Canine lymphoma is a heterogeneous disease with highly variable clinical presentation, progression, and prognosis (18,19). Most are high-grade, large cell, B- or T-cell type; however, 5-29% of lymphomas in dogs are low-grade or behave indolently (18).

Currently, canine lymphoma is classified using the World Health Organization (WHO) classification system according to the tumor site, metastasis, and presence of clinical symptoms (12). Dogs with stage V typically show peripheral blood lymphocytosis because of tumor metastasis in the bone marrow. As lymphocytosis represents the most advanced stage of the tumor, it can indicate poor prognosis in dogs with lymphoma (23). However, it is still unclear whether lymphocytosis in low-grade lymphoma is also associated with poor prognosis.

T-zone lymphoma (TZL) is an indolent type of lymphoma. It is unclear whether lymphocytosis in TZL indicates bone marrow metastasis in stage V or whether aggressive chemotherapy is beneficial for TZL (8). This case report describes a case of a mixed-breed dog with peripheral lymphocytosis and cutaneous involvement of the tumor successfully treated with chemotherapy for TZL.

A 13-year-old neutered male mixed-breed dog presented with generalized lymphadenopathy that lasted more than one month. There were no systemic symptoms at the time of admission and no prior lymph node enlargement. Antibiotics and prednisolone for ten days before referral were not responsive. Physical examination revealed bilateral enlargement of the submandibular, prescapular, inguinal, and popliteal lymph nodes. In addition, plaque-like lesions, along with redness and lichenification, were observed on both sides of the auricle. Diagnostic imaging confirmed overall intra-abdominal lymph node hypertrophy (medial iliac, internal iliac, inguinal, and splenic lymph nodes) and hyper/hypoechogenic nodules in the spleen.

Regarding cellular, immunological, and molecular characterization, the fine-needle aspirated sample was evaluated microscopically, analyzed by flow cytometry, and subjected to polymerase chain reaction (PCR) analysis. A cytological evaluation revealed a homogenous population comprising small-to-intermediate lymphoid cells with a single round nucleus and a non-prominent nucleolus. There was also an increase in small lymphocytes with cytoplasm that protruded in the form of a “hand mirror” or “comet tail,” consistent with the previously described morphology of T-zone lymphocytes (Fig. 1A) (5,8,14). Although the predominance of monomorphic lymphoid cells in cytology could suggest neoplasia, atypical reactive hyperplasia could not be ruled out. Thus, as previously reported, immunophenotyping and histopathologic analysis were performed (5). Immunophenotypic analysis revealed a homogenous population of CD3+/CD5+/CD4–/CD8–/CD21+/CD34–/CD45– cells from the left popliteal lymph node and peripheral blood (Fig. 2). Additional PCR tests for antigen receptor rearrangement performed from the right popliteal lymph node showed T-cell receptor (TCR) gene clonality (IDEXX Laboratories, Westbrook, ME, USA).

Figure 1. Cytological evaluation of the lymph node (A), peripheral blood (B), and bone marrow (C). A homogenous lymphoid population consisted of small to intermediated lymphoid cells with a non-prominent nucleolus. Small lymphocytes with cytoplasm that protruded in the form of a “hand mirror” or “comet tail” were observed (A). Small lymphocytes were predominant in the peripheral blood (B). There was a mixture of myeloid and erythroid cells in the bone marrow. Lymphocytes (arrow) comprised 1.6% of the total bone marrow cell line and were mature small lymphocytes similar to those in the lymph node (C). Diff-Quik stain, ×400 (A), ×200 (B), ×1,000 (C).

Figure 2. Immunophenotyping using flow cytometric analysis from the lymph node aspirates (A, D-F) and peripheral blood cell isolates before (B) and after treatment (C). Lymphocytes from the lymph node (A) and blood (B, C) revealed a homogenous population of CD45- consistent with T-zone lymphoma. These cells also showed CD3+/CD5+/CD21+/CD4–/CD8–/CD34– (D-F). One month after chemotherapy, a remarkable decrease of neoplastic cell population (CD21+/CD45–) in peripheral blood from 77.0% (B) to 19.1% (C) was confirmed.

Given the small cell populations in lymph nodes, CD45- immunophenotype and TCR gene clonality support a diagnosis of TZL based on the WHO classification (2,5,21).

