ARTHRITIS & RHEUMATISM Vol. 50, No. 9, September 2004, pp 3045–3048 © 2004, American College of Rheumatology CONCISE COMMUNICATIONS the last decade, high-potency sunscreens that protect against UV-induced sunburn have become widely available. However, the degree to which these sunscreens protect against other unwanted effects of UV irradiation has been called into question (4,5). Therefore, we have routinely included sunscreens in our diagnostic testing protocol for LE. A recent study involving 11 patients suggested that sunscreens protect against UV-induced LE lesions (6). In order to obtain a more general assessment of the protective capacity of sunscreens in LE in a larger cohort of patients, we performed a retrospective evaluation of 66 patients who underwent photoprovocation testing between 1999 and 2002. After informed written consent had been obtained, testing was done (according to a standard protocol) on the upper back, with a field size of 8 ⫻ 5 cm (7). The doses applied were 100 mJ/cm2 of UVA (from a Dermalight UltrA1 source; Hoenle Medizintechnik, Munich, Germany) or the 1.5-fold minimal erythema dose of UVB (from a Waldmann UVB 800 unit; Munich, Germany), or a combination of both, on 3 consecutive days. An additional field was irradiated with combined UVA/ UVB 15 minutes after the application of a commercial sunscreen that efficiently filters UVA (Anthelios W30; La RochePosay, La Roche-Posay, France). This sunscreen contained a DOI 10.1002/art.20426 Use of sunscreens to protect against ultravioletinduced lupus erythematosus Photosensitivity is the leading symptom of cutaneous and systemic lupus erythematosus (LE). The role of sunlight in the induction of LE lesions has long been recognized, based on the observation that lesions are predominantly localized to sun-exposed areas of the body, especially the face, but also the head and neck area. Moreover, LE worsens during the summer period or in the weeks following increased sun exposure (1,2). Definite proof for a causative role of ultraviolet (UV) irradiation in the induction of LE came from experimental photoprovocation testing in the early 1990s (3). Those studies unraveled the fact that the UV action spectrum that provokes LE in sensitive individuals includes both the ultraviolet B (UVB; 290–320 nm) and the UVA (320–400 nm) range. Moreover, the studies showed that LE does not develop immediately after UV exposure, but after a delay of 2–3 weeks. In view of the central role of sunlight in provoking LE, it is important for patients to establish efficient UV protection throughout the year. This is a particular challenge for areas of the skin that cannot effectively be covered by clothing. Over Figure 1. Patient with chronic discoid lupus erythematosus, showing a positive reaction to ultraviolet B (UVB) and UVA/UVB 2 weeks after photoprovocation, but no reaction to UVA alone (top). No reaction to UVA/UBV was observed in the area to which sunscreen was applied (bottom). 3045 3046 CONCISE COMMUNICATIONS Table 1. Action spectra for the induction of LE lesions* Patient group UVA UVB UVA ⫹ UVB Chronic discoid LE (n ⫽ 33) Systemic LE (n ⫽ 4) Subacute cutaneous LE (n ⫽ 2) Tumidus-type LE (n ⫽ 12) Profound LE (n ⫽ 1) 9 (27) 3 (75) 1 (50) 4 (33) 0 (0) 16 (48) 3 (75) 1 (50) 4 (33) 1 (100) 32 (97) 4 (100) 2 (100) 12 (100) 1 (100) * Values are the number (%) of patients. The ability of different wavelength spectra to induce lupus erythematosus (LE) on the upper back is described in ref. 3. Only erythematosus infiltrates at week 2 or 3 were considered positive. UV ⫽ ultraviolet. formulation of different active ingredients shielding predominantly against UVA (parsol 1789), UVB (uvinul N539, uvinul T150), or against UVA and UVB (mexoryl XL, titanium dioxide). According to the recommendations of the US Food and Drug Administration and the European Cosmetic Toiletry and Perfumery Association, 2 mg/cm2 of sunscreen was applied. Readings were performed on days 2, 3, and 4 and again 1, 2, and 3 weeks after photoprovocation. Late readings are particularly important, because cutaneous LE tends to develop primarily 2–3 weeks after irradiation (1). Criteria for a positive reaction were a palpable erythematous infiltrate clinically resembling LE that developed with a crescendo