Circulating IL-33 as a sensitive alarmin in systemic lupus erythematosus: a biomarker for disease activity and skin-specific manifestations Authors: Li Li, Xi Tan, Chen Zhao, Sijian Wen, Wenyu Li, Runge Fan, Wei Hou, Youkun Lin Published: 2026 ABSTRACT Interleukin-33 (IL-33), an alarmin predominantly released by keratinocytes and endothelial cells, plays a pivotal role in type 2 immune responses and has been linked to the pathogenesis of autoimmune diseases, including systemic lupus erythematosus (SLE). This study aimed to explore the relationship between plasma IL-33 levels and clinical features in SLE patients. A cross-sectional study was conducted involving 133 SLE patients and 81 normal controls. Plasma IL-33 levels were measured via enzyme-linked immunosorbent assay. Clinical parameters, such as SLEDAI-2K scores, photosensitivity, skin lesion distribution, laboratory indices, immunofluorescence and immunohistochemistry results were collected. SLE patients demonstrated significantly higher circulating IL-33 levels compared to healthy controls (median 342.60 pg/ml vs. 56.77 pg/ml, p < 0.0001). IL-33 levels were positively associated with SLEDAI-2K scores (r = 0.279, p = 0.001), photosensitivity (p = 0.02), and facial skin lesions (p = 0.001). The expression levels of the IL-33 receptor (ST2) in head and face cutaneous tissues were markedly higher than those in other anatomical regions and in normal control subjects (p < 0.001). Multivariate analysis revealed negative correlations with complement C3 (β = − 0.303, p < 0.001) and positive correlations with IgA (β = 0.473, p < 0.001) and hs-CRP (β = 0.310, p < 0.001). Notably, IL-33 levels were higher in male patients (p = 0.002) and treatment-naïve patients (p = 0.03). Our data associate circulating IL-33 with SLE activity, photosensitivity, and specific serological markers. For the association between IL-33 and cutaneous photosensitivity, this study provides further histological evidence. This finding provides a novel perspective on the mechanism of photosensitivity in SLE, demonstrating the potential of IL-33 as a biomarker for patient stratification management. Introduction Systemic Lupus Erythematosus (SLE) is a complex, multifactorial autoimmune disease characterized by loss of immune tolerance, production of pathogenic autoantibodies, and inflammation leading to multi-organ damage1. Many research focuses on identifying key cytokines that drive pathogenesis and serve as reliable biomarkers for disease activity and organ involvement2. Interleukin-33 (IL-33), a member of the IL-1 cytokine family, has emerged as a pivotal player in innate and adaptive immune responses. Initially identified in 2003 as Nuclear Factor from High Endothelial Venules (NF-HEV) and later characterized in 2005, IL-33 functions as an ‘alarmin’ released upon cellular damage or stress from constitutive sources like keratinocytes, epithelial cells, and endothelial cells3. A recent study has underscored its crucial role in modulating immune responses and mediating inflammatory processes4. Over the past decade, while major breakthroughs have been achieved in the approved treatments for SLE and lupus nephritis, substantial heterogeneity persists in disease progression among patients, with mortality rates remaining unacceptably high5. Additionally, the accessibility of these high-cost therapies varies considerably across global regions6. The 2025 SLE Treatment Guidelines issued by the American College of Rheumatology (ACR) emphasize the critical importance of regularly assessing disease activity and cumulative organ damage in SLE patients. Low disease activity and minimal cumulative damage are closely associated with better long-term prognosis7–9. Assessing the disease activity of SLE has long been a challenge for clinicians. The key to overcoming these challenges lies in identifying unbiased lupus biomarkers. Such biomarkers should possess dual functions: both reflecting disease activity and enabling early quantitative assessment of organ involvement10.The integration of these dual-functional molecular signatures into clinical workflows would significantly enhance the monitoring and management of SLE, thereby improving diagnostic precision and facilitating personalized treatment strategies. Previous several clinical studies have demonstrated that the expression of IL-33 is significantly upregulated in patients with SLE11–14 and is positively correlated with the SLE Disease Activity Index (SLEDAI)11. The level of sST2 (soluble receptor for IL-33) in patients with SLE and lupus nephritis was also significantly upregulated and positively correlated with SLEDAI15–17.These studies indicate that the activation of the IL-33/ST2 signaling axis is closely associated with disease severity.As a keratinocyte-derived “alarmin”, interleukin-33 (IL-33) is widely expressed in the skin tissue. Existing animal experiments have confirmed that UVB stimulation can significantly up-regulate the expression of IL-33 in skin tissue18. In addition, cellular-level studies have demonstrated that the activation of