irradiance to that for Kodacel-filtered sunlamps. The proportion of shorter UVB wavelengths (290-300 nm) emanating from Kodacel-filtered sunlamps is still significantly greater than in the solar spectrum. Thus, even with the removal of the UVC and some short UVB wavelengths, the sunlamps are not an adequate model for the UVB portion of sunlight. As mentioned above, there are different kinds of sunlamps. The investigator must be familiar with the spectral irradiance of a specific lamp and use appropriate filters for biological relevance. Finally investigators must not rely on manufacturers' data sheets for specific information (other than as a rough guide for selection). Spectroradiometric measurements need to be performed under the exact conditions in which the lamps are used. Radiometric measurements must be performed routinely as the lamp output may vary due to daily variations in line voltage.

Sunlamps were not originally developed for photobiology experiments [36]. However, they were readily adapted as perceived reasonable substitutes for sunlight many years ago. As our understanding of photochemical and photobiological effects in skin has advanced it can be appreciated that their distorted spectrum relative to sunlight means that data obtained using this convenient source, while some-what informative during the early years of modern skin photobiology, now appears to be severely deficient. Today, the number of fluorescent phosphors for light bulb construction vastly surpasses those available 30 years ago when sunlamps were first implemented in photobiology. One example is the UVA-340 fluorescent lamp from
Q-panel (Cleveland, OH). A recent publication showed that UVA-340 lamps (Q-panel, Cleveland OH) closely mimicked the UVB portion of the solar spectrum and a significant part of the UVA component of the solar spectrum [37]. Figure 1 (panel C) shows the output of the UVA-340 lamps. Even though these lamps are labeled UVA-340 they are seen to represent the UVB portion remarkably well. The near coincidence in the region 295-320 illustrates its remarkable similarity to sunlight and would suggest this lamp (or an equivalent) would be nearly an ideal replacement for fluorescent sunlamps in photobiology experiments designed to study the effects of UVR in skin or isolated skin cells. In a recent study, Brown et al. compared the ability of various sources to activate the human elastin promoter in cells containing a CAT reporter [38]. When FS lamps were employed, apparently low doses (10-20 J/cm²) were able to induce optimal CAT activity. Filtering the FS with Kodacel raised the dose delivered for optimal CAT activity to 50-60 mJ/cm². Using the more solar-like UVA-340 lamps induced lower levels of CAT activity even though the apparent doses delivered were significantly greater than for either the FS or KFS lamps. The UVB probe was employed as typically used by most users. The lights were turned on and after a 15 min warm-up period, the probe was positioned and a meter reading was taken. But there is a flaw. Meters and sources must be matched in terms of their respective output and spectral sensitivity. In the cited study, the probe best matched the output of the KFS lamps. Both the FS and UVA-340 lamps emitted photons beyond the spectral response range of the probe. For the former, these were highly energetic UVC and shorter wavelength UVB photons, while for the latter they were primarily lower energy longer wavelength UVA photons. Hence in this study to rectify this "erroneous" dosimetry, a second independent assay was performed. When DNA from parallel treated cells was analyzed for photoproduct formation by an ELISA method, it was shown that the induction of CAT activity correlated with the level of induced photoproduct formation regardless of the source employed (Table II).

It is important to remember that when FS lamps were being developed almost a half century ago, the major focus of photobiology research in dermatology was concerned with the detection of gross physiological effects in skin. In some cases this was not possible with sunlight (either because of long exposure times needed or the unpredictable availability of the sun). FS lamps with their heavy weighting of very short wavelengths of the UVB portion of the spectrum allowed the performance of photobiology experiments in a convenient time frame. Now that we have moved from the measurement of gross physiological responses of skin to UVR to more sensitive molecular biology assays, it is time to employ biologically and physiologically relevant sources that mimic natural sunlight. Some may retort "which sun" as the spectrum of the sun reaching the surface of the earth varies by geographical location, season of the year, and time of day. While this variation cannot be denied, nowhere under any set of conditions does the terrestrial solar spectrum approach the output of either the FS or KFS lamps⁵.

