55 - BULLOUS PEMPHIGOID Flashcards
most common autoimmune blistering disorder in the adult population
Bullous pemphigoid
The disease typically presents as pruritic, tense blisters often on a background of urticarial plaques. It occurs most frequently in older adults. It is mediated by autoantibodies directed against hemidesmosomal proteins BP180 and BP230, which trigger an inflammatory cascade that ultimately leads to blister formation.
Bullous pemphigoid was originally described as a subepidermal blistering disease with distinctive clinical and histologic features by Walter Lever in 1953. 1 Antibodies against the dermal–epidermal junction were first described in perilesional skin and in the serum of patients by Jordon and Beutner 14 years later, confirming the separation of bullous pemphigoid from pemphigus. 2 Over the following years, the antigenic targets were fully characterized as hemidesmosomal proteins BP180 and BP230.3-5
EPIDEMIOLOGY
Bullous pemphigoid typically occurs in patients older than 60 years of age, with a peak incidence in the 70s.6 There are exceptions in which classic bullous pemphigoid occurs in middle-aged adults and even infants and children, which is especially rare. 7-10 There is no known ethnic, racial, or sexual predilection for developing bullous pemphigoid.
The incidence of bullous pemphigoid is estimated to be 7 per 1 million per year in Germany and 14 per 1 million per year in Scotland. 6,11 Recent studies suggest that the incidence of bullous pemphigoid is increasing and may be as high as 22 to 24 per 1 million per year in the United States and France and 43 per 1 million per year in the United Kingdom.12-14
CUTANEOUS FINDINGS
The classic form of bullous pemphigoid is characterized by large, tense blisters arising on normal skin or on an erythematous or urticarial base (Fig. 54-1A). 15,16 These lesions are most commonly found on flexural surfaces, the lower abdomen, and the thighs, although they may occur anywhere. The bullae are typically filled with serous fluid but may be hemorrhagic. The Nikolsky and Asboe-Hansen signs are negative. Eroded skin from ruptured blisters usually heals spontaneously without scarring, although milia can occur, and postinflammatory pigmentation is common. Pruritus is usually intense but may be minimal in some patients.
Although the more classic presentation of bullous pemphigoid consists of tense bullae on an erythematous or urticarial base, a noninflammatory form of bullous pemphigoid may also be seen presenting as tense bullae on normal-appearing skin (Fig. 54-1B). The noninflammatory form of bullous pemphigoid may be associated with a sparser inflammatory infiltrate histologically (see Diagnosis: Pathology).
Nonbullous lesions are the first manifestation of bullous pemphigoid in almost half of patients. 17 Often, urticarial type lesions precede the more classic tense bullae early in the course of disease. Other early nonbullous findings include eczematous, serpiginous, or targetoid erythema multiforme-like lesions (Fig. 54-1C). Atypical clinical presentations can also be seen in more established disease and include erythroderma and prurigo nodularis–like or vegetative lesions.18-24
Although bullous pemphigoid is typically widespread, localized forms of disease have been reported and are becoming increasingly recognized. 14 Localized disease often presents as tense bullae restricted to localized areas of involvement, most commonly the lower legs. 18,25 Dyshidrotic dermatitis–like lesions have also been reported localized to the hands and feet.26-30 Changes induced by radiation, trauma, or surgery (colostomy, urostomy, or skin graft donor site) may precipitate localized disease in these areas. 31-39 Childhood bullous pemphigoid, although rare, most often presents as localized disease with acral and vulvar or perivulvar distribution being common. 7-10,40-42 In these cases, the diagnosis is confirmed by routine histology, direct immunofluorescence (IF), and indirect IF or enzyme-linked immuinosorbent assay (ELISA) studies. Autoantibodies from these patients show typical IF localization and bind classic pemphigoid antigens, although resuls of indirect IF and ELISA may be negative because of low levels of circulating autoantibodies.
Mucous membrane lesions occur in approximately 10% of patients and are almost always limited to the oral mucosa. 15,44-46 Intact blisters are rare, with erosions more commonly seen. The lesions heal without scarring. The presence of scarring or mucosal predominant disease in the absence of classic cutaneous findings is more suggestive of mucous membrane pemphigoid as discussed in Chap. 55.
