The Bullous Pemphigoid Disease Activity Index independently scores 2 different clinical phenotypes, including the classic presentation with tense blisters and erosions (A) and urticaria and erythema (B).
The total BPDAI score was measured in a large cohort of consecutive patients with bullous pemphigoid. Total BPDAI scores range from 0 to 240, with higher scores indicating greater disease activity. OD indicates optical density values at 450 nm. Data points indicate individual patients. Solid line indicates linear regression fit across all patients. The Spearman rank correlation coefficients and P values are given.
Disease activity was measured as the total Bullous Pemphigoid Disease Activity Index (BPDAI) score in 10 patients during the course of their disease. Follow-up occurred a median of 1.5 and 4.0 months after diagnosis. A and B. Approximate Spearman rank correlation testing showed a significant correlation of intraindividual IgE and IgG anti-BP180 NC16A levels as visualized by generally increased autoantibody levels with higher total BPDAI scores. Each patient’s data are represented by a separate curve. C and D, One representative patient with bullous pemphigoid is shown at diagnosis (time 1) and a median of 1.5 (time 2) and 4.0 (time 3) months later. The BPDAI blister/erosion and urticaria/erythema scores range from 0 to 120, and the total BPDAI score ranges from 0 to 240, with higher scores indicating greater disease activity. OD indicates optical density values at 450 nm.
Correlation of anti-BP180 NC16A IgE with the Bullous Pemphigoid Disease Activity Index (BPDAI) urticaria/erythema and BPDAI blister/erosion scores (A and B) and for the correlation of anti-BP180 NC16A IgG with the BPDAI urticaria/erythema and BPDAI blister/erosion scores (C and D) (2 outliers not included in line [open circles]). The BPDAI blister/erosion and urticaria/erythema scores range from 0 to 120, with higher scores indicating greater disease activity. OD indicates optical density values at 450 nm. Data points indicate individual patients. Solid line indicates linear regression fit across all patients. The Spearman rank correlation coefficients and P values are given.
eMethods. Patients and Spiking Experiments
eResults. Verification of Specificity of the Novel IgE anti-BP180 NC16A ELISA
eTable. Cross-reactivity of Purified IgG and IgE Secondary Antibodies
eFigure 1. Sensitivity of the Novel IgE anti-BP180 NC16A ELISA
eFigure 2. Spiking Experiments Using Human IgE
eFigure 3. Detection of IgE anti-BP180 NCi6A Reactivity
Customize your JAMA Network experience by selecting one or more topics from the list below.
van Beek N, Lüttmann N, Huebner F, et al. Correlation of Serum Levels of IgE Autoantibodies Against BP180 With Bullous Pemphigoid Disease Activity. JAMA Dermatol. 2017;153(1):30–38. doi:10.1001/jamadermatol.2016.3357
In bullous pemphigoid (BP), are serum levels of IgE autoantibodies against BP180 NC16A correlated with the disease activity or associated with the urticarial phenotype?
In this noninterventional study, anti-BP180 NC16A serum IgE was found in 47 of 117 (40.2%) BP cases, which paralleled disease activity but showed no association with urticarial or erythematous lesions. In contrast, the presence of anti-BP180 NC16A IgG was associated with the presence of erosions and blisters.
Anti-BP180 NC16A IgE appears to be of pathogenic significance, although its exact pathogenic role remains to be elucidated.
Bullous pemphigoid (BP) is by far the most frequent autoimmune blistering disease. The presence of IgE autoantibodies against the transmembrane protein BP antigen 2 (BP180, type XVII collagen) has previously been reported in 22% to 100% of BP serum samples, and the pathogenic relevance of anti-BP180 IgE has been suggested in various experimental models and by the successful use of omalizumab in individual patients with BP.
To determine the rate of anti-BP180–reactive IgE in BP, to evaluate the diagnostic relevance of anti-BP180 IgE in BP, and to correlate anti-BP180 IgE with disease activity and the clinical phenotype of patients with BP.
