To date, estimates of the prevalence of anti-Lam332 autoantibodies among MMP patients were made mainly on the basis of indirect IF on SSS, immunoprecipitation or immunoblot data from small series of patients with MMP diagnosed using direct IF and clinical findings.28,34,36- 40 In agreement with our study, anti-Lam-332 MMP rates ranged from 18% to 30% among patients with the MMP clinical phenotype; however, these studies were performed on limited series of patients. For instance, in a French retrospective study37 on 23 cicatricial pemphigoid serum samples, 5 samples (22%) showed dermal staining and another 5 samples (22%) showed combined epidermal/dermal staining by indirect IF on SSS. In a German study28 on 16 cicatricial pemphigoid serum samples, 5 samples (31%) showed dermal staining and 2 samples (13%) showed combined staining by indirect IF on SSS. In the latter study, all 5 cicatricial pemphigoid serum samples with dermal binding immunoprecipitated Lam332 from extracts and culture media. Conversely, none of 13 MMP serum samples and only 3 of 124 MMP samples showed autoantibodies to Lam332 by immunoblotting in the series of Yeh et al38 and Oyama et al,34 respectively. Here, we found that almost half of MMP serum with positive Lam332 ELISA scores had negative results of indirect IF on the dermal side of split skin (Table 4), which might be partly related to a lack of sensitivity of our indirect IF on SSS technique, which was routinely performed on a commercially available substrate. Conversely, binding of the epidermal side of the split skin in some MMP serum with positive Lam332 ELISA scores may be explained both by the low sensitivity of our indirect IF on SSS technique and by the rather frequent detection of anti-BP180 and anti-BP230 autoantibodies (Table 4). Mucous membrane pemphigoid is a clinical phenotype that encompasses several subsets, with a wide spectrum of clinical presentation, clinical course, and prognosis. In our study, determination of patients' autoantibody profiles using commercial ELISAs allowed us to confirm that BP180 is a major autoantigen in MMP, as previously demonstrated by many studies6- 9,16,27,34,40- 42 performed in the past 2 decades and based on immunoblot analysis using human epidermal, amniotic membrane, or fusion proteins. These studies showed that circulating IgG autoantibodies from patients with MMP can recognize several intracellular and extracellular BP180 epitopes, sometimes distinct from the NC16A domain.6,34,41,42 Using a panel of cell-derived and recombinant proteins covering the entire BP180 molecule, Schmidt et al41 found that 19 of 26 MMP serum samples recognized BP180, including 6 (32%) that showed only IgA reactivity to this autoantigen. By immunoblotting using human epidermal, dermal, and placental amnion proteins of a large series of MMP, Oyama et al34 similarly showed that most patients with MMP (75%) had IgG to BP180, including its soluble ectodomains. In our study, autoantibodies against the NC16A domain of BP180 were detected by ELISA in almost 40% of MMP serum samples. Other antigens targeted predominantly by IgG autoantibodies in MMP included BP230.22,27,34,43 However, autoantibodies against BP230 are detected only occasionally in MMP serum, as shown in the present study. Interestingly, serum autoantibodies directed against BP230 are more frequently detected in the serum of patients with MMP when anti-Lam332 autoantibodies are also present. This association is possibly a secondary immune phenomenon related to epitope spreading. At some stage of the pathologic progression of MMP, damage to basal epithelial cells and BMZ might result in abnormal exposure or release of BP230, which is intracellularly situated.44 The same phenomenon of epitope spreading45 possibly exists for the NC16A domain of BP180, which is highly immunogenic.16,41 Hence, the immune system of patients with MMP may produce autoantibodies against BP180 and BP230, although autoantibodies to BP180-NC16A or to BP230 did not correlate with MMP severity in our study (data not shown). This is in keeping with the fact that their specific role in the pathogenesis of the disease still is undetermined.38 Finally, it remains unclear whether the development of multiple target antigens is of pathogenic importance or a secondary phenomenon occurring within the setting of more severe or long-standing MMP.