Bullous pemphigoid may also coexist with other cutaneous diseases, notably lichen planus. Lichen planus pemphigoides describes the coexistence of bullous pemphigoid and lichen planus with typical clinical, histologic, and immunopathologic features of both diseases. 47-50 Lichen planus pemphigoides more often presents in middle-aged patients (mean age of onset, 35 to 45 years of age) and is more localized to the extremities with a less severe clinical course compared with classic bullous pemphigoid.
The Bullous Pemphigoid Disease Area Index (or BPDAI) is a recently developed and validated tool that may be used to objectively measure of clinical disease activity in patients with bullous pemphigoid.51,52 Although originally designed for use in clinical trials, it can be useful in the clinical setting as well to objectively document the level of disease activity.
NONCUTANEOUS FINDINGS
Approximately 75% of patients have elevated total serum immunoglobulin (Ig) E levels, which often correlates with titers of bullous pemphigoid IgG autoantibodies by IF. 53-55 More than half of patients have peripheral blood eosinophilia as well.
COMPLICATIONS
Complications in untreated patients include skin infection developing within denuded bullae, dehydration, electrolyte imbalance, and possibly death from sepsis. Most often complications are related to treatment with systemic corticosteroids or other immunosuppressive medications. Complications caused by bacterial (pneumonia, urinary tract infection, soft tissue infection) and viral (disseminated or localized herpes) mediated infections are common, especially among patients with low functional status and dementia, and contribute to morbidity and mortality.
ETIOLOGY AND PATHOGENESIS
Most cases of bullous pemphigoid occur sporadically without any obvious precipitating factors. However, there are reports in which bullous pemphigoid appears to be triggered by ultraviolet (UV) light, either UVB or after psoralen and ultraviolet A light (PUVA) therapy, and radiation therapy. 62-64 Certain medications have also been associated with the development of bullous pemphigoid, including penicillamine, efalizumab, etanercept, and furosemide, among others.65-71 Recent reports suggest that bullous pemphigoid may also develop after immune checkpoint blockade with anti–programmed cell death receptor 1 (anti–PD-1) treatment.72-74
IMMUNOPATHOLOGY
Remarkable advances have been made characterizing the pathophysiology of bullous pemphigoid as a multistep process involving autoantibody binding to hemidesmosomal antigens, which triggers an inflammatory cascade that ultimately results in blister formation. The development of animal models has been instrumental in demonstrating the pathogenicity of bullous pemphigoid autoantibodies and dissecting the factors that contribute to blister formation.
BULLOUS PEMPHIGOID ANTIGENS
IF techniques demonstrate that patients with bullous pemphigoid exhibit circulating and tissue-bound autoantibodies directed against antigens of the cutaneous basement membrane zone (BMZ). 2 Immunoelectron microscopy studies localize bullous pemphigoid antigens to the hemidesmosome, an organelle that is important in anchoring the basal cell to the underlying basement membrane. 75 These autoantibodies bind to both the intracellular plaque of the hemidesmosome and the extracellular face of the hemidesmosome. Bullous pemphigoid autoantibodies recognize two distinct antigens with molecular weights of 230 kDa and 180 kDa by immunoblot analysis of human skin extracts.76
The 230-kDa molecule is termed BP230, BPAG1, or BPAG1e (indicating epidermal expression).4,76-78 BP230 belongs to the plakin family of proteins. 79,80 By immunoelectron microscopy, BP230 is located in the intracellular plaque of the hemidesmosome, where keratin intermediate filaments insert. 81 Analysis of BP230-deficient mouse strain generated by transgenic knockout technology further demonstrates that the function of BP230 is to anchor keratin intermediate filaments to the hemidesmosome. 82 Mice lacking BP230 show fragility of basal cells caused by collapse of the keratin filament network but no epidermaldermal adhesion defect. Interestingly, an alternatively spliced form of BP230 is expressed in neural tissue, termed BPAG1n. BPAG1n stabilizes the cytoskeleton of sensory neurons 83,84 just as BP230 stabilizes the cytoskeleton of epidermal cells. The lack of dermalepidermal separation in the BP230-null mice indicates that pathogenic autoantibodies in bullous pemphigoid do not act simply by inhibiting the function of BP230.