Design, Setting, and Participants
This case-control cohort study examined 3 groups of patients with BP. Sixty-five patients with BP underwent an enzyme-linked immunosorbent assay for IgE antibodies against the 16th noncollagenous domain of BP180 (NC16A); 52 consecutive patients with BP underwent clinical evaluation with the Bullous Pemphigoid Disease Activity Index (BPDAI); and 36 patients with BP without anti-BP180 NC16A IgG reactivity underwent evaluation of the diagnostic importance of serum anti-BP180 IgE. In addition, 49 age-matched control individuals with noninflammatory dermatoses, 127 controls undergoing allergy testing for IgE levels, and 30 controls with pemphigus vulgaris or pemphigus foliaceus were included for comparison. Patients were seen at a university clinic from January 1, 2008, to July 31, 2014.
Main Outcomes and Measures
Serum anti-BP180 NC16A IgE and IgG levels and BPDAI scores.
Of 117 patients with BP (69 women and 48 men), anti-BP180 NC16A serum IgE was detected in 47 (40.2%) and correlated with disease activity as measured by total BPDAI (r = 0.918; P = .06). An intraindividual correlation of anti-BP180 NC16A serum levels with the total BPDAI was observed during the course of the disease in 10 randomly selected patients with BP (r = 0.983; P = .003). Although no association of circulating BP180 NC16A IgE antibodies with urticarial or erythematous lesions was observed (r = 0.481; P = .31), the presence of IgG anti-BP180 NC16A antibodies was associated with the occurrence of erosions and blisters (r = 0.985; P = .006) but not urticarial and erythematous lesions (r = 0.632; P = .23). Assaying for anti-BP180 IgE increased the diagnostic sensitivity by only 2.2% (1 of 46 serum samples) when combined with the IgG anti-BP180 enzyme-linked immunosorbent assay.
Conclusions and Relevance
Although detection of serum anti-BP180 IgE is not of diagnostic importance, it may be relevant for therapeutic decisions (eg, the use of anti-IgE treatment). The correlation of serum anti-BP180 NC16A IgE levels with disease activity in patients with BP supports the notion that anti-BP180 IgE is of pathogenic relevance. Our observation that IgG anti-BP180 antibodies are related to the occurrence of blisters and erosions may encourage further studies on the association of fine autoantibody reactivities with clinical features of BP.
Bullous pemphigoid (BP) is an autoimmune blistering disease that mainly affects elderly patients. Autoantibodies are directed against 2 hemidesmosomal proteins, the transmembrane protein BP antigen 2 (BP180, type XVII collagen) and the intracellular BP antigen 1 (BP230).1 Binding of autoantibodies leads to a complex inflammatory response involving complement activation, the infiltration of inflammatory cells, and the subsequent release of reactive oxygen species and distinct proteases that finally mediate subepidermal splitting.1 The 16th noncollagenous domain of BP180 (BP180 NC16A) has been identified as the immunodominant region of BP180,2 and IgG antibodies against BP180 NC16A (and its murine homologue) were shown to be pathogenic in several experimental models3-7 and to correlate with disease activity in patients with BP.8,9 In addition to IgG reactivity, IgA antibodies against BP180 are present in most BP serum samples,10,11 whereas data about IgE reactivity against the NC16A domain were reported in 100%, 55%, 22%, 77%, 30%, 61%, and 71% of BP serum samples.10,12-18 Depositions of IgE along the dermal-epidermal junction in patients’ skin was only observed in 25% and 41% of patients using direct immunofluorescence microscopy.19,20 The pathogenic potential of anti-BP180 IgE has been evaluated in vitro and in 2 mouse models.21-23 The following observations also point to an involvement of IgE autoantibodies in the pathogenesis of BP: (1) the frequent appearance of urticarial lesions alone or in addition to blisters and erosions24; (2) the severe pruritus in almost all patients with BP25; (3) peripheral eosinophilia in about 50% to 60% of patients with BP26; (4) elevated total IgE serum levels in 70% to 85% of patients with BP26; (5) IgE-coated mast cells in perilesional BP skin27; and (6) the successful use of the anti-IgE antibody omalizumab in individual patients with BP.28-30
Herein, we assayed serum samples from 3 large cohorts of patients with BP for IgE anti-BP180 NC16A reactivity with a novel enzyme-linked immunosorbent assay (ELISA). Furthermore, we addressed whether the levels of IgE anti-BP180 autoantibodies correlate with disease severity and are associated with a distinct clinical phenotype (ie, erythematous and urticarial lesions).
