|Home | About | Journals | Submit | Contact Us | Français|
Genetic and environmental factors may contribute to the etiology of the juvenile idiopathic inflammatory myopathies (JIIM), systemic autoimmune diseases characterized by muscle and skin inflammation. We investigated the association between ultraviolet radiation (UVR) exposure and the clinical and autoantibody expression of JIIM.
We assessed the relationship between UVR exposure in the month before symptom onset and prevalence of juvenile dermatomyositis (JDM) versus polymyositis (JPM) in 298 patients. Among JDM patients, the association between UVR exposure and myositis autoantibodies was assessed. Regression models were stratified by sex and race. The association between regional UV index in U.S. geoclimatic zones and the clinical and autoantibody subgroups was examined by weighted least squares regression analysis.
We observed increasing odds of JDM compared with JPM per unit increase in the patients’ highest UV index in the month before symptom onset in girls (OR = 1.18; 95% CI = 1.00–1.40). The average and highest UV indices were associated with increasing odds of anti-p155/140 autoantibodies, which was strongest in white males (OR 1.30 and 1.23, respectively). No association was observed between the UV index and anti-MJ autoantibodies or patients without myositis autoantibodies. Across US geoclimatic regions, the average UV index was associated with increasing odds of JDM and anti-p155/140 autoantibodies but decreasing odds of anti-MJ autoantibodies.
Short-term UVR exposure prior to illness onset may have a role in the clinical and serologic expression of juvenile myositis. Research examining mechanisms of UVR in JIIM pathogenesis is suggested from these findings.
The idiopathic inflammatory myopathies (IIM) are a heterogeneous group of systemic autoimmune diseases with the common feature of chronic muscle inflammation of unknown etiology. In children, dermatomyositis (DM), characterized by skin rashes and photosensitivity, is the most common of the IIM, and polymyositis (PM), which lacks characteristic skin rashes, constitutes about 10% of childhood-onset patients. As in adult myositis, juvenile myositis is comprised of myositis-autoantibody subgroups which define phenotypes that are each associated with different demographic, clinical and laboratory features and outcomes (1). Some of these autoantibodies are associated with DM, such as anti-Mi-2 and anti-p155/140 (TIF1γ), and others are associated with PM, such as anti-signal recognition particle (1).
Development of IIM is thought to be affected by genetic and environmental risk factors. Although HLA, cytokine polymorphisms and other immunogenetic loci have been identified as genetic risk factors for these diseases (1), little is known about their environmental risk factors. Seasonality in birth dates in subgroups of patients suggests a role for perinatal or neonatal environmental factors, including ultraviolet radiation (UVR), in the later development of these diseases (2). In registry studies, temporal associations with upper respiratory and gastrointestinal infections have been observed. Other exposures have also been observed, including development of symptoms after unusual sun exposure or sunburns (3). In adult IIM, studies in the United States and worldwide found a direct relationship between global surface UVR exposure and the proportion of patients with DM versus PM (4, 5). Seasonal differences in onset of PM/DM have been reported and are consistent with the role of UVR exposure (6). This relationship has not been examined in juvenile IIM (JIIM); therefore, we examined whether the prevalence of juvenile DM (JDM) compared with juvenile PM (JPM), as well as DM-specific autoantibodies, is associated with UVR exposure.
Four hundred thirty-six patients with probable or definite JDM or JPM diagnosed before age 18 years were enrolled in a nationwide registry, which included a physician questionnaire of demographic and clinical features and myositis autoantibody testing by validated immunoprecipitation methods (1). Patients whose first myositis-related symptoms developed before June 1, 1989 (n = 56) and those residing outside the continental United States at the time of myositis symptom onset (n = 32) were not included due to the lack of corresponding surface UVR data. Because this analysis compared JDM with JPM and myositis autoantibody subgroups, patients with overlap myositis (n = 37) or missing autoantibody testing (n = 13) were excluded; 5 patients were in more than one of the excluded categories. Thus, this study included 271 patients with JDM and 27 with JPM; of the JDM patients (Supplementary Figure), the distribution of myositis autoantibodies is detailed in Table 1. All participants signed informed consent.