the IL-33/ST2 signaling axis is closely associated with photosensitivity in lupus19. However, no clinical studies have yet been reported on the correlation between IL-33/ST2 signaling axis and lupus photosensitivity. Compared with locally expressed IL-33 in tissues, circulating IL-33 have convenient advantages to serve as an early warning indicator for reflecting the activation level of the systemic immune system induced by external stimuli such as ultraviolet B (UVB) radiation, tobacco exposure and stress. Therefore, more empirical research and clinical observations are still needed in humans. Most of the published clinical studies have focused on the correlation between IL-33 and lupus-specific serological markers. Some researchers have also found that photosensitivity in SLE patients can lead to specific skin lesions within a certain range, and there are differences in the immune cell–stromal cell regulatory networks between healthy individuals and SLE patients20. However, clinical studies regarding the association of this cytokine with different skin lesion sites and photosensitivity in patients with SLE have not been reported to date. In addition to analyzing the correlation between plasma IL-33 and common serological markers in patients with systemic lupus erythematosus (SLE), this study compared the expression levels of ST2 in tissue samples from different skin lesion sites. This finding may provide a novel perspective for the mechanism underlying photosensitivity in SLE, and demonstrate the potential of IL-33 as a biomarker for stratified management of patients’ sensitivity to environmental exposures. Materials and methods Sample size calculation and recruitment The sample size was determined a priori by utilizing the formula for comparing means between two independent groups (SLE patients vs. normal controls)21,22. Based on pre-experimental data, it was estimated that the mean plasma IL-33 level in the SLE group was 140 pg/ml, while in the control group, it was 100 pg/ml, with a common standard deviation of 75.0 pg/ml. With a significance level (α) of 0.05, a power (1 − β) of 0.8, and a 1:1 case–control ratio, the minimum sample size required for each group was calculated to be 56. Considering a potential 10% attrition rate caused by invalid data or sample contamination, a minimum of 62 subjects per group was targeted. Ultimately, 133 patients with SLE and 81 healthy controls were enrolled in this study, with the sample size of each group meeting the minimum requirement. The formula for sample size calculation is as follows.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} & n_{1} = \frac{{\left( {z_{{1 - \frac{\alpha }{2}}} + z_{1 - \beta } } \right)^{2} \left[ {\sigma_{1}^{2} + \frac{{\sigma_{2}^{2} }}{r}} \right]}}{{\Delta^{2} }} \\ & r = \frac{{n_{2} }}{{n_{1} }},\Delta = \mu_{1} - \mu_{2} \\ \end{aligned}$$\end{document} Human samples A total of 133 SLE patients and 81 age- and sex-matched NC were recruited from the Department of Dermatology and Venereology, The First Affiliated Hospital of Guangxi Medical University, between January 2023 and December 2024. This cross-sectional study was carried out in strict accordance with the Declaration of Helsinki and received approval from the Institutional Ethics Committee of The First Affiliated Hospital of Guangxi Medical University (Approval No: 2021-KY-国基-195). Written informed consent was obtained from all participants or their guardians prior to the collection of samples and clinical data. Patient exclusion criteria were meticulously defined to minimize confounding factors and included: (1) Coexisting allergic diseases; (2) Active or history of malignancy; (3) Current or past tuberculosis infection; (4) Radiological evidence of active pulmonary infection; (5) Overlap syndromes with other connective tissue diseases (e.g., rheumatoid arthritis, scleroderma); (6) Primary or secondary immunodeficiency disorders (e.g., HIV/AIDS, Wiskott-Aldrich syndrome) or patients on long-term immunosuppressants for non-SLE conditions; (7) Pregnancy or lactation; (8) Chronic active viral infections (e.g., EBV, CMV) as determined by serology. Among the 133 patients with SLE , 22 patients were treatment-naïve, meaning they had received no prior non-steroidal anti-inflammatory drugs (NSAIDs), glucocorticoid, hydroxychloroquine, immunosuppressants, or other immunomodulatory agents. The remaining 111 patients were undergoing one or a combination of these treatments at the time of enrollment. Eighty-one normal controls were recruited from the health examination center of the same hospital. They had no history of autoimmune diseases, chronic infections, or active malignancies. Clinical assessment and data collection A comprehensive collection of demographic and clinical data was gathered for all patients with Systemic Lupus Erythematosus (SLE). Trained rheumatologists quantitatively evaluated the disease activity of all patients using the Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K)23. According to the SLEDAI-2K guidelines, a score ranging from 0 to 4 indicates no disease activity, 5 to 10 represents mild to moderate disease activity, and a score of 11 or higher signifies high disease activity. Clinical data extracted included: disease duration, date of diagnosis, organ system involvement (e.g., cutaneous, renal, neuropsychiatric, hematologic), detailed documentation of cutaneous manifestations (photosensitivity, specific lesion types, and distribution across head/face, limbs, trunk, and mucosa). Photosensitivity was defined as a patient-reported history of an unusual skin rash following sunlight exposure, as per ACR criteria. Laboratory parameters collected included: complete blood count (CBC), urinalysis, renal function tests (serum creatinine, blood urea nitrogen), serological markers (anti-dsDNA antibodies, IgG, IgA, IgM), complement levels (C3, C4, C1q), erythrocyte sedimentation rate (ESR), and high-sensitivity C-reactive protein (hs-CRP). Enzyme-linked immunosorbent assay (ELISA) The ELISAs were conducted according to the manufacturer’s recommendations. To minimize in vitro degradation, all blood samples were treated with EDTA anticoagulation (EDTA inhibits metalloproteinase-mediated cleavage of IL-33) and processed within 30 min after collection. The centrifugation process was strictly maintained at 4 °C, followed by immediate storage at − 80 °C. A protease inhibitor (Thermo Fisher Scientific 87786) was added prior to storage to prevent proteolysis of IL-33. The concentrations of IL-33 in plasma were measured using specific human ELISA assay kits (Elabscience Biotechnology Inc. Wuhan, Hubei). The minimum detection concentration was 9.38 pg/mL for IL-3324,25. Immunofluorescence (IF) The tissues were fixed in formalin and embedded in paraffin for sectioning. Paraffin sections were then deparaffinized and rehydrated and the subsequently treated. The tissues were cultivated overnight with either anti-rabbit IL-1RL1/ST2 (11920-1-AP, proteintech), IgG diluted to 1:100 served as the I antibody (4 μg/ml, 4 °C). Subsequently, phosphate-buffered saline (PBS) was used for washing, and Alexa Fluor 594- (red) conjugated II antibodies (1:1000) supplied by Invitrogen, Grand Island, NY, USA, were utilized for incubation (1 h, 37 °C). Then, the nucleus staining was conducted in a mounting medium added with DAPI (4,6-diamidino-2-phenylindole). Isotype IgG antibodies of each fluorochrome were used as nucleic controls. Image acquisition was conducted with the use of a NIKON Eclipse ci microscope. The software Saiviwer supplied by Servicebio was used for the determination of positively stained areas. Four to six viewing fields were chosen randomly within each section. Three blinded pathologists performed the analysis, and the percentage of positive staining areas was calculated. Immunohistochemistry Skin tissues were collected and fixed in 4% formaldehyde. Tissue sections were deparaffinized, processed for antigen retrieval in sodium citrate buffer (pH 7.4), and blocked in 3% BSA. Sections were incubated with anti-rabbit IL-1RL1/ST2 (11920-1-AP, Proteintech), overnight at 4℃. For detection by immunohistochemistry, sections were incubated with 1:500 dilutions of biotin-labeled anti-rabbit (GB23303, Servicebio) and 3,3'-diaminobenzidine (DAB) substrate (G1212, Servicebio) according to the manufacturers’ instructions. Image acquisition was conducted with the use of a NIKON Eclipse ci microscope. The software Saiviwer supplied by Servicebio was used for the determination of positively stained areas. Four to six viewing fields were chosen randomly within each section. Three blinded pathologists performed the analysis, and the percentage of positive staining areas was calculated. Statistical analysis Statistical analyses were performed using GraphPad Prism software (version 9.0) and SPSS Statistics (version 26.0). Normality of continuous data distribution was assessed using the Shapiro–Wilk test. Data are presented as mean ± standard deviation (SD) for normally distributed variables, median with interquartile range (IQR) for non-normally distributed variables, and numbers (percentages) for categorical variables. Group comparisons for continuous variables were conducted using the Student’s t-test (parametric) or Mann–Whitney U test (non-parametric). Comparisons among more than two groups were performed using one-way ANOVA (parametric) or Kruskal–Wallis test (non-parametric), followed by appropriate post-hoc tests. Categorical variables were compared using the Chi-square test or Fisher’s exact test. Correlations between plasma IL-33 levels (non-normally distributed) and continuous/ordinal clinical variables (e.g., SLEDAI-2K score, age) were analyzed using Spearman’s rank correlation coefficient. The diagnostic performance of IL-33 for identifying active disease (SLEDAI-2K ≥ 5) was evaluated using Receiver Operating Characteristic (ROC) curve analysis, determining the area under the curve (AUC), optimal cutoff value, sensitivity, and specificity. Results Characteristics of the study population A total of 133 SLE patients and 81 normal controls (NC) were included in this cross-sectional analysis. The SLE group consisted of 118 (88.7%) females and 15 (11.3%) males, with a female-to-male ratio of approximately 7.87:1 and a mean age of 33.78 ± 13.00 years (range 8–63 years). The normal control group comprised 58 (71.6%) females and 23 (28.4%) males (ratio: 2.52:1) with a mean age of 36.11 ± 13.96 years (range 19–73 years). Both groups were predominantly female, and the 18–30 age group accounted for the largest proportion in both cohorts. The demographic data is presented in Fig. 1A, B.Fig. 1Comparison of SLE patients and normal controls. (A) The distribution of age and sex among NC group. (B) The distribution of age and sex among SLE group. (C) The comparison of plasma IL-33 between SLE group and NC group. Comparison of SLE patients and normal controls. (A) The distribution of age and sex among NC group. (B) The distribution of age and sex among SLE group. (C) The comparison of plasma IL-33 between SLE group and NC group. As shown in Fig. 1C, plasma IL-33 levels were significantly elevated in SLE patients compared with the NC group. The median IL-33 concentration in the SLE group was 342.60 pg/ml (IQR 41.98–1897.62 pg/ml), markedly higher than the median level of 56.77 pg/ml (IQR 15.84–104.95 pg/ml) observed in the NC group. Normality testing via the Shapiro–Wilk test confirmed that the data for both groups were non-normally distributed (p < 0.05 for both). Consequently, the non-parametric Mann–Whitney U test was employed for comparison, revealing a statistically significant difference (p < 0.001). As the mean values were used for sample size calculation during study design, we also calculated the statistical power of the actual data. The sample size, mean IL‑33 level, and standard deviation for the SLE group were 133, 480.89, and 534.13, respectively; those for the control group were 81, 98.51, and 80.02, respectively. The calculated statistical power was 1.0 , which exceeds the expected power of 0.8, indicating that the results are representative22,26. Among the 133 SLE patients, 58 (43.6%) had lupus nephritis, 6 (4.5%) had neuropsychiatric lupus, 30 (22.6%) had hematologic involvement, and 30 (22.6%) had concurrent infections in various organs. Cardiac involvement was noted in 6 (4.5%) patients. Photosensitivity, a key cutaneous feature under investigation, was present in 40 (30.1%) patients. Regarding the distribution of skin lesions, 46 (34.6%) patients had facial lesions, 24 (18.0%) had lesions on the limbs, 11 (8.3%) on the trunk, and 9 (6.8%) had mucosal lesions. For the 55 patients for whom the date of symptom onset was accurately recorded, the median time from symptom onset to formal diagnosis was 110 days (IQR 60, 352 days). Correlation of plasma IL-33 levels with age and sex in SLE patients Within the SLE patient cohort, we analyzed the relationship between plasma IL-33 levels and basic demographic factors. A slight increasing trend in IL-33 levels with advancing age was observed; however, Spearman correlation analysis found this trend to be statistically non-significant (p = 0.611). In contrast, a striking and statistically significant difference was found based on sex. Male SLE patients (n = 15) exhibited a significantly higher median IL-33 level compared to female patients (n = 118) (p = 0.002). It is shown as Fig. 2A–D.Fig. 2Comparison of plasma IL-33 levels in different subgroups of SLE patients. (A) Age-dependent trend of plasma IL-33 in SLE patients. (B) Correlation between plasma IL-33 and SLEDAI in SLE patients. (C) ROC curve of plasma IL-33 in SLE patients predicting clinical severity (D) Gender differences in plasma IL-33 levels in SLE patients. (E) Variation in IL-33 levels between new patients and those who had received prior immunomodulatory treatment. (F) History of photosensitivity differences in Plasma IL-33 Levels in SLE Patients (G) Patients were stratified based on the location of skin lesions to compare differences in plasma IL-33 levels. Comparison of plasma IL-33 levels in different subgroups of SLE patients. (A) Age-dependent trend of plasma IL-33 in SLE patients. (B) Correlation between plasma IL-33 and SLEDAI in SLE patients. (C) ROC curve of plasma IL-33 in SLE patients predicting clinical severity (D) Gender differences in plasma IL-33 levels in SLE patients. (E) Variation in IL-33 levels between new patients and those who had received prior immunomodulatory treatment. (F) History of photosensitivity differences in Plasma IL-33 Levels in SLE Patients (G) Patients were stratified based on the location of skin lesions to compare differences in plasma IL-33 levels. IL-33 as an indicator of disease intensity Disease activity, as measured by the SLEDAI-2K score, across the 133 SLE patients had a median value of 4 (IQR 2, 5), with a range from 1 to 31. Spearman rank correlation analysis revealed a statistically significant, positive correlation between plasma IL-33 levels and the SLEDAI-2K score (r = 0.279, p = 0.0011), indicating that higher IL-33 levels are associated with greater disease activity (Fig. 2B). To further evaluate the potential of IL-33 as a biomarker for identifying active disease, we performed ROC curve analysis using a SLEDAI-2K score of 5 as the cutoff to define active disease. The Area Under the Curve (AUC) was 0.6954 (95% CI 0.5986–0.7921, p < 0.001). The optimal cutoff value for IL-33 that maximized both sensitivity and specificity was determined to be 312.42 pg/ml. At this threshold, the sensitivity for detecting active disease was 75.0%, while the specificity was 56.2%. The result is shown in Fig. 2C. Further investigation revealed substantial variation in IL-33 levels between new patients and those who had received prior immunomodulatory treatment. The 22 new patients had a significantly higher median IL-33 level of 427.27 pg/ml (range 138.12–1897.62 pg/ml) compared to the 111 patients who had received prior immunomodulatory treatment (median: 310.34 pg/ml, range 22.86–3331.44 pg/ml). The difference was statistically significant (p = 0.03), indicating that higher IL-33 levels were associated with distinct immune activation states (Fig. 2E). Correlation of plasma IL-33 levels with photosensitivity Patients who reported a history of photosensitivity (n = 40) demonstrated a median (interquartile range, IQR) IL-33 level was 415.90 pg/ml (281.86–540.24 pg/ml). This was significantly higher than the median (IQR) level of 308.96 pg/ml (188.37–452.36 pg/ml) found in patients without photosensitivity (n = 93) (p = 0.02). The results are shown in Fig. 2F. Stratifying patients based on different locations of skin lesions yielded intriguing results. Patients with facial (head) lesions (n = 46) had significantly higher IL-33 levels (median: 407.58 pg/ml) compared to those without facial lesions (n = 87, median: 308.96 pg/ml; p = 0.001). In contrast, plasma IL-33 levels did not differ significantly according to the presence or absence of skin lesions on the limbs (n = 24 vs. n = 109; p > 0.05), trunk (n = 11 vs. n = 122; p > 0.05), or mucosa (n = 10 vs. n = 123; p > 0.05), as illustrated in Fig. 2G. ST2 as a marker of skin damage with photosensitivity We found a significant association between circulating IL-33 and Distribution of. Skin lesions in lupus patients. Therefore, we selected different skin lesion sites (including head and face, trunk, and limbs) in normal individuals and SLE patients for histological exploration of ST2 expression. As a receptor of IL-33, ST2 exhibits differential expression across various skin lesion areas, with the highest expression observed in the head and face regions where cumulative light exposure time is prolonged, suggesting IL-33’s potential role in reflecting photosensitivity damage. Median ST2 expression levels in cutaneous tissues of the head and face, and limbs in patients with SLE were 2.39 and 2.22, respectively. By Mann–Whitney U test, ST2 expression levels in head and face cutaneous tissues were significantly higher than those in other anatomical regions and in NC subjects (p < 0.001). The results are illustrated in Fig. 3.Fig. 3ST2 expression in different skin lesion sites in SLE. (A) ST2 expression in skin lesions tissue biopsies. Arrows indicate ST2-positive epithelium; (B) Bars indicate the median difference between SLE and normal control subjects.*p ≤ 0.05; **p ≤ 0.01; (C) Bars indicate the median difference differences across different regions. (D) Immunofluorescence staining of IL1RL1-positive cells shows red; nuclei appear blue after DAPI re-staining. Images are arranged with the epidermal direction as the top orientation. (Scale bar: 100 μm). ST2 expression in different skin lesion sites in SLE. (A) ST2 expression in skin lesions tissue biopsies. Arrows indicate ST2-positive epithelium; (B) Bars indicate the median difference between SLE and normal control subjects.*p ≤ 0.05; **p ≤ 0.01; (C) Bars indicate the median difference differences across different regions. (D) Immunofluorescence staining of IL1RL1-positive cells shows red; nuclei appear blue after DAPI re-staining. Images are arranged with the epidermal direction as the top orientation. (Scale bar: 100 μm). Multifactorial predictive modeling of disease parameters As illustrated in Fig. 4A, a correlation heatmap was constructed to visualize the pairwise correlations between all clinical parameters measured in the SLE cohort. The heatmap revealed several strong correlations, notably a strong negative correlation between neutrophil percentage and lymphocyte percentage (r ≈ − 0.9), a strong positive correlation between random urine protein and the urine protein-to-creatinine ratio (r ≈ 0.8), and a positive correlation between IgM and complement C1q (r ≈ 0.61). Weaker but notable positive correlations were observed between WBC and platelet counts (r ≈ 0.5) and between complement C4 and C1q (r ≈ 0.47).Fig. 4Correlation of various serological markers in SLE patients. (A) A correlation heatmap was constructed to visualize the pairwise correlations between all clinical parameters measured in the SLE cohort. (B)–(D) The trend of serological indicators (IgG, IgA, hs-CRP) significantly correlated with plasma IL-3. Correlation of various serological markers in SLE patients. (A) A correlation heatmap was constructed to visualize the pairwise correlations between all clinical parameters measured in the SLE cohort. (B)–(D) The trend of serological indicators (IgG, IgA, hs-CRP) significantly correlated with plasma IL-3. Univariate linear regression analysis identified several clinical laboratory parameters potentially associated with plasma IL-33 levels (p < 0.05), including white blood cell (WBC) count (negative correlation), complement C3 (negative correlation), IgG, IgA, hs-CRP (positive correlation), ESR (positive correlation) and SLEDAI-2K score (positive correlation). These significant variables from the univariate analysis were subsequently entered into a multiple linear regression model to identify independent predictors. The multifactorial analysis results, summarized in Table 1 and Fig. 4B–D, indicated that complement C3 maintained a significant negative correlation with IL-33 levels (β = -0.338, p < 0.001). Conversely, IgA (β = 0.509, p < 0.001) and hs-CRP (β = 0.289, p < 0.001) were independently and positively associated with plasma IL-33 levels. WBC count, IgG, and ESR did not retain independent significance in the multivariate model.Table 1Linear regression analysis of plasma IL-33 level and various clinical indicators in SLE patients.ArgumentMedian (P25, P75)Simple factor regressionMultiple regressionΒ′PΒ′PWhite blood cell count (109/L)5.03 (3.87, 6.91)− 0.2730.002− 0.1210.123Red blood cell count (1012/L)4.2 (3.78, 4.54)− 0.1480.091Hemoglobin (g/L)116.4 (101.93, 129.38)− 0.1230.161Platelets (109/L)238 (183.55, 286.88)− 0.1510.084Neutrophil percentage0.64 (0.57, 0.72)0.1080.216Percentage of lymphocytes0.24 (0.18, 0.31)− 0.1010.249Percentage of monocytes0.09 (0.07, 0.12)− 0.1230.160Randomized urine protein (mg/L)144.9 (80, 433.85)0.1040.336Random urine creatinine (μmol/L)9529 (6456, 16,645.5)0.0720.507Urea/protein/creatinine113.71 (52.96, 409.67)0.1840.088Anti-BSD antibody (IU/ml)97.20 (47.6, 283.25)0.1400.112IgG (g/L)16.18 (12.78, 21.07)0.2640.0050.0940.224IgA (g/L)2.33 (1.72, 3.08)0.5000.0000.5090.000IgM (g/L)0.87 (0.51, 1.21)− 0.0710.462Complement C3 (g/L)0.77 (0.62, 0.92)− 0.3350.000− 0.3380.000Complement C4 (g/L)0.17 (0.12, 0.22)− 0.0530.547Super sensitive C-reactive protein (mg/L)1.40 (0.50, 3.95)0.2710.0020.2890.000erythrocyte sedimentation rate (mm/h)19.00 (11.00, 34.00)0.1800.039− 0.0120.875Complement C1q (mg/L)134.60 (114.90, 151.50)0.0330.796SLEDAI-2K score4 (2, 5)0.2760.0010.0990.245*Standardized regression coefficients. Table 1Linear regression analysis of plasma IL-33 level and various clinical indicators in SLE patients.ArgumentMedian (P25, P75)Simple factor regressionMultiple regressionΒ′PΒ′PWhite blood cell count (109/L)5.03 (3.87, 6.91)− 0.2730.002− 0.1210.123Red blood cell count (1012/L)4.2 (3.78, 4.54)− 0.1480.091Hemoglobin (g/L)116.4 (101.93, 129.38)− 0.1230.161Platelets (109/L)238 (183.55, 286.88)− 0.1510.084Neutrophil percentage0.64 (0.57, 0.72)0.1080.216Percentage of lymphocytes0.24 (0.18, 0.31)− 0.1010.249Percentage of monocytes0.09 (0.07, 0.12)− 0.1230.160Randomized urine protein (mg/L)144.9 (80, 433.85)0.1040.336Random urine creatinine (μmol/L)9529 (6456, 16,645.5)0.0720.507Urea/protein/creatinine113.71 (52.96, 409.67)0.1840.088Anti-BSD antibody (IU/ml)97.20 (47.6, 283.25)0.1400.112IgG (g/L)16.18 (12.78, 21.07)0.2640.0050.0940.224IgA (g/L)2.33 (1.72, 3.08)0.5000.0000.5090.000IgM (g/L)0.87 (0.51, 1.21)− 0.0710.462Complement C3 (g/L)0.77 (0.62, 0.92)− 0.3350.000− 0.3380.000Complement C4 (g/L)0.17 (0.12, 0.22)− 0.0530.547Super sensitive C-reactive protein (mg/L)1.40 (0.50, 3.95)0.2710.0020.2890.000erythrocyte sedimentation rate (mm/h)19.00 (11.00, 34.00)0.1800.039− 0.0120.875Complement C1q (mg/L)134.60 (114.90, 151.50)0.0330.796SLEDAI-2K score4 (2, 5)0.2760.0010.0990.245*Standardized regression coefficients. Linear regression analysis of plasma IL-33 level and various clinical indicators in SLE patients. *Standardized regression coefficients. Discussion This study offers compelling evidence indicating that IL-33, a highly promising biomarker, can be employed to assist in the evaluation of sub-organ lesions in patients with SLE. Our findings not only corroborate earlier reports of elevated IL-33 in SLE patients when compared with normal controls but also substantially enhance the clinical understanding by clarifying its specific correlations with disease activity, distinct clinical phenotypes, treatment status, and key laboratory parameters. The most prominent finding of our study is the substantial increase in plasma IL-33 levels among SLE patients when compared to normal controls, which is consistent with previous research conducted on Chinese populations13. This elevation can be mechanistically elucidated through several pathways. Firstly, as an alarmin, IL-33 is released upon cellular damage. Ultraviolet B (UVB) radiation, a well-recognized trigger for SLE flares, induces keratinocyte damage and stress in the skin, leading to the passive release of IL-3327. This is corroborated by our data, which show differential expression of IL-33 receptors in skin tissue correlates with sun-exposed sites; moreover, circulating IL-33 levels are significantly elevated in photosensitive patients compared to non-photosensitive individuals. The novelty of the present study resides in histologically demonstrating differential expression of the IL‑33 receptor (ST2) across distinct skin regions in patients with SLE, with significantly higher expression observed in chronically photoexposed skin tissues. We hypothesize that IL-33 acts as an “alarm factor” in UVR-induced skin damage. The cumulative dose of UV radiation varies across different sites of skin lesions, resulting in differential ST2 levels at the histological level. The circulating IL-33 does not uniformly affect all organs, and its pathological effects may be selectively amplified in tissues with high ST2 receptor expression. This could represent one of the key mechanisms linking systemic autoimmune responses to the involvement of specific target organs (e.g. skin, joints, kidneys). This implies that IL-33 may be the molecular bridge linking UVB exposure to the subsequent systemic immune activation. Emerging evidence indicates that IL-33 is intricately involved in the formation of Neutrophil Extracellular Traps (NETs). In SLE, neutrophils undergo NETosis, and IL-33 can be complexed with dsDNA and myeloperoxidase within these NETs28. These IL-33-decorated NETs then infiltrate inflamed tissues and potently stimulate plasmacytoid dendritic cells (pDCs) via TLR9, resulting in robust IFN-α production, thereby fueling a pro-inflammatory feedback loop that exacerbates the autoimmune response29. The plasma IL-33 level likely represents the sum of such releases from various affected tissues, offering a snapshot of systemic ‘alarm’ activity. Our results offer a theoretical foundation for investigating how cutaneous photosensitivity contributes to systemic inflammation. A novel and intriguing observation from our study is the significantly higher level of IL-33 in male SLE patients. Given that SLE exhibits a strong female predominance, male patients often present with atypical and more severe disease, frequently involving the kidneys and cardiovascular system, leading to a poorer prognosis30. Some researchers performed comparative analyses of inflammatory factors in treatment-naive patients with systemic lupus erythematosus (SLE). The results demonstrated that in pediatric SLE, males (boys) exhibited significantly higher expression of tumor necrosis factor (TNF)-related inflammatory genes, whereas females (girls) showed preferential expression of interferon-α (IFN-α)-related genes31. Those findings indicate that male SLE is predominantly driven by the TNF-mediated inflammatory pathway, while female SLE is dominated by the IFN pathway, suggesting that males with SLE present a different pro-inflammatory phenotype. Additionally, Balato et al. reported that TNF-α significantly increases IL-33 mRNA expression in normal human epidermis32.Therefore, the activation of tumor necrosis factor (TNF)-related inflammatory pathways may be a contributing factor to the elevated IL-33 expression levels in male patients with systemic lupus erythematosus. Our findings suggest that IL-33 could serve as a valuable early diagnostic biomarker for this underdiagnosed and high-risk subgroup, prompting closer monitoring and earlier intervention. But we also need to be aware that, although the gender difference was significant, the sample size of male participants was only 15, which may lead to statistical errors (Type I error) due to insufficient sample size. Furthermore, we observed that treatment-naïve patients had significantly higher IL-33 levels than those who had received immunosuppressive therapies (e.g., corticosteroids, hydroxychloroquine, or immunosuppressants). This finding indicates that IL-33 levels vary across different immune states in the disease. Treatment-naive SLE patients are in a state of uncontrolled immune activation of the disease. A genetically susceptible background combined with environmental stimuli (e.g., ultraviolet radiation, viruses) leads to extensive damage to epithelial/endothelial barriers throughout the body. IL-33 is mainly stored in the nuclei of epithelial cells, endothelial cells, and fibroblasts (it binds to histones H2A/H2B to maintain barrier integrity under physiological conditions) and is passively released in large quantities into the extracellular space and circulation upon cellular necrosis/apoptosis, acting as a core alarmin that initiates and amplifies inflammatory responses. The phenomenon of significantly higher IL-33 levels in patients receiving initial treatment compared to those receiving immunosuppressive therapies such as glucocorticoids and hydroxychloroquine may be attributed to the suppression of IL-33 release and activation by immunosuppressants through multiple pathways33,34. The positive correlation between plasma IL-33 levels and the SLEDAI-2K score solidifies its role as a reliable indicator of overall disease activity. Despite their weak correlation, this is consistent with studies linking higher IL-33 to active lupus nephritis, where it correlates with proteinuria35,36. However, we also acknowledge the existing controversy in the literature, as some studies report no significant correlation with SLEDAI, instead highlighting the importance of the soluble receptor sST237. These discrepancies may arise from heterogeneity in patient populations (e.g., age, ethnicity, disease duration), differences in disease phases (active vs. chronic), variations in assay sensitivities or recall bias. Our ROC analysis proposed a cutoff value of 312.42 pg/ml for predicting disease activity (SLEDAI ≥ 5) with a sensitivity of 75% but a specificity of 56.2%. It suggests its utility as a screening tool, but the moderate specificity indicates that IL-33 should ideally be part of a panel of biomarkers rather than used in isolation. Future studies incorporating sST2 and other cytokines like IL-3438,39 in a multi-analyte model could yield a more robust diagnostic and prognostic tool. Our multivariate regression analysis unveiled profound insights into the immunopathological networks involving IL-33. The strong negative correlation with complement C3 is particularly noteworthy. Complement consumption is a cardinal feature of active SLE. IL-33 may contribute to a vicious cycle: it can activate mast cells and other inflammatory cells, potentially leading to complement activation (e.g., via the alternative pathway) and consumption40–42. Concurrently, tissue damage caused by complement-consuming immune complexes results in further IL-33 release. Of note, we also observed no significant correlation between complement C4 and SLE. This may be because C4 is primarily involved in the classical complement pathway, and its levels are susceptible to factors such as genetic polymorphisms (e.g., C4 gene copy number variations) and baseline synthetic levels, which may weaken its correlation with inflammatory markers in our study cohort. The positive correlation with IgA suggests IL-33 may directly influence B-cell class switching to IgA, potentially contributing to mucosal immunity dysregulation and renal deposition of IgA-containing complexes43. The positive association with hs-CRP, coupled with its link to platelet counts, hints at a role for IL-33 in promoting inflammation and potentially microvascular thrombosis, possibly by inducing tissue factor expression in endothelial cells44. These complex interactions position IL-33 at the nexus of inflammation, autoantibody production, complement consumption, and vascular injury in SLE. The paradoxical nature of the IL-33/ST2 pathway—being both pro-inflammatory (e.g., activating Th2 cells, mast cells, promoting NETosis) and potentially protective (e.g., inducing Tregs and Bregs, mediating tissue repair via amphiregulin)29,45,46—adds layers of complexity to its role in SLE. This duality may explain the conflicting results from different studies and underscores the context-dependent nature of its function, which may vary with disease stage, genetic background, and microenvironments within specific organs. Our study offers compelling clinical evidence that plasma IL-33 serves as a multifaceted biomarker in SLE. Its levels are correlated with global disease activity, specific clinical manifestations such as photosensitivity and facial rash, treatment status, and distinct serological abnormalities (low C3, high IgA, high hs-CRP). It holds particular promise for the early identification of male patients and for indicating active disease, potentially linked to renal or vascular complications. High IL-33 in photosensitive patients and elevated ST2 in sun-exposed skin suggest that the IL-33/ST2 axis may be a critical mediator translating UV exposure into systemic immune activation in a subset of SLE patients. In conclusion, monitoring serum levels of IL‑33 may enable early prediction of disease progression risk in naïve SLE patients, while also reflecting the intensity of the immune response to stimuli such as light exposure. Future research should focus on exploring how to modulate IL-33 release in skin tissues and elucidate its interaction mechanisms with the TNF pathway. Although our cross-sectional design firmly establishes an association, it cannot prove causation. Therefore, future longitudinal and multicenter studies are crucial for monitoring the dynamic changes of IL-33 throughout the disease course and therapeutic responses. Meanwhile, our team has conducted follow-up basic research to further elaborate the hypothesis of the IL-33/ST2 signaling pathway in the pathogenesis of SLE: in genetically susceptible individuals exposed to various environmental factors such as viruses, ultraviolet radiation, and stress, the release of IL-33 is triggered by cellular damage products resulting from impaired epithelial barriers, and how IL-33 activates and participates in the immune mechanisms of systemic inflammatory responses. It will further validate its application value in guiding personalized treatment strategies and improving long-term prognosis in SLE patients.