We assembled 6 cohorts of patients for this study. These included (1) 65 randomly selected patients with BP diagnosed in our routine autoimmune laboratory with serum BP180 NC16A IgG reactivity (patient cohort 1); (2) 52 consecutive patients with BP seen in our department before therapy was initiated (patient cohort 2); (3) 36 patients with BP without serum BP180 NC16A IgG antibodies (patient cohort 3); (4) 30 patients with pemphigus vulgaris or pemphigus foliaceus (control cohort 1); (5) 49 patients older than 75 years with noninflammatory dermatoses (mostly skin cancer) (control cohort 2); and (6) 127 patients with serum samples from our routine allergy laboratory that were sent for testing for total IgE levels (control cohort 3). In control cohort 3, total IgE serum levels were matched with those of patient cohorts 1 and 2 (total IgE levels <100 IU, 100-500 IU, and >500 IU [to convert to milligrams per liter, multiply by 0.0024]). Data were collected from January 1, 2008, to July 31, 2014. The study was performed in accordance with the Declaration of Helsinki31 and was approved by the ethics committee of the University of Lübeck, Lübeck, Germany, which waived the need for informed consent.
Bullous pemphigoid was diagnosed based on (1) a compatible clinical picture, (2) linear binding of IgG on monkey esophagus or human salt-split skin samples, and (3) ELISA reactivity with BP180 NC16A or BP230 (Euroimmun). In patient cohort 3 without NC16A IgG reactivity, all patients also had positive findings of direct immunofluorescence microscopy wherein BP230-specific IgG was detected or IgG reactivity with LAD-1, the soluble ectodomain of BP180, by immunoblotting with a conditioned medium of cultured human keratinocytes.32
In patient cohort 2 including 52 consecutive patients, the Bullous Pemphigoid Disease Activity Index (BPDAI) was determined at the time of hospitalization, when the serum sample was obtained. In some patients, the BPDAI was retrospectively obtained from photographs taken at the time of hospitalization. Scoring personnel were unaware of the presence or absence of serum IgE and/or IgG anti-BP180 autoantibodies. The BPDAI was developed by a panel of experts on BP and is composed of a score for the extent of blisters and erosions (range, 0-120 points), urticaria and erythema (range, 0-120 points), and pruritus (range, 0-30 points) (Figure 1), with higher scores indicating greater disease activity.33 The extent of the clinical presentation in this study was determined by the BPDAI blister/erosion and BPDAI urticaria/erythema scores and termed the total BPDAI score (range, 0-240 points).
Ten of the 52 patients in patient cohort 2 were examined during the course of the disease at the date of diagnosis and after a median of 1.5 and 4.0 months. Selection was based on regular follow-up visits and the availability of serum samples and BPDAI scoring. The treatment regimens for these 10 patients are detailed in the eMethods and eResults in the Supplement. Diagnosis of pemphigus in control cohort 1 was established by (1) a compatible clinical picture, (2) intercellular binding of IgG on the epithelium of monkey esophagus, and (3) ELISA reactivity with desmoglein 1 or 3 (Euroimmun).
The IgE anti-BP180 NC16A ELISA used the anti-IgE detection antibody from the total IgE ELISA with the IgG BP180 NC16A ELISA (both Euroimmun). Optimization was performed by testing (1) serial dilutions of serum (undiluted and dilutions of 1:10 and 1:100); (2) different dilution buffers; and (3) various secondary antibody dilutions (undiluted and dilutions of 1:10 and 1:50). Optimal ELISA results were obtained by incubation of BP180 NC16A 4 × ELISA plates (Euroimmun) with 50 µL of undiluted serum for 30 minutes at room temperature followed, after washing, by incubation with 100 µL of murine peroxidase-conjugated anti-IgE antibody (Euroimmun) diluted 1:10 in distilled water. After washing, the wells were incubated with 100 µL of a 3,3′,5,5′-tetramethylbenzidine–hydrogen peroxide solution chromogen and stopped with 100 µL of a 0.5M sulfuric acid solution. Autoantibody levels were determined as mean optical density (OD) values (at 450 nm) after subtraction of the blank OD value.
