Oral Squamous Cell Carcinoma in a Red Deer (Cervus elaphus)

The red deer (Cervus elaphus) is a common wildlife species in Europe, Asia, and North Africa, and a frequent subject of ecologic science and popular hunting reports (Ludt et al. 2004). However, little is known about the type and frequency of its naturally occurring noninfectious diseases, especially neoplasms. The most frequently reported tumors of red deer are papillomavirus (PV) -associated dermal fibropapillomas and papillomas (Boch and Schneidawind 1988; Erdelyi et al. 2009). Further reports include hepatic carcinoid, auricular chondromas, oral fibrosarcomas, testicular fibroma, metastatic cavernous hemangiosarcoma of bone, subcutis, and lung, testicular interstitial cell tumor, dermal malignant melanoma, anaplastic oligodendroglioma, malignant schwannoma, and ovarian teratoma (Boch and Schneidawind 1988; Perez et al. 1998; Hofle et al. 2004; Scandrett and Wobeser 2004; Gregoire et al. 2008; Peters and Wohlsein 2008; Zele et al. 2011).

In August 2010, an emaciated, free-ranging, adult, female red deer was killed by a forest ranger in Lower Saxony, Germany. The age of the deer, determined by counting annual cementum layers in both first incisors (I1; Azorit et al. 2004) was 13 yr. The head and neck were submitted for pathologic investigation of a grossly evident mass in the right maxilla.

An infiltrative, multilobulated, white to grey, rubbery to firm, 12×6×6-cm mass extended from the oral cavity into the right maxilla, nasal turbinates, and nasal septum (Fig. 1). The right upper jaw displayed focally extensive osteolysis and loss of all premolars and the first and second molars. The right retropharyngeal lymph node was moderately enlarged.

Histopathology revealed an infiltrative, moderately cellular mass extending from the oral squamous mucosa into the underlying soft tissues and bone. The mass consisted of cords and nests of epithelial cells within an abundant collagenous stroma, interpreted as desmoplasia (Fig. 2A). The epithelial cells had distinct borders, a variable diameter of ∼25 µm, a moderate to abundant amount of eosinophilic cytoplasm, and oval nuclei with coarsely granulated chromatin and one–three nucleoli. There was moderate anisocytosis and anisokaryosis. There were up to two mitotic figures per high-power (400×) field. A few scattered cells displayed cytoplasmic hypereosinophilia combined with nuclear condensation and loss, interpreted as dyskeratosis (Fig. 2B). The oral mucosa was multifocally ulcerated and the adjacent mass displayed a focally extensive infiltration of mainly neutrophils, eosinophils, fewer macrophages, and granulation tissue. There was a mild, diffuse infiltration of lymphocytes and fewer macrophages and plasma cells. The cortical sinus of the right retropharyngeal lymph node contained a cluster of densely packed polygonal epithelial cells resembling those described for the oral mass, interpreted as metastasis.

Immunohistochemical (IHC) staining was performed with the avidin–biotin–peroxidase complex method (Vector Laboratories, Burlingame, California, USA) with the use of two monoclonal anti-cytokeratin antibodies (clone AE1/AE3, reacts with human cytokeratins 1, 2, 3, 4, 5, 6, 7, 8, 10, 13, 14, 15, 16, and 19; clone MNF116, reacts with human cytokeratins 5, 6, 8, 17, and probably 19), a monoclonal anti-vimentin antibody (clone V9; Ulrich et al. 2009), and a polyclonal rabbit anti-bovine PV (BPV) antibody (Teifke et al. 2003; all from Dako Deutschland GmbH, Hamburg, Germany). The specificity of the IHC reactions was confirmed with the use of normal red deer oral mucosa and BPV 1-induced bovine cutaneous fibropapilloma. As a negative control, serial sections were treated with Balb/c mouse ascites fluid or rabbit hyperimmune-serum directed against the major capsid protein pUL19 of pseudorabies virus (Klupp et al. 2000), instead of the primary monoclonal or polyclonal antibodies, respectively. Most of the neoplastic epithelial cells in the oral mass and the lymph node displayed a variably strong cytoplasmic immunoreactivity employing both pan-cytokeratin markers. In general, more cells were immunoreactive and the label was of stronger quality with clone MNF116 (Fig. 2C). Although most spindloid stromal fibroblasts and vessels displayed moderate cytoplasmic vimentin immunoreactivity, neoplastic epithelial cells were uniformly negative. No PV immunoreactivity was detected.

