Pemphigoid Group

Humoral and cellular responses

Almost all patients with BP have circulating IgG autoantibodies that bind to BP180. More specifically, it is the non-collagenous NC16A domain, a region of BP180 located extracellularly but close to the transmembrane domain, that constitutes the immunodominant region (see Fig. 31.9 ) . However, additional antigenic sites exist within both the extracellular and intracellular domains of BP180, and they are recognized by up to 70% of BP sera . Patients with BP also exhibit significant autoreactivity to BP230 , and BP230-reactive autoantibodies bind predominantly to the C-terminal region of this autoantigen . The presence of several antigenic sites throughout BP180 and BP230 most likely results from the phenomenon known as “epitope spreading” (see below) . This phenomenon may also account for the finding that patients' sera rarely contain autoantibodies targeting additional components of the BMZ.

Patients with BP develop an autoreactive T-cell response to BP180 and BP230, and this is probably crucial for stimulating B cells to produce autoantibodies . The responsiveness of anti-BP180 autoreactive T cells is restricted by certain HLA class II alleles (e.g. HLA-DQB1*0301), which are prevalent in BP patients. These T lymphocytes, whose major relevant epitopes seem to be harbored within the NC16 domain, have a CD4 ⁺ phenotype and produce both Th1 (e.g. interferon-γ) and Th2 cytokines (e.g. interleukin [IL]-4, IL-5 and IL-13; see Ch. 4 ) . Th2, as well as more recently identified Th17, cytokines may be particularly relevant to the pathophysiology of BP ; they predominate within lesional tissue and in patients' sera. Also, the IgG4 subclass, the secretion of which is regulated by Th2 cytokines, is one of the major isotypes of anti-BP180 autoantibodies.

Upon binding of autoantibodies to their target antigens, subepidermal blister formation results from a cascade of events in which Fc receptor-mediated mechanisms, via BP180–anti-BP180 antibody complexes, are critical . The latter involve complement activation and recruitment of inflammatory cells, with liberation of their proteases including matrix metalloproteinase-9, neutrophil elastase, and mast cell proteases . These proteinases proteolytically degrade various extracellular matrix proteins as well as BP180. Infiltrating innate immune cells (macrophages, neutrophils, mast cells, and eosinophils) contribute to tissue damage via the release of proinflammatory cytokines, e.g. IL-4, -5, -8, -17, and eotaxin, which further enhance the inflammatory response . Autoantibodies to BP180 can also amplify the inflammatory response by directly stimulating keratinocytes to express inflammatory cytokines ( Fig. 30.1 ) . Finally, IgG autoantibodies reduce hemidesmosomal BP180 content and thereby contribute in a complement-independent manner to weakening of dermal–epidermal cohesion .

Various animal models have provided strong evidence that autoantibodies against BP180 are pathogenic. When passively transferred to neonatal mice (in which BP180 had been fully or partially humanized by genetic engineering), human autoantibodies against the NC16A domain were able to induce a blistering disorder that reproduced all the key features of BP . In a different murine model, anti-BP230 antibodies induced an inflammatory reaction and subepidermal blistering . Nonetheless, the currently accepted pathomechanism is that antibodies against the ectodomain of BP180 are critical for disease initiation, while the development of antibodies against BP230 represents a secondary event that contributes to tissue damage .

Non-bullous pemphigoid

The cutaneous manifestations of BP can be extremely polymorphic ( Figs 30.2–30.6 ) . In the prodromal, non-bullous phase of the disease, signs and symptoms are frequently nonspecific, with mild to severe intractable pruritus alone or in association with excoriated, eczematous, papular and/or urticarial lesions that may persist for several weeks or months (see Figs 30.3–30.5 ). Importantly, these nonspecific skin findings may remain as the only signs of the disease. More recent studies have noted that at least 20% of patients have neither obvious blisters nor erosions due to ruptured bullae at the time of diagnosis .

Bullous phase

The bullous stage of BP is characterized by the development of vesicles and bullae on apparently normal or erythematous skin, together with urticarial and infiltrated papules and plaques that occasionally assume an annular or figurate pattern. The blisters are tense, up to 1–4 cm in diameter, contain a clear fluid, and may persist for several days, leaving eroded and crusted areas (see Fig. 30.2 ). Occasionally, the blister fluid becomes blood-tinged. The lesions frequently have a symmetrical distribution pattern, and they predominate on the flexural aspects of the extremities and the lower trunk, including the abdomen. Within intertriginous zones, vegetating plaques may be observed (pemphigoid vegetans; see Fig. 30.6C ). The extent and degree of disease activity can be assessed by employing the BP disease area index (BPDAI) or a daily blister count.

Residual postinflammatory changes include hyper- and hypopigmentation and, only occasionally, milia. Involvement of the oral cavity is observed in 10–30% of patients. The mucosae of the eyes, nose, pharynx, esophagus, and anogenital region are more rarely affected. In ~50% of patients, a peripheral blood eosinophilia is noted.

