LAUSR

Dear Editor Pemphigus vulgaris (PV) is an autoimmune blistering disease involving mainly the mucous membrane with or without cutaneous erosions. In recent years, there have been emerging discussions among the cutaneous-type PV (cPV), which was previously considered to be very rare. Patients with cPV have no mucosal involvement, while they may histologically present with suprabasal acantholysis as in typical PV and serologically possess auto-antibodies against both desmoglein 1 (Dsg1) and desmoglein 3 (Dsg3). We retrospectively reviewed patients with pemphigus treated in ourhospital since2008.Thediagnosisofpemphiguswasconfirmed by clinical presentations, characteristic pathological findings, presenceof IgGand/orC3 indirect immunofluorescence studyand evidence of circulating auto-antibodies using indirect immunofluorescence and/or enzyme-linked immunosorbent assay (ELISA) to detect the presence of auto-antibodies toDsg1 andDsg3 (MBLCo. Ltd,Nagoya,Japan)duringtheactivestageoftheirdiseases. Among our patients, 6 were classified as cPV (Table 1, Fig. 1). All patients showed positive results for both anti-Dsg1 and antiDsg3 auto-antibodies, with higher levels of anti-Dsg1 at the diagnosis of cPV. Four of them were newly diagnosed, with cPV as their initial presentation, while the other two had disease recurrence in the pattern of cPV. Among their cutaneous findings, some patients presented with skin lesions similar to pemphigus foliaceus instead of those in typical mucocutaneous PV (mcPV). We also found that our cPV patients all had prominent scalp involvements regardless of overall severity. One limitation of this study was lacking validation of the auto-antibody titers with a second ELISA kit from EUROIMMUN. However, given the high interrater agreement of the two kits as previously reported, we consider that re-testing by another ELISA kit will not change the conclusion of this report. Yoshida et al. first described four cPV patients with remarkable titers of both anti-Dsg1 and anti-Dsg3 IgG auto-antibodies. They proposed an extended desmoglein compensation theory assuming the weak anti-Dsg3 auto-antibodies are only potent enough to block the Dsg3 function in the skin but not the mucosa, where Dsg3 is enriched. This concept could be preliminarily supported by a study that the anti-Dsg3 auto-antibodies from cPV patients were unable to induce keratinocyte separation in vitro, indicating their lower pathogenicity. Findings from our patients are consistent with Yoshida et al.’s results and could be explained by their hypothesis. One of the aspects regarding the pathogenicity of auto-antibodies lies in their predominant immunoglobin (Ig) subclass. Previous studies have demonstrated higher levels of anti-desmoglein IgG4 during the active stage of pemphigus, followed by IgG1. Nonetheless, when classifying patients with their morphology, most cPV patients had significantly low titers of anti-Dsg3 IgG4, compared to those of mcPV and mucosal PV (mPV). In addition, M€ uller et al. demonstrated that sera from mPV and mcPV patients, but not cPV, reacted well with the NH2-terminus of Dsg3, which was considered to be associated with active PV. Such difference in IgG reactivity against the ectodomain of Dsg3 during the clinical course of each patient is known as epitope spreading. Moreover, one recent report identified desmogleinspecific memory B-cells in pemphigus with capabilities of ongoing affinity maturation, clonal selection and synergistic pathogenicity, supporting that the pathogenic potential of each