946
The 180-kDa bullous pemphigoid autoantigen is termed BP180, BPAG2, or type XVII collagen.3,85,86 BP180 is a transmembrane protein of the collagen family with an intracellular amino-terminal domain and an extracellular carboxy-terminal domain that spans the lamina lucida and projects into the lamina densa of the basement membrane. 87-91 Its intracellular or cytoplasmic domain is located in the plaque of the hemidesmosome, and its extracellular domain is linked to anchoring filaments. 92-94 The extracellular domain of BP180 contains a series of 15 collagen regions interrupted by 16 noncollagen sequences. 89 The noncollagen region 16A, also known as the NC16A domain, is adjacent to the membrane-spanning region and harbors the major autoantibody-reactive epitopes.95,96 ELISA to measure antibodies against the BP180 NC16A domain is both sensitive and specific for a diagnosis of bullous pemphigoid, 97-99 and its titers correlate with disease activity. 98,100 Further evidence that BP180 mediates dermal–epidermal adhesion comes from analysis of the gene defect in patients with the inherited junctional subepidermal blistering disease, non-Herlitz junctional epidermolysis bullosa (JEB-nH), previously known as generalized atrophic benign epidermolysis bullosa. These patients have recessively inherited mutations in the BP180 gene that result in a missing or dysfunctional protein.
PATHOPHYSIOLOGY OF SUBEPIDERMAL BLISTERING
The distinctive feature of bullous pemphigoid is the presence of circulating and tissue-bound autoantibodies against BP180 and BP230. Autoantibodies of various Ig isotypes and IgG subclasses are present in bullous pemphigoid sera with IgG being predominant followed by IgE. 104-106 Serum levels of anti-BP180NC16A IgG correlate well with disease activity in bullous pemphigoid patients.85,86,100
Inflammatory cells are present in the upper dermis and bullous cavity, including eosinophils (the predominant cell type), neutrophils, lymphocytes, and monocytes and macrophages. Both intact and degranulating eosinophils, neutrophils, and mast cells (MCs) are found in the dermis. 107-110 Local activation of these cells may occur via the multiple inflammatory mediators present in the lesional skin or blister fluid.54,111-118 Several proteinases are found in bullous pemphigoid blister fluid, including plasmin, collagenase, neutrophil elastase, and matrix metalloproteinase (MMP)-9,119-126 which may play a crucial role in subepidermal blister formation by their ability to degrade extracellular matrix (ECM) proteins.
Both in vitro and in vivo data demonstrate that autoantibodies, particularly those against BP180, are pathogenic suggesting that the autoantibodies trigger the entire inflammatory cascade that ultimately leads to tissue injury and blister formation. In vitro studies using normal human skin sections indicate that bullous pemphigoid IgG is capable of generating dermalepidermal separation in the presence of complement and leukocytes. 127,128 Early attempts to demonstrate the pathogenicity of patient autoantibodies by a passive transfer mouse model were unsuccessful because bullous pemphigoid anti-BP180-NC16A autoantibodies fail to cross-react with the murine BP180. 129 To overcome this difficulty, rabbit antibodies were raised against the epitope on mouse BP180. Passive transfer of these rabbit antibodies to neonatal mice induces blisters that show some of the key features of human bullous pemphigoid, including in situ deposition of rabbit IgG and mouse C3 at the BMZ, dermal–epidermal separation, and an inflammatory cell infiltrate. 129 These and other studies demonstrate that experimental blistering in
animals requires activation of the classical pathway of complement system, MC degranulation, and neutrophil infiltration. 130-134 A well-orchestrated proteolytic event occurs during the disease progression. Plasmin activates proenzyme MMP-9 and activated MMP-9 then degrades α1-proteinase inhibitor, the physiological inhibitor of neutrophil elastase. Unchecked neutrophil elastase degrades BP180 and other ECM components, resulting dermal–epidermal junction separation 135-138 (Fig. 54-2). To directly test the pathogenicity of anti-BP180 IgG autoantibodies from bullous pemphigoid patients, humanized BP180 mouse strains were generated, in which the human BP180 or NC16A domain replaces the murine BP180 or corresponding domain. 139,140 These humanized mice, upon injection with anti-BP180 IgG from bullous pemphigoid patients, develop subepidermal blisters. 139,140 Like the rabbit antimurine BP180 IgGinduced model, the humanized NC16A mouse model of bullous pemphigoid also requires complement, MCs, and neutrophils (Fig. 54-3).139
IgE anti-BP180 autoantibodies may also play a role in blister formation. Human skin grafts onto immunedeficient mice injected with an IgE hybridoma to the extracellular portion of BP180 or total IgE from bullous pemphigoid patients’ sera exhibit histologic dermalepidermal separation, 141,142 suggesting that anti-BP180 IgE antibodies may also participate in pathogenesis of bullous pemphigoid through activating MCs and recruitment of eosinophils.