The UV index consists of an integration of the UVR action spectrum–weighted UVA and UVB irradiances over the 290–400 nm range. The U.S. National Weather Service calculates UV index using a computer model that relates the ground-level strength of solar UVR to forecasted stratospheric ozone concentration, forecasted cloud cover, and elevation (http://www.epa.gov/sunwise/uvicalc.html). Each patient’s residential location (city and state) at onset of myositis symptoms was linked to the UV index (Table 1), which was obtained from the National Weather Service UV Index Cities Forecast Archive (ftp://ftp.cpc.ncep.noaa.gov/long/uv/cities). The average UV index was calculated for each patient by using daily UV index data for 30 days prior to the date of myositis symptom onset, based on the closest city monitored by the National Oceanic and Atmospheric Administration (NOAA). The maximum UV index in the 30 days before symptom onset was also noted. When two or more cities monitored by NOAA were equidistant to the patient’s residence, the one closest in latitude was used. UV index data for 154 patients diagnosed between June 1, 1989 and May 31, 1994 were extrapolated using 1994 data due to lack of available data before 1994. A 30-day interval before symptom onset was selected due to the importance of short-term UVR as a risk factor for other autoimmune diseases such as lupus (7), and known variation of UVR with season, cloud cover and ozone. We also examined the average annual UV index for the median year prior to diagnosis for all patients in the study, which was 1995, to test for associations utilizing methodology similar to that of a prior adult IIM study (5).
We modeled the logarithm of the odds for JDM versus JPM, stratified by gender and race, based on gender and race differences observed in adult IIM (5). UV index was examined as a continuous variable, and the odds ratios represent the odds of an outcome per unit increase in UV index. Separate logistic regression models were also fitted for the presence of anti-p155/140 autoantibodies, anti-MJ autoantibodies and no myositis autoantibody; for these analyses only JDM patients were used. We also examined the association between regional UV index in nine U.S. geoclimatic zones (http://www.ncdc.noaa.gov/temp-and-precip/us-climate-regions.php) and the proportion of JIIM patients with JDM, as well as the proportion of JDM patients with anti-p155/140 or anti-MJ autoantibodies. Weighted least squares regression analysis using individual patient data was used to examine the associations of average or highest regional UV index for the month before symptom onset with the prevalence of juvenile DM, anti-p155/140 or anti-MJ autoantibodies in the nine geoclimatic regions. Weighted least squares regressions were repeated for whites, and the coefficient of determination was calculated. We assessed the R2 value to determine the fraction of variance explained by the regression model. An R2 value > 0.6 was considered strong, 0.3 to 0.6 moderate and < 0.3 weak. SAS version 9.1 (SAS Institute, Cary, NC) was used for all analyses.
Seventy-two percent of the patients were female; 69.1% were white, 15.6% black, 6.5% Hispanic and 8.8% other races. Non-whites were significantly more likely than whites to have JPM (15.5% vs. 6.5%, P = 0.02). White patients were significantly more likely to have anti-p155/140 autoantibodies than non-white patients (37.4% vs. 14.3%, P = 0.0001). There were no significant gender differences in the clinical or autoantibody groups, and there were no racial differences in clinical subgroup when stratified by gender. Patients in this analysis resided in 36 states (Table 1) at the time of myositis symptom onset. Most patients resided in the northeast (n = 105), followed by the south (n = 50) and central (n = 41) regions.
The average UV index for the month before symptom onset was highly correlated with the highest daily UV index over the same period (r = 0.98, P < 0.001). The effects of UVR on clinical group and myositis autoantibodies were assessed in the total JIIM population and by gender and race (Table 2). The average UV index in the month before myositis onset was positively associated with the proportion of JDM patients, but this was not significant (OR = 1.12, P = 0.17). There was an increasing odds of JDM compared with JPM per unit increase in the patients’ highest UV index in the month before symptom onset in girls (OR = 1.18, 95% CI 1.00–1.40). The patient’s highest UV index was associated with increasing odds of anti-p155/140 autoantibodies (OR 1.10, 95% CI 1.10–1.22). The association between the average and highest UV index the month before symptom onset and the proportion of JDM patients with anti-p155/140 autoantibodies was stronger in whites (OR = 1.16, 95% CI 1.03–1.30; OR 1.11, 95% CI 1.01–1.23) and strongest in white males (OR = 1.30, 95% CI 1.02–1.67; OR 1.23, 95% CI 1.00–1.50). There was no association between average or highest UV index in the month before symptom onset and the proportion of JDM patients with anti-MJ autoantibodies (Table 2) or those who were negative for myositis autoantibodies (data not shown).
Loess plots demonstrated that the significant regression models had linear relationships. There were no significant associations between average or highest UV index by racial group in females with JDM compared to JPM or in patients with anti-p155/140 or anti-MJ autoantibodies versus those who were autoantibody negative. There was no association between average UV index and children over age 7 versus those under age 7 years. We also looked at whether there was an association when examining UV index data for the median year before diagnosis for the entire study population, which was 1995, similar to the methods of Love et al (5), and found no significant associations with those variables.
