Diagnostic challenge: Juvenile bullous pemphigoid with a negative BP180 ELISA

Abstract

Bullous pemphigoid (BP) is an autoimmune blistering disease primarily affecting the elderly, whereas cases of juvenile BP are rare. Both types of BP are typically mediated by autoantibodies targeting the NC16A region of BP180; however, a small subset of adult patient sera react to other regions of the protein. The incidence of a similar occurrence in juvenile BP is unknown. This case of juvenile BP with a negative BP180 ELISA highlights diagnostic pitfalls that can lead to a delay in the correct diagnosis in the pediatric population.

1. INTRODUCTION

Bullous pemphigoid (BP) is an autoimmune blistering disease with pathogenesis primarily mediated by autoantibodies targeting BP180 (also termed collagen XVII). ¹ , ² These antibodies target the extracellular NC16A region of BP180 in ~90% of patients, ¹ , ² while the serum of the remaining patients targets other areas of the BP180 protein, known as “atypical or non‐NC16A BP.” ³ This is a diagnostic issue since the CLIA‐approved BP180 ELISA screens only for reactivity to NC16A and will result in a negative test in this patient subset.

Juvenile BP is much less common but shares the same phenotype with sub‐epidermal blisters and deposition of IgG and C3 at the dermo‐epidermal junction (DEJ). ⁴ , ⁵ Autoantibodies in juvenile BP also target BP180 and the disease runs a limited course, usually resolving within 2–5 years. ⁴ , ⁵ Due to the rarity of juvenile BP, the incidence of the atypical form is unknown. ⁶ We report a case of juvenile BP with a negative serum BP180 ELISA test that highlights the need for a complete immunodiagnostic workup in these patients.

2. CASE REPORT

A 6‐year‐old girl was seen by pediatric dermatology for an itchy eruption on her trunk and extremities and oral erosions. Topical clobetasol (0.05%) and hydrocortisone (2.5%) had been previously prescribed. Her oral lesions combined with an ANA (1:640) and lymphopenia raised concern for bullous SLE; however, SM, RNP, SS‐A, and SS‐B antibodies, cardiolipin testing, and complement levels were all normal and the patient did not have arthritis, photosensitivity, pancytopenia, renal involvement, neurologic abnormalities, or a malar/discoid rash. Her serum IgE was elevated at 198 IU/mL (normal 0–90 IU/mL).

A perilesional punch biopsy showed sub‐epidermal blistering with mixed acute and chronic inflammation featuring predominant lymphocytes and some neutrophils. Direct immunofluorescence (IF) revealed linear deposition of IgG and C3 at the dermo‐epidermal junction but no IgM, IgA, or fibrinogen. A serration pattern (n or u) was not noted, and no archival tissue remained. Indirect IF was not performed. The patient was diagnosed with neutrophilic subepidermal vesicular dermatosis, which responded to prednisone (1 mg/kg/d) and dapsone (25 mg/d). After about 4 months, initiation of prednisone taper resulted in new skin and oral blisters. The patient continued taking dapsone and the skin and oral blister were resolving (without prednisone) when she returned to the clinic a month later. The patient was lost to follow‐up for approximately 2 years.

Two years later, the patient was referred by an outside physician to our Blistering Disease Clinic with a two‐year history of intermittent blistering involving the scalp, trunk, extremities, and, more recently, oral mucosa. The oral blistering interfered with swallowing and necessitated a visit to the emergency department; however, no laryngeal involvement was noted. Her skin involvement was extensive with erythematous plaques with tense blisters persisting on her face, trunk, and extremities. Older lesions showed residual hyperpigmentation and extensive milia (Figure 1A). Indirect IF was strongly positive for IgG on the roof of salt‐split skin (1:2560, epidermal pattern); IgA was negative. IgG ELISAs for BP180 (NC16A), BP230, and collagen VII were all negative (ARUP, Salt Lake City, UT). Her course was complicated by barriers that hindered follow‐up and led to lapses in her therapy and workup. During one lapse, trouble swallowing prompted a visit to an outside emergency room and a subsequent endoscopic examination of her GI tract that included biopsies of friable lesions in the upper esophagus (Figure 1B). Histologic examination revealed a subepithelial split with minimal chronic inflammation. ⁷ , ⁸ Again, no DIF was performed on esophageal tissue, and archival tissue was unavailable. The patient was prescribed a 20‐day prednisone taper and further work‐up was recommended.

FIGURE 1.

Clinical presentation of juvenile bullous pemphigoid (BP). (A) Widespread erythematous plaques and older lesions with residual hyperpigmentation and extensive milia on the leg. Difficulty swallowing led to an examination of her GI tract. A fluoroscopic swallow study noted strictures of the proximal esophagus and (B) endoscopic views noted small (<1 cm) esophageal strictures with friable, sloughing mucosa.

Two months later, additional testing was conducted through our Program for Cutaneous Autoimmunity Lab. Serum antibodies for BP180 (NC16A) and BP230 were confirmed negative by ELISA (≤9 Index, MBL International, Japan). Due to her mucosal involvement, reactivity to collagen VII and laminin 332 (Biochip, Euroimmun US, New Jersey) was evaluated with negative results. Serum reactivity to areas of BP180 outside of the NC16A region was tested using recombinant proteins by immunoblot using published methods. ³ Proteins included the entire extracellular domain of BP180 (sec180 ⁹ ) and smaller fragments ³ including the NC16A and NC4 domains (Figure 2A). Also included were lysates made from normal human keratinocytes and BP180 negative keratinocytes, derived from a patient with generalized atrophic benign epidermolysis bullosa (GABEB ¹⁰ ).

FIGURE 2.

Recombinant fusion proteins used to screen serum for reactivity to regions of BP180 (Collagen XVII). (A) Representation of the full‐length BP180 protein, the extracellular domain (referred to as sec180), and smaller regions of known immunoreactivity arbitrarily numbered 1 to 6. The amino acids (AA) encoding each region are indicated as the non‐collagenous 16A (NC16A) region and the non‐collagenous NC4 region of BP180. The NC16A region is immunodominant in both adult and juvenile BP. (B) Immunoblot summary data showing reactivity of serum from this case of juvenile BP vs. serum from a well‐characterized patient with classic BP to potential target antigens including native BP180 expressed by keratinocytes and recombinant proteins encoding sec180 and BP180 fusion proteins 1 to 6. Although the juvenile BP serum was negative on the diagnostic BP180 (NC16A) ELISA, immunoblot confirmed reactivity to BP180, specifically to the NC16A and NC4 fusion proteins. ND = not done. Created with Biorender.com.

Despite the negative ELISA test, the summarized immunoblot results (Figure 2B) show that the juvenile BP serum was reactive to both the sec180 protein and the BP180‐positive keratinocyte lysate. Further screening against BP180 protein fragments identified reactivity to the NC16A (+) and the NC4 (++) domains. As expected, control serum from a patient with classic BP was strongly reactive to the BP180 positive keratinocyte lysate, and both the sec180 and NC16A proteins. ¹ , ³ Images of the immunoblots using sera from this juvenile BP case and a patient with classic BP are shown in Figure 3.

FIGURE 3.

Immunoblots showing serum IgG reactivity to native and recombinant BP180. In the left immunoblot, juvenile BP serum was screened for reactivity to the immunodominant NC16A‐GST fusion protein or equimolar amounts of GST control protein, the recombinant full‐length extracellular domain of BP180 (sec180), and lysates of immortalized normal BP180‐expressing human epithelial keratinocytes (iHEKC) or BP180 negative keratinocytes obtained from a patient with generalized atrophic benign epidermolysis bullosa (GABEB KC). The vertical white line indicates where the blot was cut for incubation with the indicated serum. In the right immunoblot, further screening of juvenile BP serum against recombinant proteins representing immunoreactive regions (numbered 1–6 in Figure 2) of BP180 outside NC16A. In both panels, blots are co‐stained with commercial antibodies (Ab) specific for BP180 or GST (red) to confirm the position of target protein bands. White arrowheads highlight bands of reactivity and orange bands indicate a strong overlap of serum (green) and commercial Ab reactivity (red). As additional controls, identical blots are probed with a well‐characterized BP serum and healthy control serum (background). Molecular weight (MW) markers are indicated with numerical weight in kDa. Antibodies used: Anti‐GST (Cell Signaling CS2624S, Danvers, MA), anti‐BP180 (Abcam 184,996, Waltham, MA), IRDye 800CW Goat Anti‐human IgG (Licor 926‐32232, Lincoln, NE), IRDye 680RD Goat anti‐mouse IgG (Licor 926‐68070), IRDye 680RD Goat anti‐rabbit IgG (Licor 926‐68071). The protein bands were visualized by using a Licor Odyssey IR system.

The patient returned 6 months later with active blistering, widespread scarring, and milia formation but no oral lesions. She was prescribed topical clobetasol (0.05%), hydrocortisone (2.5%), and tacrolimus (0.03%). Due to her prior esophageal lesions and difficulties in frequent follow‐up, the required testing (CBC, AST/ALT, hepatitis, HIV) for rituximab was obtained. In January 2024, she received 1 cycle of rituximab (375 mg/m² per week for 4 weeks). After completion of the four doses and cessation of prednisone, the esophageal lesion had resolved, and the skin lesions were much improved.

3. DISCUSSION

This report emphasizes that the CLIA‐approved ELISA alone is not adequate to rule out pemphigoid and its variants. ⁷ , ⁸ In the case of a negative BP180 serum ELISA, diagnosis of both adult and pediatric patients is based on subepidermal blistering, direct IF showing linear deposition of immune reactants at the DEJ, and/or indirect IF with staining localized to the roof of salt‐split skin. ⁷ , ⁸ Here, a diagnosis of BP over MMP was based on the high titer antibodies reactive to the roof of salt‐split skin, prominent and sustained skin involvement, and reactivity to BP180, despite occasional oral involvement. ¹² In line with this, the mucosae (oral and genital) and hands and feet are more commonly involved in the juvenile form of BP. ⁴ , ⁵ , ⁶ Additionally, EBA and bullous SLE are characterized by IIF reactivity on the base of salt‐split skin, or occasionally both the base and the roof. ⁷ , ¹¹

Few specialized labs can test for serum reactivity to specific regions of BP180 outside of NC16A, usually by immunoblot. ¹ , ³ In our 2016 report of four adult non‐NC16A patients, one serum had weak reactivity to recombinant NC16A by immunoblot, ³ like the pediatric case presented here. This observation may be due to the increased sensitivity of immunoblot or unmasking of antigenic sites by the denaturing conditions. In this pediatric case, serum reactivity to NC4 appeared stronger than NC16A, which suggests NC4 may be the primary pathogenic epitope. Reactivity to the NC4 domain (with or without reactivity to NC16A) is also observed in adult BP patients. ³ , ¹³

Adults with atypical BP have a variable phenotype; some look like typical BP while others have unusual features such as significant oral involvement and/or nail involvement. ³ It has been hypothesized that disease presentation is influenced by the identity of the immunodominant epitope for each patient. For example, reactivity to the C‐terminal areas of BP180 is associated with oral lesions in mucous membrane pemphigoid. ¹⁴ This same phenomenon may also apply to juvenile BP; however, additional patients are needed to address this.

In summary, we report the first case of juvenile BP with a negative serum BP180 ELISA test. Physicians caring for these patients should know that a negative test is inadequate to exclude BP and that further workup (including IIF) is warranted.

FUNDING INFORMATION

University of Iowa Department of Dermatology.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

CONSENT STATEMENT

Written consent was obtained from the parent/guardian granting permission to publish this case and include patient images.

Fairley JA, Berrebi KG, Poggemiller AM, Varzavand A, Messingham KN. Diagnostic challenge: Juvenile bullous pemphigoid with a negative BP180 ELISA . Pediatr Dermatol. 2025;42(2):342‐345. doi: 10.1111/pde.15741

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.