Total serum IgE levels were determined using a total IgE ELISA (Euroimmun) following the manufacturer’s instructions. Eosinophil counts were analyzed during routine laboratory testing on the same day when serum samples were taken.
For statistical analysis, we used Gnu R open-access statistical software (version 3.1; R Development Core Team, R Foundation for Statistical Computing [https://cran.r-project.org/]) with contributed packages “base,” “irr,” and “rocplus” for receiver operating characteristic curve analysis; “coin” for improved nonparametric tests (Spearman rank correlation analysis and Wilcoxon rank sum testing were used for 1-sided P values); ”MASS” for miscellaneous graphics and analyses; and “psych” for conversion of Spearman z to r values. All correlations were determined by approximate Spearman rank correlation test, if not outlined otherwise in the text. For all statistical analyses, 1-sided P < .05 was regarded as significant.
Patient cohort 1 included 65 patients with BP (40 women [61.5%]; 25 men [38.5%]; mean [SD] age, 74.6 [13.9] years); patient cohort 2, 52 patients with BP (29 women [55.8%]; 23 men [44.2%]; mean [SD] age, 78.2 [9.7] years); and patient cohort 3, 36 patients with BP (14 women [38.9%]; 22 men [61.1%]; mean [SD] age, 74.5 [11.1] years). Control cohort 1 included 30 patients with pemphigus vulgaris or pemphigus foliaceus (16 women [53.3%]; 14 men [46.7%]; mean [SD] age, 54.4 [12.3] years); control cohort 2, 49 patients with noninflammatory dermatoses older than 75 years of age (28 women [57.1%]; 21 men [42.9%]; mean [SD] age, 81.6 [4.6] years); and control cohort 3, 127 patients with serum samples from our routine allergy laboratory (70 women [55.1%]; 57 men [44.9%]; mean [SD] age, 50.3 [21.1] years).
In initial experiments, the anti-IgE detection antibody was shown to be specific in 3 different approaches. Elimination of IgG by affinity chromatography using protein G from BP serum samples with high anti-BP180 NC16A IgG levels resulted in about the same anti-BP180 NC16A IgE levels compared with IgE levels in the naive (nondepleted) BP serum samples. In addition, no cross-reactivity was seen between purified IgG and anti-IgE secondary antibodies (eTable in the Supplement). Furthermore, the 30 serum samples from controls with pemphigus vulgaris or pemphigus foliaceus did not reveal any IgE anti-BP180 NC16A reactivity. The interassay variability assessed by coefficient of variation was 7.4% (4 ELISA analyses performed on different days).
In subsequent experiments using a large number of control serum samples with total serum IgE levels matched with those of our BP cohorts, the sensitivity of the novel IgE anti-BP180 NC16A ELISA was shown to depend on total IgE serum levels (eFigure 1 in the Supplement). This finding was corroborated by spiking experiments using human IgE (eFigure 2 in the Supplement). In these experiments, the addition of increasing amounts of human IgE led to increased OD values in 4 of 6 serum samples. With cutoffs set in relation to total IgE serum levels (<100 IU, 100-500 IU, and >500 IU) (eFigure 3 in the Supplement), IgE anti-BP180 NC16A reactivity was detected in 25 of 65 serum samples (38.5%) from patient cohort 1 and in 22 of 52 serum samples (42.3%) from patient cohort 2 (combined sensitivity, 40.2%; 47 of 117 BP serum samples), with specificities ranging from 97.9% to 100%.
In patient cohort 2 (52 consecutive BP serum samples), 10 samples (19.2%) did not contain IgG anti-BP180 NC16A antibodies. In only 1 of these 10 IgG anti-BP180 NC16A–negative serum samples and in none of the 36 serum samples from patient cohort 3 (patients with BP without IgG anti-BP180 NC16A ELISA reactivity) was IgE anti-BP180 NC16A detected. Based on these results, the increase of the diagnostic sensitivity by the novel IgE anti-BP180 NC16A ELISA, in addition to the available IgG ELISA, can be estimated to 2.2% in BP.
