Abstract
Background:
Antinuclear antibody (ANA) prevalence in the U.S. population increased from 1988–2012, especially in white and more educated individuals. In adults ages 20–39 years from the National Health and Nutrition Examination Survey (NHANES) 2003–2004 and 2011–2012, ANA prevalence was previously associated with urinary concentrations of a common sunscreen ingredient, benzophenone 3, measured in winter. Because spot urines may not capture relevant chronic exposures, we examined whether ANA was related to sunscreen use.
Methods:
In a cross-sectional study of adults ages 20–59 (N=416 ANA positive, 2656 ANA negative, by Hep-2 immunofluorescence, 1:80 dilution), we examined associations of ANA with reported sunscreen use when in the sun for 1 hour or more. Logistic regression was used to calculate covariate-adjusted prevalence odds ratios (POR) and 95% Confidence Intervals (CI), overall and stratified by demographic factors, season, and vitamin D. We explored associations and joint effects with other sun protective behaviors and sunburn in the past 12 months.
Results:
The association of ANA with sunscreen differed by age (interaction p=0.004): for ages 20–39, we saw an exposure response (POR 2.61, 95% CI 1.50, 4.24 for using sunscreen always or most of the time, and POR 1.85; 1.12, 3.05 for less frequent versus never-use; trend p <0.001). These associations were more apparent in females (interaction p=0.082), non-Hispanic white and black participants (vs. other race/ethnicity, interaction p=0.023), and those with sufficient serum vitamin D (≥50 vs. <50nmol/L, interaction p=0.001). ANA was not associated with other protective behaviors and not confounded or modified by these behaviors or recent sunburn.
Conclusions:
These cross-sectional findings showed frequent sunscreen was associated with ANA in younger adults, supporting the need for replication, and longitudinal studies with detailed exposure histories.
Keywords: Antinuclear antibodies, autoimmunity, cross-sectional studies, oxybenzone, sunburn, sunscreen
1.0. Introduction
1.1. ANA trends
Data from the the National Health Examination and Nutrition Survey (NHANES), showed that ANA prevalence in the U.S. population significantly increased from 1998–1994 to 2011–2012 [1]. Trends were seen in all age groups but were more pronounced at younger ages, among non-Hispanic whites, and those with higher education and income, and were not explained by changes in sample characteristics, such as obesity and smoking [1].
1.2. Sunscreen and ANA
We sought to identify environmental factors that could relate to these increases, including exposures that tend to be more common among white individuals and those of higher socioeconmic status. Sunscreen use has increased in recent decades and is differentially patterned by age, race, and socioeconoimc status. Increased xenobiotic chemicals produced from 1970–2010 have introduced new active ingredients used in the personal care and cosmetic industry, including organic sun filters benzophones, camphors, and cinnimates [2]. Several of these dermally applied chemicals are rapidly absorbed in the bloodstream [3–5]; the most abundant, benzophenone-3 (BP3), has been identified for its potential adverse human and environmental effects [4, 6]. By 2003–04, most of the U.S. population had measurable urinary BP-3, with greater concentrations the subsequent decade, mirroring increased sunscreen use, including in cosmetics and higher sun-protection factor (SPF)-products [6–11].
Experimental studies suggest sunscreens, including BP-3, may have immunomodulatory effects [12–16]. Given their concurrently increasing trends and potential for adverse impacts on the immune system, we investigated urinary BP-3 concentrations and ANA in a sample of 1,785 NHANES participants ages 12–39 years in 2003–4 and 2011–12; we found that elevated BP-3 levels in winter were associated with higher ANA prevalence, especially among those ages 20–39 and with greater sunscreen use [17]. Because BP-3 concentrations in spot urines are influenced by recent sunscreen applications [3, 5], regular use across the seasons may be a better indicator of chronic exposures contributing to the development of ANA. Other sun-protective behaviors and sunburn could impact development of ANA through shared or interacting pathways. Sunburn is also more common in younger adults, whites, and sunscreen users, and less common in those practicing other sun protective behaviors [18]. Burn-associated damage may stimulate ANA through release of self-antigens and inflammation, while sun avoidance may result in lower vitamin D levels or reduced immunosuppression due to UV exposure [19].
1.3. Study Aims
In the present study we examined the association of ANA on the frequency of sunscreen use in a larger sample of adults ages 20–59 years from the NHANES cycles with available data on both sunscreen use and ANA (2003–2004 and 2011–2012). Sunscreen use was not collected at younger ages (12–19 years), but the current sample included those ages 40–59, given increasing ANA in middle-aged as well as younger adults. We also considered the impact of sunburn and other protective measures, potential susceptibility factors, and ANA phenotype data on intensity and staining patterns.
2.0. Methods
2.1. Design and sample
The NHANES is a nationally representative sample of the non-institutionalized U.S. population, collecting data in consecutive two-year cycles since 1999. This cross-sectional study sample included 3,072 adults ages 20–59 years in NHANES 2003–2004 and 2011–2012 cycles with data on serum ANA assessed by HEp-2 immunofluorescence [1]. Questionnaire data on sunscreen and blood specimens were collected during an in-person exam following a seasonal pattern with states at higher latitude sampled in the summer (May-October), and lower-latitude and west-coast states sampled in the winter (November-April). The NHANES protocol was approved by the Institutional Review Board of the Centers for Disease Control and Prevention (CDC). All participants provided informed consent.
1.2. ANA assessment
ANA data were generated as previously described [1]. Serum samples were stored at −80°C until they were assessed at the 1:80 dilution using HEp-2 indirect immunofluorescence assay NOVA Lite HEp-2 ANA slide with DAPI kit (INOVA Diagnostics, San Diego, CA), detected by a highly specific fluorescein isothiocyanate (FITC)-conjugated secondary antibody (goat anti-human IgG), and evaluated using the NOVA View automated fluorescence microscope system (INOVA Diagnostics). Staining intensity was rated on a scale of 0 to 4, and ANA positivity defined as any signal above zero. International consensus criteria were used in evaluating nuclear, cytoplasmic, and mitotic ANA patterns (1). One laboratory assayed all the specimens, including a random replicate sample rated with over 98% concordance. Two independent reviewers agreed on >95% of ratings, resolving differences by consensus or a third reviewer.
2.3. Data on exposures and covariates
Participant characteristics and exposures were collected by questionnaire. Questions asked about usual sun protective behaviors, i.e., “when you go out on a very sunny day for more than an hour how often do you use sunscreen/stay in the shade/wear a hat/wear a long-sleeved shirt (never, rarely, sometimes, most of the time, always).” Classification of these responses for analyses was based on response frequencies to achieve sufficiently sized referent groups and reduce small cell sizes (see Table 1 for frequencies). Specifically, sunscreen use was grouped for analyses as never, infrequent (rarely/sometimes), and frequent (most of the time/always). Other sun protective behaviors were grouped by the most frequent responses for staying in the shade and long-sleeved shirt use, e.g., always wore long sleeves and sometimes stayed in the shade were grouped as “always,” and those who never or rarely practiced these behaviors were grouped due to small numbers reporting “never”. Questions asked about sunburn in the past year, including how many times. Sunburn was grouped as “any” and “one” or “two or more.”
Table 1.
