Distinct Clinical Profiles of IgE and IgG Autoantibodies to Thyroid Peroxidase in Chronic Spontaneous Urticaria Authors: Jae-Hyuk Jang, Yi-Kui Xiang, Eun-Mi Yang, Ji-Hyun Kim, Boyoun Choi, Hae-Sim Park, Marcus Maurer, Young-Min Ye Published: 2024 ABSTRACT AbstractPurposeIn chronic spontaneous urticaria (CSU), autoimmune thyroid disease is the most common autoimmune comorbidity, and many CSU patients have immunoglobulin (Ig)E or IgG autoantibodies to thyroid peroxidase (TPO). It remains unclear how anti-TPO IgE and IgG autoantibodies are linked to each other and are associated with CSU features, activity, and therapeutic responses.MethodsCSU patients (n = 146, 92 females, mean age 42.9 years) and healthy normal controls (NCs, n = 30) were assessed for anti-TPO IgE and IgG by enzyme-linked immunosorbent assay, and the clinical and laboratory profiles, disease activity (UAS7), and response to 3 months of H1-antihistamine treatment were evaluated for anti-TPO IgE-positive and/or IgG-positive patients.ResultsAmong 146 CSU patients, 67 (46%, NCs: 6.6%) had elevated anti-TPO IgE and/or IgG, and 32 (22%), 35 (24%), and 5 (3%) had elevated anti-TPO IgE, anti-TPO IgG, and both, respectively. Of the patients with anti-TPO IgE and/or IgG, 44% (n = 27) had anti-TPO IgE but not IgG, 48% (n = 30) had anti-TPO IgG but not IgE, and only 8% (n = 5) had anti-TPO IgE and IgG. Compared to anti-TPO IgE-negative patients, anti-TPO IgE-positive ones had greater rates of atopy and antihistamine responses and lower disease activity (UAS7). Anti-TPO IgG-positive patients had greater rates of angioedema and elevated levels of anti-thyroglobulin IgG than anti-TPO IgG-negative patients.ConclusionsForty-six percentages of CSU patients have autoantibodies to TPO; most have either IgE or IgG autoantibodies but not both. Anti-TPO IgE and anti-TPO IgG come with distinct CSU profiles, including treatment responses.Keywords: Autoimmunity; urticaria; thyroid peroxidase; immunoglobulin E; mast cells, IgG INTRODUCTION Chronic spontaneous urticaria (CSU) is a common skin disease characterized by recurrent wheals and/or angioedema for more than 6 weeks.1 Skin mast cells, which undergo recurrent activation and degranulation, play a crucial role in the pathogenesis of CSU.2 Autoimmunity has emerged as a significant underlying cause of mast cell activation in CSU.3 Two distinct autoimmune endotypes have been identified in CSU. Type I autoimmunity, also called autoallergy, is characterized by the presence of immunoglobulin (Ig)E autoantibodies targeting self-antigens such as thyroid peroxidase (TPO), thyroglobulin (TG), interleukin (IL)-24, tissue factor, and dsDNA. In contrast, type IIb autoimmunity involves IgG autoantibodies that target and activate mast cells directly by binding to IgE or its high-affinity receptor, FcεRIα.2,3 Several autoimmune diseases, including thyroid diseases, rheumatoid arthritis, and type I diabetes mellitus, are reportedly associated with CSU, particularly in the context of type IIb autoimmune CSU.3 In addition, studies have reported a higher prevalence of hypothyroidism in patients with CSU than in individuals without CSU.4 Sánchez and coworkers5 demonstrated that IgE anti-TPO autoantibodies can induce mast cell degranulation and basophil activation. In contrast, there is no evidence, and it is unlikely that IgG autoantibodies against thyroid antigens, such as TPO and TG, can activate and degranulate mast cells in CSU patients. CSU is a complex and multifactorial condition, and the involvement of autoantibodies in its pathogenesis can vary among individuals. The co-occurrence or differential expression of the two different autoimmune endotypes (type I and IIb autoimmunity) against the same autoantigens in patients with CSU and their distinct contributions to clinical phenotypes of CSU remain unclear. In this study, our primary objective was to determine the positivity rate of autoantibodies, particularly focusing on the most common autoantigen, TPO, in patients with CSU. Additionally, we aimed to investigate the associations between these autoantibody profiles and clinical characteristics/treatment responses in CSU patients MATERIALS AND METHODS Study design The present study aims to evaluate the clinical characteristics of CSU patients according to the presence of autoantibodies, utilizing cut-off values defined as the means plus two standard deviations of healthy controls in a cross-sectional format. In total, we investigated 146 CSU patients and 40 healthy normal controls (NCs). Sera from NCs were supplied by the Ajou University Human Bio-Resource Bank, and those were confirmed as not having any previous history of inflammatory or allergic skin disease or urticaria. Due to insufficient data collection, the total number of NCs for comparison is 30. The secondary objectives were to evaluate the severity of urticaria and treatment response associated with the presence of autoantibodies. The urticaria activity score over 7 days (UAS7) was used to assess disease activity. Patients with a UAS7 value of ≥ 28 were categorized as having severe CSU disease activity.6,7 To evaluate the control status of urticarial symptoms, the Urticaria Control Test (UCT) was utilized.6,8 Patients participating in the study had serum samples drawn at initial registration. Serum samples from