Diagnostic criteria of BP: 1) Intensely pruritic blisters of the skin, 2) Positive linear IgG and/or C3c staining along the BMZ by direct immunofluorescence (DIF), 3) Positive IgG staining on the epidermal side of 1M-NaCl-split normal human skin by indirect immunofluorescence (ssIIF), 4) Positive IgG enzyme-linked immunosorbent assay (ELISA) of BP180 NC16A and/or BP230 (MBL, Nagoya, Japan, >9 U/mL). BP was diagnosed if the first criterion and at least two of the last three criteria were met. Exclusion criteria included drug-induced BP, patients with BP who were followed up for less than 3 months, and those who had used other biological agents within half a year before dupilumab treatment.

Sixty-nine BP patients diagnosed between June 2021 and June 2023 in the Hospital for Skin Diseases, Institute of Dermatology, Chinese Academy of Medical Sciences & Peking Union Medical College were involved in this study, including 13 patients treated with dupilumab monotherapy (dupilumab group, Dupi group), 24 patients treated with dupilumab combined with corticosteroids (dupilumab and corticosteroids group, Dupi + CS group), and 32 patients treated with monotherapy of corticosteroids (Corticosteroids group, CS group). In addition, all patients were initially prescribed 0.05% clobetasol propionate cream twice a day on the area of lesional skin and discontinued when the skin lesions had healed. In the present study, the main reasons of patients for dupilumab treatment include 1) Some patients cannot tolerate long-term or large-dose corticosteroids therapy due to underlying diseases (advanced age, hypertension, diabetes, kidney disease, hepatitis, tuberculosis, etc.); 2) Traditional treatment has proved ineffective; 3) Some patients refused to use corticosteroids therapy. Based on the regimen of dupilumab for the treatment of atopic dermatitis,29 patients in the Dupi group and Dupi + CS group first received 600 mg of dupilumab (induction dose) and then 300 mg every other week via subcutaneous injection.

In this study, all BP patients were followed up for 3–12 months. Whole blood samples were collected from all BP patients at baseline, 2–3 weeks after treatment, and 1–3 months after treatment. The Bullous Pemphigoid Disease Area Index (BPDAI) and the Numeric Rating Scale (NRS) score of itching were performed at baseline and 2–4 weeks after treatment. The BPDAI was determined by a panel of experts on BP and is composed of a score for the degree of erosions/blisters (range, 0–120 points), urticaria/erythema (range, 0–120 points) and mucosal erosions/blisters (range, 0–120 points), with higher scores indicating greater disease activity.30 BPDAI total score (range, 0–360 points) is the sum of three components including the BPDAI erosion/blister, BPDAI urticaria/erythema and BPDAI mucosal erosion/blister. Pruritus was evaluated by itching NRS score, which ranges from 0 (no itching) to 10 points (worst insupportable itching).30

At each follow-up, the occurrence of disease control, complete remission during tapering, relapse and adverse events were assessed. Disease control is defined as the point at which new lesions or pruritic symptoms cease to form and established lesions begin to heal.30 The disease control period is the period from baseline to disease control. Complete remission during tapering is defined as the absence of non-transient lesions while the patient is receiving more than minimal therapy.30 Relapse is defined as the appearance of 3 or more new lesions a month (blisters, eczematous lesions, or urticarial plaques) or at least one large (10 cm diameter) eczematous lesion or urticarial plaque that does not heal within 1-week, or the extension of established lesions or daily pruritus in a patient who has achieved disease control.30 Additionally, since the present study does not involve dosage reduction of Dupilumab, the concept of complete remission with medicine is exclusively used in the Dupi group, meaning the absence of non-transient lesions.

The changes of all clinical and laboratory parameters were evaluated by reduction rate, the percentage of decrease in the parameters after treatment compared to those before treatment.

The present study was approved by the Ethics Committee of the Institute of Dermatology and Hospital for Skin Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College (2017-KY-022). Written informed consents were obtained from all involved patients and controls.