Characteristics and weighted proportions by ANA status in U.S. adults ages 20–59 years (NHANES cycles 2003–2004, 2011–2012)
| ANA | ||
|---|---|---|
| Positive | Negative | |
| N = 416 N (%) |
N = 2,656 N (%) |
|
| Age (Years) | ||
| 20 to 29 | 90 (20.6) | 753 (25.3) |
| 30 to 39 | 104 (21.1) | 682 (25.2) |
| 40 to 49 | 99 (24.3) | 647 (26.7) |
| 50 to 59 | 123 (34.0) | 574 (22.8) |
| Sex | ||
| Male | 132 (29.8) | 1,365 (52.3) |
| Female | 284 (70.2) | 1,291 (47.7) |
| Race/Ethnicity | ||
| White | 158 (67.7) | 1,092 (66.5) |
| Black | 110 (13.6) | 563 (10.6) |
| Other | 148 (18.7) | 1,001 (22.8) |
| NHANES cyclea | ||
| 2003–2004 | 85 (42.7) | 693 (53.1) |
| 2011–2012 | 331 (57.3) | 1,963 (46.9) |
| Season of Blood Collection | ||
| Winter (November-April) | 198 (40.9) | 1,236 (40.4) |
| Summer (May-October) | 218 (59.1) | 1,420 (59.6) |
| Vitamin D (nmol/L) | ||
| <50 | 160 (26.8) | 964 (27.0) |
| 50–75 | 133 (35.9) | 1,013 (40.0) |
| ≥75 | 123 (37.3) | 679 (33.0) |
| Body Mass Index (kg/m2) | ||
| Normal/underweight (<25) | 120 (28.8) | 901 (35.6) |
| Overweight (25–30) | 125 (28.5) | 825 (31.3) |
| Obese (>30) | 171 (42.7) | 930 (33.1) |
| Sunscreen use | ||
| Never | 156 (26.5) | 1,187 (35.6) |
| Rarely | 53 (16.2) | 361 (16.4) |
| Sometimes | 81 (23.6) | 483 (20.3) |
| Most of the time | 59 (16.5) | 326 (15.1) |
| Always | 67 (17.2) | 299 (12.7) |
| Other sun protective behaviors | ||
| Never | 25 (5.6) | 229 (8.6) |
| Rarely | 53 (15.7) | 352 (16.1) |
| Sometimes | 157 (41.9) | 964 (38.9) |
| Most of the time | 113 (24.4) | 687 (25.3) |
| Always | 68 (12.4) | 424 (11.0) |
| Sunburns past year | ||
| None | 259 (50.2) | 1,622 (51.2) |
| One | 82 (26.8) | 543 (25.2) |
| Two or more | 75 (23.0) | 491 (23.6) |
ANA obtained on 1/3rd of 2003–2004 sample, resulting in greater weighted proportions.
Participants were also asked, “If, after several months of not being in the sun, you went out in the sun without sunscreen or protective clothing for half an hour, what would happen to your skin?” Sun sensitivity was dichotomized as yes (severe sunburn with blisters or peeling) and no (all other). Participants were also asked if they had ever been diagnosed with psoriasis (yes/no).
Covariates included age (continuous for adjustment, categorical for stratified models – ages 20–39, 40–59 years), gender (female, male), race/ethnicity (white, African American, other), education (<high school, high school grad/equivalent, some college or above), and body mass index (BMI: underweight/normal [<25 kg/m2], overweight [25>30 kg/m2)], and obese [30+ kg/m2]) from measured height and weight, and season of NHANES visit (winter, summer).
Serum 25(OH)D levels were determined in 2003–2004 samples by radioimmunoassay (Diasorin, Stillwater, MN), and by liquid chromatography / tandem mass spectroscopy (LC-MS/MS) in 2011–2012: (NHANES 2011–2012: Vitamin D Data Documentation, Codebook, and Frequencies (cdc.gov). Radioimmunoassay levels, harmonized to LC-MS/MS equivalents [20], were grouped as: <50 nmol/L (20ng/ml), and 50 nmol/L and above (generally regarded as sufficient)[20].
2.4. Statistical analyses
All analyses accounted for survey sampling weights (SAS 9.4, Cary, NC, USA). We first generated descriptive frequencies by ANA status for exposures and covariates, using logistic regression to calculate age-adjusted prevalence odds ratios (POR) and 95% confidence intervals (CI). Then, we examined ANA associations with frequency of sunscreen use, other sun protection, and sunburn, evaluating potential covariate and main effect interactions. Final models adjusted for age, survey cycle, gender, race/ethnicity, BMI, and BMI x age. Due to an interaction observed between age and sunscreen use (described in the results text), we ran initial overall models and then stratified by the mid-point of age, optimizing sample size for statistical power. Frequency of sunscreen use was not strongly correlated with other sun protective behaviors and sunburn (r=0.14 and 0.16, respectively; Supplemental Table 1); therefore, these were not considered as potential confounders. Instead, we evaluated the joint effects of sunscreen use with higher or lower levels of other sun protections and in those with or without sunburn in the past 12 months, testing interactions by age comparing −2 Log Likelihoods in models with and without product terms. Due to the significant interaction of age X sunscreen use, all analyses stratified by age.
In sensitivity analyses, we excluded those reporting sun sensitivity (N=38 ANA positive [11%] and 284 ANA negative [14%] or psoriasis (N=18 ANA -positive [6%] and 66 ANA-negative [3%]), conditions that could be related to ANA or other autoimmune diseases influencing sunscreen use. We then explored the association of ANA with sunscreen use, stratified by age and key covariates: gender, race/ethnicity, sample season. Greater sunscreen use was correlated with higher vitamin D levels (Supplemental Table 1), so we also ran stratified models to explore patterns of association by vitamin D sufficiency.
Finally, to better understand the potential clinical implications of these findings, we also explored the overall associations by staining intensity (1, 2, 3–4) and nuclear ANA pattern (FS, fine speckled; DFS, dense fine speckled; and other) collapsing categories as needed to achieve sufficient cell sizes across categorical variables and using polytomous regression models for the different levels and types of patterns.
3.0. Results
3.1. Sample characteristics
Table 1 shows covariate and sunscreen frequencies by ANA status: ANA positive individuals (N=416) were more likely than ANA-negative individuals (N=2,656) to be older, female, from the 2011–2012 NHANES cycle, and obese (BMI>30 vs. <25 kg/m2). ANA prevalence did appear not vary by race/ethnicity, season, or vitamin D levels. ANA-positive participants more frequently reported sunscreen use but did not appear to differ by other sun protective behaviors and sunburn. Supplemental Table 2 shows covariates, sunscreen use, other protective behavior, and sunburn by ANA, stratified by age categories (20–39 and 40–59).
3.2. Main effects for sunscreen, other protective behaviors, and sunburn
In multivariable models (Table 2), ANA prevalence was associated with sunscreen use in the overall sample, for both infrequent (sometimes or rarely, POR 1.52; 95% CI 1.05, 2.19) and frequent (always or most of the time, POR 1.49; 1.03, 2.16) use compared to never use. In age-stratified models, we saw an exposure-response trend in the association between sunscreen use and ANA (POR 1.85; 1.12, 3.05 for less frequent and POR 2.61, 1.50, 4.24 for frequent use, trend p <0.001) in younger adults (ages 20–39 years). In older adults (ages 40–59), ANA was not associated with sunscreen use, representing a significant age by sunscreen use interaction (p=0.004). We saw no associations of other sun protective behaviors or sunburn, overall or by age, except for a slightly elevated POR for sunburn at older ages (POR 1.29; 1.01, 1.90).
Table 2.