time course, becoming most pronounced 2–3 weeks after photoprovocation. Approximately 75% of our LE patients showed a positive reaction to the UV challenge (Figure 1). Only those patients who showed photosensitivity (n ⫽ 52) were analyzed further. Most of these patients had discoid or tumidus-type LE. As described previously (2,3), we found the combined irradiation with UVA and UVB to be most effective in triggering cutaneous LE (98%). In fact, 30% of subjects reacted to the combined irradiation only and to neither UVA nor UVB when applied alone, suggesting a synergistic effect of the combination in some patients. In 19% of the patients, both UVA and UVB induced LE lesions. UVA alone induced LE in only 17 patients (33%). UVB induced LE in 48% of the patients. One patient showed a positive reaction exclusively to UVB and was excluded from further analysis (Table 1). In selected patients, histologic examination was performed to confirm the clinical diagnosis. Of 51 patients who had developed LE in the UVA/ UVB field, 49 (96%) were entirely protected by the sunscreen against the development of LE or LE-like lesions. Protected patients showed either no reaction or pigmentation only (Table 2). Thus, the capacity of the sunscreen to prevent UVinduced LE was higher than the protective action against UV-induced pigmentation. In a small number of patients, we investigated in a third area a high- potency sunscreen based primarily on physical protection (TiO2) and obtained similar results. Our data show that modern sunscreens that efficiently protect against both UVA and UVB are very effective in protecting against UV-induced LE. Preliminary data suggest that daily application of high-potency sunscreens also protects patients against UV-induced LE by natural sunlight. Our personal experience is mainly limited to only 2 sunscreens, but other high-potency sunscreens that protect against UVA and Table 2. Protective effect of sunscreen in patients who developed LE lesions at the site of combined UVA/UVB irradiation* Patient group LE lesion No reaction Pigmentation only Chronic discoid LE (n ⫽ 33) Systemic LE (n ⫽ 4) Subacute cutaneous LE (n ⫽ 2) Tumidus-type LE (n ⫽ 12) Profound LE (n ⫽ 1) Total 2 (6) 0 (0) 0 (0) 0 (0) 0 (0) 2 (4) 15 (47) 2 (50) 1 (50) 6 (50) 0 (0) 26 (47) 15 (47) 2 (50) 1 (50) 6 (50) 1 (100) 25 (49) * Values are the number (%) of patients. Sunscreen (2 mg/cm2) was applied to a closely adjacent test site 15 minutes before irradiation with ultraviolet A (UVA) and UVB. LE ⫽ lupus erythematosus. UVB might be similarly effective. However, a study comparing 3 different sunscreen formulations in a set of 11 patients found these sunscreens to vary in their protective capacity against LE provocation, though all of the formulations used contained various compounds that filter UVA and UVB (6). The best protection was achieved by the only formulation that contained Mexoryl XL, a rather novel compound filtering in both the UVA and UVB range. Interestingly, this compound was also contained in the sunscreen used in our present study. The fact that our sunscreen protected against LE in 98% of patients but still allowed for pigmentation in 49% suggests that the effectiveness of a compound to shield against LE cannot necessarily be extrapolated from its capacity to prevent other biologic responses elicited by sunlight. Dr. Röcken’s work was supported in part by grant RO764/8-1 from the Deutsche Forschungsgemeinschaft. Thomas Herzinger, MD Gerd Plewig, MD Ludwig-Maximillian-University Munich, Germany Martin Röcken, MD Eberhard-Karl-University Tübingen, Germany 1. Millard TP, Hawk JL, McGregor JM. Photosensitivity in lupus. Lupus 2000;9:3–10. 2. Kuhn A, Sonntag M, Ruzicka T, Lehmann P, Megahed M. Histopathologic findings in lupus erythematosus tumidus: review of 80 patients. J Am Acad Dermatol 2003;48:901–8. 3. Lehmann P, Holzle E, Kind P, Goerz G, Plewig G. Experimental reproduction of skin lesions in lupus erythematosus by UVA and UVB radiation. J Am Acad Dermatol 1990;22:181–7. 4. Haywood R, Wardman P, Sanders R, Linge C. Sunscreens inadequately protect against ultraviolet-A-induced free radicals in skin: implications for skin aging and melanoma? J Invest Dermatol 2003;121:862–8. 