Next, anti-BP180 IgE reactivity in patient cohort 2 (52 consecutive patients with BP) was related to the corresponding BPDAI score. We found an insignificant correlation of IgE anti-BP180 NC16A serum levels with the total BPDAI score (r = 0.918; P = .06) (Figure 2A). The presence of IgE anti-NC16A autoantibodies only was not significantly associated with a higher BPDAI score (z = −1.4364; P = .08, approximate Wilcoxon rank sum test). As expected, the total BPDAI score was significantly associated with anti-BP180 NC16A IgG serum levels (r = 0.979; P = .01, Figure 2B) and the presence of circulating anti-NC16A IgG autoantibodies (z = −1.5975; P = .046, approximate Wilcoxon rank sum test).
When we subsequently followed up the selected 10 patients with BP during the course of their disease, IgE anti-BP180 NC16A levels significantly paralleled disease activity as determined by total BPDAI score (r = 0.989; P = .003, with the approximate Spearman rank correlation test stratified by patient). During the course of the disease, IgE anti-BP180 reactivity became negative at the third time point (after a median of 4.0 months) in 8 patients; in 4 patients, anti-BP180 IgE was already undetectable at the second time point (after a median of 1.5 months). As expected, the same correlation was seen for anti-BP180 NC16A IgG (r = 0.997; P < .001; Figure 3A and B). Significant correlations of anti-BP180 serum antibodies were also observed with the BPDAI urticaria/erythema score (anti-BP180 NC16A IgE, r = 0.953; P = .03; anti-BP180 NC16A IgG, r = 0.983; P = .001) and the BPDAI blister/erosion score (anti-BP180 NC16A IgE, r = 0.989; P = .002; anti-BP180 NC16A IgG, r = 0.996; P < .001) (Figure 3C).
Reactivity of IgE against BP180 NC16A was not associated with urticarial or erythematous lesions (Figure 1) as scored by the BPDAI (r = 0.481; P = .31). Also, no association of anti-BP180 NC16A IgE with the classic phenotype, that is, the occurrence of blisters and erosions, was observed (r = 0.866; P = .09). In contrast, the level of IgG anti-BP180 NC16A autoantibodies significantly correlated with the BPDAI blister/erosion score (r = 0.985; P = .006) but not with the BPDAI urticaria/erythema score (r = 0.632; P = .23) (Figure 4 and Table).
In contrast to levels of BP180 NC16A–specific IgE, total serum IgE levels correlated with the BPDAI urticaria/erythema score (r = 0.930; P = .049) but not the BPDAI blister/erosion score (r = 0.560; P = .26) (Table). Furthermore, the extent of blood eosinophilia was associated with the total BPDAI score (r = 0.941; P = .04) and BPDAI blister/erosion score (r = 0.955; P = .03), but not the BPDAI urticaria/erythema score (r = 0.285; P = .39) (Table).
The severity of pruritus measured by the BPDAI correlated with levels of total serum IgE (r = 0.971; P = .02) but not with levels of circulating BP180 NC16A–specific IgE (r = 0.339; P = .36) or BP180 NC16A–specific IgG (r = −0.220; P = .58) or with the extent of blood eosinophilia (r = 0.252; P = .40). Furthermore, pruritus correlated with the BPDAI urticaria/erythema score (r = 0.956; P = .03) but not with the BPDAI blister/erosion score (r = −0.804; P = .86).
The pathogenic importance of IgE anti-BP180 NC16A antibodies has been demonstrated previously in various experimental approaches21-23 and with the successful use of the anti-IgE antibody therapy omalizumab in individual patients with BP.28-30 However, the following relevant clinical research questions remained unanswered: (1) The percentage of patients with BP and IgE anti-BP180 NC16A autoantibodies is unclear, with the reported frequency of positive IgE reactivity against B180 ranging from 22% to 100% in different reports10,12-18,20,34,35; (2) the diagnostic relevance of serum anti-BP180 IgE has not been elucidated; (3) the correlation of IgE anti-BP180 serum levels with disease activity has been reported13,14,18,36 but not thoroughly explored using a standardized clinical score; and (4) no data about the association of anti-BP180 IgE reactivity with a distinct clinical presentation has been addressed systematically.