Association of ANA with sunscreen use, sun protective behaviors, and sunburn in U.S. adults ages 20–59 years (NHANES cycles 2003–2004, 2011–2012): overall, and stratified by age.
| Overall aPORa,b (95% CI) |
Age 20–39 aPORa (95% CI) |
Age 40–59 aPORa (95% CI) |
|
|---|---|---|---|
| Sunscreen use | |||
| Never | 1.00 (referent) | 1.00 (referent) | 1.00 (referent) |
| Sometimes or rarely | 1.52 (1.05, 2.19) | 1.85 (1.12, 3.05) | 1.44 (0.93, 2.22) |
| Always or most of the time | 1.49 (1.03, 2.16) | 2.61 (1.60, 4.24) | 1.05 (0.64, 1.72) |
| p-trend | 0.038 | <0.001 | 0.898 |
| Other sun protective behaviors | |||
| Never or rarely | 1.00 (referent) | 1.00 (referent) | 1.00 (referent) |
| Sometimes | 1.20 (0.83, 1.73) | 1.61 (0.96, 2.69) | 0.99 (0.60, 1.65) |
| Always or most of the time | 0.94 (0.60, 1.49) | 1.33 (0.70, 2.50) | 0.75 (0.43, 1.31) |
| p-trend | 0.666 | 0.447 | 0.250 |
| Sunburns past year | |||
| None | 1.00 (referent) | 1.00 (referent) | 1.00 (referent) |
| Any | 1.23 (0.96, 1.59) | 1.08 (0.68, 1.73) | 1.39 (1.01, 1.90) |
| One | 1.23 (0.90, 1.68) | 1.11 (0.70, 1.76) | 1.33 (0.82, 2.18) |
| Two or more | 1.24 (0.84, 1.84) | 1.04 (0.58, 1.86) | 1.46 (0.97, 2.18) |
| p-trend | 0.196 | 0.856 | 0.032 |
Models adjusted for sampling weights, age, gender, race/ethnicity, season, BMI, and BMI*Age calculated prevalence odds ratios (POR) and 95% confidence intervals.
Interaction assessed for product of age group by main effects: p(age*sunscreen)=0.004, p(age*sun protection)=0.795, p(age*sunburn)=0.361.
3.3. Joint effects of sunscreen, other sun protective behaviors, and sunburn
Table 3 shows associations for ANA with sunscreen use relative to other protective behaviors and sunburn in the past 12 months. No statistically significant interactions were noted (p>0.20) and results were similar to main the effects for frequent sunscreen use among participants ages 20–39 years, e.g., elevated POR in those with versus those with and without sunburn (POR 2.34 vs. 2.66, with CIs excluding the null) versus non-users without sunburn. Results also showed that among those aged 40–59 years, ANA was associated with infrequent sunscreen use among those with sunburn (POR 1.95), a similar magnitude as seen among younger participants (POR 1.86). Notably, other sun protective measures were not associated with ANA among those who did not use sunscreen in both younger (POR 1.26) and older individuals (POR 0.57).
Table 3.
Prevalence of joint exposure to sunscreen and sunburn or other sun protective behaviors, and associations with ANA, stratified by age: NHANES cycles 2003–2004 and 2011–2012.
| Age 20–39 | Age 40–59 | |||||
|---|---|---|---|---|---|---|
| ANA | ANA | |||||
| Positive N (%) | Negative N (%) | aPORa (95% CI) | Positive N (%) | Negative N (%) | aPORa (95% CI) | |
| Sunscreen use stratified by other sun protection | ||||||
| Lower sun protection | ||||||
| Sunscreen frequency | ||||||
| Never | 36 (11.7) | 356 (24.1) | 1.00 (referent) | 57 (20.0) | 303 (20.2) | 1.00 (referent) |
| Sometimes/rarely | 40 (24.7) | 362 (28.6) | 1.92 (1.06, 3.45) | 39 (30.0) | 218 (25.1) | 1.26 (0.72, 2.21) |
| Always/most of the time | 37 (27.2) | 177 (14.4) | 3.38 (1.80, 6.36) | 26 (12.9) | 129 (14.8) | 0.86 (0.47, 1.54) |
| Higher sun protection | ||||||
| Sunscreen frequency | ||||||
| Never | 26 (8.9) | 247 (12.9) | 1.23 (0.55, 2.75) | 37 (10.7) | 281 (13.9) | 0.57 (0.25, 1.29) |
| Sometimes/rarely | 25 (11.8) | 139 (9.0) | 2.20 (1.02, 4.74) | 30 (12.2) | 125 (10.6) | 1.02 (0.52, 1.99) |
| Always/most of the time | 30 (15.7) | 154 (10.9) | 2.18 (1.07, 4.45) | 33 (14.2) | 165 (15.4) | 0.86 (0.41, 1.81) |
| Sunscreen use stratified by sunburn in the past year | ||||||
| No sunburn | ||||||
| Sunscreen frequency | ||||||
| Never | 47 (15.1) | 439 (23.3) | 1.00 (referent) | 78 (22.7) | 458 (24.7) | 1.00 (referent) |
| Sometimes/rarely | 30 (13.5) | 222 (13.1) | 1.58 (0.82, 3.05) | 38 (17.1) | 196 (17.8) | 1.12 (0.62, 2.02) |
| Always/most of the time | 30 (17.1) | 135 (9.1) | 2.66 (1.24, 5.69) | 36 (13.6) | 172 (14.4) | 1.05 (0.56, 1.99) |
| Any sunburn | ||||||
| Sunscreen frequency | ||||||
| Never | 15 (5.4) | 164 (13.8) | 0.83 (0.38, 1.84) | 16 (8.0) | 126 (9.4) | 1.14 (0.51, 2.56) |
| Sometimes/rarely | 35 (23.0) | 279 (24.4) | 1.86 (0.92, 3.77) | 31 (25.1) | 147 (17.9) | 1.95 (1.23, 3.08) |
| Always/most of the time | 37 (25.8) | 196 (16.3) | 2.35 (1.12, 4.94) | 23 (13.5) | 122 (15.7) | 1.14 (0.64, 2.03) |
Models adjusted for sampling weights, age, gender, race/ethnicity, season, BMI, and BMI*Age calculated prevalence odds ratios (POR) and 95% Confidence intervals. Interaction assessed for product of sunscreen by sunburn (any) and sunscreen by sun protection (lower=not always/most of the time), Ages 20–39: p(sunscreen*sunburn)=0.768 p(sunscreen*sun protection)=0.327; Ages 40–59: p(sunscreen*sunburn)=0.620, p(sunscreen*sun protection)=0.518.
3.4. Sensitivity analysis
We saw no substantial impact of excluding individuals with sun sensitivity (tendency to burn) or self-reported psoriasis (total N after exclusions: 360 ANA-positive and 2,306 ANA-negative) for the effects of sunscreen and other protective behaviors or sunburn at younger ages (e.g., POR 2.83; 95%CI 1.65, 4.86 for frequent use, and 1.69; 0.99, 2.90 for infrequent versus never use, Supplemental Table 3). The POR for sunburn at older ages was similar, but less precise (1.25; 0.85, 1.89). Looking at joint effects, we saw no joint effects for younger or older adults for sunscreen use together with other sun protection practices or sunburn in the past year.
3.5. Covariate-stratified models (Figure 1)
Figure 1.
Association of ANA with frequent and infrequent versus no sunscreen use in U.S. adults ages 20–59 years (NHANES cycles 2003–2004, 2011–2012): stratified by demographic factors, season, and vitamin D sufficiency. Calculated in logistic regression models adjusted for sampling weights, age, gender, race/ethnicity, season, BMI, and BMI*Age calculated prevalence odds ratios (OR) and 95% (CI) Confidence intervals.