Although most animal model studies clearly show that complement deposition and the subsequent inflammatory cascade triggered by BP180-specific antibodies is essential for blister formation, direct interference of hemidesmosome-mediated cell–cell matrix adhesion by anti-BP180 autoantibodies and BP180 depletion via the ubiquitin/proteasome pathway may represent complement-independent mechanisms of anti-BP180 autoantibody pathogenicity. 143,144 Interestingly, recent studies have shown that deletion of the dominant BP180 epitope domain results in the development of anti-BP180 antibodies, blistering and itching in mice, suggesting a more complex and expanding role for this region. 145 Involvement of anti-BP230 autoantibodies in bullous pemphigoid blistering is also implicated in some animal model studies146,147 ; however, direct evidence in humans is lacking.
Autoreactive T lymphocytes that recognize BP180 are present in addition to autoreactive B lymphocytes,148-151 supporting the concept that bullous pemphigoid is a T cell–dependent antibody-mediated skin autoimmune disease. As in most autoimmune diseases, the initial trigger for induction of autoreactive lymphocytes and autoantibody production in bullous pemphigoid remains unknown.
Several other subepidermal blistering diseases also show autoimmune responses to BP180. These include pemphigoid gestationis (or herpes gestationis), cicatricial pemphigoid (or mucous membrane pemphigoid), linear IgA bullous dermatosis, and lichen planus pemphigoid. 152-162 It is possible they may share some common immunopathological mechanisms with bullous pemphigoid.
Proposed mechanism of subepidermal blister formation in mouse model of bullous pemphigoid (BP).
Subepidermal blistering is an inflammatory process that involves the following steps: (1) anti-BP180 IgG binds to the pathogenic epitope of BP180 antigen on the surface of basal keratinocytes (BK). (2) The molecular interaction between BP180 antigen and anti-BP180 IgG activates the classical pathway of the complement system (C’) (3) C′ activation products C3a and C5a cause mast cells (MCs) to degranulate. (4) Tumor necrosis factor-α and other proinflammatory mediators released by MCs recruit neutrophils (PMNs). (5) Infiltrating PMNs bind to the BP180-anti-BP180 immune complex via the molecular interaction between Fcγ receptor III (FcγRIII) on neutrophils and the Fc domain of anti-BP180 IgG. (6) The interaction between Fc and FcγRIII activates PMNs to release neutrophil elastase (NE), gelatinase B (MMP-9), plasminogen activators (PAs), and reactive oxygen species (ROS). (7) Proteolytic enzymes and ROS work together to degrade BP180 and other extracellular matrix proteins, leading to subepidermal blistering.
RISK FACTORS
No specific environmental risk factors have been identified for bullous pemphigoid. In terms of genetic risk factors, certain human leukocyte antigens (HLA) alleles have been associated with bullous pemphigoid. HLA-DQB1 ∗ 0301 has been associated with classic bullous pemphigoid, as well as cicatricial pemphigoid in whites. 163,164 HLA-DRB1 ∗ 04, DRB1 ∗ 1101, and DQB1 ∗ 0302 alleles are associated with an increased risk of bullous pemphigoid among patients of Japanese descent.165
Neurologic disease is seen frequently in bullous pemphigoid patients. Recent studies have shown that patients with neurologic disease, including dementia, stroke, and Parkinson disease, have a significantly higher risk of developing bullous pemphigoid than those without neurological disease.166,167
Although there have been many case reports of bullous pemphigoid associated with malignancy, case-control studies have revealed conflicting data regarding the frequency of malignancy in bullous pemphigoid patients compared with age-matched control participants. 168-171 Most studies have shown no increased risk of malignancy in bullous pemphigoid patients. 172,173 However, recent evidence suggests that hematologic malignancies may be increased in patients with bullous pemphigoid. 174 Although a thorough review of systems and symptom guided workup is indicated in patients with a new diagnosis of bullous pemphigoid, extensive screening for an asymptomatic malignancy is not necessary.
