| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Abstract
Paraneoplastic pemphigus (PNP) is an autoimmune blistering skin disease of humans that consists of characteristic skin lesions associated with concurrent neoplasia. In this study we provide histologic and serologic evidence to support a diagnosis of PNP in a dog with splenic sarcoma. Skin lesions consisted of widespread erosions involving haired skin, mucocutaneous junctions, and oral mucosa. Microscopic examination of skin and mucosae revealed lesions consistent with both pemphigus vulgaris and erythema multiforme. Immunoprecipitation confirmed that circulating IgG autoantibodies from this patient recognized five distinct antigens, presumed to represent epidermal plakins. Clinical, histopathologic, and immunologic findings in this patient were similar to those observed in human patients with PNP. The splenic neoplasia in this dog was diagnosed as a phenotypically variable spindle cell sarcoma. To date, only one other dog has been reported with PNP. This is the second reported case of canine PNP and the first patient in whom skin lesions were identified in association with splenic neoplasia.
Key words: Autoantibodies; autoimmunity; blistering skin disease; canine; erythema multiforme; immunoblotting; immunofluorescence; immunoprecipitation; neoplasia; pemphigus; skin; spleen.
Reported in humans2 and one dog,6 paraneoplastic pemphigus (PNP) is a rare and severe autoimmune mucosal and cutaneous blistering disease associated with hematopoietic or solid neoplasia. In both species, PNP exhibits histologic characteristics of both pemphigus vulgaris (PV) (suprabasal clefting with acantholysis) and erythema multiforme (EM) (cell-rich interface dermatitis with lymphocyte-mediated keratinocyte apoptosis).5,6
In human patients, PNP has been associated with lymphoma, chronic lymphocytic leukemia, undifferentiated or spindle cell sarcoma, squamous cell sarcoma of the lung, or thymoma.8 In the only reported case of canine PNP, this syndrome was attributed to the presence of a thymic lymphoma.6 In human individuals with PNP, circulating autoantibodies identify several protein antigens of the plakin family that include plectin (> 500 kd), desmoplakin-I (250 kd), bullous pemphigoid antigen I (230 kd), desmoplakin-II (210 kd), envoplakin (210 kd), periplakin (190 kd), and an unknown 170-kd antigen.4 Pathogenic autoantibodies directed against desmosomal cadherins, such as desmoglein-3 (130 kd; PV antigen)1 and desmoglein-1 (160 kd; pemphigus foliaceus antigen), also have been identified.7 In the dog previously reported with PNP, autoantibodies were shown to target desmoplakin-I and -II, envoplakin, and periplakin.3
A 7-year-old male castrated Golden Retriever was presented with a 1-month duration of widespread erosions involving haired skin, mucocutaneous junctions, and oral mucosa. A large multinodular splenic mass was detected by ultrasound, and it was surgically removed. The patient developed cardiac abnormalities and died approximately 12 hours after surgery.
Gross examination revealed facial lesions that were bilaterally symmetrical with alopecia and deep erosions on the rostral, dorsal, and lateral muzzle, the dorsomedial ocular canthi, and the oral commissures (Fig. 1). Erosions extended into the proximal nasal cavity and were also present on the gums and soft palate. Severe erosions oozing serum were observed on the concave pinnae (Fig. 2) and auditory orifices, as well as anus and prepuce. In many areas, pressure exerted along the edge of the lesions lifted the epidermis (Nikolskiy's sign). Haired skin lesions consisted of large erosions on the ventral thorax, axillae, sternum, groin, and pressure points of the limbs. Most of the footpad epidermis had sloughed, therefore leaving underlying erosions.