Among those ages 20–39 years, ANA appeared more strongly associated with frequent sunscreen use in females (POR 3.32 vs. 1.13 in males, interaction p=0.082), and in non-Hispanic White and Black participants (4.01 and 3.76, respectively, vs. 1.73 in other racial/ethnic groups; interaction p=0.023). Associations were similar when stratified by season of sampling wave; but the association with frequent use was primarily seen in those with sufficient vitamin D concentrations, i.e., ≥50nmol/L (3.52 vs. 1.03 with <50nmol/L; interaction p=0.001). Infrequent use was associated with ANA regardless of vitamin D sufficiency (1.92 in those with Vitamin D ≥50nmol/L vs. 2.26 <50nmol/L). Among those ages 40–59 years, some POR were elevated (>1.5) for infrequent use (in White and Black participants, the summer sample, and those with sufficient vitamin D), but we saw no interactions (p>0.20).
3.6. ANA staining intensity and patterns
The main effects for associations with staining intensity and patterns among those ages 20–39 as shown in Table 4. The association of more frequent sunscreen use was increased for those with greater intensity of ANA, e.g., for using sunscreen all or most of the ORs ranged from 1.77 for level 1 intensity, to 3.4 and 6.45 for those with level 2 or level 3–4 intensity respectively (the latter two manifesting significant trends for increasingly frequent use compared with never use (p-values 0.001 and 0.024, respectively). Reporting a sunburn in the past 12 months was also significantly associated with higher intensity ANA (levels 3–4; POR 3.61;95% CI 1.78, 7.32) and negatively associated with lower intensity ANA (level 1; 0.54; 0.29, 1.00). These patterns were not seen among those ages 40–59 (Supplemental Table 4). However, level 1 intensity ANA appeared to be negatively associated with greater use of other sun protective behaviors (p-trend=0.038; POR for always/most of the time versus never/rarely 0.48 (0.23–1.02), and positively associated with reporting one sunburn in the past year (p=trend=0.189; POR 1.82; 95%CI 1.00–3.30).
Table 4.
ANA intensity level in relation to sunscreen use, sun protective behaviors, and sunburn in U.S. adults ages 20–39 years (NHANES cycles 2003–2004, 2011–2012).
| Level 1 | Level 2 | Levels 3–4 | ||||
|---|---|---|---|---|---|---|
| N (%) | aPORa (95% CI) | N (%) | aPORa (95% CI) | N (%) | aPORa (95% CI) | |
| Sunscreen use | ||||||
| Never | 31 (31) | 1.00 (referent) | 22 (14) | 1.00 (referent) | 9 (10) | 1.00 (referent) |
| Sometimes/rarely | 26 (33) | 1.20 (0.58, 2.52) | 28 (40) | 2.63 (1.44, 4.81) | 11 (37) | 4.37 (1.65, 11.39) |
| Always/most of the time | 25 (36) | 1.77 (0.85, 3.68) | 25 (46) | 3.41 (1.95, 5.97) | 17 (52) | 6.45 (2.11, 19.73) |
| p-trend | 0.148 | 0.001 | 0.024 | |||
| Other sun protective behaviors | ||||||
| Never/rarely | 15 (22) | 1.00 (referent) | 13 (18) | 1.00 (referent) | 5 (26) | 1.00 (referent) |
| Sometimes | 31 (38) | 1.38 (0.64, 2.95) | 32 (49) | 2.08 (0.75, 5.79) | 17 (39) | 1.18 (0.32, 4.45) |
| Always/most of the time | 36 (40) | 1.38 (0.68, 2.81) | 30 (34) | 1.34 (0.50, 3.61) | 15 (35) | 1.08 (0.20, 5.86) |
| p-trend | 0.400 | 0.655 | 0.941 | |||
| Sunburns past year | ||||||
| None | 50 (63) | 1.00 (referent) | 39 (36) | 1.00 (referent) | 18 (24) | 1.00 (referent) |
| Any | 32 (37) | 0.54 (0.29, 1.00) | 36 (64) | 1.48 (0.64, 3.41) | 19 (76) | 3.61 (1.78, 7.32) |
| One | 16 (16) | 0.46 (0.20, 1.06) | 18 (36) | 1.61 (0.69, 3.73) | 11 (47) | 3.91 (1.66, 9.21) |
| Two or more | 16 (21) | 0.64 (0.31, 1.33) | 18 (28) | 1.31 (0.50, 3.49) | 8 (30) | 3.12 (1.03, 9.47) |
| p-trend | 0.158 | 0.522 | 0.030 |
Models adjusted for sampling weights, age, gender, race/ethnicity, season, BMI, and BMI*Age calculated prevalence odds ratios (POR) and 95% confidence intervals.
Looking at ANA staining patterns among those ages 20–39 (Table 5), increasing frequency of sunscreen use was associated with the DFS pattern, i.e., POR 2.39; 95%CI 1.07–5.31 for using sunscreen sometimes or rarely, and 5.42; 2.32–12.64 for use always or most of the time versus never; trend-p<0.0001). Having any sunburn in the past year was associated with the FS pattern (POR 1.96;95%CI 1.08, 3.58), and a clear negative trend was observed in the association of other ANA staining patterns with sunburn (2 or more; 0.38; 0.18, 0.83, p-trend=0.005). The FS pattern was associated with more frequent use of other sun protective behaviors, with greater PORs, albeit with CI excluding the null (p-trend=0.05). These patterns were not seen among those ages 40–59 (Supplemental Table 5).
Table 5.
ANA staining pattern in relation to sunscreen use, sun protective behaviors, and sunburn in U.S. adults ages 20–39 years (NHANES cycles 2003–2004, 2011–2012).
| Fine speckled | Dense fine speckled | Other | ||||
|---|---|---|---|---|---|---|
| N (%) | aPORa (95% CI) | N (%) | aPORa (95% CI) | N (%) | aPORa (95% CI) | |
| Sunscreen use | ||||||
| Never | 25 (31) | 1.00 (referent) | 15 (10) | 1.00 (referent) | 22 (24) | 1.00 (referent) |
| Sometimes/rarely | 23 (45) | 1.90 (1.07, 3.39) | 23 (30) | 2.39 (1.07, 5.31) | 19 (38) | 1.61 (0.69, 3.73) |
| Always/most of the time | 15 (25) | 1.51 (0.82, 2.81) | 29 (60) | 5.42 (2.32, 12.64) | 23 (37) | 1.72 (0.77, 3.83) |
| p-trend | 0.122 | <0.001 | 0.243 | |||
| Other sun protective behaviors | ||||||
| Never/rarely | 6 (12) | 1.00 (referent) | 18 (20) | 1.00 (referent) | 9 (30) | 1.00 (referent) |
| Sometimes | 22 (38) | 2.42 (0.85, 6.90) | 29 (46) | 1.86 (0.89, 3.89) | 29 (43) | 1.18 (0.41, 3.34) |
| Always/most of the time | 35 (50) | 3.11 (0.95, 10.19) | 20 (34) | 1.33 (0.51, 3.48) | 26 (27) | 0.70 (0.28, 1.77) |
| p-trend | 0.050 | 0.636 | 0.432 | |||
| Sunburns past year | ||||||
| None | 35 (40) | 1.00 (referent) | 30 (36) | 1.00 (referent) | 42 (63) | 1.00 (referent) |
| Any | 28 (60) | 1.96 (1.08, 3.58) | 37 (64) | 1.33 (0.59, 3.00) | 22 (37) | 0.54 (0.31, 0.93) |
| One | 18 (37) | 2.24 (1.20, 4.20) | 14 (27) | 1.11 (0.46, 2.69) | 13 (25) | 0.67 (0.33, 1.37) |
| Two or more | 10 (23) | 1.56 (0.73, 3.32) | 23 (37) | 1.62 (0.62, 4.22) | 9 (13) | 0.38 (0.18, 0.83) |
| p-trend | 0.130 | 0.342 | 0.005 | |||
Models adjusted for sampling weights, age, gender, race/ethnicity, season, BMI, and BMI*Age calculated prevalence odds ratios (POR) and 95% confidence intervals.