This article would be deficient if it did not touch on the issue of artificial tanning. This is another area where more heat than light has been generated in recent years. As intuitively obvious as it may seem, the message to the public to avoid actinic damage by limiting sun exposure and/or the use of artificial tanning sources has fallen on deaf ears for a majority of the population. What could be the reason for this? In this era of heightened health consciousness, it would be expected that such a straightforward public health message would be adhered to by the vast majority. Could it be that scare tactics have turned into the "boy-who-cried-wolf" syndrome? After all, everyone has relatives or friends who have spent a lot of time in the sun (may be even frequenting tanning salons) who never get a skin cancer! For example, it appears that outdoor workers become acclimated to sun exposure by a variety of mechanisms. In fact most skin adapts to chronic (not necessarily excessive) sun exposure. Recent skin cancer statistics indicate that skin cancers of any kind were found in less than 15% of a screened population [39]. Certainly many of the 85% without skin cancers may have avoided sun (at least in their adult years), yet many (most likely the vast majority) have probably spent a considerable amount of time in the sun during their lifetime. Another factor that may explain the artificial tanning fad was recently proposed by Gee and Dufresne who analyzed season tanning patterns and found that the users suffered from moderate to severe symptoms of season affective disorder (SAD) [40]. Thus it would appear that some users of tanning beds are self-medicating themselves. It is important note while there may be some positive effect derived from tanning beds in terms of relief of SAD, the action spectrum for mood effects lies in the visible region [41] thus the beneficial effects could be obtained using other sources that would not risk the health of the patients' skin.

In recent years an entire industry has grown up around the concept of acquiring and/or maintaining tanned skin by exposure to artificial sources of UVA. There is limited direct data indicating any long term ill outcome in terms of skin cancer although there is evidence of significant actinic-type damage. Almost certainly excessive tanning will lead to photoaged skin in these frequent tanners. In addition an unknown percentage will have an increased incidence of skin cancers however, it will not be a simple task to deconvolute the relative contributions of natural solar exposure and artificial tanning. At this time, actual data is not available to determine whether the incidence of skin cancer in artificial tanners will be significantly greater than the general population and the extent to which artificial tanning made a contribution.

<sup>____________________________

1</sup> It has been estimated that fewer than 0.1% of actinic keratoses develop into SCC, other AK appear to undergo spontaneous remission [16].

² There are many manufacturers of these products. The most commonly referred to sunlamp is one that emits radiation from 270-375 nm.

³ The number refers to lamp wattage.

⁴ Some advocate the use of 280 nm as the boundary between UVC and UVB. It is important to note that this boundary was defined on the basis of optical filter transmissions more than half a century ago. As we learn more about skin photobiology, their inadequacy has become apparent. Therefore, in this paper we have adopted Coohill's approach to defining the UVC/UVB boundary as 290 nm since this is close to the shortest wavelengths detected at the surface of the earth (see ref. 33). Furthermore it should be noted that the NSF biosphere monitoring of solar UV uses 290 nm as the cutoff between UVC and UVB.

5 Figure 1 (panel A - dotted line) also shows the solar irradiance on a typical day for Mauna Loa, the site of the greatest daily UVB dose on earth. The greater amount of UVB (compared to CT shoreline) is evident, yet this extreme difference barely moves the spectrum towards that of FS lamps.

⁶ For a recent example, see ref. 42 and references therein.

CONCLUSION

There will always be an interest in using UV and visible radiation to treat skin diseases because the skin is so easily accessible to light treatment. Furthermore it is clear that there is much more to be learned about the response of human skin to sunlight. To make these advances it will be necessary to employ artificial sources of UVR in laboratory settings. While no one could have asked pioneering photobiologists from a half century ago to see beyond the limits of their own time, modern photobiologists must set higher standards so that today's results will remain meaningful when examined by future generations. Finally, more rigorous standards must be set and adhered to by journal editors and reviewers. The use of inappropriate sources and non-physiological doses must be avoided. Most importantly, incorrect interpretations must be met with a response from the field. Erroneous studies, once published, are difficult to remedy⁶ [42]. Perhaps in the bourgeoning use of the internet in all aspects of science, there will develop a feedback system on the usefulness of published data.

Article accepted on 6/3/00

Acknowledgements

We thank Barry A. Bodhaine, NOAA-CMDL, for the solar spectrum from Mauna Loa (Fig. 1A).

REFERENCES

1. Brash DE, Ziegler A, Jonason AS, Simon JA, Kunala S, Leffell DJ. Sunlight and sunburn in human skin cancer: 53, apoptosis, and tumor promotion. J Invest Dermatol (Symposium Proceedings) 1996; 4: 136-42.

2. Johnson RL, Rothman AL, Xie J, Goodrich LV, Bare JW, Bonifas JM, Quinn AG, Myers RM, Cox DR, Epstein EH Jr, Scott MP. Human homolog of patched, a candidate gene for the basal cell nevus syndrome. Science 1996; 272: 1668-71.