Histopathology of bullous pemphigoid
Biopsy of an early small vesicle is diagnostic with histology revealing a subepidermal blister with a superficial dermal infiltrate consisting of eosinophils, neutrophils, lymphocytes, and monocytes and macrophages (Fig. 54-4). 15 The infiltrate ranges from intense (classic) to sparse (cell poor) but characteristically contains eosinophils and neutrophils, which may also be seen in the blister cavity. The blister roof is usually viable without evidence of necrosis. Histology of urticarial lesions may only show a superficial dermal infiltrate of lymphocytes, monocytes and macrophages, and eosinophils with papillary dermal edema or eosinophilic spongiosis.175
To directly test the pathogenicity of anti-BP180 IgG autoantibodies from bullous pemphigoid patients, humanized BP180 mouse strains were generated, in which the human BP180 or NC16A domain replaces the murine BP180 or corresponding domain. 139,140 These humanized mice, upon injection with anti-BP180 IgG from bullous pemphigoid patients, develop subepidermal blisters. 139,140 Like the rabbit antimurine BP180 IgGinduced model, the humanized NC16A mouse model of bullous pemphigoid also requires complement, MCs, and neutrophils (Fig. 54-3).
Figure 54-3 Humanized BP180NC16A mouse model of bullous pemphigoid (BP). A, Human BP180 (top panel) is a transmembrane protein of basal keratinocytes. It contains a single transmembrane domain. The extracellular region is consisted of 15 interrupted collagen domains (yellow bars) and 16 noncollagen domains (black lines). The NC16A domain (red line) harbors immunodominant epitopes recognized by BP autoantibodies. The extracellular region of mouse BP180 (middle panel) contains 13 collagen domains (blue bars) and 14 noncollagen domain (black lines). In humanized BP180NC16A mice, the mouse BP180NC14A domain was replaced by the human NC16A domain (lower panel). B, Neonatal NC16A mice injected intradermally (i.d.) with BP180NC16A-specific immunoglobulin (Ig) G autoantibodies developed clinical blistering (a). Direct immunofluorescence showed basement membrane zone (BMZ)deposition of human IgG (b) and murine C3 (c). Histologic sections of lesional skin showed dermal–epidermal separation (d). Examination of toluidine blue–stained skin sections revealed degranulating mast cells (MCs) (e). Hematoxylin and eosin staining showed infiltrating neutrophils (PMNs) in the upper dermis (400× magnification). (f).Arrows in b to d indicate basal keratinocytes. D, dermis; E, epidermis; V, vesicle.