In our initial experiments, receiver operating characteristics curve analyses resulted in greatly different sensitivities of 75% using the age-matched controls and 14% using the total serum IgE-matched cohort, whereas specificities were comparable. To verify the specificity of our anti-IgE detection antibody, we applied 2 different approaches. First, IgG was removed from selected BP serum samples. This procedure did not result in higher anti-BP180 NC16A ELISA values, excluding false-negative reactivity by serum IgG. This approach is of particular relevance because IgG is usually found in an approximately 10 000-fold excess compared with total serum IgE (IgG, 10 mg/mL; IgE, 100 IU [0.0003 mg/mL]), and IgG autoantibodies otherwise may have blocked binding sites of IgE anti-BP180 NC16A. Cross-reactivity between anti-IgE secondary antibody and IgG was excluded because no IgE reactivity was detected when purified IgG was assayed in the IgE BP180 NC16A ELISA. In line with this approach, assaying of BP serum samples after removal of IgG for IgE anti-BP180 ELISA reactivity revealed similar values compared with values in naive BP serum samples. The high specificity of the applied anti-IgE secondary antibody is also reflected by its wide use in commercial assays for the detection of total and specific IgE reactivity in allergies.
We then confirmed the effect of high IgE levels on the novel IgE anti-BP180 NC16A ELISA by spiking experiments of selected serum samples with human IgE. In these experiments, addition of increasing amounts of human IgE to 6 BP and control serum samples resulted in increased OD values in 4 samples when assayed using the novel IgE anti-BP180 NC16A ELISA. This phenomenon is known in allergy diagnostics, where low allergen-specific IgE levels associated with a high total serum IgE level are often not clinically relevant.37 Consequently and in line with our approach, in some allergy assays, the ratio between specific and total serum IgE levels is given, which was shown to be particularly valuable in serum samples with very low and high total serum IgE levels.37 To this end, the ratio of specific to total serum IgE was shown to predict the efficacy of anti-IgE treatment when measured by basophil activation (eg, in cat allergy).38,39
To account for the effect of total serum IgE levels, we determined anti-BP180 IgE levels in relation to the corresponding total IgE levels, resulting in a sensitivity of the novel IgE anti-BP180 NC16A ELISA of 38.5% and specificities ranging from 97.9% to 100%. A similar sensitivity (42.5%) was revealed when patient cohort 2 (52 consecutive patients with BP) underwent analysis. Comparable sensitivities were reported by Döpp et al13 and Ishiura et al,15 who used immunoblotting and ELISA, respectively. Divergent sensitivities may result from the use of other assay systems, such as immunoblotting, different detection antibody, and analyses independent of total serum IgE levels.
To evaluate the diagnostic use of the novel ELISA, the 10 BP serum samples that were found to be negative for IgG anti-NC16A in patient cohort 2 and the 36 BP serum samples without anti-BP180 NC16A IgG in patient cohort 3 (all positive by direct immunofluorescence microscopy) were analyzed and revealed IgE anti-BP180 reactivity in only 1 sample (2.2%). Thus, the IgE BP180 NC16A ELISA does not significantly increase the diagnostic sensitivity of the corresponding IgG ELISA but may be useful to identify patients with treatment-refractory disease before anti-IgE therapy (eg, omalizumab) is initiated. Whether the specific to total IgE antibody ratio remains unchanged in BP, as in anti-IgE treatment for allergic asthma,39 is yet to be determined.
In the next set of analyses, we showed that levels of circulating anti-BP180 IgE parallel disease activity during the course of the disease. The relatively weak association of IgE and IgG anti-BP180 NC16A serum levels with the disease activity in this large cohort of patients with BP is reflected by the clinical observation that a similar disease extent in different patients with BP is not necessarily accompanied by similar anti-BP180 ELISA values. More important and relevant for the monitoring of serum autoantibody levels during the course of the disease is a close intraindividual correlation of serum anti-BP180 antibodies with disease severity. Similar data have previously been reported for IgG anti-BP180 NC16A antibodies,8,9 leading to the recommendation of regular monitoring of anti-BP180 NC16A IgG serum levels during the course of the disease in recent guidelines of the European Academy of Dermatology and Venereology40 and the German Society of Dermatology.41 Previous studies have already suggested an association between IgE anti-BP180 reactivity and disease activity. Iwata et al18 found anti-BP180 IgE predominantly in patients with BP and extensive disease; Messingham et al42 observed a correlation of BP180 NC16A–specific IgE with disease activity in patients with BP and high total IgE levels; and Delaporte et al35 and Kalowska et al34 found a decrease of IgE anti-BP180 autoantibody levels when patients entered remission. These data further support the notion that IgE anti-BP180 NC16A antibodies are of pathogenic relevance in BP.