CSU patients were collected after a minimum 5-day pause in antihistamine treatment and stored at −70°C. After more than six months of antihistamine treatment, the treatment response and the use of other medications were collected. After initial visit with blood sampling, H1-antihistamine (H1AH) responders were defined as patients who showed a significant response to antihistamine treatment, up to fourfold, within 3 months. This included patients who achieved a UCT score of ≥ 12 with H1AH treatment alone and those who experienced a reduction in H1AH doses while still maintaining control of urticaria. On the other hand, patients with H1AH refractory CSU were identified as those requiring additional treatments such as oral corticosteroids, cyclosporine, and/or omalizumab due to a lack of response to H1AH (up to fourfold for more than a month) or a UCT score < 12 within the initial 3 months after serum collection. After this period, anti-TPO was measured from the stored serum samples that had been collected at the beginning of the study. The present study was a cross-sectional observational study, with all data analyzed retrospectively. The response to antihistamine treatment was not assessed concurrently with the initial blood sampling. Instead, it was retrospectively evaluated through a comprehensive review of medical records at the time of autoantibody measurement. This approach enabled simultaneous analysis of autoantibody status and treatment outcomes at a single time point (Supplementary Fig. S1). Atopy was defined as a positive skin prick test response to at least one common inhalant allergen (pollens of alder, birch, oak, grass mixture, mugwort, ragweed, cat or dog fur, Dermatophagoides pteronyssinus, D. farinae, Aspergillus niger, or Alternaria spp.). Thyroid autoantibodies (anti-TPO and anti-TG) were checked. Serum total IgE levels were measured using the ImmunoCAP system (Thermo Fisher Scientific, Waltham, MA, USA). The autologous serum skin test (ASST) was performed and interpreted in accordance with the EAACI/GA2LEN task force ASST consensus report.9 Detection of autoantibodies Serum levels of anti-TPO IgE were measured by a modified direct ELISA system.10 Briefly, a commercial 96-well microplate coated with recombinant TPO (Origene, Rockville, MD, USA) was blocked with 1% BSA in PBS-T for 3 hours at room temperature. After blocking, the plates were gently incubated with 1:1 diluted serum overnight at room temperature with shaking, followed by incubation with HRP-conjugated human anti-IgE antibody (Invitrogen, Carlsbad, CA, USA) for 1 hour and then with streptavidin-HRP (diluted 1:5,00 in PBS containing 1% BSA and 1 mM MgCl2) for 2 hours. Finally, the enzymatic dye reaction (TMB solution, BD bioscience, Franklin Lakes, NJ, USA) was stopped with 2 NH2SO4 solution, and the optical density was measured at 450 nm and 620 nm. Between each step, the samples were vigorously washed with 0.05% Tween 20 in PBS more than 5 times. Purified chimeric human anti-TPO IgE from the supernatant of SP1.4-IgE hybridoma cells was used as a standard curve and positive control.11 The levels of serum IgG to TPO and TG were measured using a commercial ELISA kit (Zeus Scientific Inc., Branchburg, NJ, USA). Anti-IL-24 IgE was measured by site-directed ELISA according to a previously described method.11 MaxiSorp black 384-well plates (Thermo Fisher Scientific) along with Tris-buffered saline containing 0.05% Tween 20 were used for ELISA plates and as the washing solution, respectively. The IgE antibody (Clone MHE-18; BioLegend, San Diego, CA, USA) was immobilized on two plates, which were then stored overnight at 4°C. The following day, the cells were blocked with fetal bovine serum for two hours, followed by an overnight incubation at 4°C with various 1:5 serum dilutions. On the final day, the plates were incubated for one hour in the presence or absence of recombinant human IL-24 (Miltenyi Biotec, Bergisch Gladbach, Germany) to detect free or complexed anti-IL-24 IgE or complexed anti-IL-24 IgE, respectively. Anti-IL-24 IgE was detected using a biotinylated anti-IL-24 antibody (R&D Systems, Minneapolis, MN, USA) and subsequent incubation with streptavidin-HRP (Thermo Fisher Scientific) and ECL (GE Healthcare, Chicago, IL, USA). The light intensity emitted from each plate was measured using a Perkin Elmer VICTOR X5 microplate reader. Data from CSU patients and NCs were normalized by a standard curve set up by recombinant IgE-anti-TPO and expressed in arbitrary units (AUs).11 IgG autoantibodies against FcεRI or IgE were detected via direct ELISA as described previously.11 MaxiSorp black 384-well plates (Thermo Fisher Scientific) were used for the ELISA. The washing solution comprised PBS mixed with 0.5 M NaCl, 0.01% Triton, and 0.05% Tween 20, while 10% protein-free blocking buffer (LI-COR Biosciences, Lincoln, NE, USA) dissolved in the washing solution served as the assay diluent. Recombinant human FcεRI (ACROBiosystems, Newark, DE, USA) or human myeloma IgE (Merck Millipore, Burlington, MA, USA) was coated onto 2 plates overnight at 4°C. The next day, the cells were blocked for one hour at room temperature with the assay diluent. After blocking, 1:100 serum dilutions were incubated for one hour at room temperature. Finally, the plates were incubated for 30 minutes at room