Confirmatory histopathological examination of the right popliteal lymph node revealed that the architecture of the cortex was effaced by diffuse infiltrates of monomorphous lymphoid cells measuring 6-8 μm in diameter. This population proliferated adjacent to the follicle and shifted the original structure toward the capsule. The mitotic count was very low, and only one mitotic figure was identified in ten high-power fields. Immunohistochemical staining (IHC) revealed neoplastic proliferation with immunolabeling of CD3 lymphocytes (T) but was negative for PAX-5 (B) (Fig. 3).

Figure 3. Histopathologic and immunohistochemistry examination of the right popliteal lymph node. On hematoxylin and eosin staining, the architecture of the cortex was effaced by diffuse infiltrates of monomorphous lymphoid cells measuring 6-8 µm in diameter (A). Immunohistochemical staining revealed neoplastic proliferation with immunolabeling of CD3 lymphocytes (B) but was negative for PAX-5 (C). This population proliferated adjacent to the follicle and shifted the original structure toward the capsule.

Regarding cutaneous lesions, a punch biopsy was also performed from the auricle to determine whether the lesion had resulted from lymphoid metastasis. This lesion contained extensive infiltration of small lymphocytes, similar to the cells in the lymph nodes. These lymphoid cells formed small nodules oriented around the adnexal structures. Compared to canine cutaneous lymphoma, epitheliotropism of the lymphocytes, such as Pautrier’s microabscess, was not observed. The results of the IHC test of these cells were also CD3 positive and PAX-5 negative, suggestive of metastasis of neoplastic lymphoid cells (Fig. 4).

Figure 4. Histopathologic and immunohistochemistry examination of the auricular cutaneous lesion. On hematoxylin and eosin staining, extensive infiltrates of small lymphocytes measuring 7-8 μm in diameter, similar to cells in the lymph nodes, are noted (A). Immunohistochemical staining also revealed neoplastic proliferation with immunolabeling of CD3 lymphocytes (B) but was negative for PAX 5 (C).

A blood test was conducted to confirm the presence of paraneoplastic syndrome and its effect on systemic organs. Laboratory analysis showed lymphocytosis (lymphocyte count, 14,144 cells/μL [reference interval, 1,050-5,100 cells/μL]) in the peripheral blood. Small lymphocytes were also the predominant cells in the blood, as previously observed in lymph nodes (Fig. 1B). To investigate the possibility of bone marrow metastasis, a bone marrow aspiration cytology was performed from the right proximal humerus. The examination revealed a heterogeneous appearance with numerous cell lines, including myeloid and erythroid cells, in an estimated 1:1 ratio, and both lineages appeared complete with orderly maturation. Lymphocytes comprised only 1.6% of the total bone marrow cell line, which is below the normal range. The observed lymphocytes had the appearance of homogenous mature small lymphocytes; however, the hand-mirror-shaped cells previously described by cytology were not confirmed (Fig. 1C). No evidence of primary bone marrow origin lymphocytic leukemia or infiltrating small cell lymphoma was found, and active hematopoiesis with no strong evidence of lymphocytic infiltration was diagnosed.

The dog was diagnosed with TZL with cutaneous metastasis and lymphocytosis but without bone marrow metastasis. Chemotherapy was initiated with chlorambucil at a dose of 4 mg/m2 and prednisolone (1 mg/kg) once a day, as recommended for asymptomatic lymphoma in dogs, considering lymphocytosis with a circulating lymphocyte count of ≥9,200 cells/L, as well as cutaneous infiltration of neoplastic lymphoid cells (10). After one month of medication, the sum of the longest diameters of peripheral lymph nodes decreased from 192.5 mm to 64 mm. The peripheral lymphocytosis was also resolved (lymphocyte count, 3,524 cells/μL), and the plaque-like lesions and pruritus in both auricles were almost resolved. After four months of treatment, ultrasonographic examination revealed a reduction in the size of the overall intra-abdominal lymph nodes, and immunophenotypic analysis of peripheral blood showed a significant decrease in the CD45- cell population (Fig. 2C). The dog is healthy, without any signs of progression of the indolent lymphoma and side effects of chemotherapy until eight months after the initial diagnosis at present.

Canine TZL is categorized as indolent lymphoma. TZL shows indolent behavior with a long survival period despite generalized lymphadenopathy and poor response to chemotherapy. Dogs with TZL have a median survival time ranging from 14-33.5 months (4,6,8,13,22). One retrospective study of 51 TZL in dogs reported that a few dogs died within a short time (median survival time, 760 days), despite of high prevalence of stage V (93%) (6); bone marrow infiltration was found in 40/43 dogs with TZL from that study.