One important aim of the present study was the analysis of BP180 NC16A–specific IgE in relation to the clinical phenotype of patients with BP. We hypothesized that the presence of IgE anti-BP180 reactivity is associated with the occurrence of urticarial lesions and erythema based on the following lines of evidence: (1) anti-BP180 IgE induced pruritic erythema and microscopic but not macroscopic blistering in 2 mouse models22,23; (2) intense pruritus is found in atopic dermatitis, a dermatosis, like BP, associated with high serum IgE levels and eosinophilia25,43; and (3) mast cell activation, as shown in BP, may trigger urticarial lesions, such as in IgE-mediated type I allergies.27 However, no significant association of anti-BP180 IgE serum reactivity and urticarial or erythematous lesions was found in a clinically and immunopathologically well-characterized cohort of 52 consecutive patients with BP (patient cohort 2). This surprising result is in line with 2 other findings of this study. First, IgG anti-BP180 NC16A antibodies were strongly related to the classic phenotype (ie, blisters and erosions) but not to urticarial lesions and erythema, and IgG and IgE anti-BP180 antibodies highly correlated with each other, making an association of anti-BP180 IgE levels with the urticarial or erythematous phenotypes unlikely. Total serum IgE levels may be more relevant for the expression of urticarial or erythematous lesions because their levels correlated with this phenotype but not with the presence of blisters or erosions.
In accordance with Messingham et al,42 blood eosinophilia correlated with the overall disease activity and the extent of erosions and blisters, which suggests a pathogenic role of eosinophils in BP. This assumption is supported by the recent finding that the high-affinity FcεRI is expressed on peripheral and skin eosinophils of patients with BP, allowing anti-BP180 NC16A IgE to trigger eosinophil degranulation.42
A limitation of our study is that the data were gathered from patients at a single center. Including serum samples from more than 1 center would have strengthened our results.
The novel ELISA allows the reliable detection of serum IgE anti-BP180 NC16A antibodies in relation to total serum IgE levels. Although not of diagnostic relevance, the novel ELISA may be helpful to identify patients with BP who are suitable candidates for anti-IgE therapy. The clear correlation of IgE anti-BP180 NC16A serum levels with disease activity further suggests a pathogenic role of IgE autoantibodies in BP. Although IgE anti-BP180 reactivity was not associated with a distinct clinical presentation, IgG autoantibodies against BP180 NC16A were related to the classic blistering phenotype. Further studies are needed to define the pathogenic impact of anti-BP180 IgE and its relation to activation and/or degranulation of mast cells and eosinophils in BP.
Corresponding Author: Enno Schmidt, MD, PhD, Department of Dermatology and Lübeck Institute of Experimental Dermatology, University of Lübeck, Ratzeburger Allee 160, D-23538 Lübeck, Germany (firstname.lastname@example.org).
Accepted for Publication: July 21, 2016.
Correction: This article was corrected on May 24, 2017, to fix an error in Figure 4D.
Published Online: November 9, 2016. doi:10.1001/jamadermatol.2016.3357
Author Contributions: Drs van Beek and Schmidt had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: van Beek, Schmidt.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: van Beek.
Critical revision of the manuscript for important intellectual content: Lüttmann, Huebner, Recke, Karl, Schulze, Zillikens, Schmidt.
Statistical analysis: van Beek, Huebner, Recke, Zillikens.
Obtained funding: Schmidt.
Administrative, technical, or material support: Lüttmann, Recke, Schulze, Zillikens, Schmidt.
Study supervision: Zillikens, Schmidt.
Conflict of Interest Disclosures: Drs Zillikens and Schmidt report having scientific cooperation with and research grants from Euroimmun. No other disclosures were reported.
Additional Contributions: Medical technical assistants Vanessa Krull and Ingeborg Atefi, University of Lübeck, Lübeck, Germany, contributed technical assistance, for which they did not receive additional compensation.
Create a personal account or sign in to: