Pax5 and CD3 immunophenotyping of lymphoma in 2 central bearded dragons

Abstract

Two central bearded dragons (Pogona vitticeps), a 3-y-old male and a 5-y-old female, were diagnosed with different manifestations of lymphoma at the Kansas State Veterinary Diagnostic Laboratory between 2019 and 2020. The 3-y-old male was presented for postmortem evaluation and was in poor body condition. Microscopically, nearly all examined organs contained variable numbers of neoplastic round cells. Neoplastic cells in the stomach and liver had moderate immunoreactivity to CD3 consistent with multicentric T-cell lymphoma, and non-neoplastic lymphocytes infiltrating the stomach mass had strong immunoreactivity to Pax5. The 5-y-old female had an ulcerated oral mass located in the right lingual gingiva submitted as an excisional biopsy. Microscopically, the mass was composed of large numbers of neoplastic round cells in the epithelium and connective tissue that were strongly and diffusely positive for CD3 and frequently positive for Pax5, consistent with a dual-positive, localized, epitheliotropic T-cell lymphoma. Neoplastic and non-neoplastic lymphocytes did not stain with CD20 or CD79a. Neoplasms are increasingly reported as a cause of morbidity and mortality in reptiles. Our 2 cases illustrate various presentations of T-cell lymphoma and the effectiveness of CD3 and Pax5 immunohistochemistry in bearded dragons.

Neoplasia has historically been considered uncommon in reptiles, but in recent years has been identified as a relatively common cause of morbidity and mortality in saurians.^{5,12 –14} In a 3-y study of disease prevalence in bearded dragons (Pogona vitticeps) that evaluated 529 animals, neoplasia accounted for only 7 cases (1.3%). ¹⁶ A more recent retrospective study of lizards over a 10-y period identified a 22.9% prevalence of neoplasia, with bearded dragons and panther chameleons (Furcifer pardalis) representing a higher relative proportion of neoplasia diagnoses compared with other lizard species. ¹⁴ Bearded dragons may be over-represented in case reports of neoplasia given the fact that they are popular household pets with more frequent veterinary visits during times of morbidity. ² A retrospective study of neoplasia in reptilian species identified an 8.5% prevalence of neoplasia in reptiles overall, with a 4.8% prevalence in iguanids. ⁵ In that report, lymphoid neoplasia predominated in central bearded dragons (syn. inland bearded dragon; a prevalence of 1.6% in agamids); most cases were multicentric and of a blastic phenotype. ⁵ Neoplasia in reptiles is often associated with nonspecific clinical signs including lethargy, anorexia, and weight loss; anemia and leukocytosis evident on CBC; and multi-organ neoplastic cellular infiltration identified histologically. ⁶ We describe here different presentations of T-cell lymphoma in 2 central bearded dragons.

Case 1, a 3-y-old male central bearded dragon, was submitted by a client to the Kansas State Veterinary Diagnostic Laboratory (KSVDL; Manhattan, KS, USA) for autopsy. The minimal available history indicated that the bearded dragon had initially been presented to the referring clinic 4 mo prior to death with ocular crusting and edema, and was prescribed 2 different courses of topical ophthalmic antibiotics, as well as a course of oral cephalexin (doses and durations unknown). The bearded dragon’s weight dropped by 50% during the treatment period, and he developed hemorrhagic diarrhea that was treated with fenbendazole, sulfadimethoxine, and injectable ceftazidime (dose and duration unknown). One month later, the animal was euthanized and submitted for postmortem examination.

Major gross lesions included severe emaciation (Fig. 1A) with coelomic adipose tissue atrophy. The gastrointestinal tract was grossly unremarkable and was placed intact in 10% neutral-buffered formalin to allow adequate fixation. Histologic examination revealed large numbers of a moderately pleomorphic population of round cells with a nuclear diameter of 5–10 µm, with occasional karyomegaly that effaced and expanded ~60% of the gastric mucosa and submucosa, and extended into the tunica muscularis (Fig. 1B, 1C). Neoplastic cells had marked anisocytosis and anisokaryosis, with a mitotic count of 51 in 2.37 mm². Moderate-to-abundant numbers of similar neoplastic cells variably effaced, expanded, or infiltrated the following tissues: hepatic sinusoids and spaces of Disse; pulmonary, renal, epicardial, myocardial, and testicular interstitium; coelomic adipose tissue; and transmurally throughout the remaining alimentary tract from the esophagus to the colon. Several other incidental findings included esophageal and gastric Cryptosporidium and moderate glomerulosclerosis.

Figure 1.

T-cell lymphoma in a 3-y-old male bearded dragon (case 1). A. Emaciation evident on autopsy. B. Neoplastic round cells efface and expand the gastric mucosa, submucosa, and tunica muscularis. H&E. C. Neoplastic cells below the gastric epithelium display anisokaryosis, anisocytosis, and mitotic activity. H&E. D. Neoplastic lymphocytes have positive cytoplasmic immunoreactivity to CD3 (brown); scattered non-neoplastic lymphocytes have positive intranuclear immunoreactivity to Pax5 (red). Dual CD3 and Pax5 immunohistochemistry, hematoxylin counterstain.

