Paraneoplastic pemphigus is a rare and severe autoimmune blistering disease characterized by a recalcitrant and severe mucositis, and polymorphic cutaneous lesions, associated with benign and malignant neoplasms. Paraneoplastic pemphigus is caused by production of autoantibodies against various epidermal proteins involved in cell adhesion. Bronchiolitis obliterans (BO) is one of the leading causes of mortality in these patients. Recent advances have associated the presence of anti-epiplakin antibodies with the development of BO in adult patients. Here we describe the first pediatric patient in whom the association of anti-epiplakin antibodies and BO have been reported so far.

Paraneoplastic pemphigus (PNP), also named paraneoplastic autoimmune multiorgan syndrome (PAMS), occurs in the presence of an underlying malignancy, most commonly Castleman Disease (CD) in children. Intractable mucocutaneous lesions may be the first sign of PNP. Early diagnosis, detection, and surgical resection of the underlying tumor improve mucosal and cutaneous lesions, but mortality rates are high related to the pulmonary involvement leading to bronchiolitis obliterans (BO).1 

A potential autoantibody-clinical phenotype correlation has been described in adults with PNP and PAMS, where desmoglein (Dsg) 3 and epiplakin (EPPK) antibodies are associated with the development of BO.2,3  We report the first case of BO associated with positive anti-EPPK antibodies in a pediatric patient with PNP and PAMS related to CD.

A 16-year-old boy with a history of recurrent painful oral ulcers, cutaneous blisters, and significant weight loss was referred to our dermatology department with a clinicopathological diagnosis of pemphigus vulgaris under treatment with prednisone (1 mg/kg) for the past 7 months. Despite previous treatment, physical examination revealed severe oral mucositis, disseminated erosions, and flaccid blisters on the face, trunk, and extremities, some covered by honey-colored hematic crusts (Fig 1 A and B). He also presented nail abnormalities, such as ungual dystrophy and onychomadesis (Fig 1 C and D). No other pathologic findings were found on physical examination. Routine hematology and biochemistry tests were within normal limits. On reappraisal PNP was suspected due to recalcitrant painful mucosal erosions, polymorphous cutaneous lesions, and suprabasal acantholysis.

FIGURE 1

Clinical pictures. A and B, Oral mucositis, disseminated erosions and flaccid blisters on the face, trunk, and extremities, some covered by honey-colored hematic crusts. C, Ungual dystrophy. D, Onychomadesis. E and F, Healed mucocutaneous lesions after surgical removal of the tumor.

FIGURE 1

Clinical pictures. A and B, Oral mucositis, disseminated erosions and flaccid blisters on the face, trunk, and extremities, some covered by honey-colored hematic crusts. C, Ungual dystrophy. D, Onychomadesis. E and F, Healed mucocutaneous lesions after surgical removal of the tumor.

Close modal

The chest x-rays showed mediastinal widening, and on a thoracoabdominal CT scan, 3 enhancing mediastinal nodal masses were observed; no other lymph node enlargement was noted. The diagnosis of hyaline-vascular CD was established after a tumor biopsy. Indirect immunofluorescence (IIF) was positive in the intercellular spaces, titer 1:10240 and 1:2560 in cow nose and 1:320 in rat bladder (Fig 2 A and D). In addition, the patient’s samples were positive in 1:439 and 1:532 in a sandwich ELISA (enzyme-linked immunosorbent assay) containing EPPK1 protein. The immunoblot assay with EPPK1 protein yielded circulating IgG autoantibodies against a ∼500 kDa band that corresponded to EPPK (Fig 3A). A Western blot assay using rat bladder extract demonstrated autoantibodies against envoplakin and desmoplakin II (210 kDa), periplakin (190 kDa), α-2-macroglobulin-like protein 1 (A2ML1, 170 kDa), Dsg1 (160 kDa), and other PAMS antigens (56 kDa) (Fig 3B).