5. Poon TS, Barnetson RS, Halliday GM. Prevention of immunosuppression by sunscreens in humans is unrelated to protection from erythema and dependent on protection from ultraviolet A in the face of constant ultraviolet B protection. Invest Dermatol 2003;121: 184–90. 6. Stege H, Budde MA, Grether-Beck S, Krutmann J. Evaluation of the capacity of sunscreens to photoprotect lupus erythematosus patients by employing the photoprovocation test. Photodermatol Photoimmunol Photomed 2000;16:256–9. 7. Walchner M, Messer G, Kind P. Phototesting and photoprotection in LE. Lupus 1997;6:167–74. CONCISE COMMUNICATIONS DOI 10.1002/art.20475 Lack of linkage of IL1RN genotypes with ankylosing spondylitis susceptibility The interleukin-1 receptor antagonist gene (IL1RN), a natural modulator existing in the body, can bind IL-1R and block its effects. Recently, a variable-number tandem repeat polymorphism in intron 2 of IL1RN was associated with ankylosing spondylitis (AS) in 2 European population groups (1,2). Here, we investigated the role of IL1RN genes in AS susceptibility in North American families with 2 or more siblings meeting the 1984 modified New York criteria for AS (3). Informed consent was obtained from each family member. This study was approved by the Committee for the Protection of Human Subjects at the University of Texas Health Science Center at Houston. In total, 229 pedigrees with at least 1 affected individual were included for family-based association analyses, and 244 affected sibpairs derived from 180 pedigrees were included for linkage analyses after removing the pedigrees in which the self-reported relationship was not consistent with the genetic data. Of these families, 94% were Caucasian, and 97% of the probands were HLA–B27 positive. The genetic variation of IL1RN was ascertained by sequencing its complete coding sequence, all intron/exon junction regions, 3⬘- and 5⬘-untranslated regions, and the promoter region in 102 patients and 48 ethnically matched controls. In total, 25 single-nucleotide polymorphisms (SNPs) were identified (data not shown). The 5 most informative SNPs (rs1794065, rs419598, rs315952, rs315951, and rs895495) were selected for further association and linkage studies, along with 6 microsatellite markers (D2S2216, D2S373, D2S340, D2S160, D2S121, and D2S347) that are mapped to this region (see Table 1 for the location of the markers). Family-based association analyses were performed using the pedigree disequilibrium test (PDT) (4) and the transmission disequilibrium test (TDT) (5). Neither the TDT nor the PDT approach showed significant association for individual markers (Table 1) and for the haplotypes based on 5 SNPs (data not shown). We further examined the linkage of the region containing IL1RN using the 11 markers. Nonparametric linkage (NPL) analyses were performed with all 5 SNPs and 3 microsatellite markers (D2S2216, D2S160, and D2S347) using 244 affected sibpairs derived from 180 pedigrees. Both parametric (LOD; logarithm of odds) and nonparametric (NPL and ASM) analyses were conducted for linkage analyses, using GeneHunter-Plus software (6) for LOD, and NPL and ASM for an allele-sharing–based approach (7) (see Table 1). Three other markers, including D2S373, D2S340, and D2S121, in the same region were also typed in a subset of the pedigrees (171 affected sibpairs derived from 119 pedigrees). No linkage was observed with any of the 11 markers studied. In a recent collaborative study of a large cohort of Canadian patients with AS, using 3 SNPs of IL1RN identified in this study (rs419598, rs315952, and rs315951), we found an association of AS with the SNPs of IL1RN and their haplotypes (8). TDT analysis of a limited sample of AS families from western Canada demonstrated significant differences in transmission of an IL1RN SNP haplotype that was diseaseassociated in the case–control evaluation. In this study, we were not able to validate the association in our cohort, although our sample size for the TDT and PDT analyses was much larger. The discrepancy may reflect clinical differences in the patient populations studied, i.e., differences in disease 3047 severity or disease phenotypes (such as uveitis or peripheral arthritis, the frequencies of which were not specified in other series) (1,8) due to differing referral patterns between