4.0. Discussion
4.1. Overview of findings in context
Results from this cross-sectional study, showing that frequent sunscreen use is associated with ANA in U.S. adults ages 20–39 years, are consistent with the possibility that some ingredients in sunscreen may contribute to the development or expression of ANA as seen in our prior findings for BP-3 and ANA in the same age group [17]. The findings were primarily seen for those with higher intensity staining and the nuclear dense fine speckled (DFS) pattern, which has been associated with inflammatory conditions and are often observed in the absence of autoimmune disease (1–3). The association of ANA with frequent sunscreen use in younger adults was robust in joint effect models and did not depend on or interact with other sun protective behaviors and sunburn. The possibility that cellular damage due to sun-exposure might trigger ANA was not supported by evidence for sunburn in the past 12 months at younger ages overall, but did appear among those with higher intensity staining and the FS pattern. The lack of association of ANA with sunscreen use among those ages 40–59, suggests a more complex picture.
4.1.1. Age differences
The age-difference in the ANA association with sunscreen was unexpected but not surprising. Prior findings for urinary BP-3 concentrations and ANA prevalence also showed age differences, with an association among those ages 20–39 years and not ages 12–19 [12]. Analyses of cross-sectional data cannot capture the temporal relationship of sunscreen and incident ANA, and reported sunscreen use may (or may not) reflect longer-term patterns and past use; so, one explanation for age differences may be past exposures to different types of sunscreens or amounts of active ingredients at the time ANA first developed [21, 22]. Alternatively, other factors could contribute more to the elevated ANA prevalence seen with aging [23, 24]. For example we noted an age interaction with overweight/obesity when evaluating covariates, with an association among those ages 40–59 years but not at younger ages (frequencies seen in Supplemental Table 2). Among older adults with recent sunburn, ANA was associated with infrequent sunscreen use (vs. those without sunburn who didn’t use sunscreen), which could be due to other factors not addressed in these analyses (see sections 4.1.2 and 4.2.2).
4.1.2. Other sun protective behaviors and sunburn
We did not see any overall ANA associations with other sun protective behaviors, regardless of sunscreen use or with more frequent sunburn in the past 12 months. While recent sunburn was associated with ANA in adults ages 40–59 years, there was no exposure response trend and the difference by age was not significant. In joint effects models, the elevated POR for sunburn in older adults was most apparent among those reporting sometimes or rarely using sunscreen, but the interaction was not significant. These findings could be affected by exposure misclassification if underlying relationships depend on cumulative sunburn history or the time since sunburn. The effects of ultraviolet (UV) exposure on dermal and systemic immunity are complex, informing multiple pathways by which sunburn could induce or promote ANA and differences in susceptibility, as seen in lupus animal models [19, 25, 26]. Sunburn, the acute damage caused by intense UV exposure, may lead to the release of self-antigens by apoptotic keratinocytes. At the same time, in the absence of burn, UV exposure has been shown to damage Langerhans cells, i.e., antigen-presenting cells, contributing to an antibody-specific immune response along with localized immunosuppression and systemic immunosuppression induced through other mechanisms, including soluble mediators [27, 28].
4.1.3. Differences by gender and other covariates
The association of ANA with sunscreen use was more apparent among females, which could reflect more cumulative use, e.g., daily use of sunscreen-containing cosmetics, or the use of higher SPF products when modeling use and applying products to children. On the other hand, some sunscreen ingredients could have differential endocrine-mediated effects. Cell-based reporter gene assays indicate that the sunscreen BP-3, and its metabolite BP-1, have both estrogenic and anti-androgenic effects, and experimental models have demonstrated negative effects on female reproductive endpoints and on sperm concentrations, as reviewed by Mustieles et al., 2023 [29]; however, evidence in human studies is inconsistent and does not typically include direct comparisons of effects in females compared to males, except for studies of pre-natal or early life childhood. Females also may be less sensitive to UV-induced immunosuppression [30, 31], which could influence potential effects of sunscreen on autoimmunity when exposed to sunlight. We saw an interaction by race/ethnicity, with elevated PORs for frequent sunscreen use in only among non-Hispanic White and Black participants. However, limited sample size precluded more detailed examination of other racial and ethnic subgroups.
Notably, we saw evidence of effect modification of the ANA/sunscreen association by vitamin D sufficiency at the time the sample was taken; the association with frequent sunscreen use was seen only among those with sufficient vitamin D levels. Among those with insufficient levels, infrequent (but not frequent) sunscreen use was associated with ANA. The effects of vitamin D on the immune system are complex, however growing evidence suggests vitamin D may influence systemic autoimmunity in healthy individuals and patients with autoimmune diseases [32–35] Vitamin D sufficiency could be a marker for sun exposure, dietary intake, and supplement use. No differences were seen by sample season, which also relates to geographic location (e.g., winter sampling in more southern states), peak and cumulative UV exposure.
4.1.4. Variation across staining intensity and nuclear patterns
In adults ages 20–39s, the association of ANA with more frequent sunscreen use was most apparent for higher intensity staining and those with the nuclear dense fine speckled (DFS) pattern, which has been associated with inflammatory conditions and in healthy individuals as well as patients with autoimmune diseases [36–38]. This combination of higher intensity and DFS staining may be the most common among healthy individuals, specifically including but not limited to anti-dense fine speckled 70 (anti-DFS70) antibodies. However, long-term follow-up studies are lacking and there is some evidence of gender differences in the concordance of the DFS pattern with anti-DFS70 and other types of autoantibodies [39, 40].
Sunburn in the past 12 months was associated with higher intensity ANA and but also the fine speckled (FS) staining pattern and was negatively associated with low intensity ANA and other staining patterns besides speckled nuclear (e.g., cytoplasmic and mitotic). The FS pattern has been associated with systemic autoimmune diseases and specific antibodies such as Ro and La [36], also linked to cutaneous lupus erythematosus [41]. We are unaware of prior studies showing a links of the FS pattern to sunburn but note suggestive evidence of a possible contribution of sunburn to the underlying pathology and possible development of cutaneous or systemic lupus [42–45]. Given the cross-sectional nature of the data, we cannot infer a temporal relationship, and prospective research is needed on the role of sunburn in the development of autoimmune disease.
4.2. Limitations and strengths
4.2.1. Study design and approach
These cross-sectional analyses of prevalent ANA cannot show a temporal relationship, however reverse causality seems unlikely as most autoimmune diseases are rare and study participants are unlikely to know about their ANA status and change their behaviors. Moreover, findings were unchanged in sensitivity analyses excluding those reporting psoriasis (the most common dermatologic autoimmune condition) and high sun sensitivity (which may include those with photosensitivity due to conditions such as systemic lupus or dermatomyositis). We did not mutually adjust for sunburn and other sun protective behaviors, since they were not strongly correlated and could in fact be causally related, e.g., greater sunscreen use leading to more risk of sunburn due to greater exposures and fewer protective behaviors. However, joint effect models suggest the observed sunscreen associations were not explained by nor modified by either. Neither did we adjust for current Vitamin D levels, which was not associated with ANA but may mediate some immune effects of UV exposure [19, 27].