FIGURE 2

Antiepithelial antibodies by indirect immunofluorescence. A and B, Nose cow. C and D, Murine bladder epithelia. Counterstain with propidium iodide.

FIGURE 2

Antiepithelial antibodies by indirect immunofluorescence. A and B, Nose cow. C and D, Murine bladder epithelia. Counterstain with propidium iodide.

Close modal
FIGURE 3

Immunoblotting assays. A, EPPK1 protein. Patient serum recognizes a ∼500 kDa band (EPPK). B, Murine bladder extract showing antibodies against envoplakin and desmoplakin II (210 kDa), periplakin (190 kDa), α-2-macroglobulin-like protein 1 (A2ML1, 170 kDa), Dsg1 (160 kDa). NHS = normal human serum (negative control); MWS = molecular weight standard.

FIGURE 3

Immunoblotting assays. A, EPPK1 protein. Patient serum recognizes a ∼500 kDa band (EPPK). B, Murine bladder extract showing antibodies against envoplakin and desmoplakin II (210 kDa), periplakin (190 kDa), α-2-macroglobulin-like protein 1 (A2ML1, 170 kDa), Dsg1 (160 kDa). NHS = normal human serum (negative control); MWS = molecular weight standard.

Close modal

The patient met both Anhalt et al4  (clinical, histopathology, IIF, immunoprecipitation) and Svoboda et al5  (3 major and 1 minor criteria) diagnostic criteria for PNP (Tables 1 and 2). Because of the difficult surgical approach for CD, a trial with rituximab was attempted to reduce tumor size, achieving a 20% tumoral decrease. Further surgical removal of the residual tumor resulted in complete resolution of the skin and mucosal lesions (Fig 1 E and F).

TABLE 1

Diagnostic Criteria for Paraneoplastic Pemphigus by Anhalt et al4 

Clinical Painful mucosal erosions with or without accompanying polymorphous cutaneous lesions in the setting of underlying malignancy. 
Histopathology Suprabasal (intraepidermal) acantholysis, interface dermatitis, and keratinocyte necrosis. 
Direct immunofluorescence Intercellular and basement membrane zone deposition of IgG and complement (C3). 
Indirect immunofluorescence Detection of autoantibodies targeting intracellular proteins found in transitional or stratified squamous epithelia. 
Immunoprecipitation Precipitation of characteristic proteins: desmoplakin 1 (250 kD), bullous pemphigoid antigen (230 kD), desmoplakin 2 (210 kD), periplakin (190 kD), α-2-macroglobulin-like 1 antigen (170 kD). 
Clinical Painful mucosal erosions with or without accompanying polymorphous cutaneous lesions in the setting of underlying malignancy. 
Histopathology Suprabasal (intraepidermal) acantholysis, interface dermatitis, and keratinocyte necrosis. 
Direct immunofluorescence Intercellular and basement membrane zone deposition of IgG and complement (C3). 
Indirect immunofluorescence Detection of autoantibodies targeting intracellular proteins found in transitional or stratified squamous epithelia. 
Immunoprecipitation Precipitation of characteristic proteins: desmoplakin 1 (250 kD), bullous pemphigoid antigen (230 kD), desmoplakin 2 (210 kD), periplakin (190 kD), α-2-macroglobulin-like 1 antigen (170 kD). 
TABLE 2