For DIF, skin biopsy specimens of BP patients were embedded in the optimum cutting temperature compound. Cryosections (6 µm thick) of these specimens were incubated for 35-min at 37 °C with fluorescein isothiocyanate (FITC)-conjugated sheep anti-human IgG antibody (ZF0306, ZSGB-BIO, Beijing, China) or FITC conjugated rabbit anti-human C3c antibody (ZF0301, ZSGB-BIO, Beijing, China) both at a 1:100 dilution with Phosphate-Buffered Saline (PBS). The results were viewed independently using a fluorescent microscope (Eclipse 50i, Nikon, Tokyo, Japan) by at least two trained dermatologists.

The skin tissues taken from normal individuals were washed with aseptic saline solution and then placed into a 50 mL sealed centrifuge tube with 1 moL/L NaCl (20°‒30 °C) for about 24 h, until the surface of the skin wrinkled, which indicated that the epidermal and dermal layers had been separated. This salt-split skin was next embedded with ornithine carbamyl transferase and cut into frozen sections 6 µm thick.31,32 Serum samples of BP patients were diluted at 1:40 with PBS, and incubated with salt-split skin sections at 37 °C for 30-min. After thoroughly washing with PBS (10 mM, Ph 7.3), the sections were incubated with FITC-conjugated sheep anti-human IgG antibody (GTX26866, GeneTex, San Antonio, USA) at a 1:500 dilution with PBS at 37 °C for 30-min. The fluorescence intensity was assessed by at least two trained dermatologists.

IgG antibodies against BP180 NC16A, BP230, Desmoglein 1 (Dsg1) and Dsg3 were measured by MESACUP BP180/BP230/Dsg1/Dsg3 TEST (MBL, Nagoya, Japan) according to the manufacturer's instruction. The serum samples were diluted to 1:101. The cut-off values for anti-BP180 NC16A, BP230, Dsg1 and Dsg3 antibodies (IgG) in serum samples were 9 U/mL, 9 U/mL, 20 U/mL and 20 U/mL, respectively.

The routine blood test of BP patients was measured using the colter cellular analysis system (UniCel DxH 600, Beckman Coulter, California, USA). Whole blood count and percentage for eosinophil were used for the analyses in the present study and the cut-off values were 0.02∼0.52*109/L and 0.4∼8.0%, respectively.

The level of serum total IgE, which was included in a screening panel of nutritive and inhalant allergens, was performed by Immuno CAP (Phadia/Thermofisher, Uppsala, Sweden). The cut-off value for total IgE in serum samples was 60 KU/L.

Wilcoxon signed-rank test, Mann-Whitney U test, Kruskal-Wallis test and Fisher’s exact test was conducted depending on the type of analyses. The Steel-Dwass test was used for post-hoc pairwise comparisons between three subgroups after significant Kruskal-Wallis test. All data were presented as mean ± SD for descriptive purposes. The p-values were two-sided and performed with the appropriate statistical tests using GraphPad Prism software 9.4.0 and R software 4.1.2.33 A statistically significant difference was considered to be present at p < 0.05.

Sixty-nine BP patients were involved in this study, including 13 patients treated with dupilumab monotherapy (Dupi group), 24 patients treated with Dupilumab combined with Corticosteroids (Dupi + CS group), and 32 patients treated with monotherapy of Corticosteroids (CS group). The average age of BP patients in Dupi, Dupi + CS and CS groups were 73.2, 67.5 and 68.7 years old, respectively; and the ratio of male patients in these three groups was 61.5%, 58.3% and 40.6%, respectively. Clinical and immunological features of all patients at baseline were shown in Table 1.

For immunofluorescence detection in Dupi, Dupi + CS, and CS groups, the positive rates of DIF for linear IgG deposition to BMZ were 40.0% (2 of 5 cases), 73.3% (11 of 15 cases) and 71.4% (15 of 21 cases), respectively; the positive rates of DIF for linear C3c deposition to BMZ were 40.0% (2 of 5 cases), 60.0% (9 of 15 cases) and 81.0% (17 of 21 cases), respectively; the positive rates of ssIIF for IgG reactivity with epidermal side were 66.7% (8 of 12 cases), 76.2% (16 of 21 cases) and 83.3% (25 of 30 cases), respectively (Table 1).

By ELISA, all the serum samples (69 cases) of BP patients in the three groups were positive for anti-BP180 and/or anti-BP230 autoantibodies (IgG) at baseline. The data about serum total IgE level, blood eosinophil count and eosinophil percentage in BP patients of the three groups were also collected (Table 1).