Clinical variants

Several clinical variants of BP have been described and are outlined in Table 30.2 . Pemphigoid gestationis (or gestational pemphigoid ) is also a variant of BP, which typically occurs during pregnancy (see Ch. 27 ).

While the individual lesions of BP in infants and in older children ( infantile and childhood BP ) are similar to those observed in older adults, the sites of involvement can differ ( Fig. 30.7 ). In infants, bullae often first appear acrally and then can generalize to other sites, including the face. Involvement of the genital region, e.g. vulvar childhood pemphigoid , as well as other mucosal sites has been observed in children .

Drug-induced bullous pemphigoid

In certain patients, systemic medications can lead to the development of BP . The implicated drugs are numerous, including diuretics (e.g. furosemide, spironolactone), NSAIDs (e.g. ibuprofen, topical diclofenac), antibiotics (e.g. amoxicillin, ciprofloxacin), ACE inhibitors, TNF inhibitors, potassium iodide, vaccines, and more recently, dipeptidyl peptidase-4 inhibitors (e.g. vildagliptin) and checkpoint inhibitors (e.g. pembrolizumab). Reproduction of BP lesions following drug rechallenge has been observed with some medications (e.g. furosemide), but for others, the association is based on less evidence. Case–control studies assessing the drugs used on a long-term basis prior to onset of the disease found that two classes of drugs, diuretics and neuroleptics, were used more frequently by BP patients than by control subjects. With regard to diuretics, the risk was variably linked to aldosterone antagonists and loop diuretics . Hence, in all patients, a careful drug history is mandatory to exclude the triggering effect of a drug, since its prompt discontinuation may result in improvement.

It is likely that the drugs act as triggers in patients with an underlying genetic susceptibility by either modifying the immune response or altering the antigenic properties of the epidermal basement membrane.

Light microscopy

In the non-bullous phase or in atypical variants of BP, routine histology may provide less specific information, since only eosinophilic spongiosis and/or dermal infiltrates of eosinophils may be seen ( Fig. 30.9 ). In biopsy specimens of an early bulla, a subepidermal blister accompanied by a dermal inflammatory infiltrate composed of eosinophils and mononuclear cells is typically observed. The infiltrate favors the uppermost dermis, and the cavity of the bulla contains a net of fibrin with a variable inflammatory infiltrate ( Fig. 30.10 ). Electron microscopy studies have shown that subepidermal blister formation occurs at the level of the lamina lucida.

Direct immunofluorescence (DIF) microscopy

In almost all patients, DIF microscopy of perilesional, uninvolved skin will characteristically demonstrate the presence of fine, linear, continuous deposits of IgG and/or C3 (and, more rarely, other Ig classes) along the epidermal basement membrane (see Ch. 27 ). IgG4 and IgG1 are the predominant IgG subclasses. Two additional studies are helpful in distinguishing BP from other autoimmune blistering disorders: (1) close analysis of the linear fluorescence pattern at the BMZ in order to determine if it is n-serrated (BP and linear IgA bullous dermatosis) versus u-serrated (epidermolysis bullosa acquisita) ; and (2) the salt-split skin assay in which perilesional skin is examined after treatment with 1 M NaCl (see Fig. 28.7 ). In BP, immune deposits are found in the epidermal side (roof) or in both the epidermal and dermal sides of the split. Although not routinely available, the computer-aided fluorescence overlay antigen mapping (FOAM) technique allows one to determine more precisely the localization of deposited immunoreactants .

Indirect immunofluorescence (IIF) microscopy

For IIF studies in subepidermal blistering diseases, salt-split normal human skin is the substrate of choice, rather than intact normal human skin or monkey esophagus (see Table 29.5 ). Circulating anti-basement membrane autoantibodies of the IgG class, and, less frequently, of the IgA and IgE classes, are detectable in 60–80% of patients . These autoantibodies typically bind to the epidermal side or, less frequently, to both the epidermal and dermal sides of saline-separated normal human skin ( Fig. 30.11 ) . A BIOCHIP IIF technique has also been introduced in which the substrates are recombinant tetrameric BP180-NC16A and transfected cells expressing BP230 .

ELISA

An increasing number of autoantibodies associated with autoimmune bullous diseases can be detected via ELISA (see Table 29.6 ). Target antigens include the NC16A domain of BP180 and the C-terminus (+/− the N-terminus) of BP230 . In the case of BP, these tests have been found to be fairly specific (≥90%); occasionally, low-titer, false-positive results are observed in healthy subjects and elderly patients with pruritic cutaneous eruptions . When performed in unselected BP patients, the overall sensitivity of the BP180-NC16A ELISA is comparable to that of IIF (with salt-split skin as a substrate) . Combining the ELISA for BP230 with the BP180-NC16A ELISA increases the overall sensitivity by ~10% so the former is only recommended in the setting of a negative BP180 ELISA . In contrast to immunoblotting, ELISA antigens are tested under native conditions, and, as a result, binding activity against conformational antigens is not lost.