We performed a PCR reaction with the use of the previously described consensus primer pair set IFNR2 and IDNT2, which allows the detection of a 102 base-pair fragment of the L1 gene of multiple PVs, including BPV1, BPV2, ovine PV (OPV)1, OPV2, European elk PV, deer PV, canine oral PV, and the PV of Mastomys coucha in formalin-fixed, paraffin-embedded tissues (Teifke et al. 2003). Cloned genomic DNA of BPV1 (kindly provided by P.M. Howley, Maryland) was used as positive control and a sample without DNA was included in each run to check for contamination. However, no PV DNA was detectable within the tumor tissue of the present case.

Squamous cell carcinoma (SCC) is a locally invasive and occasionally metastatic malignant neoplasm of the epidermal cells of the skin and the squamous epithelia of the mucous membranes with varying degree of keratinocyte (squamous cell) differentiation. Etiologic risk factors for SCCs include ultraviolet light, PV infection, bracken fern ingestion, and carcinogens in tobacco, coal tar, soot, arsenic, and smegma (Brown et al. 2007). There has been a marked increase in the incidence of human SCC over the last 50 yr, with UV light representing the most important risk factor (Dal et al. 2008), and PV infection as a possible cofactor (Munday and Kiupel 2010). To our knowledge SCC has not been previously reported in red deer. However, SCC has been reported in other Cervidae, including the skin of the front leg of an 18-yr-old female Père David's deer (Elaphurus davidianus; Agrimi et al. 1993) and the skin of the front leg with metastasis to the ipsilateral prescapular lymph node and lungs of a 9-yr-old female Indochina sika deer (Cervus nippon pseudaxis; Ensley et al. 1980). In addition, an oronasal SCC with metastasis to the ipsilateral retropharyngeal lymph node was described in a 3.5-yr-old white-tailed deer (Odocoileus virginianus; Stroud and Amundson 1983). Although SCCs of the oropharynx, esophagus, and forestomachs of cattle are rare, they are relatively common in certain geographic locations in England, Scotland, Brazil, and Bolivia. Suspected causes for this regional disparity include BPV4 infection and bracken fern ingestion (Brown et al. 2007). There is growing evidence that PVs are involved in the pathogenesis of SCCs in multiple species including humans, cats (Felis catus), dogs (Canis lupus familiaris), sheep (Ovis aries), rabbits (Sylvilagus floridanus and Oryctolagus cuniculus), western barred bandicoots (Perameles bougainville), Natal multimammate mice (Mastomys natalensis; Munday and Kiupel 2010), a ferret (Mustela putorius furo; Rodrigues et al. 2010), and horses (Equus ferus caballus; Knight et al. 2011). No cytopathic evidence of PV infection and no PV antigen or DNA were detectable in our case. This does not, however, rule out the possibility that PVs can be involved in the induction of SCCs in red deer, because PV-induced neoplastic transformation is often associated with restricted and nonproductive infection that may not be detectable in all cases (Teifke et al. 2003; Munday and Kiupel 2010; Knight et al. 2011). Many well-differentiated SCCs display formation of concentric lamellae of keratin (keratin pearls). In contrast, many poorly differentiated tumors, such as our case, only show keratinization of individual cells (Goldschmidt and Hendrick 2002). Immunohistochemistry for cytoskeletal intermediate filaments like cytokeratin and vimentin are commonly applied tools for the identification of the histogenetic origin of poorly differentiated tumors (Ulrich et al. 2009). Cytokeratin immunoreactivity of the neoplastic epithelial cells in the oral mass and lymph node metastasis prove the epithelial histogenesis of this tumor (Erdelyi et al. 2009).

Continuous wildlife disease monitoring and documentation through published reports is essential to help identify environmental hazard-induced (e.g., radiation, carcinogens, and toxins) increases in disease frequencies among exposed sentinel species.

We thank Günter Schröder, Enercity Forestry Office, Wedemark for submission of this case.