Revised Diagnostic Criteria for Paraneoplastic Pemphigus by Svoboda et al5 

Major criteria 
 Mucous membrane lesions with or without cutaneous involvement 
 Concomitant internal neoplasm 
 Evidence of anti-plakin antibodies,a including but not limited to IP, IB, ELISA, IIF on transitional epithelium 
Minor criteria 
 Acantholysis and/or lichenoid interface dermatitis observed on histopathology, +/− necrotic keratinocytes 
 DIF displaying intercellular and/or basement membrane staining 
Diagnosis of paraneoplastic pemphigus requires 3 major criteria, or 2 major and both minor criteria. 
Major criteria 
 Mucous membrane lesions with or without cutaneous involvement 
 Concomitant internal neoplasm 
 Evidence of anti-plakin antibodies,a including but not limited to IP, IB, ELISA, IIF on transitional epithelium 
Minor criteria 
 Acantholysis and/or lichenoid interface dermatitis observed on histopathology, +/− necrotic keratinocytes 
 DIF displaying intercellular and/or basement membrane staining 
Diagnosis of paraneoplastic pemphigus requires 3 major criteria, or 2 major and both minor criteria. 

DIF, direct immunofluorescence; ELISA, enzyme-linked immunosorbent assay; IP, immunoprecipitation; IB, immunoblot; IIF, indirect immunofluorescence.

a

Includes antibodies to α-2-macroglobulin-like-1.

However, during the course of the disease the patient developed respiratory symptoms (dyspnea, cough, and desaturation), hyperinflation in chest radiographs, a pattern of mosaic attenuation in high-resolution CT scanning, and a severe obstructive ventilatory pattern in spirometry, all compatible with the diagnosis of BO. He received treatment with inhaled fluticasone (250 μg/day), prednisone (0.1–0.5 mg/kg perday) and mycophenolate mofetil (1–2 g/day) for 2 years, with steady progression of symptoms and worsening of pulmonary function tests (reduction of the forced expiratory volume in the first second [FEV1] from 20% to 12% and forced vital capacity [FVC] from 50% to 35% without significant response to bronchodilator). At the age of 18 he was transferred to an adult health care facility with the plan of a lung transplantation in the near future.

The term PNP was originally proposed 30 years ago by Anhalt et al.4  It is a rare autoimmune blistering disease of the skin and mucous membranes where a neoplasm, benign or malignant, elicits the production of autoantibodies against various epidermal proteins involved in cell adhesion, which leads to the blistering. The pathogenesis is not yet completely defined.6,7 

The incidence of PNP is unknown.6  There are ∼500 cases described in the literature.7  It mostly affects adults between 45 and 70 years of age but may also be found in younger patients.7,8  Adolescents are the most affected group among pediatric patients.1 

Unlike adults, PNP in pediatric patients is most commonly caused by CD rather than by non-Hodgkin’s lymphoma. CD consists of a heterogeneous group of benign lymphoproliferative disorders.6  It may be localized to a single lymph node region (unicentric [UCD]), or occur systemically (multicentric [MCD]).9  PNP is more frequent in the context of UCD, as in our patient.10 

The pathogenic IgG autoantibodies in PNP are polyclonal, directed against multiple antigens: Dsg3 and/or Dsg1; Desmocollin (Dsc)1, Dsc2, Dsc3; proteins of the plakin family (envoplakin, periplakin, desmoplakin I, demosplakin II, EPPK, plectin, and BP230), in addition to a protease inhibitor, A2ML1.6 

Patients with PNP usually present with a consumptive syndrome (malaise, weakness, weight loss) due to underlying disease or decreased food intake associated to an extensive and recalcitrant stomatitis, which is the most frequent mucocutaneous problem and the least-responsive symptom to treatment.3,6  In fact, both the stomatitis and the consumptive syndrome raised the suspicion of PNP in our patient. Oral lesions consist of chronic painful erosions and crusting, and they may occur throughout the entire buccal mucosa.6  The genital (62%) and ocular (59%) mucosae may also be involved.6  A wide range of skin manifestations may present, including blisters, erosions, macules, papules, and plaques. The distinct clinical presentations may coexist or evolve during the course of the disease.8 