|
Indirect immunofluorescence was used to determine whether circulating anti-keratinocyte antibodies were present. Serum from this patient and a dog with PV were serially diluted and applied onto 5-µm-thin frozen sections of normal canine gingiva and urinary bladder as described previously.9 Circulating IgG autoantibodies from this patient were found to bind keratinocyte membranes at all levels of both gingival and bladder epithelia (e.g., intercellular fluorescence pattern). In contrast, IgG from a dog with PV bound to keratinocytes on gingival but not bladder sections (data not shown). A pattern of oral epithelial intercellular IgG autoantibody binding is present in all subsets of pemphigus, but antibody recognition of keratinocyte membrane of bladder transitional epithelium, as seen in our patient, is deemed highly suggestive of PNP.2
Immunoprecipitation was done using 14C-labeled human keratinocytes incubated with sera from our patient, a normal dog, and a human individual with PNP.2 Circulating IgG autoantibodies from the dog and human patients recognized a "ladder" of antigens identified as desmoplakin-I (250 kd), bullous pemphigoid antigen I (230 kd), a superposition of envoplakin (210 kd), and desmoplakin-II (210 kd), as well as periplakin (190 kd). A yet uncharacterized 170 kd unidentified antigen was detected by human PNP serum IgG, but it was not precipitated by autoantibodies from our patient (data not shown).
Gross examination of the spleen revealed a firm, multi-nodular, solitary mass, measuring 15 x 10 x 7 cm, variegated dark red, pink, and tan on cut surface, with multifocal friable yellow necrotic regions. Histologic evaluation of the mass revealed diffuse lymphoid and stromal hyperplasia with extramedullary hematopoiesis admixed with multifocal regions of an unencapsulated, infiltrative, and densely cellular neoplastic spindle cell population arranged in rows, swirls, and interlacing bundles. One section of the mass selected for histologic evaluation had three distinct populations of neoplastic spindle cells based on morphologic and immunologic phenotypes (Figs. 5–9). The cells in one region had a moderate amount of pale eosinophilic to amphophilic fibrillar cytoplasm with indistinct cell borders and oval nuclei with stippled, marginated chromatin and a mitotic index of 3 per 10 random high-power fields (hpf) (Fig. 5). These cells were characterized by nuclear and cytoplasmic staining with S100 (Fig. 6), cytoplasmic staining with vimentin, and a lack of staining with smooth muscle actin (SMA) and desmin (data not shown). The cells in the second region had a pattern similar to those in Fig. 5; however, the spindle cells were larger, the nuclei were larger and more pleomorphic, and there was abundant eosinophilic fibrillar cytoplasm (Fig. 7). These cells were characterized by positive staining with SMA (Fig. 8), positive staining with vimentin, and a lack of staining with S100 and desmin (data not shown). The cells in the third region were histologically more similar to the cells in Fig. 5; however, this region was more densely cellular and had a high mitotic index of 26/10 hpf (Fig. 9). The cells from this region did not stain with any of the four immunohistochemical markers (S100, SMA, vimentin, desmin) (data not shown). Our interpretation of these observations was that the splenic tumor, classified as a spindle cell sarcoma, had at least three distinct regions with different phenotypes, including smooth muscle differentiation.
|
Our observations, coupled to those of the previously published case of canine PNP,6 suggest that most characteristics of canine PNP closely correlate to those of the human counterpart, making canine PNP a relevant homologue of the human disease. The pathogenic mechanisms that lead to the development of concurrent neoplasia and skin lesions consistent with PV and EM have not yet been determined. In humans, skin lesions often precede clinical signs attributed to the neoplasia, and they are considered a "marker" of internal disease. These skin lesions often are resistant to therapy, and most patients die of complications caused by skin lesions or bronchiolitis obliterans but not of neoplastic invasion or metastasis. Although the efficacy of immunosuppressive therapy and prognosis of this disease has not been determined in dogs, PNP should be identified as early as possible to permit early tumor removal and initiation of immunosuppression.
Acknowledgments
We are grateful to Dr. Peter Moore for helpful histologic review of the splenic tumor and to Dr. Dave Rotstein for assistance with gross necropsy and subsequent histologic evaluation of collected tissues.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