temperature with HRP-conjugated anti-human IgG antibody (Jackson ImmunoResearch, West Grove, PA, USA), immediately followed by detection using ECL (GE Healthcare). The light intensity emitted from each plate was measured using a Perkin Elmer VICTOR X5 microplate reader. The data from CSU patients and NCs were normalized to the standard curves generated with recombinant anti-FcεRI IgG or anti-IgE IgG (Miltenyi Biotec) and expressed in AU. The positive cutoff value for these autoantibodies was determined as the mean of each value plus twice the standard deviation in 30 healthy controls. Statistical analysis Descriptive statistics were used to summarize demographic and clinical characteristics. The positivity rate of different autoimmune endotypes was determined, and associations with clinical features were assessed using the χ2 test or Fisher’s exact test. The Kruskal–Wallis test or t-test was applied to compare continuous variables. The Dunn test was used for post hoc analysis after the Kruskal–Wallis test. Continuous variables were standardized before regression. Logistic regression analysis was used to identify predictors for H1AH nonresponders. Factors with P values less than 0.05 in the univariate analysis were included in the multivariate logistic regression analysis and were removed in a stepwise manner. The significance level was set to P < 0.05. Analyses were carried out using SPSS for Windows (ver. 25.0; IBM Corp., Armonk, NY, USA), Prism (ver. 9.0; GraphPad Software Inc., San Diego, CA, USA) software, and R 4.2.2 (R Core Team, 2022). P < 0.05 was considered to indicate statistical significance. Ethics statement Written informed consent was obtained from all subjects. Biospecimens and data of healthy subjects were provided by the Biobank of Ajou University Hospital, which is a member of the Korea Biobank Network. The study was approved by Ajou University Hospital Institutional Review Board (AJOUIRB-MDB-2017-221, AJIRB-BMIR-KSP-21-075). Written informed consent was obtained from each patient. The study was conducted in accordance with the tenets of the Declaration of Helsinki. RESULTS Clinical characteristics of the study subjects A total of 146 CSU patients and 30 NCs were enrolled in this study (Table 1). The mean age of the CSU group was greater than that of the NC group (42.9 ± 13.2 years vs. 34.7 ± 10.2 years, P = 0.002), but there was no difference in gender distribution. Table 1Clinical characteristics and immunologic findings of the study subjectsVariablesCSU (n = 146)NC (n = 30)P valueAge (yr)42.9 ± 13.234.7 ± 10.20.002Female sex92 (63.0)16 (53.3)0.432Urticaria duration (mon)25.1 ± 37.7UAS719.3 ± 10.9UAS7 ≥ 2837/131 (28.2)UCT6.03 ± 2.7No. of H1Ahs (loratadine 10 mg Eq)3.1 ± 1.2H1AH refractory CSU80 (54.8)ANA26 (17.8)4 (13.3)0.790* ASST positivity83 (58.5)Atopy74 (50.7)9 (30.0)0.062Total IgE (kU/L)180.5 ± 179.653.3 ± 45.2< 0.001IgE to TPO (pg/mL)69.4 ± 150.922.6 ± 33.50.001IgE to TPO32 (21.9)1 (3.3)0.019* IgE to IL2412/110 (10.9)5/20 (25.0)0.140* IgG to TPO (IU/mL)31.6 ± 50.310.6 ± 5.0< 0.001IgG to TPO35 (24.0)1 (3.3)0.011* IgG to TG (IU/mL)36.8 ± 71.617.5 ± 14.60.015IgG to TG35 (24.0)3 (10.0)0.141* IgG to FcεRIα8/108 (7.4)1/24 (4.2)1.000* IgG to IgE4/108 (3.7)0 (0.0)1.000* Values are presented as number (%) unless otherwise indicated.CSU, chronic spontaneous urticaria; NC, normal control; UAS7, the sum of urticaria activity score over 7 days, including the daily number of wheals and the itch severity score; UCT, urticaria control test; H1AH, H1 antihistamines; ANA, anti-nuclear antibody; ASST, autologous serum skin test; Ig, immunoglobulin; FcεRIα, high affinity IgE receptor; TPO, thyroid peroxidase; TG, thyroglobulin.*Fisher’s exact test. Clinical characteristics and immunologic findings of the study subjects Values are presented as number (%) unless otherwise indicated. CSU, chronic spontaneous urticaria; NC, normal control; UAS7, the sum of urticaria activity score over 7 days, including the daily number of wheals and the itch severity score; UCT, urticaria control test; H1AH, H1 antihistamines; ANA, anti-nuclear antibody; ASST, autologous serum skin test; Ig, immunoglobulin; FcεRIα, high affinity IgE receptor; TPO, thyroid peroxidase; TG, thyroglobulin. *Fisher’s exact test. The mean total IgE levels were significantly higher in CSU patients than in NCs, but the percentage of atopic subjects did not significantly differ between the CU and NC groups (180.5 ± 179.6 kU/L vs. 53.3 ± 45.2 kU/L, P < 0.001; 50.7% vs. 30.0%, P = 0.062). In patients with CSU, the mean duration of urticaria was 25 months, and the mean UAS7 and UCT scores were 19.3 and 6.0, respectively. Among the CSU patients, 54.8% were classified as antihistamine refractory; 28.2% had a UAS7 score ≥ 28 at enrollment. Additionally, 58.5% of these patients tested positive for ASST. Positivity rate of anti-TPO IgE and anti-TPO IgG in CSU patients The levels of autoantibodies, including anti-TPO IgE (69.4 ± 150.9 pg/mL vs. 22.6 ± 33.5 pg/mL, P = 0.001), anti-TPO IgG (31.6 ± 50.3 IU/mL vs. 10.6 ± 5.0 IU/mL, P < 0.001), and anti-TG IgG (36.8 ± 71.6 IU/mL vs. 17.5 ± 14.6 IU/mL, P = 0.015), were significantly higher in the CSU patients than in the NCs (Table 1 and Fig. 1). Since there was an age difference between the CSU and NC groups, age matching was performed to exclude the influence