the North American Ankylosing Spondylitis Gene Consortium centers and other centers where associations have been described. Variable patterns of association across populations may also be attributable to population-specific mechanisms of genetic susceptibility. Finally, these data do not exclude a minor role for IL1RN genes. The number of sibpairs that would be necessary to detect a gene with a small genetic effect (s ⱕ1.5) would be ⬎1,000. If such a small effect were present, our study would lack power to determine this. Thus, we present evidence that may lead to the exclusion of this region on chromosome 2q and, more specifically, IL1RN as having a significant impact on AS susceptibility. However, an impact on disease severity cannot be ruled out by this study. This is currently under further investigation. Li Jin, PhD Ge Zhang, MS Joshua M. Akey, PhD Jingchun Luo, MM Juwon Lee, PhD University of Cincinnati Cincinnati, OH Michael H. Weisman, MD Cedars–Sinai Medical Center Los Angeles, CA Jane Bruckel, BSN Spondylitis Association of America Sherman Oaks, CA Robert D. Inman, MD Millicent A. Stone, MD University of Toronto Toronto, Ontario, Canada Muhammad A. Khan, MD Case Western Reserve University Cleveland, OH H. Ralph Schumacher, MD University of Pennsylvania Philadelphia, PA Walter P. Maksymowych, MD University of Alberta Edmonton, Alberta, Canada Maren L. Mahowald, MD University of Minnesota Minneapolis, MN Allen D. Sawitzke, MD University of Utah Salt Lake City, UT Frank B. Vasey, MD University of South Florida Tampa, FL David T. Y. Yu, MD University of California, Los Angeles John D. Reveille, MD University of Texas Health Science Center at Houston 1. McGarry F, Neilly J, Anderson N, Sturrock R, Field M. A polymorphism within the interleukin 1 receptor antagonist (IL-1Ra) gene is associated with ankylosing spondylitis. Rheumatology (Oxford) 2001;40:1359–64. 3048 CONCISE COMMUNICATIONS Table 1. Linkage and family-based linkage disequilibrium analyses of markers near and within IL1RN* Marker Position NPL P for NPL ASM LOD P for TDT P for PDT D2S2216 D2S373* D2S340* D2S160 rs1794065 rs419598 rs315952 rs315951 rs895495 D2S121* D2S347 111.21 114.21 120.29 122.96 123.26 123.27 123.27 123.27 123.27 123.49 131.51 ⫺0.422 0.149 0.629 0.790 0.019 ⫺0.001 0.675 0.574 1.252 ⫺0.457 0.341 0.668 0.438 0.260 0.205 0.491 0.499 0.242 0.275 0.097 0.678 0.360 0.000 0.012 0.230 0.254 0.208 0.214 0.210 0.202 0.221 0.000 0.041 0.763 0.287 0.412 0.343 0.384 0.439 0.347 0.333 0.674 0.148 0.568 0.722 0.622 0.522 0.924 0.294 0.370 0.409 0.377 0.469 0.878 0.569 * D2S373, D2S340, and D2S121 were typed only in a subset of pedigrees. The genetic distances of the single-nucleotide polymorphisms from 2pter were inferred from their genomic coordinates, based on National Center for Biotechnology Information Human Genome Build 33. LOD ⫽ logarithm of odds. NPL ⫽ nonparametric linkage; TDT ⫽ transmission disequilibrium test; PDT ⫽ pedigree disequilibrium test. 2. Van der Paardt M, Crusius JB, Garcia-Gonzalez MA, Baudoin P, Kostense PJ, Alizadeh BZ, et al. Interleukin-1␤ and interleukin-1 receptor antagonist gene polymorphisms in ankylosing spondylitis. Rheumatology (Oxford) 2002;41:1419–23. 3. Van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis: a proposal for modification of the New York criteria. Arthritis Rheum 1984;27:361–8. 4. Martin ER, Monks SA, Warren LL, Kaplan NL. A test for linkage and association in general pedigrees: the pedigree disequilibrium test. Am J Hum Genet 2000;68:146–54. 5. Spielman RS, McGinnis RE, Ewens WJ. Transmission test for linkage disequilibrium: the insulin gene region and insulin- dependent diabetes mellitus (IDDM). Am J Hum Genet 1993; 52:506–16. 6. Markianos K, Daly MJ, Kruglyak L. Efficient multipoint linkage analysis through reduction of inheritance space. Am J Hum Genet 2001;68:963–77. 7. Kong A, Cox NJ. Allele-sharing models: LOD scores and accurate linkage tests. Am J Hum Genet 1997;61:1179–88. 8. Maksymowych WP, Reeve JP, Reveille JD, Akey JM, Buenviaje H, O’Brien L, et al. High-throughput single-nucleotide polymorphism analysis of the IL1RN locus in patients with ankylosing spondylitis by matrix-assisted laser desorption ionization–time-of-flight mass spectroscopy. Arthritis Rheum 2003;48:2011–8.