However, although peripheral lymphocytosis was noted, bone marrow metastasis was not observed in our case. Peripheral blood lymphocytosis accounts for 48-64% of canine TZL (4,5,14,15). Two hypotheses have been proposed to explain this peculiar phenomenon of peripheral lymphocytosis without bone marrow involvement. One hypothesis is that neoplastic cells are released into the bloodstream without invading the bone marrow or returning to the lymph nodes. Second, the lack of CD45 may impair tumor cell motility and prevent them from returning to the lymph nodes (6,16,17). However, the exact mechanism has not yet been identified. While there may be some academic debate regarding the nomenclature ascribed to the question of whether to call a disease stage V lymphoma with peripheral blood involvement or leukemia, these findings emphasize the importance of not using the finding of positive lymphocytosis as a screening tool for bone marrow metastasis. Hematologic abnormalities do not always indicate bone marrow involvement, and bone marrow infiltration may occur even in the absence of hematologic abnormalities or cytopenia. Therefore, BM evaluation should be included in routine staging to improve classification (7). In terms of the short survival time in later stages, the utility of the WHO system for staging indolent canine lymphoma is questionable (6). More prognostic variables should be investigated in TZL in the future.

In this case, cutaneous signs were observed. Cutaneous involvement was found in 7 of 101 TZL cases in a previous study (2). A specific phenotype was proposed for the presence of cutaneous symptoms in lymphoma: 60% of skin symptoms in 217 canine lymphomas had a CD4-CD8-phenotype (2). Approximately 20% of canine epitheliotropic T cells express neither CD4 nor CD8 (11). Therefore, this specific phenotype may be associated with cutaneous manifestations of lymphoproliferative diseases. More cases are needed to prove whether this phenotype is related to the development of cutaneous symptoms and to elucidate whether the CD4-CD8-phenotype does not represent a true phenotypic aberrancy in canine cutaneous lymphoma but rather a neoplastic proliferation of a resident population characterized by this peculiar phenotype.

The dog was responsive to chemotherapy. Watchful waiting without treatment is generally recommended for TZL because of the long survival time and lack of clinical symptoms associated with disease progression (19). However, some cases may experience tumor-related symptoms, such as dyspnea caused by enlarged submandibular lymph nodes (1,6). Although there is no consensus on the criteria for initiating treatment in dogs with TZL, several standards have been suggested in recent review articles on the treatment of T-cell lymphoma (10).

Chemotherapy for TZL ranges from prednisone alone to a combination of prednisone and chlorambucil, doxorubicin, or cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP)-based chemotherapy protocols similar to those used for aggressive lymphoma (4,20). In the study by Valli et al. (20), the longest surviving dog with TZL was diagnosed with stage IIIa, was given the CHOP protocol, and survived 990 days. It may also be an effective option because long-term survivors are included in the group treated with CHOP. However, an oral chemotherapy protocol combining chlorambucil and prednisone is more effective than a commonly used multidrug protocol (e.g., CHOP) (3,4,8,9). A longer survival rate and lower lymphoma-related mortality were observed in the group treated with chlorambucil and prednisolone. Continuous delivery of a slow alkylating agent is considered a biologically rational approach for indolent lymphoma treatment (4,9). Therefore, if chemotherapy is administered, chlorambucil and prednisone could be the most appropriate first-line therapy protocol (4). To determine which treatment method is most appropriate, TZL should be classified according to their staging (including the presence or absence of symptoms) in a treatment trial.

Unfortunately, a core biopsy could not be performed in this case. Bone marrow aspiration is the most basic method for determining bone marrow status, although bone marrow biopsy is known to detect tumor invasion more sensitively. For a more accurate evaluation, further investigation would be necessary to confirm immunomorphological and molecular levels of bone marrow invasion through bone marrow core biopsy or flow cytometry.

The present case showed cutaneous symptoms associated with tumor metastasis. Chemotherapy was initiated based on the number of circulating lymphocytes exceeding 9,200 cells/μL. Within a month, the lymph nodes in the whole body had decreased in size, and the lesions in the ear had almost completely resolved. The dog regained an excellent quality of life and had no side effects related to the use of chlorambucil. The present case is a rare instance of TZL canine patients who present with cutaneous involvement and lymphocytosis without bone marrow infiltration and respond successfully to chemotherapy involving chlorambucil and prednisolone.

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (2020R1C1C1008675).

The authors have no conflicting interests.