of age. The analysis showed no differences in anti-TPO IgE and IgG levels, however, only 10 people remained in each group after matching, which lacked statistical power. Additionally, correlation analysis showed no significant relationships between age and autoantibodies. The positive cutoff values for anti-TPO IgE and anti-TPO IgG were 90 pg/mL and 25 IU/mL, respectively (Fig. 1). The levels of anti-TPO IgE (A) and IgG (B) in the serum of CSU patients and NCs. TPO, thyroid peroxidase; CSU, chronic spontaneous urticaria; NC, normal control. The positivity rate of anti-TPO IgE/IgG was greater in CSU patients than in NCs (21.9% vs. 3.3%, P = 0.019 for IgE and 24.0% vs. 3.3%, P = 0.011 for IgG). Among the 146 CSU patients, 62 (42%) had IgE/ IgG to TPO, and 32 (22%), 35 (24%), and 5 (3%) had anti-TPO IgE, anti-TPO IgG, and both, respectively (Fig. 2). Among the patients with anti-TPO autoantibodies, only 5 (8.1%) had both anti-TPO IgE and IgG, and 54.3% of those with anti-TPO IgG also had anti-TG IgG. The positivity rates of IgG autoantibodies to FcεRIα (7.4%) and IgE (3.7%) were very low in CSU patients, with no statistically significant differences compared to those in NCs. Euler diagram depicting serum autoantibody profiles in CSU patients. TG, thyroglobulin; CSU, chronic spontaneous urticaria; TPO, thyroid peroxidase. Clinical profiles according to the presence of anti-TPO autoantibodies in CSU patients Table 2 shows the comparison of clinical profiles according to the presence of anti-TPO autoantibodies. CSU patients positive for anti-TPO IgE showed significantly lower disease activity, as assessed by the UAS7, than those who were negative (15.4 ± 10.3 vs. 20.4 ± 10.8, P = 0.025, Table 2). The rate of antihistamine-refractory CSU was also lower in CSU patients positive for anti-TPO IgE than in those negative for anti-TPO IgE (37.5% vs. 59.6%, P = 0.043). In contrast, according to the anti-TPO IgG results, there was no difference in either the mean UAS7 score or the percentage of patients with antihistamine-refractory CSU (18.6 ± 10.9 vs. 19.5 ± 10.9, P = 0.686; 23, 65.7% vs. 57, 51.4%, P = 0.196; respectively). Table 2Comparisons of clinical and immunologic findings according to the presence of anti-TPO autoantibodiesVariablesAnti-TPO IgEP valueAnti-TPO IgGP valuePositive (n = 32)Negative (n = 114)Positive (n = 35)Negative (n = 111)Clinical parametersAge (yr)43.8 ± 12.442.7 ± 13.40.68342.7 ± 12.743.0 ± 13.40.885Urticaria duration (mon)28.5 ± 48.124.2 ± 34.70.65823.3 ± 33.125.6 ± 39.20.766UAS7 (0–42)15.4 ± 10.320.4 ± 10.80.02518.6 ± 10.919.5 ± 10.90.686UCT (0–16)6.1 ± 3.06.0 ± 2.70.8726.1 ± 2.96.0 ± 2.70.942Atopy25 (78.1)49 (43.0)< 0.00116 (45.7)58/111 (52.3)0.631ASST positivity16 (50.0)67 (60.9)0.36923/33 (69.7)60/109 (55.0)0.195Angioedema11 (34.4)48 (42.1)0.55921 (60.0)38 (34.2)0.012H1AH refractoriness12 (37.5)68 (59.6)0.04323 (65.7)57 (51.4)0.196Blood eosinophil (%)2.5 ± 1.51.9 ± 1.60.0341.9 ± 1.62.0 ± 1.60.796Eosinopenia (< 500/µL)2/31 (6.5)23/108 (21.3)0.0667/33 (21.2)18/106 (17.0)0.769Blood basophil (%)0.61 ± 0.30.54 ± 0.30.3180.60 ± 0.40.54 ± 0.30.485Basopenia (< 100/µL)1/31 (3.2)5/108 (4.6)1.0000/336/106 (5.7)0.336* Total IgE (kU/L)179.7 ± 167.9180.7 ± 183.40.978171.5 ± 170.0183.4 ± 183.10.733Total IgE < 40 kU/L1 (3.1)17 (14.9)0.1245 (14.3)13 (11.7)0.769* IgE to IL-24 (AU)24.6 ± 24.612.8 ± 17.90.02815.0 ± 20.215.9 ± 20.50.849Autoallergic measurementIgE to TPO5 (14.3)27 (24.3)0.249* IgE to IL-245/27 (18.5)7/83 (8.4)0.2693/24 (12.5)9/86 (10.5)0.722* Any of IgE autoantibody positivity32/32 (100)7/83 (8.4)< 0.00119/30 (63.3)65/101 (64.4)1.000Autoimmune measurementANA2 (6.2)24 (21.1)0.067* 8 (22.9)18 (16.2)0.521IgG to FcεRIα3/27 (11.1)5/81 (6.2)0.410* 1/21 (4.8)7/87 (8.0)0.959IgG to IgE1/27 (3.7)3/81 (3.7)1.000* 0/214/87 (4.6)0.721IgG to TG7 (21.9)28 (24.6)0.93619 (54.3)16 (14.4)< 0.001IgG to TPO5 (15.6)30 (26.3)0.249* Any of IgG autoantibody positivity11/30 (36.7)62/99 (62.7)0.02135/35 (100)38/94 (40.4)< 0.001Values are presented as number (%) unless otherwise indicated.TPO, thyroid peroxidase; Ig, immunoglobulin; UAS7, the sum of urticaria activity score over 7 days including the daily number of wheals and the itch severity score; UCT, urticaria control test; ASST, autologous serum skin test; H1AH, H1 antihistamines; IL, interleukin; ANA, anti-nuclear antibody; FcεRIα, high affinity IgE receptor; TG, thyroglobulin.*Fisher’s exact test. Comparisons of clinical and immunologic findings according to the presence of anti-TPO autoantibodies Values are presented as number (%) unless otherwise indicated. TPO, thyroid peroxidase; Ig, immunoglobulin; UAS7, the sum of urticaria activity score over 7 days including the daily number of wheals and the itch severity score; UCT, urticaria control test; ASST, autologous serum skin test; H1AH, H1 antihistamines; IL, interleukin; ANA, anti-nuclear antibody; FcεRIα, high affinity IgE receptor; TG, thyroglobulin. *Fisher’s exact test. The percentage of CSU patients with angioedema was higher in patients with anti-TPO IgG than in those without (60.0% vs. 34.2%, P = 0.012). The percentage of eosinophils in the blood was higher in CSU patients with anti-TPO IgE than in those without (2.5 ± 1.5% vs. 1.9 ± 1.6%, P=0.034). The prevalence of eosinopenia, defined as blood eosinophils under 500/µL, was numerically higher in