3. McGregor JM, Young AR. Sunscreens, suntans and skin cancer. BMJ 1996; 312: 1621-2.

3. a) Bissett DL, McBride JF, Hannon DP, Patrick LF. Time-dependent decrease in sunscreen protection against chronic photodamage in UVB-irradiated hairsless mouse skin. J Photochem Photobiol B Biol 1991; 9: 323-34. b) Reeve VE, Bosnic M, Boehm-Wilcox C. Effect of ultraviolet (UV) radiation and UVB-absorbing sunscreen ingredients on 7,12 dimethylbenz(a)anthracene-initiated skin tumorigenesis in hairless mice. Photoderm Photoimmunol Photomed 1990; 7: 222-7. c) Sunscreen Photobiology (Edited by FP Gasparro), Landes Press, Georgetown TX.

4. Lowe NJ, Meyers DP, Wieder JM, Luftman D, Borget T, Lehman MD, Johnson AW, Scott IR. Low doses of repetitive ultraviolet A induce morphologic changes in human skin. J Invest Dermatol 1995; 105: 739-43.

5. Wolf P, Kripke M. In: Sunscreen Photobiology. Gasparro FP, ed. Landes 1997. b) Halliday GM, Bestak R, Yuen KS, Cavanagh LL, Barnetson RS. UVA-induced immunosuppression. Mutat Res 1998; 422: 139-45.

6. Davenport V, Morris JF, Chu AC. Immunologic protection afforded by sunscreens in vitro. J Invest Dermatol 1997; 108: 859-863. b) Gasparro FP. Photobiology 101. J Invest Dermatol 1998; 110: 183.

7. El-Ghorr AA, Norval M. The effect of UV-B irradiation on primary and secondary HSV-1 infections in interleukin-4 knockout mice. Arch Dermatol Res 1999; 291: 459-65.

8. Young AR, Walker SL. Sunscreens: photoprotection of non-erythema endpoints relevant to skin cancer. Photodermatol Photoimmunol Photomed 1999; 15: 221-5.

9. Damian DL, Barnetson RS, Halliday GM. Measurement of in vivo sunscreen immune protection factors in humans. Photochem Photobiol 1999; 70: 910-5.

10. Weinstock MA. Sunscreens for melanoma prevention. Photodermatol Photoimmunol Photomed 1999; 15: 209-11.

11. Autier P, Doré JF, Cattaruzza MS, Renard F, Luther H, Gentiloni-Silverj F, Zantedeschi E, et al. Sunscreen use, wearing clothes, and number of nevi in 6- to 7-year-old European children. J Natl Cancer Inst 1998; 90: 1873-80.

12. Bigby M. Welcome to the "Evidence-based dermatology" section in the Archives of Dermatology. Arch Dermatol 1999; 135: 1512-3.

13. Bigby M. The sunscreen and melanoma controversy. Arch Dermatol 1999; 135: 1526-7.

14. Yound AR, Walker SL. Protoprotection form UVR-induced immunosuppression. In: Photoimmunology. Krutmann J, Elmets CA, eds. Blackwell Science, Cambridge MA, 1995: 22.

15. Thompson SC, Jolley D, Marks R. Reduction of solar keratoses by regular sunscreen use. N Engl J Med 1993; 329: 1147-51.

16. Cohn BA. From sunlight to actinic keratosis to squamous cell carcinoma. J Am Acad Dermatol 2000; 42: 143-4.

17. Stern RS, Thibodeau LA, Kleinerman RA, et al. Risk of cutaneous carcinoma in patients treated with oral methoxsalen photochemotherapy for psoriasis. N Engl J Med 1979; 300: 809-13.

18. Stern RS, Nichols KT and Väkevä LH for the PUVA follow-up study. Malignant melanoma in patients treated for psoriasis with methoxsalen (psoralen) and ultraviolet A radiation (PUVA). N Engl J Med 1997; 336: 1041-5.

19. Wang XM, McNiff JM, Klump V, Asgari M, Gasparro FP. An unexpected spectrum of p53 mutations from squamous cell carcinomas in patients treated with PUVA. Photochem Photobiol 1997; 66: 194-9.

20. Gasparro FP. Extracorporeal Photochemotherapy : clinical aspects and the molecular basis for efficacy. RG Landes (Medical Intelligence Unit), 1994.

21. Rook AH, Suchin KR, Kao DM, Yoo EK, Macey WH, DeNardo BJ, Bromely PG, Geng Y, Junkins-Hopkins JM, Lessin SR. Photopheresis: clinical applications and mechanism of action. J Investig Dermatol Symp Proc 1999; 4: 85-90.

22. Schwartz J, Gonzalez J, Palangio M, Klainer AS, Bisaccia E. Extracorporeal photochemotherapy in progressive systemic sclerosis: a follow-up study. Int J Dermatol 1997; 36: 380-5.

23. Greinix HT, Volc-Platzer B, Rabitsch W, Gmeinhart B, Guevara-Pineda C, Kalhs P, Krutmann J, Honigsmann H, Ciovica M, Knobler RM. Successful use of extracorporeal photochemotherapy in the treatment of severe acute and chronic graft-versus-host disease. Blood 1998; 92: 3098-104.

24. Salerno CT, Park SJ, Kreykes NS, Kulick DM, Savik K, Hertz MI, Bolman RM 3rd. Adjuvant treatment of refractory lung transplant rejection with extracorporeal photopheresis. J Thorac Cardiovasc Surg 1999; 117: 1063-9.

25. Fritsch C, Goerz G, Ruzicka T. Photodynamic therapy in dermatology. Arch Dermatol 1998; 134: 207-14.

26. Mechcatie EFDA. Approves first drug-device combo for AKs. Skin & Allery News 2000.

27. Woodburn KW, Fan Q, Kessel D, Luo Y, Young SW. Photodynamic therapy of B16F10 murine melanoma with lutetium texaphyrin. J Invest Dermatol 1998; 110: 746-51.

28. Krutmann J. High-dose ultraviolet A1 (UVA1) phototherapy: does it work? Photodermatol Photoimmunol Photomed 1997; 13: 78-81.

29. Stege H, Berneburg M, Humke S, Klammer M, Grewe M, Grether-Beck S, Boedeker R, Diepgen T, Dierks K, Goerz G, Ruzicka T, Krutmann J. High-dose UVA1 radiation therapy for localized scleroderma. J Am Acad Dermatol 1997; 36: 938-44.

30. Plettenberg H, Stege H, Megahed M, Ruzicka T, Hosokawa Y, Tsuji T, Morita A, Krutmann J. Ultraviolet A1 (340-400 nm) phototherapy for cutaneous T-cell lymphoma. J Am Acad Dermatol 1999; 41: 47-50.

31. Elkind MM, Han A, Chang-Liu CM. "Sunlight"-induced mammalian cell killing: a comparative study of ultraviolet and near-ultraviolet inactivation. Photochem Photobiol 1977; 27: 709-15.

32. Woollons A, Kipp C, Young AR, Petit-Frere C, Arlett CF, Green MH, Clingen PH. The 0.8% ultraviolet B content of an ultraviolet A sunlamp induces 75% of cyclobutane pyrimidine dimers in human keratinocytes in vitro. Br J Dermatol 1999; 140: 1023-30.

33. Coohill TP. Action spectroscopy: Ultraviolet radiation In: CRC Handbook of Photochemistry and Photobiology. Hoorspool WM, Song PS, eds. CRC Press, Boca Raton, FL, 1995: 1267-75.

34. Drouin R, Therrien JP. UVB-induced cyclobutane pyrimidine dimer frequency correlates with skin cancer mutational hot spots in p53. Photochem Photobiol 1997; 66: 719-26.

35. Learn DB, Beard J, Moloney JS. The ultraviolet C energy emitted from FS lamps contributes significantly to the induction of human erythema and murine ear edema. Photodermatol Photoimmunol Photomed 1993; 9: 147-53.

36. Urbach F. Sources of radiant energy for medical uses. J Invest Dermatol 1959; 32S: 167-9.

37. Beasley DG, Beard J, Stanfield JW, Roberts LK. Evaluation of an economical sunlamp that emits a near solar UV power spectrum for conducting photoimmunological and sunscreen immune protection studies. Photochem Photobiol 1996; 64: 303-9.

38. Douglas B Brown, Peritz AE, Mitchell DL, Chiarello S, Uitto J, Gasparro FP. Fluorescent sunlamps: an inappropriate substitute for sunlight (submitted).

39. The year book on dermatology and dermatologic surgery. Thiers BH, ed. Mosby. NY NY, 1999, Table 19: 23.

40. Gee S, Dufresne R. The assessment of seasonal affective disorder in the tanning bed population: a survey of tanning bed users. J Invest Dermatol 1999; 112: 519A.

41. Rosenthal NE, Moul DE, Hellekson CJ, Oren DA, Frank A, Brainard GC, Murray MG, Wehr TA. A multicenter study of the light visor for seasonal affective disorder: no difference in efficacy found between two different intensities. Neuropsychopharmacology 1993; 8: 151-60.

42. Gasparro FP. Photobiology. J Invest Dermatol 1998; 110: 183.