UV index data were aggregated into nine regions of the U.S. and displayed graphically in weighted linear regression analyses. There was a moderate association between average UV index for the 30 days before myositis symptom onset and the proportion of JDM patients (R2 = 0.42, P = 0.06, Figure 1A). A similar relationship was seen for the highest UV index in the 30 days before symptom onset and the proportion of JDM cases (R2 = 0.35, P < 0.001). The association between average or highest UV index and presence of anti-p155/140 autoantibodies was weak (R2 = 0.15 and 0.08, P = 0.05 and 0.44, Figure 1B) for the month before symptom onset. We saw moderate to strong negative relationships between the proportion of anti-MJ autoantibody-positive patients’ average and highest UV index (R2 = 0.61 and 0.59, respectively), but the associations were not significant (P = 0.50 and 0.70) (Figure 1C).
This study examined the variations in JIIM clinical features and serologic expression among certain demographic groups in relation to exposure to surface UVR. As the average and highest UV index increased for the month before onset of myositis, white children had higher odds of developing anti-p155/140 autoantibodies, which was particularly true for white boys. For girls, the odds of JDM increased as the highest UV index increased for the month before symptom onset. When looking at variations across U.S. geographic regions, UVR intensity for the month before symptom onset was moderately associated with the proportion of JDM relative to JPM and weakly associated with anti-p155/140 autoantibodies. These data showed that certain regions within the U.S., including the south, southwest, southeast and west, had higher prevalence of patients with JDM and anti-p155/140 autoantibodies. In contrast, for anti-MJ autoantibodies, there was a higher prevalence in northern latitudes with lower UVR exposure. These findings suggest that variations may exist for myositis autoantibodies in the U.S. and are related, directly or inversely, to UVR intensity. As in adult myositis in the U.S. (5), we found in juvenile myositis that white race was associated with the effect of UVR in the expression of clinical and autoantibody phenotypes.
Exposure to solar UVR is recognized to have both beneficial and harmful effects on human health. With regard to immune responses, it can suppress immunity and the synthesis of vitamin D, a hormone that alters both innate and adaptive immunity (8). The consequences in both adults and children of such UVR-induced changes are considerable. The positive outcomes include protection against some T cell-mediated autoimmune diseases, such as multiple sclerosis (9), and the negative outcomes include higher risk of skin cancer and less effective control of several infectious diseases (10). Overexposure to UVR may suppress proper functioning of the body’s immune system and the skin’s natural defenses, e.g., by increasing and exacerbating certain autoimmune diseases, such as systemic lupus erythematosus (SLE) (11). Geographic variation in exposure to UVR has also been associated with differences in mortality in SLE (12).
Our analysis complements that of Love et al (5), as we looked at whether short-term UVR exposure, as measured by the average and highest UV indices 30 days before symptom onset, was associated with the proportion of JDM patients, in particular those with anti-p155/140 or anti-MJ autoantibodies. We may have also used a more relevant window of exposure. For example, risk of SLE has been associated with outdoor work in the year before diagnosis, particularly among those who are prone to sunburn and among persons who had serious sunburns before age 20 years (7). Thus, childhood exposure to UVR resulting in sunburn may be important for the later development of some autoimmune diseases.
Studies in adult myositis examined the influence of UVR on anti-Mi-2 autoantibodies, which are associated with adult DM (4, 5). Because few JIIM patients had anti-Mi-2 autoantibodies, we could not examine the impact of UVR on anti-Mi-2; instead we looked at the impact on the myositis autoantibodies frequently associated with JDM (anti-p155/140 and anti-MJ autoantibodies). We saw an association between UVR and the frequency of anti-p155/140 autoantibodies that was similar to those found for anti-Mi-2 autoantibodies (4, 5). We also saw a negative relationship between UVR exposure and MJ autoantibodies.
Similar to the study by Love et al (5), we found evidence of racial and gender influences on associations with UVR in autoimmune disorders, including an association between UVR exposure and JDM in girls only. In contrast, we found a male predominance in the association of UVR with anti-p155/140 autoantibodies. The reasons for these gender effects with JDM and UVR remain unclear, and we may have been underpowered to detect differences in some subgroups in the JIIM study. Love et al suggested differential impacts of UVR in males and females based on studies in mice (5). Previous studies have shown gender and racial differences in vitamin D metabolism (13). Of note, skin cancer is more prevalent in men than women, and men are prone to greater UVR-induced immune suppression (14). How UVR might result in anti-p155/140 autoantibodies is also unknown. UVR induces a cascade of events involving type I interferons (5), which have been associated with DM. The tripartite motif-containing (TRIM) family proteins, of which the p155/140 autoantigen is a member, are notably upregulated by interferons (15).