One of the most frustrating aspects of PNP is that even with a successful removal of the neoplasm with remission of the mucocutaneous lesions, it can lead to respiratory compromise such as BO, resulting in high morbidity and mortality rates.3  Pulmonary disease, when present, is irreversible despite aggressive therapy.8  BO is a form of obstructive lung disease that follows inflammatory and fibrotic changes primarily affecting the small airways and may occur in 30% of CD-associated PNP patients. Progressive dyspnea and hypoxemia are the most frequent and warning signs of BO and should prompt an immediate examination by a pulmonologist.11 

The pathogenesis of PNP-related BO is currently unknown. Nousari et al detected autoantibodies that reacted to plakins in respiratory epithelium and the concomitant presence of acantholysis of the bronchial epithelium in 2 patients, suggesting that autoantibodies may have a role in the mechanism of pulmonary injury, but causation has yet to be proved.12  It also appears that cellular responses mediated by CD8+ lymphocytes may play a key role in the progression of bronchiolitis.13 

More recent evidence in adult populations highlights potential autoantibody-clinical phenotype correlations, with EPPK autoantibodies associated with the development of BO. Tsuchisaka et al found that the presence of anti-EPPK antibodies significantly correlated with the development of BO and increased mortality in adult Japanese patients with PNP, but this correlation was not observed in the European population, possibly caused by a different genetic background, including human leukocyte antigens.3  Anti-EPPK antibodies injected into mice induce a disruption of bronchial epithelium integrity and mononuclear cell inflammation, providing evidence of its pathogenic role in BO. Antibodies in PNP sera may react with EPPK in lung tissues and cause BO. Our patient is the first pediatric patient in whom the association of anti-EPPK antibodies and BO has been reported to date.1,3 

The mainstay of therapy for paraneoplastic manifestations of CD is definitive surgical treatment of the neoplasm.14  However, several nonsurgical treatments have proven to be effective in achieving less morbid surgical resections or reducing symptoms in patients with inoperable neoplasms. In this case, rituximab was used to reduce the tumor size and surgical risks, but also because of its proven efficacy for refractory autoimmune bullous diseases including PNP.15  Additionally, Kaibuchi-Ando et al very recently reported the successful use of combined rituximab, intravenous immunoglobulin, and betamethasone to decelerate and stabilize respiratory function in patients with BO and PNP; however, this has not been achieved in all cases.13,16 

The overall prognosis of PNP is poor, with a mortality of 75% to 90%, and respiratory failure the main cause of death.8  The detection of anti-EPPK antibodies may lead to more aggressive surveillance and treatment to detect and try to avoid the development of fatal PNP-related BO. As far as we know, this is the first pediatric PNP case where the relationship between anti-EPPK antibodies and BO has been searched for. More studies are needed to evaluate this phenomenon in children.

Drs Rivas-Calderon and Sáez-de-Ocariz conceptualized and designed the study, drafted the initial manuscript, reviewed and revised the manuscript, designed the data collection instruments, and collected data; Drs Yamazaki-Nakashimada, Orozco-Covarrubias, and Durán-McKinster participated in the analysis and interpretation of data and revised the article critically for intellectual content; Drs Avalos-Díaz and Pacheco-Tovar designed the data collection instruments, conducted the laboratory analyses, acquired, analyzed, and interpreted the data, and critically reviewed the manuscript for important intellectual content; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

FUNDING: No external funding.