The BPDAI was used to measure the disease activities of BP. The mean of BPDAI Total Score in the three groups at baseline were 48.1, 60.1 and 51.1, respectively (Table 1). In addition to BPDAI Total Score, other scores including BPDAI Erosion/Blister Score, BPDAI Urticaria/Erythema Score, BPDAI Mucosal Score and Itching NRS Score were also summarized in Table 1. In the present study, there were 6 of 13 (46.2%), 8 of 24 (33.3%) and 13 of 32 (40.6%) patients with oral mucosa lesions in Dupi, Dupi + CS and CS groups, respectively; and the mean of BPDAI Mucosal score at baseline were 4.9, 3.5 and 4.1, respectively (Table 1). Among these patients, 0% (0 of 5 cases), 42.9% (3 of 7 cases) and 25% (3 of 12 cases) were positive for herpes simplex virus type 1 by oral mucosal virus detection, respectively. These patients also underwent fungal detection on the erosive surface of the oral mucosa, but all were negative.

For Dupi + CS and CS groups, the median dose of oral prednisone used at baseline was 40 mg (range: 10–75 mg) and 50 mg (range: 20–80 mg), respectively; the median cumulative dosage of prednisone at the time of disease control were 225 mg (range: 30–675 mg) and 280 mg (range: 90–960 mg), respectively; and the median period of time to initiate the tapering were 7 days (range: 5–28 days) and 14 days (range: 7–28 days), respectively.

Clinical and laboratory differences between pre-treatment and post-treatment in the three groups of BP patients were summarized in Table 1.

In Dupi group, the BPDAI Total (p < 0.01), Erosion/Blister (p < 0.01), Urticaria/Erythema (p < 0.05) and Itching NRS (p < 0.01) scores were all reduced significantly after 2–4 weeks’ treatment, but there was no significant decrease in the BPDAI Mucosal Score after 2–4 weeks’ treatment (Table 1 and Fig. 1). The mean serum total IgE level was elevated from 1246.6 KU/L at baseline to 1710.2 KU/L after 2–3 weeks’ treatment (n = 5). The mean serum levels of anti-BP180 (n = 4) and anti-BP230 (n = 3) autoantibodies (IgG) were changed from 85.5 U/mL and 8.8 U/mL at baseline to 82.1 U/mL and 31.7 U/mL after 1–3 months’ treatment, respectively. The blood eosinophil count and percentage were not reduced significantly after 2–3 weeks of treatment (Table 1).

Fig. 1.

Comparison of clinical parameters before and after 2‒4 weeks’ treatment in Dupi group of BP patients.

Five clinical scores for evaluation of BP patients. Statistical analyses were performed with Wilcoxon signed rank test. *p < 0.05; **p < 0.01; ns, not significant.

(0.12MB).

In the Dupi + CS group, all the clinical parameters including the BPDAI Total (p < 0.001), Erosion/Blister (p < 0.001), Urticaria/Erythema (p < 0.001), Mucosal (p < 0.05) and Itching NRS (p < 0.001) scores were reduced significantly after 2–4 weeks’ treatment (Table 1). The serum level of anti-BP180 autoantibodies (IgG) was reduced significantly (p < 0.01) after 1–3 months treatment (Table 1). The blood eosinophil count and percentage were reduced significantly (both p < 0.001) after 2–3 weeks treatment (Table 1). However, there was no significant decrease in the serum total IgE level after 2–3 weeks’ treatment; and no significant decrease in the serum anti-BP230 autoantibodies (IgG) level was observed after 1–3 months’ of treatment (Table 1).

In the CS group, all the clinical parameters including the BPDAI Total (p < 0.001), Erosion/Blister (p < 0.001), Urticaria/Erythema (p < 0.001), Mucosal (p < 0.01) and Itching NRS (p < 0.001) scores were reduced significantly after 2–4 weeks’ treatment (Table 1). Both the serum levels of anti-BP180 (p < 0.001) and anti-BP230 (p < 0.01) autoantibodies (IgG) were reduced significantly after 1–3 months’ of treatment (Table 1). The blood eosinophil count and percentage were reduced significantly (both p < 0.001) after 2–3 weeks treatment (Table 1). But no significant decrease in the serum total IgE level was observed after 2–3 weeks’ of treatment (Table 1).