Direct immunofluorescence of bullous pemphigoid perilesional tissue
Direct IF of perilesional skin shows linear IgG (usually IgG1 and IgG4, although all IgG subclasses and IgE have been reported) and C3 along the basement membrane 2,44,55,176,177 (Fig. 54-5). In approximately 70% of patients, there are circulating IgG autoantibodies that bind the BMZ on normal human skin or monkey esophagus by indirect IF. 44,55,57,105,176,178-180 The use of 1-M NaCl split skin, which separates the epidermis from the dermis at the lamina lucida, as the substrate for indirect IF testing is more sensitive for the detection of circulating anti-BMZ autoantibodies. 181,182 In addition to being more sensitive, the other advantage of the 1-M NaCl-split skin substrate is that it allows for the differentiation between bullous pemphigoid and epidermolysis bullosa acquisita (EBA) antibodies. Whereas bullous pemphigoid antibodies bind the epidermal side of the artificially induced blister (ie, the bottom of the basal cells), antibodies from EBA bind the dermal side of split skin (Fig. 54-6). In contrast to pemphigus,
indirect IF antibody titers do not usually correlate with disease extent or activity in bullous pemphigoid.183
ELISA testing has also proven to be useful in both clinical and research settings for the detection of circulating antigen-specific autoantibodies. Commercial kits are available for detection of both BP-180 (NC16A and total) and BP-230 IgG antibodies. A sensitivity of 89% and specificity of 98% when used with appropriate cutoff values are reported with these assays.97
As many as 75% of patients also have antigen specific IgE with anti-BP180 and anti-BP230 IgE antibodies detectable by IF and ELISA.104,106,179,184-187 Antigen-specific IgE antibodies may correlate with disease severity and could play a role in recruiting eosinophils to skin lesions.141,184,188
Approximately 7% of the normal population has anti-BP180 antibodies detectable by ELISA in the absence of clinical and histologic features of disease without age or gender predilection. The relevance of this positivity is not known because long-term followup data are not available. However, this finding underscores the importance of using ELISA in appropriate clinical settings and not as a screening tool in patients who lack other features of disease.189,190
In addition to being more sensitive, the other advantage of the 1-M NaCl-split skin substrate is that it allows for the differentiation between bullous pemphigoid and epidermolysis bullosa acquisita (EBA) antibodies. Whereas bullous pemphigoid antibodies bind the epidermal side of the artificially induced blister (ie, the bottom of the basal cells), antibodies from EBA bind the dermal side of split skin (Fig. 54-6).
Figure 54-6 Indirect immunofluorescence on normal skin previously incubated in 1-M NaCl to induce a split through the lamina lucida of the dermal epidermal junction. A, Immunoglobulin (Ig) G antibodies from bullous pemphigoid serum binds to the epidermal side (roof) of the artificial blister (BP180 and BP230 of hemidesmosomes). B, IgG antibodies from epidermolysis bullosa acquisita (EBA) serum binds to the dermal side (floor) of the split (collagen VII of anchoring fibers).
Comparison of Bullous Pemphigoid and Pemphigus Vulgaris
Differential Diagnosis of Bullous Pemphigoid
Treatments for Bullous Pemphigoid
Treatment depends on multiple factors, including extent of disease and patient comorbidities. 206 Patients with localized bullous pemphigoid often can be treated successfully with topical corticosteroids alone (Table 54-3).55,176,207
Patients with more extensive disease are usually treated with oral prednisone. 207-209 Despite the lack of randomized controlled trials, oral prednisone remains the mainstay of therapy. Studies also suggest that potent topical steroids, such as clobetasol proprionate cream 0.05% applied twice daily, are also effective in both moderate and severe bullous pemphigoid and may be safer than oral prednisone. 203 High-potency topical treatment does result in significant systemic absorption and therefore may act via local and systemic effects.210 Such topical therapy can be expensive and difficult to apply, which may prove prohibitive in many patients.
In older adult patients, the complications of systemic glucocorticoid therapy (eg, osteoporosis, diabetes, and immunosuppression) may be especially severe.211 Therefore, it is important to try to minimize the total dose and duration of therapy with oral glucocorticoids. Starting doses of prednisone of 0.75 to 1.0 mg/kg/day
or even less may be adequate for disease control. 212 In addition, immunosuppressive agents such as methotrexate, azathioprine, and mycophenolate mofetil are often used in conjunction with prednisone for their potential steroid-sparing effects, 207,209,213-222 although very few controlled trials have addressed this common approach to therapy. After the development of blisters has been arrested (consolidation phase), a careful tapering of the prednisone is recommended. 51 A weekly lowering of 5 mg to reach 30 mg is recommended, after which an alternating-day tapering regimen is typically initiated. Prednisone tapering must be done according to the clinical response and side effects of the patient. The majority of disease may be controlled with small amounts of prednisone and immunosuppressive drugs.
Sulfones may be effective in a minority of patients. Dapsone and sulfapyridine have been reported to control disease activity in 15% to 44% of patients with bullous pemphigoid. 209,223-225 Reports have described successful treatment of some bullous pemphigoid patients with tetracycline and nicotinamide or variations on this theme, such as erythromycin and nicotinamide or tetracycline alone. 226-228 Other effective therapies include plasmapheresis, 229 intravenous immunoglobulins, 230-232 omalizumab, 233,234 and rituximab.235,236
