Gamma/delta T-cell lymphoma in a dog

Abstract

A 15-month-old dachshund was presented for examination because of a cough. Thoracic radiographs revealed the presence of a mass in the upper mediastinum. A diagnosis of γδ T-cell lymphoma was made by biopsy and flow cytometry analysis. The dog was treated with chemotherapy and remains asymptomatic after 24 months.

Malignant lymphoma is the most common hematopoietic malignancy in the dog. It occurs in multicentric, alimentary, mediastinal, and extranodal forms. Immunophenotype is a well established prognostic marker for canine lymphoma (1,2). T-cell lymphoma is generally associated with a worse prognosis compared with the more common B-cell lymphoma (2,3). The incidence of lymphoma originating in γδ T-cells is very low in both humans (4,5) and dogs (6), and the prognosis is generally poor.

Case description

In January 2006, a 15-month-old dachshund weighing 6 kg was presented to the Clinic of Dog and Cat Diseases of The University of Veterinary and Pharmaceutical Sciences in Brno, The Czech Republic. The primary clinical complaint was a history of coughing lasting for 1 day. Abnormalities noted on the initial physical examination included enlargement of the prescapular lymph nodes and a slight expiratory dyspnea. Thoracic radiographs revealed the presence of an upper mediastinal mass (5 times the width of a thoracic vertebra) elevating the trachea dorsally and pushing the tracheal bifurcation towards the 6th intercostal space (Figure 1). Abdominal ultrasound did not reveal any abnormalities. Cytologic examination of the left prescapular lymph node showed a predominance of small to medium-sized lymphoblasts with scant basophilic cytoplasm. The nuclei were round or slightly irregular in shape and contained multiple conspicuous nucleoli. The mitotic index was high, with 4 mitotic figures per 5 fields under a 40× objective lens (Figure 2).

Figure 1.

Thoracic radiograph showing the presence of an upper mediastinal mass pushing the tracheal bifurcation towards the 6th intercostal space.

Figure 2.

Cytologic appearance of the left prescapular lymph node. There is a predominance of small- to medium-sized lymphoblasts with scant basophilic cytoplasm. The nuclei are round or slightly irregular in shape and contain multiple nucleoli (Dip Quick Stain; magnification 1000×).

The initial laboratory tests included a complete blood cell count (CBC) and a serum chemistry panel. The CBC revealed leukopenia (3.7 × 10⁹/L; reference range: 6 to 17 × 10⁹/L); neutropenia (1.4 × 10⁹/L; reference range: 3 to 11.5 × 10⁹/L) and thrombocytopenia (59 × 10⁹/L; reference range: 200 to 500 × 10⁹/L). Serum chemistry showed that there was hypoproteinemia (53 g/L; reference range: 54 to 74 g/L) without hypoalbuminemia, increased bilirubin (0.8 mg/dL; reference range: 0 to 0.4 mg/dL), and elevated alanine aminotransferase (124.6 U/L; reference range: 19 to 70 U/L), aspartate aminotransferase (147.6 U/L; reference range: 15 to 43 U/L), alkaline phosphatase (1303.4 U/L; reference range: 15 to 127), and lactate dehydrogenase (1431.4 U/L; reference range: < 228 U/L).

Prolonged prothrombin time (PT) (27.2 s; reference range: 12 to 17 s) and markedly prolonged activated partial thromboplastin time (aPTT) (++++; reference range: 12 to 18 s) were abnormalities noted on the coagulation profile. The dog was crossmatched and received a whole blood transfusion (20 mL/kg, IV), Ringer’s solution (Ringeruv roztok; Braun Melsungen AG, Melsungen, Germany; Braun Medical S.A., Rubi, Spain) 3.5 mL/kg/h, IV, vitamin K1 (Kanavit; Hoechst-Biotika, Martin, Slovak Republic) 5 mg/kg, SQ, q24h, and hemostatics etamsylate (Dicynone; OM Portuguesa, Amadora, Portugal; Sanofi Winthrop Industrie, Quetigny, France) 10 mg/kg, IV, q12h. Following stabilization, the patient underwent ultrasound-guided fine-needle aspiration biopsy of the mediastinal mass under general anesthesia. This procedure was complicated by intrathoracic bleeding and a Tru-cut biopsy was not attempted. Fine-needle aspiration biopsy of the left prescapular lymph node for flow cytometry analysis and Tru-cut biopsy for histologic assesment were also performed. Bone marrow aspiration biopsy from the proximal humerus of the left forelimb was performed to rule out bone marrow infiltration.