Difference analyses of the reduction rates of clinical and laboratory characteristics after corresponding treatment among the three groups of BP patients were summarized in Figs. 2, 3, S2, and S3.

Fig. 2.

Comparison of reduction rates of two clinical parameters after 2‒4 weeks’ corresponding treatments among Dupi, Dupi + CS and CS groups.

(A–E) Reduction rates of BPDAI Total Score (A), and BPDAI Mucosal score (B). Statistical analyses were performed with Steel-Dwass test for post-hoc pairwise comparisons between three groups after significant Kruskal-Wallis test. **p < 0.01; ns, not significant.

(0.14MB).

Fig. 3.

Comparison of reduction rates of blood eosinophil level after corresponding treatments among Dupi, Dupi + CS and CS groups.

(A–B) Reduction rates of blood eosinophil count (UniCel DxH 600, Beckman Coulter, California, USA, 0.02∼0.52*109/L) after 2–3 weeks’ treatment (A), and blood eosinophil percentage (UniCel DxH 600, Beckman Coulter, California, USA, 0.4∼8.0%) after 2–3 weeks’ treatment (B). Statistical analyses were performed with Steel-Dwass test for post-hoc pairwise comparisons between three groups after significant Kruskal-Wallis test. **p < 0.01; ***p < 0.001; ns, not significant.

(0.14MB).

For clinical indicators, the reduction rates of both the BPDAI Total and Mucosal scores in Dupi group were significantly lower than those in Dupi + CS (both p < 0.01) and CS (both p < 0.01) groups, respectively, with no significant difference between the latter two groups (Fig. 2A and B). There were no significant differences in the reduction rates of BPDAI Erosion/Blister, Erythema, and Itching NRS scores among the three groups (Figs. S1‒S2).

For laboratory parameters, the reduction rates of the blood eosinophil count and percentage in the Dupi group were significantly lower than those in Dupi + CS (both p < 0.01) and CS (p < 0.001, p < 0.01) groups, respectively, with no significant difference between the latter two groups (Fig. 3A and B). However, there were no significant differences in the reduction rates of serum levels of total IgE, anti-BP180, and anti-BP230 autoantibodies (IgG) among the three groups (Figs. S3‒S4).

In the Dupi group, only 5 of 13 cases achieved disease control and the mean disease control period was 14.8 days (9–21 days); among them, 2 patients achieved complete remission with medicine at week 8. There were no significant differences in the disease control period, percentage of patients achieving complete remission during tapering at week 8, and relapse rates within one year between Dupi + CS and CS groups (Figs. S5‒S6). However, the dose of prednisone at baseline (37.3 ± 16.9 mg vs. 54.2 ± 18.4 mg, p < 0.01) and cumulative dosage of prednisone at the time of disease control (238.1 ± 162.3 mg vs. 351.7 ± 207.9 mg, p < 0.05) administered in Dupi + CS group were significantly less than those in CS group, respectively (Fig. 4A and B).

Fig. 4.

Comparison of dosage of prednisone administered between Dupi + CS and CS groups.

(A) Dose of prednisone used at baseline. (B) Cumulative dosage of prednisone at the time of disease control. Statistical analyses were performed with Mann-Whitney U test and fisher’s exact test. *p < 0.05; **p < 0.01.

(0.15MB).

As summarized in Table 2, a total of 23 adverse events occurred during the follow-up period. Four patients experienced transient peripheral blood eosinophilia, including three (23.1%, 3 of 13 cases) in Dupi group and one (4.2%, 1 of 24 cases) in Dupi + CS group. One patient (4.2%, 1 of 24 cases) in Dupi + CS group experienced conjunctivitis after two dupilumab injections. Compared with patients in the CS group (15.6%, 5 of 32 cases), patients in Dupi + CS group (8.3%, 2 of 24 cases) showed a decreased overall risk of infections (including serious or opportunistic infections). More cases of osteoporosis (6.3%, 2 of 32 cases), hypertension (9.4%, 3 of 32 cases), diabetes (6.3%, 2 of 32 cases) and hypodynamia (9.4%, 3 of 32 cases) were seen in the CS group. No deaths occurred during the follow-up period. No patients in the Dupi and Dupi + CS groups were discontinued due to adverse events.