Immunohistochemistry (IHC) was performed on affected sections of stomach and liver and on the spleen as an internal control using the following antibodies: anti-CD3 (mouse anti-human monoclonal; Bond ready-to-use primary antibody CD3 [LN10]; Leica), anti-CD20 (rabbit polyclonal, 1:400 dilution; CD20 epitope–specific rabbit antibody; Thermo Fisher), anti-CD79a (mouse monoclonal, 1:600 dilution; anti-CD79a antibody [HM57]; Abcam), and anti-Pax5 (rabbit monoclonal, B-cell lineage–specific activator protein ready-to-use primary antibody; VeteMab [CHMP2015]; ChampDx). Staining was performed at the KSVDL using standard procedures and manufacturer’s instructions. Moderate cell membrane immunoreactivity of neoplastic cells was observed in the stomach and liver, and in non-neoplastic lymphocytes in the spleen using anti-CD3, consistent with a diagnosis of CD3+, multicentric lymphoma (Fig. 1D). Non-neoplastic small lymphocytes, scattered within the neoplastic cell population in the stomach from this case, and moderate-to-abundant numbers of frequently white pulp–associated lymphocytes in a control spleen from a non-affected bearded dragon, had strong, nuclear immunoreactivity to anti-Pax5 antibody. Neoplastic and non-neoplastic lymphocytes had no immunoreactivity to either anti-CD79a or anti-CD20 antibodies.

Case 2, a 5-y-old, female central bearded dragon was evaluated by the Exotic Animal Medicine Service at KSU because of respiratory signs characterized by open-mouth breathing with increased effort, and intermittent inappetence and lethargy for 1 mo prior to presentation. She had not produced eggs for at least 1 year prior to presentation. The physical examination was unremarkable except for a 4-mm flat, soft, white lesion of unknown duration that extended from the buccal to the lingual aspect of the right mandibular gingiva (Fig. 2A). CBC revealed marked monocytosis (5.9 × 10⁹/L, RI: 0–0.5 × 10⁹/L), marked leukocytic toxic change, and occasional polychromasia. ³ The plasma chemistry panel was unremarkable. ¹⁸

Figure 2.

T-cell lymphoma in a 5-y-old female bearded dragon (case 2). A. A 3-mm, flat, soft, white lesion (arrow) of unknown duration on the right mandibular gingiva. B. Neoplastic round cells infiltrate the subgingival connective tissue and focally efface overlying surface gingival epithelium. H&E. C. Neoplastic cells with marked epitheliotropism. H&E. D. Neoplastic lymphocytes have strong positive cytoplasmic immunoreactivity for CD3. CD3 immunohistochemistry (IHC), hematoxylin counterstain. E. Neoplastic lymphocytes have strong positive nuclear immunoreactivity for Pax5. Pax5 IHC, hematoxylin counterstain. F. Neoplastic lymphocytes have dual-positive immunoreactivity to CD3 and Pax5. Dual CD3 and Pax5 IHC, hematoxylin counterstain.

Ten days after initial presentation, physical examination was unchanged except for a palpable mass in her left caudal coelom. A whole-body CT scan demonstrated a 1.9 × 1.8 × 1.1-cm, ovoid, soft-tissue mass and 0.7-mm diameter, round nodules in the caudal coelom. The gingival mass was surgically excised, fixed in formalin, and submitted to the KSVDL.

Histologic examination of the oral mass revealed large numbers of neoplastic, monomorphic round cells with a nuclear diameter of 4–6 µm within the subgingival connective tissue and both surface and glandular epithelium, suggestive of epitheliotropic lymphoid neoplasia (Fig. 2B, 2C). The neoplastic cells had moderate anisocytosis and anisokaryosis, with 5 mitotic figures identified in 2.37 mm². Giemsa staining, performed to rule out mast cell tumor, did not reveal metachromatic granules in neoplastic cells. Ulcerative and necrotizing gingivitis and stomatitis with bacteria were noted in other areas of the biopsy. IHC was performed with anti-CD3, anti-Pax5, anti-CD79a, and anti-CD20 antibodies as described in case 1. Neoplastic round cells had strong and nearly diffuse cytoplasmic and cell membrane immunoreactivity for anti-CD3 antibody, and ~80% of these cells also had strong, nuclear immunoreactivity for anti-Pax5 antibody. Rare, non-neoplastic lymphocytes at the periphery of the neoplasm also had strong, nuclear immunoreactivity to anti-Pax5 antibody. Neoplastic cells had no immunoreactivity for CD79a or CD20. These findings were consistent with dual Pax5+ and CD3+ epitheliotropic lymphoma (Fig. 2D–F).