Histology of the prescapular lymph node showed non-Hodgkin’s high-grade lymphoblastic lymphoma (Figure 3). Morphologic examination of the bone marrow detected normal cellularity, however, with infiltration of a uniform population of small blasts with rich basophilic cytoplasm and fine chromatin. The small to medium-sized blasts (Figure 4) comprised 55% of the bone marrow cells, other cell lines were significantly reduced. Immunophenotype was determined by flow cytometry on a sample obtained by fine-needle aspiration biopsy of the left prescapular lymph node. Cells were stained using the indirect immunofluorescence technique as described elsewhere (7). Primary monoclonal antibodies against CD45, CD3, CD4, CD8, CD21, γδ-TCR, CD90, MHC-II (provided by Dr. P.F. Moore, U.C. Davis, California, USA) and CD14 (DAKOCytomation) were used to characterize the tumor cells (Table 1; Figure 5). Data were acquired on a standard FACSCalibur flow cytometer (Becton Dickinson, Mountain View, California, USA) operated by the CELLQuest software.

Figure 3.

Histological appearance of the left prescapular lymph node, showing a high-grade lymphoblastic lymphoma (Tru-cut biopsy specimen, hematoxylin and eosin; magnification 400×).

Figure 4.

Bone marrow aspirate. There is infiltration with small- to medium-sized blasts with basophilic cytoplasm and fine chromatin (May-Grünwald, Giemsa Romanowski; magnification 1000×).

Table 1.

The primary monoclonal antibodies used for the flow cytometry analysis and the presence of cell surface antigens on lymphoblasts from the prescapular lymph node

Subset mAb	Specifity	Prescapular lymph node Blasts (%)
CA17.2A12	CD3	97.3
CA13.1E4	CD4	2.2
CA9.JD3	CD8	0.7
CA12.10C12	CD45	100
CA20.8H1	γδ-TCR	95.4
TUK4	CD14	0
CA1.4G8	Thy-1	5
CA2.1C12	MHC-II	15
CA2.1D6	CD 21	2.1

Figure 5.

Results of a flow cytometry analysis of a sample from the prescapular lymph node. On the basis of light scatter characteristics, lymphocytes and malignant cells can be distinguished. Presence of cell surface antigens on lymphocytes (left columns) and malignant cells (right columns) are shown in the respective histograms.

The malignant cells were distinguished from lymphocytes on the basis of typical light scatter characteristics — slightly larger size (characterized as a forward scatter) and an intensive side scatter signal — together with a lack of CD14 (Figure 5). The immunophenotype of malignant cells was CD45⁺CD3⁺ γδ-TCR⁺CD4⁻CD8⁻ (Table 1; Figure 5). No lymphoblasts were detected in the peripheral blood by morphologic examination and flow cytometric analysis.

The patient was clinically staged as stage Vb, according to the WHO staging system for lymphoma in domestic animals (8). The dog was started on a modified version of the Wisconsin-Madison University protocol for malignant lymphoma in dogs (9). The protocol uses vincristine, cyclophosphamide, doxorubicin, prednisone and L-asparaginase. L-asparaginase was not administered and was reserved for a rescue situation. We recommended feeding with polyunsaturated fatty acids (PUFAs) of the n-3 series. These fatty acids sensitize cancer cells to doxorubicin cytotoxicity (10). The N-3 PUFAs also normalize elevated blood lactic acid, resulting in an increase in disease-free interval and survival time for dogs with lymphoma (11). Clinical examination, CBC, and a serum chemistry panel were performed before each application of chemotherapy. Cardiac function was assessed by echocardiography and electrocardiography before each administration of doxorubicin.

Two days after the first administration of vincristine (0.7 mg/m²) the dog developed septic shock. He was hyperhydrated with infusions of Ringer’s solution, 4 mL/kg/h, IV, and he received amoxicillin/clavulanate (Augmentin; SmithKline Beechman Pharmaceuticals, Worthing, United Kingdom) 25 mg/kg, IV, q8h, enrofloxacin (Enroxil; Krka, Novo Mesto, Slovenia) 5 mg/kg, IV, q12h, the antiemetic agent thietylperazin (Torecan; Krka, Novo Mesto, Slovenia) 0.3 mg/kg, SQ, q8h, and the H₂-blocker ranitidine (Ranital; Lek Pharmaceuticals, Ljubljana, Slovenia) 2.5 mg/kg, SQ, q12h. His clinical status improved within 2 d. The dose of vincristine was subsequently reduced by 25%. As the dog tolerated this dose well, it was increased to 80% of the primary dose, but severe gastrointestinal toxicity occured. He did well thereafter and the therapy led to a complete clinical remission together with normalization of altered laboratory and radiographic findings within 8 wk. At week 10 of the protocol, an aspiration biopsy and flow cytometry of the bone marrow were performed, and no malignant cells were detected. The dog finished the chemotherapy protocol and in January 2008 (24 mo after the first presentation) the dog was very well and in complete remission. The patient was scheduled to be monitored at routine appointments including clinical examination, CBC, serum chemistry panel, and thoracic ultrasound every 3 mo for 1 y and then every 6 mo for the following 2 y.