Because we could not assess individual exposure to UVR, we used data from the city closest to where the patient lived. We also extrapolated UVR data to 1994 values for patients with onset of symptoms within 5 years earlier. UVR intensities can vary based on weather and altitude, and that information also was not available. Future studies with individual UVR monitoring would be helpful to confirm the findings of this study. As expected, our JIIM cohort had a lower prevalence of PM than in the adult IIM study, which contributed to a low power to detect effects. Nevertheless, overall our findings were similar to those of Love et al, although the effect was smaller. The stronger effect of adult DM, including female susceptibility to UVR exposure, may be related to an interaction of UVR with hormones (5), which we were unable to examine. We also did not account for differences in use of photoprotective measures or in the frequency of sunburns prior to onset of JIIM, which are frequent in teenage children (16).
Despite its limitations, this study adds important information about the impact of UVR on autoimmune muscle disease. The association of UVR with JDM and with anti-p155/140 autoantibodies has implications for photoprotective prevention measures, as well as for research regarding the role of UVR in the pathogenesis of myositis in children.
We thank Drs. Glinda Cooper and Satoshi Okada for valuable comments after their critical reading of the manuscript. We thank Mona Shah’s Ph.D. thesis committee of the George Washington University Department of Epidemiology and Biostatistics for their valuable feedback on this work, including Dr. Sean Cleary and committee members Drs. Heather Young and Yinglei Lai and examiners Drs. Mark Gourley and Margaret Ulfers.
Funding: This work was supported in part by the Intramural Research Programs of NIEHS, NIH (project number ES101074), NIAMS, NIH, and CBER, Food and Drug Administration. Ira Targoff is a consultant to the Oklahoma Medical Research Foundation Clinical Immunology Laboratory.
Members of the Childhood Myositis Heterogeneity Collaborative Study Group who contributed to this study:
Leslie S. Abramson, Daniel A. Albert, Bita Arabshahi, Alan N. Baer, Imelda M. Balboni, C. April Bingham, William P. Blocker, John F. Bohnsack, Gilles Boire, Gary R. Botstein, Suzanne Bowyer†, Jon M. Burnham, Ruy Carrasco, Victoria W. Cartwright, Gail D. Cawkwell, Chun Peng T. Chao, Randy Q. Cron, Marietta M. DeGuzman, Anne Eberhart, John F. Eggert, Andrew H. Eichenfield, Melissa E. Elder, Terri H. Finkel, Robert C. Fuhlbrigge, Christos A. Gabriel, Vernon F. Garwood, Abraham Gedalia, Stephen W. George, Harry L. Gewanter, Ellen A. Goldmuntz, Donald P. Goldsmith, Gary V. Gordon, Alexia C. Gospodinoff, Beth Gottlieb, Thomas A. Griffin, Brandt P. Groh, Hillary M. Haftel, Michael Henrickson, Gloria C. Higgins, George Ho, Mark F. Hoeltzel, J Roger Hollister, Russel J. Hopp, Lisa Imundo, Jerry C. Jacobs†, Laura James-Newton, Anna Jansen, Rita Jerath, Olcay Y. Jones, Lawrence K. Jung, Thomas V. Kantor, Ildy M. Katona, James D. Katz, Yukiko Kimura, Daniel J. Kingsbury, Steven J. Klein, W. Patrick Knibbe, David K. Kurahara, Andrew Lasky, Julia Lee, Johanan Levine, Carol B. Lindsley, Gulnara Mamyrova, Paul L. McCarthy, John J. Miller III, Stephen R. Mitchell, Hamid Jack Moallem, Chihiro Morishima, Terrance O’Hanlon, Judyann C. Olson, Elif A. Oral, Lauren M. Pachman, Ramesh Pappu, Murray H. Passo, Maria D. Perez, Donald A. Person, Karin S. Peterson, Paul H. Plotz, Marilyn G. Punaro, C. Egla Rabinovich, Charles D. Radis, Ann M. Reed, Robert M. Rennebohm, Peter D. Reuman, Rafael F. Rivas-Chacon, Deborah Rothman, Kenneth N. Schikler, Donald W. Scott, Bracha Shaham, David D. Sherry, Edward Sills, Sara H. Sinal, Robert P. Sundel, Ilona S. Szer, Simeon I. Taylor, Richard K. Vehe, Scott A. Vogelgesang, Larry B. Vogler, Steven Wall, Carol A. Wallace, Jennifer C. Wargula, Patience H. White, M. Jack Wilkenfeld, Andrew P. Wilking, Lan Wu, Christianne M. Yung, Lawrence S. Zemel.
Conflicts of Interest: The authors do not have any conflicts of interest related to this work.