BO

Bronchiolitis obliterans

CD

Castleman Disease

Dsc

Desmocollin

Dsg

Desmoglein

EPPK

Epiplakin

MCD

Multicentric Castleman Disease

PAMS

Paraneoplastic autoimmune multiorgan syndrome

PNP

Paraneoplastic pemphigus

UCD

Unicentric Castleman Disease

1
Schultz
B
,
Hook
K
.
Bullous diseases in children: a review of clinical features and treatment options
.
Paediatr Drugs
.
2019
;
21
(
5
):
345
356
.
2
Ohzono
A
,
Sogame
R
,
Li
X
, et al
.
Clinical and immunological findings in 104 cases of paraneoplastic pemphigus
.
Br J Dermatol
.
2015
;
173
(
6
):
1447
1452
.
3
Tsuchisaka
A
,
Numata
S
,
Teye
K
, et al
.
Epiplakin is a paraneoplastic pemphigus autoantigen and related to bronchiolitis obliterans in japanese patients
.
J Invest Dermatol
.
2016
;
136
(
2
):
399
408
4
Anhalt
GJ
,
Kim
SC
,
Stanley
JR
, et al
.
Paraneoplastic pemphigus. An autoimmune mucocutaneous disease associated with neoplasia
.
N Engl J Med
.
1990
;
323
(
25
):
1729
1735
.
5
Svoboda
SA
,
Huang
S
,
Liu
X
,
Hsu
S
,
Motaparthi
K
.
Paraneoplastic pemphigus: Revised diagnostic criteria based on literature analysis
.
J Cutan Pathol
.
2021
;
48
(
9
):
1133
1138
.
6
Han
SP
,
Fu
LS
,
Chen
LJ
.
Masked pemphigus among pediatric patients with Castleman’s disease
.
Int J Rheum Dis
.
2019
;
22
(
1
):
121
131
.
7
Maruta
CW
,
Miyamoto
D
,
Aoki
V
,
Carvalho
RGR
,
Cunha
BM
,
Santi
CG
.
Paraneoplastic pemphigus: a clinical, laboratorial, and therapeutic overview
.
An Bras Dermatol
.
2019
;
94
(
4
):
388
-
398
.
8
Tirado-Sánchez
A
,
Bonifaz
A
.
Paraneoplastic pemphigus. A life-threatening autoimmune blistering disease
.
Actas Dermosifiliogr
.
2017
;
108
(
10
):
902
910
.
doi:10.1016/j.ad.2017.04.024
9
Kumar
S
,
Bishnoi
K
,
Murugavaithianathan
P
,
Panwar
VK
.
Castleman’s disease presenting as localized abdominal mass and paraneoplastic pemphigus
.
J Clin Imaging Sci
.
2016
;
6
:
6
.
10
Szalat
R
,
Munshi
NC
.
Diagnosis of Castleman disease
.
Hematol Oncol Clin North Am
.
2018
;
32
(
1
):
53
64
.
11
Gu
L
,
Ye
S
.
Tocilizumab cannot prevent the development of bronchiolitis obliterans in patients with Castleman disease-associated paraneoplastic pemphigus
.
J Clin Rheumatol
.
2019
;
25
(
5
):
e77
e78
.
12
Nousari
HC
,
Deterding
R
,
Wojtczack
H
, et al
.
The mechanism of respiratory failure in paraneoplastic pemphigus
.
N Engl J Med
.
1999
;
340
(
18
):
1406
1410
.
13
Hoffman
MA
,
Qiao
X
,
Anhalt
GJ
.
CD8+ T lymphocytes in bronchiolitis obliterans, paraneoplastic pemphigus, and solitary Castleman’s disease
.
N Engl J Med
.
2003
;
349
(
4
):
407
408
.
14
Raza
HA
,
Nokes
BT
,
Rosenthal
AC
, et al
.
Unicentric castleman disease complicated by paraneoplastic bronchiolitis obliterans and pemphigus
.
Respir Med Case Rep
.
2018
;
25
:
129
132
15
Bandera
B
,
Ainsworth
C
,
Shikle
J
,
Rupard
E
,
Roach
M
.
Treatment of unicentric Castleman disease with neoadjuvant rituximab
.
Chest
.
2010
;
138
(
5
):
1239
1241
16
Kaibuchi-Ando
K
,
Sugiura
K
,
Muro
Y
, et al
.
Successful treatment with i.v. immunoglobulin and rituximab for bronchiolitis obliterans associated with paraneoplastic pemphigus
.
J Dermatol
.
2020
;
47
(
10
):
e368
e370
.

Competing Interests

CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no potential conflict of interest to disclose.