Discussion

Malignant lymphoma is the most common hematopoietic malignancy in the dog (12). The annual incidence is approximatelly 33 cases per 100 000 dogs (13).

Immunophenotyping has been shown to indicate the prognosis for canine lymphomas (14,15). T-cell lymphoma is generally associated with a worse prognosis. Dogs with T-cell lymphoma treated with multiagent chemotherapy have reported median survival times of < 8 mo, whereas dogs with B-cell lymphoma treated with the same protocol have median survival times exceeding 15 mo.

Flow cytometry is a suitable method for the immunophenotypic analysis of canine lymphoma (16). This methodology is accepted for the classification of the T-cell receptor (TCR), which consists of αβ or γδ heterodimers expressed in association with the CD3 antigen (17). The majority of peripheral T-lymphocytes in humans and dogs express the αβ TCR, while only a small distinct subpopulation express the γδ TCR (18). Murine monoclonal antibodies (Mab) specific for canine TCR-1 (TCR- γδ) were developed by Moore et al (19) for use in the diagnosis of epidermotropic cutaneous T-cell lymphoma. The γδ TCR bearing T-lymphocytes represent an early stage in phylogenic and ontogenic development of the immune system but their function has not been fully elucidated. They are a normal subset of postthymic T cells with preferential homing to the sinusoidal areas of the spleen and some epithelial-rich tissues (20). These γδ T-cells have cytotoxic capabilities (20) and provide a first line of defense in the epidermal and mucosal epithelial linings (21).

Faldyna et al (22) characterized lymphocyte subsets in the peripheral blood, spleen, mesenteric and popliteal lymph nodes in adult dogs. The highest relative numbers of γδ T-cells were found in the spleen (7.2 and 6.4% mean and median, respectively) and very few γδ T-lymphocytes were present in lymph nodes (mean and median 0.9 and 1.0%, respectively). Faldyna et al (7) reported a nonsignificant age-related decrease in lymphocytes bearing γδ TCR in the peripheral blood [2.4 ± 1.1%/mean ± standard deviation (s) in dogs of age 5–6 d and 1.7 ± 1.6% for dogs > 5 y]. Some γδ T-lymphocytes in human spleen and peripheral blood are CD8⁺ (22), but most γδ T-cells have neither CD8 nor CD4 on their surface (23). In our patient, CD8 and CD4 were not expressed by γδ TCR-positive cells.

There is a small number of γδ T-lymphocytes in both humans and dogs and the incidence of lymphomas originating in γδ-T cells is low (4,6). The most common anatomical site of γδ T-lymphoma is the skin, liver, or spleen. In human medicine, hepatosplenic γδ T-cell lymphoma is a well-defined clinicopathologic entity with a typical immunophenotype and aggressive clinical course. This entity was first recognized by Farcet et al (24). It occurs predominantly in young male adults and is characterized by hepatosplenomegaly without lymphadenopathy. The liver, spleen, and bone marrow are infiltrated by monomorphic small to medium-sized tumor cells (25). Circulating malignant cells are noted in more than 25–50% of human patients during the course of the disease (26). In our patient, circulating malignant cells were not detected.

Only a few cases of nonhepatosplenic γδ T-cell lymphoma have been described in human medicine (27). Mediastinal lymphadenopathy was described in 1 Chinese patient with this type of lymphoma (28).

Fry et al (6) described the first case of hepatosplenic γδ T-cell lymphoma (HSTL) in the dog. The history and physical examination findings in our patient were completely different from those in the dog with hepatosplenic γδ T-cell lymphoma as described by Fry et al (6). This is due to the different location of the tumor. Thrombocytopenia, hypoproteinemia and abnormalities in liver parameters were in accordance with findings described in their report. Thrombocytopenia and neutropenia secondary to bone marrow involvement in our canine patient was confirmed by aspiration biopsy. Sex and age (young adult male) in our patient were in accordance with findings in human medicine.

Most patients described in the human literature have been treated with CHOP-based (cyclophosphamide, doxorubicin/hydroxydoxorubicin, oncovin, prednisone) chemotherapy but still the mean survival is less than a year (4). With respect to poor prognosis and aggressive clinical course described in human medicine, we opted for a multiagent chemotherapy protocol to achieve remission. L-asparaginase was not administered as this has recently been shown to not affect overall prognosis when used in this protocol (29,30).

It is generally agreed that systemic signs (Substage b WHO), bone marrow involvement (Stage V WHO), T-cell phenotype, and cranial mediastinal lymphadenopathy are associated with a poorer outcome (3,31). Despite the negative prognostic indicators and the poor prognosis of γδ T-cell lymphoma in humans and dogs, the patient has remained in complete remission for 24 mo. To our knowledge, this is the first case of canine γδ T-cell lymphoma in other than hepatosplenic or cutaneous location that has been described.