A recheck appointment 2 d after surgical excision of the oral mass found the patient lethargic with minimal truncal lift. Coelomic ultrasonography revealed a large pocket of anechoic fluid in the coelom; coelomocentesis produced yellow, cloudy fluid (a low-cellularity, high-protein exudate). Empiric prednisolone (2 mg/kg PO q24h) and cyclophosphamide (3 mg/kg PO q24h) therapy was initiated, but the patient died 48 h later. The body was submitted for postmortem examination, which revealed localized egg yolk–induced coelomitis. Neither gross nor histologic postmortem examination identified neoplasia at the oral surgical site or within other examined tissues or organs. Therefore, oral epitheliotropic lymphoma was diagnosed, with complete antemortem excision; egg yolk coelomitis was considered the cause of death.

We have described here different presentations consistent with T-cell lymphoma in 2 captive central bearded dragons. Blood work was only available for case 2, and CBC did not reveal anemia as reported with lymphoma in a bearded dragon and savannah monitor (Varanus exanthematicus), nor leukocytosis as reported with lymphoma in a bearded dragon and a green iguana (Iguana iguana).^12,19 Lymphoid and myelogenous malignancies are generally multicentric or disseminated in lizards, ¹⁴ including reports of lymphocytic leukemia, ⁶ chronic monocytic leukemia,^8,12 lymphoblastic leukemia, ¹⁷ and myelogenous leukemia. ¹⁹ Interestingly, a retrospective study of lymphoma in reptiles noted that oral lesions were the most common manifestation of multicentric lymphoma, suggesting that lymphoma should be considered as a differential diagnosis for oral lesions in addition to stomatitis, gingivitis, or other neoplasms (e.g., squamous cell carcinoma). ¹¹ Case 2 underscores the importance of pursuing surgical biopsy for diagnosis of oral lesions in bearded dragons, given that it may significantly alter diagnostic and treatment plans.

A study reported that lymphoma was diagnosed on autopsy of 9.3% of lizards and 1.6% of agamids. ⁵ However, immunophenotyping was scarcely pursued or reported in cases of lymphoid or bone marrow neoplasia in bearded dragons. ⁵ Of the 7 reported cases of lymphoid neoplasia in bearded dragons, IHC was pursued in only 3 cases: 1) a CD79a+ case of lymphocytic leukemia, ⁶ 2) a case of chronic monocytic leukemia negative for all markers tested (CD3, CD79a, CD68, Mac387), ⁸ and 3) a case of CD3-chronic monocytic leukemia diagnosed as B-cell phenotype based on cellular morphology. ¹³ The other 4 cases of lymphoid neoplasia in bearded dragons were diagnosed morphologically as myeloid leukemia, lymphoblastic leukemia, leukemia, and leukemia with disseminated lymphoma.^16,17,19

When IHC is used in an attempt to obtain definitive diagnoses in cases of neoplasia in reptiles, results should be interpreted cautiously. The markers commonly used in veterinary medicine were initially developed and validated for use in mammals, and these markers have not been validated in reptilian tissues.^9,12 The most common mammalian T-cell marker, CD3, has been used to identify T-cells in reptiles; common mammalian B-cell markers (e.g., CD20, CD79a) are considered less definitive. ¹³ This perhaps was illustrated by complete lack of staining for both CD20 and CD79a within the control tissue (spleen) in case 1. Pax5, a nuclear transcription factor of B-cells, has been reported as a B-cell lineage indicator in canine, feline, and psittacine lymphoma. ⁷ Given the reported success in identifying B-cell lineage lymphoma in psittacines, anti-Pax5 was used in both bearded dragon cases described here, and demonstrated strong nuclear immunoreactivity in control tissue lymphocytes (spleen) and as a dual marker of CD3+ neoplastic cells in case 2. Our findings suggest that Pax5 is a better immunohistochemical stain than either CD20 or CD79a when attempting to classify lymphoma or leukemia in reptiles.

Case 2 illustrates the limited understanding of lymphocyte markers in reptiles. Dual CD3+ and Pax5+, non-epitheliotropic lymphomas have been reported in humans and psittacines.^4,7,10 Dual CD3+ and CD20+ epitheliotropic lymphoma has also been described in humans and a dog, although in the single canine case and several human cases, the neoplastic cells were CD3+, CD20+, and Pax5–.^1,15 Many of these reported cases suggest that neoplastic lymphocytes expressing B- and T-cell antigens may be the result of loss of, or ectopic expression of, transcription factors in neoplastic cells.^7,10

It has been proposed that lymphoid neoplasia in lizards may not fall into the traditional B- and T-cell categories that are assigned to most mammalian lymphoid malignancies. ¹² In a case series of 8 Egyptian spiny-tailed lizards (Uromastyx aegypticus) with lymphoid neoplasia, IHC using CD3 and CD79a markers was unsuccessful in all 8 cases on both neoplastic tissue and normal lymphocytes, further suggesting that these markers may not be appropriate for use in all reptilian species. ⁹ BLA36 and MUM 1 have also been reported to reliably identify B cells in mammals, reptiles, and psittacines, but were not used in either of our cases.^7,9,12

Our 2 cases represent different manifestations of suspected T-cell lymphoma in bearded dragons (disseminated disease with comorbidities vs. localized neoplasia in the oral cavity) diagnosed through histologic examination and IHC. Our cases also support Pax5 as a useful IHC marker in characterizing reptile lymphoma and leukemia.