4.2.2. Exposures
Because the NHANES data do not include information on the frequency and duration of sun exposure, including daily and periodic exposures at the intensity described in the question on sunscreen use, the reported frequency of use does not represent a cumulative exposure. While no data were available on historic sunscreen use, participants’ report of usual use may reflect longer term patterns. Interpretation of the frequency of sunscreen use was subjective; though comparisons of “never” and “always/most of the time,” are likely to reflect substantial differences in habitual use. Although our exploratory analyses were generally underpowered to detect exposure-ANA associations in stratified models or effect measure modification, we saw suggestive differences among younger adults by gender and race/ethnicity. Differences by demographic factors may be related to different patterns of use or types of sunscreen [46–49], along with other underlying susceptibility factors, providing context for future studies.
Our results do not address potential immunomodulatory ingredients in different types of sunscreens, which besides BP-3 may include other active ingredients, such as organic filters avobenzone, camphor, and cinnamate derivatives, as well as mineral sunscreens like zinc oxide. In addition to changes in the use of sunscreen in recent decades, more products have been released with higher sun protection factors (typically derived from combinations of filters), as well as those designed for daily use [22]. Together these differences may have implications for potential mechanisms relevant to autoimmunity. For example, products could have differential effects by blocking UVA wavelengths as well as UVB. One study found that sunscreens blocking UVA, at the same SPF level, more effectively prevented UV-induced suppression of both local and distant immune responses [50]. Spray-on products may contain propellants and potential inhaled exposures [22, 51]. In 2022, a consumer product screen detected benzene contamination in more sprayed-on sunscreens than in lotions [52]. A cross-sectional NHANES study found no evidence of increased blood benzene concentrations by frequency of overall sunscreen use in 2003–2006 and 2009–2018, however the half-life of benzene is short, and data did not include timing or type of sunscreen use [53]. Sunscreen lotions contain other compounds, such as parabens and phthalates, with potential endocrine effects; however, prior analyses showed no evidence of associations of these chemicals with elevated ANA [17, 54]. Other sunscreen combined ingredients may include insect-repellents or pigments in cosmetics.
4.3. ANA as pre-clinical biomarkers
Our findings address possible associations with pre-clinical autoimmunity, but should not be interperted as causal in relation to disease risk. While the incidence of some autoimmune diseases may be increasing, U.S. research has been limited by a lack of disease registries, further complicated by changes in disease definitions and diagnostic practices over time [55–59]. Serologic markers of autoimmunity, such as ANA, are sensitive biomarkers that can be detected years before disease onset (e.g., systemic lupus erythmatosus, SLE)[60, 61], enabling research on trends and risk factors for pre-clinical autoimmunity. The natural history of autoimmunity is not well understood, especially in asymptomatic individuals in the general population, and autoimmune pathologies and diseases such as SLE are uncommon, arising from the intersection of diverse environmental and genetic risk factors [62]. ANA in our study were assayed in a single lab using standardized methods and quality control protocols [1]. Specific disease-related autoantibodies were not assayed, though these are likely to be only a small fraction of ANA. Prospective studies are needed to better understand the relationship of sunscreen and related expousres in with preclinical autoimmunity and disease among populations at elevated risk, such as family members of affected patients. Our novel findings relating sunburn to the FS pattern among ANA positive individuals warrant further investigation. The DFS pattern includes those with anti-DFS70, an ANA antibody of debated clinical signficance, comprised a larger fraction of ANA in our sample (about 2–3% overall)[63].
4.4. Conclusions
Our novel findings of an association of sunscreen use and ANA prevalence in younger adults may have implications for risk of developing disease-specific autoantibodies and autoimmune diseases in some individuals. However, given the lack of clear associations of sunscreen and ANA at older ages and observed differences in various subgroups, these results should be cautiously interpreted. Most individuals with ANA do not develop a clinical autoimmune diseases and ANA cannot be considered strong pre-clinical markers of disease in the general population. Thus, our findings do not support the cessation of sunscreen use. More data are needed to rule out the role of damaging sun exposure, especially given our findings for specific patterns of ANA. Altogether, these findings indicate a need for larger studies with longitudinal data on ANA and related immune biomarkers, along with detailed data on sun exposure habits and use of sunscreen-containing products with diverse active ingredients.
Sunscreens play an important role in protecting against skin cancer, along with a range of other factors that protect against skin cancer, including physical barriers (e.g., hats and clothing, or staying in the shade). However, with a growing number of chemical sunscreens used broadly across the population, the ease of transdermal absorption, and a lack of robust data on human health effects [64], immune endpoints such as ANA may warrant greater attention in evaluating product safety.
Supplementary Material
Highlights.
Anti-nuclear antibodies (ANA) prevalence in the U.S. population has increased.
Sunscreen use and concentrations of active ingredients have also increased.
In U.S. adults ages 20–39, frequent sunscreen use was associated with ANA.
Related risk factors, including sunburn, did not increase sunscreen-associated ANA.
Studies of sunscreen safety should also include autoimmune endpoints and patients.
Acknowledgements
This research was supported in part by the Intramural Research Program of the NIH, National Institute of Environmental Health Sciences (Z01-ES049028, Z01-ES101074, and P30-ES001247) and contract HHSN273201600011C to Social & Scientific Systems, and in part by the Shaw Scientist Award from the Greater Milwaukee Foundation.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The authors declare they have nothing to disclose.
References
- [1].Dinse GE, Parks CG, Weinberg CR, Co CA, Wilkerson J, Zeldin DC et al. Increasing Prevalence of Antinuclear Antibodies in the United States. Arthritis Rheumatol, 2022;74:2032–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Egeghy PP, Judson R, Gangwal S, Mosher S, Smith D, Vail J et al. The exposure data landscape for manufactured chemicals. Sci Total Environ, 2012;414:159–66. [DOI] [PubMed] [Google Scholar]
- [3].Matta MK, Florian J, Zusterzeel R, Pilli NR, Patel V, Volpe DA et al. Effect of Sunscreen Application on Plasma Concentration of Sunscreen Active Ingredients: A Randomized Clinical Trial. Jama, 2020;323:256–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Benson HA. Assessment and clinical implications of absorption of sunscreens across skin. Am J Clin Dermatol, 2000;1:217–24. [DOI] [PubMed] [Google Scholar]
- [5].Matta MK, Zusterzeel R, Pilli NR, Patel V, Volpe DA, Florian J et al. Effect of Sunscreen Application Under Maximal Use Conditions on Plasma Concentration of Sunscreen Active Ingredients: A Randomized Clinical Trial. Jama, 2019;321:2082–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Kim S, Choi K. Occurrences, toxicities, and ecological risks of benzophenone-3, a common component of organic sunscreen products: a mini-review. Environ Int, 2014;70:143–57. [DOI] [PubMed] [Google Scholar]
- [7].Calafat AM, Wong LY, Ye X, Reidy JA, Needham LL. Concentrations of the sunscreen agent benzophenone-3 in residents of the United States: National Health and Nutrition Examination Survey 2003−−2004. Environ Health Perspect, 2008;116:893–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Ferguson KK, Colacino JA, Lewis RC, Meeker JD. Personal care product use among adults in NHANES: associations between urinary phthalate metabolites and phenols and use of mouthwash and sunscreen. J Expo Sci Environ Epidemiol, 2017;27:326–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Zamoiski RD, Cahoon EK, Michal Freedman D, Linet MS. Self-reported sunscreen use and urinary benzophenone-3 concentrations in the United States: NHANES 2003–2006 and 2009–2012. Environ Res, 2015;142:563–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Trepanowski N, Huang L, Hartman RI. Sunscreen-using U.S. adults are using higher sun protection factor sunscreens: Data from the National Health Interview Survey 2005–2015. J Am Acad Dermatol, 2022;87:1383–6. [DOI] [PubMed] [Google Scholar]
- [11].Berger KP, Kogut KR, Bradman A, She J, Gavin Q, Zahedi R et al. Personal care product use as a predictor of urinary concentrations of certain phthalates, parabens, and phenols in the HERMOSA study. J Expo Sci Environ Epidemiol, 2019;29:21–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Nowak K, Jabłońska E, Ratajczak-Wrona W. Immunomodulatory effects of synthetic endocrine disrupting chemicals on the development and functions of human immune cells. Environ Int, 2019;125:350–64. [DOI] [PubMed] [Google Scholar]
- [13].Weiss V, Gobec M, Jakopin Ž. In vitro investigation of immunomodulatory activities of selected UV-filters. Food Chem Toxicol, 2023;174:113684. [DOI] [PubMed] [Google Scholar]
- [14].Rachoń D, Rimoldi G, Wuttke W. In vitro effects of benzophenone-2 and octyl-methoxycinnamate on the production of interferon-gamma and interleukin-10 by murine splenocytes. Immunopharmacol Immunotoxicol, 2006;28:501–10. [DOI] [PubMed] [Google Scholar]
- [15].Dong F, Zheng M, Wang H, Jing C, He J, Liu S et al. Comparative transcriptome analysis reveals immunotoxicology induced by three organic UV filters in Manila clam (Ruditapes philippinarum). Mar Pollut Bull, 2022;185:114313. [DOI] [PubMed] [Google Scholar]
- [16].Ao J, Yuan T, Gao L, Yu X, Zhao X, Tian Y et al. Organic UV filters exposure induces the production of inflammatory cytokines in human macrophages. Sci Total Environ, 2018;635:926–35. [DOI] [PubMed] [Google Scholar]
- [17].Parks CG, Meier HCS, Jusko TA, Wilkerson J, Miller FW, Sandler DP. Benzophenone-3 and antinuclear antibodies in U.S. adolescents and adults ages 12–39 years. Front Immunol, 2022;13:958527. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Holman DM, Ding H, Guy GP Jr., Watson M, Hartman AM, Perna FM. Prevalence of Sun Protection Use and Sunburn and Association of Demographic and Behaviorial Characteristics With Sunburn Among US Adults. JAMA Dermatol, 2018;154:561–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Hart PH, Gorman S, Finlay-Jones JJ. Modulation of the immune system by UV radiation: more than just the effects of vitamin D? Nat Rev Immunol, 2011;11:584–96. [DOI] [PubMed] [Google Scholar]
- [20].Schleicher RL, Sternberg MR, Lacher DA, Sempos CT, Looker AC, Durazo-Arvizu RA et al. The vitamin D status of the US population from 1988 to 2010 using standardized serum concentrations of 25-hydroxyvitamin D shows recent modest increases. Am J Clin Nutr, 2016;104:454–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Ma Y, Yoo J. History of sunscreen: An updated view. J Cosmet Dermatol, 2021;20:1044–9. [DOI] [PubMed] [Google Scholar]
- [22].Wang SQ, Tanner PR, Lim HW, Nash JF. The evolution of sunscreen products in the United States--a 12-year cross sectional study. Photochem Photobiol Sci, 2013;12:197–202. [DOI] [PubMed] [Google Scholar]
- [23].Meier HCS, Parks CG, Liu HB, Sandler DP, Simonsick EM, Deane K et al. Cellular aging over 13 years associated with incident antinuclear antibody positivity in the Baltimore Longitudinal Study of Aging. J Autoimmun, 2019;105:102295. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Meier HCS, Sandler DP, Simonsick EM, Weng NP, Parks CG. Sex differences in the association between antinuclear antibody positivity with diabetes and multimorbidity in older adults: Results from the Baltimore Longitudinal Study of Aging. Exp Gerontol, 2020;135:110906. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Maverakis E, Miyamura Y, Bowen MP, Correa G, Ono Y, Goodarzi H. Light, including ultraviolet. J Autoimmun, 2010;34:J247–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Wolf SJ, Estadt SN, Theros J, Moore T, Ellis J, Liu J et al. Ultraviolet light induces increased T cell activation in lupus-prone mice via type I IFN-dependent inhibition of T regulatory cells. J Autoimmun, 2019;103:102291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Bernard JJ, Gallo RL, Krutmann J. Photoimmunology: how ultraviolet radiation affects the immune system. Nat Rev Immunol, 2019;19:688–701. [DOI] [PubMed] [Google Scholar]
- [28].Mai ZM, Byrne SN, Little MP, Sargen MR, Cahoon EK. Solar UVR and Variations in Systemic Immune and Inflammation Markers. JID Innov, 2021;1:100055. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [29].Mustieles V, Balogh RK, Axelstad M, Montazeri P, Marequéz S, Vrijheid M, et al. Benzophenone-3: Comprehensive Review of the Toxicological and Human Evidence with Meta-Analysis of Human Biomonitoring Studies. Environ Int 2023; 173:107739. [DOI] [PubMed] [Google Scholar]
- [29].Scinicariello F, Buser MC. Serum Testosterone Concentrations and Urinary Bisphenol A, Benzophenone-3, Triclosan, and Paraben Levels in Male and Female Children and Adolescents: NHANES 2011–2012. Environ Health Perspect, 2016;124:1898–904. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Damian DL, Patterson CR, Stapelberg M, Park J, Barnetson RS, Halliday GM. UV radiation-induced immunosuppression is greater in men and prevented by topical nicotinamide. J Invest Dermatol, 2008;128:447–54. [DOI] [PubMed] [Google Scholar]
- [31].Widyarini S, Domanski D, Painter N, Reeve VE. Estrogen receptor signaling protects against immune suppression by UV radiation exposure. Proc Natl Acad Sci U S A, 2006;103:12837–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Meier HC, Sandler DP, Simonsick EM, Parks CG. Association between Vitamin D Deficiency and Antinuclear Antibodies in Middle-Aged and Older U.S. Adults. Cancer Epidemiol Biomarkers Prev, 2016;25:1559–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [33].Ritterhouse LL, Crowe SR, Niewold TB, Kamen DL, Macwana SR, Roberts VC et al. Vitamin D deficiency is associated with an increased autoimmune response in healthy individuals and in patients with systemic lupus erythematosus. Annals of the rheumatic diseases, 2011;70:1569–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [34].Young KA, Munroe ME, Guthridge JM, Kamen DL, Niewold TB, Gilkeson GS et al. Combined role of vitamin D status and CYP24A1 in the transition to systemic lupus erythematosus. Annals of the rheumatic diseases, 2017;76:153–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [35].Cutolo M, Smith V, Paolino S, Gotelli E. Involvement of the secosteroid vitamin D in autoimmune rheumatic diseases and COVID-19. Nat Rev Rheumatol, 2023;19:265–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [36].Damoiseaux J, Andrade LEC, Carballo OG, Conrad K, Francescantonio PLC, Fritzler MJ, et al. Clinical relevance of HEp-2 indirect immunofluorescent patterns: the International Consensus on ANA patterns (ICAP) perspective. Annals of the rheumatic diseases. 2019;78(7):879–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [37].Ochs RL, Mahler M, Basu A, Rios-Colon L, Sanchez TW, Andrade LE, et al. The significance of autoantibodies to DFS70/LEDGFp75 in health and disease: integrating basic science with clinical understanding. Clin Exp Med. 2016;16(3):273–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [38].Mariz HA, Sato EI, Barbosa SH, Rodrigues SH, Dellavance A, Andrade LE. Pattern on the antinuclear antibody-HEp-2 test is a critical parameter for discriminating antinuclear antibody-positive healthy individuals and patients with autoimmune rheumatic diseases. Arthritis Rheum. 2011;63(1):191–200. [DOI] [PubMed] [Google Scholar]
- [39].Conrad K, Röber N, Andrade LE, Mahler M. The Clinical Relevance of Anti-DFS70 Autoantibodies. Clin Rev Allergy Immunol. 2017;52(2):202–16. [DOI] [PubMed] [Google Scholar]
- [40].Carbone T, Pafundi V, Tramontano G, Gilio M, Padula MC, Padula AA, D’Angelo S. Prevalence and serological profile of anti-DFS70 positive subjects from a routine ANA cohort. Sci Rep. 2019;9(1):2177. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [41].Yu C, Chang C, Zhang J. Immunologic and genetic considerations of cutaneous lupus erythematosus: a comprehensive review. J Autoimmun. 2013;41:34–45. [DOI] [PubMed] [Google Scholar]
- [43].Pickering MC, Fischer S, Lewis MR, Walport MJ, Botto M, Cook HT. Ultraviolet-radiation-induced keratinocyte apoptosis in C1q-deficient mice. J Invest Dermatol. 2001;117(1):52–8. [DOI] [PubMed] [Google Scholar]
- [44].Wolf SJ, Estadt SN, Gudjonsson JE, Kahlenberg JM. Human and Murine Evidence for Mechanisms Driving Autoimmune Photosensitivity. Front Immunol. 2018;9:2430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [45].Hile GA, Kahlenberg JM. Immunopathogenesis of skin injury in systemic lupus erythematosus. Curr Opin Rheumatol. 2021;33(2):173–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [46].Buller DB, Cokkinides V, Hall HI, Hartman AM, Saraiya M, Miller E et al. Prevalence of sunburn, sun protection, and indoor tanning behaviors among Americans: review from national surveys and case studies of 3 states. J Am Acad Dermatol, 2011;65:S114–23. [DOI] [PubMed] [Google Scholar]
- [47].Bandi P, Cokkinides VE, Weinstock MA, Ward E. Sunburns, sun protection and indoor tanning behaviors, and attitudes regarding sun protection benefits and tan appeal among parents of U.S. adolescents-1998 compared to 2004. Pediatr Dermatol, 2010;27:9–18. [DOI] [PubMed] [Google Scholar]
- [48].Calderón TA, Bleakley A, Jordan AB, Lazovich D, Glanz K. Correlates of sun protection behaviors in racially and ethnically diverse U.S. adults. Prev Med Rep, 2019;13:346–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [49].Robinson JK, Rigel DS, Amonette RA. Trends in sun exposure knowledge, attitudes, and behaviors: 1986 to 1996. J Am Acad Dermatol, 1997;37:179–86. [DOI] [PubMed] [Google Scholar]
- [50].Moyal DD, Fourtanier AM. Broad-spectrum sunscreens provide better protection from solar ultraviolet-simulated radiation and natural sunlight-induced immunosuppression in human beings. J Am Acad Dermatol, 2008;58:S149–54. [DOI] [PubMed] [Google Scholar]
- [51].Teplitz RW, Glazer AM, Svoboda RM, Rigel DS. Trends in US sunscreen formulations: Impact of increasing spray usage. J Am Acad Dermatol, 2018;78:187–9. [DOI] [PubMed] [Google Scholar]
- [52].Hudspeth A, Zenzola N, Kucera K, Wu Q, Light D. Independent Sun Care Product Screening for Benzene Contamination. Environ Health Perspect, 2022;130:37701. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [53].Chang MS, Moore KJ, Trepanowski N, Koru-Sengul T, Hartman RI. Sunscreen use is not associated with increased blood concentrations of benzene among adults in the United States: Data from the National Health and Nutrition Examination Survey 2003–2006 and 2009–2018. J Am Acad Dermatol, 2022;87:440–3. [DOI] [PubMed] [Google Scholar]
- [54].Dinse GE, Co CA, Parks CG, Weinberg CR, Xie G, Chan EKL et al. Expanded assessment of xenobiotic associations with antinuclear antibodies in the United States, 1988–2012. Environ Int, 2022;166:107376. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [55].Uramoto KM, Michet CJ Jr., Thumboo J, Sunku J, O’Fallon WM, Gabriel SE. Trends in the incidence and mortality of systemic lupus erythematosus, 1950–1992. Arthritis Rheum, 1999;42:46–50. [DOI] [PubMed] [Google Scholar]
- [56].Almallouhi E, King KS, Patel B, Wi C, Juhn YJ, Murray JA et al. Increasing Incidence and Altered Presentation in a Population-based Study of Pediatric Celiac Disease in North America. J Pediatr Gastroenterol Nutr, 2017;65:432–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [57].Myasoedova E, Crowson CS, Kremers HM, Therneau TM, Gabriel SE. Is the incidence of rheumatoid arthritis rising?: results from Olmsted County, Minnesota, 1955–2007. Arthritis Rheum, 2010;62:1576–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [58].Icen M, Crowson CS, McEvoy MT, Dann FJ, Gabriel SE, Maradit Kremers H. Trends in incidence of adult-onset psoriasis over three decades: a population-based study. J Am Acad Dermatol, 2009;60:394–401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [59].Cooper GS, Bynum ML, Somers EC. Recent insights in the epidemiology of autoimmune diseases: improved prevalence estimates and understanding of clustering of diseases. J Autoimmun, 2009;33:197–207. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [60].Arbuckle MR, McClain MT, Rubertone MV, Scofield RH, Dennis GJ, James JA et al. Development of autoantibodies before the clinical onset of systemic lupus erythematosus. N Engl J Med, 2003;349:1526–33. [DOI] [PubMed] [Google Scholar]
- [61].Ma WT, Chang C, Gershwin ME, Lian ZX. Development of autoantibodies precedes clinical manifestations of autoimmune diseases: A comprehensive review. J Autoimmun, 2017;83:95–112. [DOI] [PubMed] [Google Scholar]
- [62].Woo JMP, Parks CG, Jacobsen S, Costenbader KH, Bernatsky S. The role of environmental exposures and gene-environment interactions in the etiology of systemic lupus erythematous. J Intern Med, 2022;291:755–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [63].Dinse GE, Zheng B, Co CA, Parks CG, Weinberg CR, Miller FW et al. Anti-dense fine speckled 70 (DFS70) autoantibodies: correlates and increasing prevalence in the United States. Front Immunol, 2023;14:1186439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [64].Pantelic MN, Wong N, Kwa M, Lim HW. Ultraviolet filters in the United States and European Union: A review of safety and implications for the future of US sunscreens. J Am Acad Dermatol, 2022. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.