CSU patients who were negative for anti-TPO IgE than in those who were positive (21.3% vs. 6.5%, P = 0.07), but this difference was not statistically significant. Basophil counts and the prevalence of basopenia did not differ between groups depending on the presence of autoantibodies to TPO. There were no notable differences in serum total IgE levels between CSU patients with and without anti-TPO IgE or between those with and without anti-TPO IgG. There were more atopic subjects (78.1% vs. 43%, P < 0.001) and higher anti-IL-24 IgE levels (24.6 vs. 12.8, P = 0.028) among CSU patients with anti-TPO IgE. However, anti-TPO IgG positivity was not related to atopy. The percentage of CSU patients positive for anti-TG IgG was greater among those who were positive for anti-TPO IgG than among those without anti-TPO IgG (54.3% vs. 14.4%, P < 0.001). The positive rates for anti-IL24 IgE, anti-FcεRIα IgG, and anti-IgE IgG were comparable in patients with and without anti-TPO IgE or anti-TPO IgG. In patients who tested positive for ANA, IgG to FcεRIα, IgG to IgE, IgG to TG, or IgG to TPO, the positivity rate was significantly higher in CSU patients who were negative for anti-TPO IgE (62.6%) than in those positive for anti-TPO IgE (36.7%, P = 0.021, Table 2). Exploring the clinical characteristics of four autoimmune endotypes based on anti-TPO IgE and IgG results in patients with CSU To further compare the clinical characteristics of CSU patients according to the presence of anti-TPO IgE and/or IgG, CSU patients were classified into 4 groups (Table 3). Patients with positive anti-TPO IgE and negative anti-TPO IgG (EPGN, n = 27), patients with negative anti-TPO IgE and positive anti-TPO IgG (ENGP, n = 30), patients with positive anti-TPO IgE and positive anti-TPO IgG (EPGP, n = 5), and patients with negative anti-TPO IgE and negative anti-TPO IgG (ENGN, n = 84) were included. The demographic factors, including sex, age, height, and weight, were comparable across the groups. Urticaria activity, ASST positivity, and disease duration also showed no differences among the four groups. However, notable differences were found in the comparison of the presence of angioedema, atopic state, and treatment response to antihistamines. The ENGP group exhibited a greater incidence of angioedema than did the ENGN group (P = 0.009). Moreover, atopy was more common in the EPGN group than in both the ENGN (P = 0.003) and ENGP groups (P = 0.02). Patients who were responsive to antihistamine tended to be more prevalent in the group with positive anti-TPO IgE than in the other groups (EPGN, 59.3%; EPGP, 80.0%; ENGP, 26.7%; ENGN, 45.2%). The white blood cell count did not significantly differ (P = 0.500), while the percentage of eosinophils significantly differed between the ENGN and EPGP groups (P < 0.05) or between the EPGP and ENGP groups (P = 0.016). Basophil percentages in blood also showed significant differences between the EPGP and ENGP groups (P = 0.04). Total IgE did not differ across groups. Table 3Comparisons of clinical and immunologic findings among the 4 different groupsVariablesEPGN (n = 27)ENGP (n = 30)EPGP (n = 5)ENGN (n = 84)P valuePost hoc (adjusted P)Female14 (51.9)20 (66.7)4 (80.0)54 (64.3)0.514Age (yr)42.0 [35.0–53.0]38.0 [31.0–52.0]45.0 [35.0–45.0]40.0 [33.0–51.5]0.917UAS713.0 [10.0–23.0]18.0 [11.5–28.0]8.0 [6.0–25.0]21.0 [12.0–29.0]0.166Duration (mon)7.0 [3.0–27.0]7.0 [3.5–31.0]6.0 [6.0–36.0]8.0 [3.0–30.0]0.999Angioedema10 (37.0)20 (66.7)1 (20.0)28 (33.3)0.010ENGN-ENGP, 0.009Positive ASST13 (48.1)20 (71.4)3 (60.0)47 (57.3)0.366Atopy22 (81.5)13 (43.3)3 (60.0)36 (42.9)0.004ENGN-EPGN, 0.003ENGP-EPGN, 0.020H1AH response16 (59.3)8 (26.7)4 (80.0)38 (45.2)0.033WBC (*103/μL)6.7 [5.7–7.8]6.8 [5.8–8.4]5.9 [4.7–6.7]6.5 [5.1–8.2]0.500Eosinophil (%)2.0 [1.4–3.0]1.1 [0.6–2.2]4.0 [3.1–4.7]1.6 [0.8–2.2]0.009ENGN-EPGP, < 0.05ENGP-EPGP, 0.016Basophil (%)0.5 [0.4–0.7]0.5 [0.3–0.7]1.0 [0.6–1.4]0.5 [0.3–0.7]0.065EPGP-ENGP, 0.040Total IgE (kU/L)132.0 [74.0–191.0]102.5 [45.0–238.0]124.0 [106.0–155.0]110.0 [68.0–247.0]0.862Anti-TPO IgE (pg/mL)159.7 [135.5–319.6]0.0 [0.0–23.1]220.3 [128.5–287.1]0.0 [0.0–32.3]< 0.001ENGN-EPGN, < 0.001ENGN-EPGP, < 0.001EPGN-ENGP, < 0.001EPGP-ENGP, < 0.001Anti-IL-24 IgE (AU)11.0 [3.5–41.4]6.2 [2.3–11.0]29.0 [16.0–48.1]4.8 [2.2–14.0]0.115Anti-FcεR1α IgG (AU)7.0 [0.0–14.9]1.5 [0.0–8.9]4.2 [2.1–23.3]2.4 [0.0–16.1]0.760Anti-IgE IgG (AU)0.0 [0.0–20.9]0.0 [0.0–0.0]77.9 [38.9–130.0]0.0 [0.0–22.4]0.277Anti-TPO IgG (IU/mL)9.7 [8.4–16.0]114.1 [33.2–126.0]125.9 [34.3–126.0]9.5 [7.2–12.8]< 0.001ENGN-EPGP, < 0.001ENGN-ENGP, < 0.001EPGN-EPGP, < 0.010EPGN-ENGP, < 0.001Anti-TG IgG (IU/mL)5.6 [0.6–9.5]106.0 [3.4–189.0]130.5 [ 3.9–161.2]2.8 [0.0–8.4]0.001ENGN-ENGP, < 0.001Data with normal distribution were expressed as means ± standard deviations, while nonparametric data were presented as median and interquartile ranges. Values are presented as number (%) unless otherwise indicated. P value was evaluated using K-W test and Fisher’s exact test, and post hoc analysis was done using Dunn test.TPO, thyroid peroxidase; Ig, immunoglobulin; EPGN, anti-TPO IgE positive and anti-TPO IgG negative; ENGP, anti-TPO IgE negative and anti-TPO IgG positive; EPGP, anti-TPO IgE positive and anti-TPO IgG positive; ENGN, anti-TPO IgE negative and anti-TPO IgG negative; UAS, urticaria activity score; ASST, autologous serum skin test; H1AH Res, H1 antihistamine responder; WBC, white blood cell count; IL, interleukin; FcεRIα, high affinity IgE receptor; TG, thyroglobulin. Comparisons of clinical and immunologic findings among the 4 different groups Data with normal distribution were expressed as means ± standard deviations, while nonparametric data were presented as median and interquartile ranges. Values are presented as number (%) unless otherwise indicated. P value was evaluated using K-W test and Fisher’s exact test, and post hoc analysis was done using Dunn test. TPO, thyroid peroxidase; Ig, immunoglobulin; EPGN, anti-TPO IgE positive and anti-TPO IgG negative; ENGP, anti-TPO IgE negative and anti-TPO IgG positive; EPGP, anti-TPO IgE positive and anti-TPO IgG positive; ENGN, anti-TPO IgE negative and anti-TPO IgG negative; UAS, urticaria activity score; ASST, autologous serum skin test; H1AH Res, H1 antihistamine responder; WBC, white blood cell count; IL, interleukin; FcεRIα, high affinity IgE receptor; TG, thyroglobulin. A total of 29 patients were finally treated with omalizumab in this study populations (18 ENGN, 5 EPGN, 6 ENGP, no EPGP). When we analyzed the response to omalizumab retrospectively, 8 of 18 (44.4%) ENGN, 1 (20.0%) of 5 EPGN, and all 6 ENGP showed early and complete response to omalizumab. The sample size was too small to determine the effects of CSU endotypes on the response to omalizumab in this study. The presence of anti-TPO IgE and IgG is linked to H1AH treatment responses The H1AH nonresponders had high UAS7 values at baseline and a high rate of comorbid angioedema (Table 4). H1AH nonresponders exhibited a greater prevalence of high urticaria activity (UAS7 ≥ 28, 43.5% vs. 11.3%, P < 0.001) and had lower UCT scores (5.0 ± 2.5 vs. 7.1 ± 2.6, P < 0.001). Gender distribution, age, disease duration, atopic status, and rates of ASST positivity were not significantly different depending on the antihistamine treatment response. However, the incidence of angioedema was higher in the H1AH refractory group than in the H1AH responder group (41, 51.2% vs. 18, 27.3%, P = 0.006). The basophil percentage was higher in H1AH responders than in H1AH nonresponders (0.63 ± 0.4 vs. 0.49 ± 0.3, P = 0.013). The percentage of anti-TPO IgE-positive patients was higher in the H1AH responder group than in the H1AH nonresponder group (30.3% vs. 15.0%, P = 0.043). Logistic regression analysis revealed that angioedema, disease activity and the presence of anti-TPO IgE were associated with antihistamine refractoriness (odds ratio, 2.8, P = 0.004; 6.04, P < 0.001; 0.41, P = 0.029; Table 5). The ratio of anti-TPO IgG/total IgE, known for omalizumab treatment response, showed no association with antihistamine treatment response in this study. Table 4Comparisons of clinical and immunologic findings between H1AH responders and nonrespondersVariablesH1AH responder (n = 66)H1AH nonresponder (n = 80)P valueAge (yr)42.7 ± 14.343.1 ± 12.20.824Female40 (60.6)52 (65.0)0.584UAS7 ≥ 287/62 (11.3)30/69 (43.5)< 0.001UAS715.2 ± 9.622.9 ± 10.7< 0.001UCT7.1 ± 2.65.0 ± 2.5< 0.001Numbers of daily H1RAs2.3 ± 1.23.7 ± 0.8< 0.001Urticaria duration (mon)27.6 ± 43.122.6 ± 31.90.441Angioedema18 (27.3)41 (51.2)0.006Atopy33 (50.0)41 (51.2)1.000ASST positivity44/66 (66.7)39/76 (51.3)0.093ANA positivity10 (15.2)16 (20.0)0.586Blood neutrophil (%)58.0 ± 9.160.8 ± 10.10.083Blood eosinophil (%)2.03 ± 1.42.0 ± 5.70.834Blood basophil (%)0.6 ± 0.40.5 ± 0.30.013Total IgE (kU/L)166.2 ± 172.4192.4 ± 185.50.382IgE to TPO20 (30.3)12 (15.0)0.043IgE to IL246/52 (11.5)6/58 (10.3)1.000IgG to FcεRIα4/50 (8)4/58 (6.9)1.000IgG to TG17 (25.8)18 (22.5)0.792IgG to TPO12 (18.2)23 (28.8)0.196IgG to TG or TPO32 (48.5)28 (35)0.139Values are presented as number (%) unless otherwise indicated.H1AH, H1 antihistamines; UAS, urticaria activity score; UCT, urticaria control test; H1RA, Histamine H1 receptor antagonist; ASST, autologous serum skin test; ANA, anti-nuclear antibody; Ig, immunoglobulin; TPO, thyroid peroxidase; IL, interleukin; FcεRIα, high affinity IgE receptor; TG, thyroglobulin. Comparisons of clinical and immunologic findings between H1AH responders and nonresponders Values are presented as number (%) unless otherwise indicated. H1AH, H1 antihistamines; UAS, urticaria activity score; UCT, urticaria control test; H1RA, Histamine H1 receptor antagonist; ASST, autologous serum skin test; ANA, anti-nuclear antibody; Ig, immunoglobulin; TPO, thyroid peroxidase; IL, interleukin; FcεRIα, high affinity IgE receptor; TG, thyroglobulin. Table 5Logistic regression analysis for identifying predictors for antihistamine-refractory CSU patientsFactorsUnivariate analysisMultivariate analysisOdds ratio (95% CI)P valueOR (95% CI)P valueAge1.00 (0.98–1.03)0.820Female1.21 (0.61–2.37)0.584UAS7 ≥ 286.04 (2.53–16.25)< 0.0014.53 (1.71–13.34)0.004Urticaria duration1.00 (0.99–1.01)0.437Angioedema2.8 (1.41–5.72)0.0043.64 (1.62–8.54)0.002ASST positivity0.53 (0.26–1.04)0.065Atopic status1.05 (0.55–2.02)0.880Total IgE (kU/L)1.00 (0.99–1.00)0.381WBC (*103/μL)1.21 (1.03–1.45)0.026Blood neutrophil (%)1.03 (1.00–1.07)0.085Blood eosinophil (%)0.98 (0.79–1.21)0.833Blood basophil (%)0.24 (0.07–0.72)0.016Lymphocyte (%)0.96 (0.92–1.00)0.078Anti-TPO IgE positivity0.41 (0.18–0.9)0.0290.30 (0.10–0.80)0.020Anti-TPO IgG positivity1.82 (0.84–4.11)0.139Factors with P values less than 0.05 in the univariate analysis were included in the multivariate logistic regression analysis and were removed in a stepwise manner. Continuous variables were standardized before regression.UAS, urticaria activity score; ASST, autologous serum skin test; Ig, immunoglobulin; WBC, white blood cell count; TPO, thyroid peroxidase. Logistic regression analysis for identifying predictors for antihistamine-refractory CSU patients Factors with P values less than 0.05 in the univariate analysis were included in the multivariate logistic regression analysis and were removed in a stepwise manner. Continuous variables were standardized before regression. UAS, urticaria activity score; ASST, autologous serum skin test; Ig, immunoglobulin; WBC, white blood cell count; TPO, thyroid peroxidase. DISCUSSION In this study, we discovered distinct clinical features in CSU patients with autoantibodies to TPO, specifically the IgE and IgG isotypes. We found associations of anti-TPO IgE positivity with decreased urticaria activity scores and improved outcomes following antihistamine therapy. In contrast, anti-TPO IgG positivity was more frequently associated with angioedema. Remarkably, only 5 of 35 patients with anti-TPO IgG also had IgE autoantibodies to this antigen, whereas 19 of 35 patients with anti-TPO IgG were positive for another thyroid autoantigen, TG. Higher levels of anti-IL24 IgE autoantibodies were observed in patients with anti-TPO IgE. These findings suggest that the immune environment influencing the type of autoantibody produced is more crucial for CSU patients than specific autoantigen exposure. Additionally, the atopic status of CSU patients was significantly associated with anti-TPO IgE positivity. IgE provides the link between antigen recognition and the effector functions of mast cells and basophils in skin and mucosal tissues.12,13 Autoimmune diseases, especially thyroid dysfunctions such as hypothyroidism and Hashimoto’s thyroiditis, are more common in CSU patients than in the general population.4 Additionally, treating hypothyroidism often improves urticaria symptoms, highlighting a probable connection.4,14,15 In this study, CSU patients demonstrated greater levels of autoantibodies, including anti-TPO IgE, anti-TPO IgG, and anti-TG IgG, than NCs. TPO has been recognized as the common target of both autoallergy and autoimmunity. In the general population, the positivity rate of anti-TPO IgG is 3%–6%, while in CU patients, it increases to 17.7%–29%.16 This study revealed that 24% of CSU patients had anti-TPO IgG, consistent with prior results.17 Mozena et al.18 demonstrated that the incubation of patient sera with TPO did not affect the detection of anti-FcεRIα antibodies or leukotriene secretion by mast cells in CSU patients who had both anti-FcεRIα and anti-TPO antibodies. Their results suggested that the elevated anti-TPO antibodies in CSU patients are unlikely due to epitopic cross-reactivity between two different autoantigens. Instead, a genetic predisposition to autoimmune diseases may be a potential mechanism. Furthermore, we did not identify a significant association between anti-TPO IgG positivity and anti-FcεRIα antibodies or between anti-TPO IgG positivity and ASST results in CSU patients. It is worth noting that the positivity rate of anti-FcεRIα antibodies was relatively low in our study population compared to other ethnicities. The pathogenic significance of anti-TPO IgE has been well established, as it can activate basophils and induce hives in healthy individuals.5,10,19 The positivity rate of anti-TPO IgE was 21.9% in this study, which was lower than that reported in previous studies (40.5%, 54%).11,20 However, there are studies where anti-TPO IgE levels did not significantly differ from those in control groups.21 While previous studies reported the presence of 2 thyroid autoantibodies in CSU patients, the relationship between anti-TPO IgE and anti-TPO IgG remains unclear. In a study of 23 patients with CU, anti-TPO IgG was detected in the serum, but no patient tested positive for anti-TPO IgE.21 Another study demonstrated that the presence of anti-TPO IgG may not significantly influence the levels of anti-TPO IgE.17 Conversely, a high prevalence of anti-TPO IgE was observed in patients with anti-TPO IgG, with significant correlations between the two antibodies.22 Moreover, anti-TPO IgE-positive CSU patients exhibited significantly higher anti-TPO IgG levels.20 However, our study suggested a more independent relationship between anti-TPO IgE and IgG than between anti-TPO IgG and anti-TG IgG. Additionally, the co-occurrence of anti-TPO IgG and anti-TG IgG, as found in previous research,4 hints at an autoimmune response triggered by the disruption of the normal structure of the thyroid gland.22 Animal studies suggest that anti-TG antibodies develop prior to anti-TPO antibodies, indicating an initial breakdown of immune tolerance toward TG followed by a later shift toward TPO.23 Consequently, it appears that anti-TG antibodies have a more pronounced association with thyroid function and autoimmune thyroid diseases than does anti-TPO IgG, which is more relevant for CSU. The levels of anti-TPO IgE are altered during urticaria exacerbations.19 In addition, cross-reactivity with eosinophil peroxidase was