I commend the authors of the Primer on pemphigus (Pemphigus. Nat. Rev. Dis. Primers 3, 17026 (2017))1 for their detailed review of this disease, which is an excellent and highly defined model to study human autoimmunity. In the Primer, the authors included paraneoplastic pemphigus (PNP) as a subtype of pemphigus; however, it should be noted that a different classification of PNP has been described, based on differences in the underlying pathophysiological mechanisms. The term PNP was coined after identifying shared serological immunopathological features with pemphigus2. Clinically, PNP has a wide spectrum of morphological presentations: it can mimic erythema multiforme, lichen planus pemphigoides, bullous pemphigoid, graft-versus-host disease or pemphigus3. In fact, a lichenoid presentation has more commonly been described than a pemphigus presentation4. Furthermore, the clinical course of PNP can be distinct from that of other pemphigus subtypes like pemphigus vulgaris or pemphigus foliaceus. For example, PNP does not always respond to treatments that usually resolve pemphigus, the spectrum of autoantibodies in patients with PNP is different from that observed in pemphigus vulgaris or pemphigus foliaceus with the presence of non-skin autoantibodies, and PNP by definition is associated with malignancy or Castleman disease. For example, in patients with Castleman disease and a PNP phenotype, removal of the tumours resulted in the resolution of PNP5. Thus, PNP would be more appropriately considered a single pemphigus-like mucocutaneous phenotype of paraneoplastic autoimmune multi-organ syndrome (PAMS)6. Here, I present some evidence supporting this classification.

Clinically, PNP has many distinctions from pemphigus vulgaris and pemphigus foliaceus, supporting the notion that PNP is an unrelated condition. The cutaneous manifestations of pemphigus subtypes frequently involve the scalp and usually do not present on palms and soles, whereas, as the authors note, in PNP there is “involvement of palms and soles but not the scalp.” Indeed, the involvement of the palms and soles is common in pemphigoid diseases7. When PNP affects the oral mucosa, the PNP is far more diffuse and severe than that observed in pemphigus vulgaris, resulting in erosive stomatitis8.

In marked contrast to pemphigus vulgaris and pemphigus foliaceus, in which mortality does not exceed 10%9, a fatal outcome is observed in 90% of patients with PNP10. The authors of the Primer state that this mortality could be due to underlying malignancy, infection or bronchiolitis obliterans (BO), which could result from fibrotic changes in the bronchioles induced by desmoglein 3-specific CD4+ T cells that target desmoglein 3-expressing cells in squamous metaplasia in the lung11. However, although metaplasia is found in the affected pulmonary epithelial lining of patients with PNP with pulmonary involvement, metaplastic human bronchial epithelia do not express desmoglein 39. Thus, fibrosis and BO are probably mediated by T cells and autoantibodies against epiplakin, which is expressed by human pulmonary epithelia6,12.

Similarly, fibrosis is not the hallmark of pemphigus vulgaris or pemphigus foliaceus. The presence of anti-desmoglein 3 autoantibody-induced pulmonary fibrosis in mice is not consistent with humans with pemphigus vulgaris, who do not develop pulmonary fibrosis. As desmoglein 3 is present at other sites, namely the oesophagus, pharynx, anal canal and vagina, it is unclear why fibrosis in patients with PNP only occurs in the lung. Whether lung-specific metaplasia precipitates BO remains unclear. Indeed, sloughing of bronchial epithelial cells can contribute to occlusion of the small airways, providing a potential mechanism for the respiratory failure in patients with PNP6. Finally, metaplasia is not typically observed in autoimmune conditions but is a feature of precancerous or cancer pathology, which may account for the over-representation of Castleman disease in patients with PNP and BO13.

How to cite this article

Amber, K.T. Paraneoplastic autoimmune multi-organ syndrome is a distinct entity from traditional pemphigus subtypes. Nat. Rev. Dis. Primers. 4, 18012 (2018).