found in autoantibodies to TPO in CSU and AD patients, implying that conditions related to eosinophils may influence the measurement of thyroid autoantibodies24 Furthermore, molecular mimicry between TPO and other endogenous peroxidase or viral pathogens may lead to cross-reactivity with anti-TPO IgG. These factors, along with the diverse characteristics of the studies and patient demographics, can influence the variability of thyroid autoantibody results. Therefore, it is essential to consider these aspects when measuring IgE anti-TPO in future research. Methodological differences in measurement across studies have also contributed to instances where IgE anti-TPO was unmeasurable or indistinct. Despite the lower positivity rate of IL-24 IgE autoantibodies observed in this study, their presence is indicative of an autoallergic phenotype in CSU.14,25 The variance in the positivity rate of anti-FcεRIα IgG and anti-IgE IgG between CSU patients and healthy individuals in our research deviates from that reported in prior studies,11,26 necessitating further investigation to understand the role of these autoantibodies in CSU more clearly.11 In the present study, we not only identified distinct profiles among CSU patients based on the presence of anti-TPO IgE and anti-TPO IgG autoantibodies but also observed variability in urticaria severity and treatment response to antihistamines depending on the positivity of each autoantibody. These observations suggest potential differences in the underlying pathophysiology and immune mechanisms driving CSU in patients with different autoantibody profiles. Our findings revealed a distinct pattern: CSU patients lacking anti-TPO IgE exhibited greater urticaria severity and increased antihistamine refractoriness than those with anti-TPO IgE. This is underpinned by observations that high disease activity predicts refractoriness to antihistamines.27 However, anti-TPO IgE and anti-TPO IgG were suggested to be independent predictors of antihistamine refractoriness associated with lower levels of total IgE.22 Interestingly, the anti-TPO IgG/total IgE ratio, which has been previously associated with omalizumab response,28,29 showed no association with antihistamine treatment response in this study. Nonetheless, all 6 ENGP showed a favorable response to omalizumab, implying that even in patients with anti-TPO IgG whose urticaria is not controlled with antihistamines alone, omalizumab add-on can be effective. Our analysis indicated that the presence of anti-TPO IgE, rather than anti-TPO IgG, helps predict the response to antihistamine treatments alongside known factors, disease severity, and angioedema. This study has several limitations. First, the measurement of autoantibodies is not yet standardized globally, leading to potential variability in methodologies across different studies. Secondly, the variation in the prevalence rates of autoantibodies appears to reflect geographical and ethnic differences. The presence of thyroid disease, treatment status, and thyroid function can especially influence the prevalence of thyroid autoantibodies.16,19 The present study design did not fully control for these factors, highlighting the need for future research to incorporate such considerations. Furthermore, the definition of endotypes based on the presence of autoantibodies was made to compare those groups with or without autoantibodies and currently, it has no unified consensus. Despite these limitations, our single-center study includes the measurement of six different autoantibodies within the same patient cohort for the first time. Specifically, it includes assessments of 2 autoreactive IgEs (anti-TPO and anti-IL24 IgE) and 4 autoreactive IgGs (anti-TPO, anti-TG, anti-FcεRI, and anti-IgE IgG), which are linked to established or possible autoimmune markers in patients with CSU who were receiving antihistamine treatment. By assessing the presence or absence of these autoantibodies, we elucidated their clinical characteristics and the potential to predict the response to antihistamine therapy. In conclusion, determining endotypes in CSU patients primarily hinges on the detection of both IgE and IgG autoantibodies against common autoantigens. Forty-six percentages of CSU patients have autoantibodies to TPO, and most have either IgE or IgG autoantibodies but not both. The distinct clinical profiles associated with anti-TPO IgE and anti-TPO IgG, including differential antihistamine responses, illuminate the heterogeneous nature of CSU. ACKNOWLEDGMENTS Biospecimens from healthy normal controls were provided by the Ajou University Human Bio-Resource Bank (AHBB), which is a member of the Korea Biobank Network, which is supported by the Ministry of Health and Welfare. This work was supported by a grant from the National Research Foundation of Korea funded by the Korean government (NRF-2022R1A2C2006607) and partly funded by a grant from the Korea Health Technology R&D Project (RS-2024-00439277) through the Korean Health Industry Development Institute, funded by the Ministry of Health and Welfare. Footnotes Disclosure: There are no financial or other issues that might lead to conflict of interest. SUPPLEMENTARY MATERIAL Supplementary Fig. S1 The study design. References Associated Data Supplementary Materials Supplementary Fig. S1 The study design.