Cutaneous lymphoid hyperplasia mimicking cutaneous lymphoma in a hyperthyroid cat

Abstract

A 12-year-old neutered male domestic shorthair cat presented for chronic, localized, swelling and crusting of the left upper lip, weight loss, sporadic vomiting, and focal alopecia between the scapulae was diagnosed with hyperthyroidism and regional eosinophilic lymphadenitis. Treatment with methimazole exacerbated an underlying hypersensitivity disorder leading to marked generalized lymphadenopathy that histologically mimicked lymphoma.

A 12-year-old, 4.85-kg, neutered male, domestic shorthair cat was presented to the Veterinary Medical Center at the Western College of Veterinary Medicine (WCVM), for chronic, localized swelling and crusting of the upper left lip, weight loss, sporadic vomiting, and interscapular alopecia. The lesion on the upper left lip had been present for years but had worsened in the last 6 mo. The focal area of interscapular alopecia had become apparent in the last month. The cat was nonpruritic and had lost 1.25 kg of body weight over the last 8 mo despite normal activity and appetite. The owner felt the cat was also breathing more deeply after exercise at home and open-mouth panting was observed prior to physical examination. The cat was not taking any medications, was housed strictly indoors and had no significant travel history.

Case description

On physical examination the heart rate, respiratory rate, and rectal temperature were 270 beats/min (bpm), 70 breaths/min, and 39.7°C, respectively. The upper left lip was swollen and covered by a serosanguinous crust (Figure 1). Mild bilateral popliteal lymphadenomegaly, a cervical thyroid slip and a 1-cm diameter, mobile, subcutaneous mass on the right lateral thorax were also detected.

Figure 1.

Focal lip lesion in a cat ultimately diagnosed with cutaneous lymphoid hyperplasia mimicking cutaneous lymphoma.

A complete blood (cell) count (CBC), serum biochemistry profile, serum total T4 concentration, and serum enzyme-linked immunosorbent assay (ELISA) tests for feline leukemia virus (FeLV) and feline immunodeficiency virus (FIV) were evaluated. Abnormalities included a mild eosinophilia [1.692 × 10^9/L; reference interval (RI): 0.1 to 1.5 × 10^9/L], a moderate decrease in creatinine (55 μmol/L; RI: 78 to 178 μmol/L), a mild decrease in total calcium (2.16 mmol/L; RI: 2.26 to 2.86 mmol/L), and a mild increase in serum alkaline phosphatase activity (ALP; 95 U/L; RI: 11 to 56 U/L). The serum total T4 concentration was markedly elevated (227 nmol; RI: 13 to 50 nmol/L), confirming hyperthyroidism. The cat was negative for both FeLV and FIV. Systolic hypertension was documented (average reading 200 mmHg), leading to possible concerns for hypertension-induced end organ damage (1). The cytological diagnoses from fine-needle aspirates of the popliteal lymph nodes and the mass on the lateral right thorax were eosinophilic lymphadenitis and mixed lymphadenitis, respectively. Skin scrapings were negative for ectoparasites. Impression smear cytology of the upper left lip lesion revealed suppurative inflammation.

The weight loss, vomiting, systemic hypertension, mild elevation in serum ALP activity, mild hypocalcemia, focal interscapular hair loss, and the cervical thyroid slip were all attributed to hyperthyroidism. Differential diagnoses for the upper left lip lesion included idiopathic hypereosinophilic syndrome, neoplasia, or a hypersensitivity reaction to a contact, inhalant, food, or insect bite allergen. A longstanding hypersensitivity reaction was considered most likely, given the chronicity of the lesion, the peripheral eosinophilia, and eosinophilic lymphadenitis. However, parasitism, hyperthyroidism, or a paraneoplastic eosinophilia could not be excluded. Methimazole (Paladin Labs, Montreal, Quebec), 2.5 mg, PO, q12h for hyperthyroidism; atenolol (Teva Pharmaceutical Industries, Toronto, Ontario), 6.25 mg, PO, q24h for presumed secondary hyperthyroid related tachycardia, and cephalexin (Nu Pharm, Richmond Hill, Ontario), 25 mg/kg body weight, PO, q12h for 14 d for the pyoderma, were prescribed. The cat was prophylactically dewormed with praziquantel and pyrantel (Bayer, Toronto, Ontario) for possible gastrointestinal parasites and the owner was instructed to transition the cat to a hypoallergenic diet over 10 d.

Two weeks later (day 14 after initial presentation) the cat was presented for widespread cutaneous erythema and scaling and crusting of the dorsum of the neck, interscapular area, and pinnae. The swelling of the upper left lip was substantially worse and the cat had developed intermittent diarrhea. In the last 48 h the cat had become markedly pruritic. A drug reaction to methimazole was suspected as the cause of the worsening of the cutaneous lesions, pruritus, and the diarrhea. The cat remained tachycardic (270 bpm) and the prescapular, axillary, popliteal, and submandibular lymph nodes were markedly enlarged. A repeat CBC was unremarkable, and the mild increase in serum ALP activity (81 U/L) persisted. Urine collected by cystocentesis revealed a urine specific gravity (USG) of 1.034, 1 + proteinuria, and 3 + hematuria; the hematuria and proteinuria were attributed to the sampling method. The cat’s serum total T4 concentration was still elevated (124 nmol/L) but showed a downward trend.

Fine-needle aspiration cytology for multiple lymph nodes revealed a heterogeneous population of lymphocytes in a moderately hemodilute background (Figure 2). Lymphocyte cytoplasmic fragments (lymphoglandular bodies) were numerous. The results of a 200-cell differential cell count were as follows: 53% small lymphocytes, 25% large lymphocytes and 22% medium lymphocytes. Mitotic figures were frequent. Eosinophils, neutrophils, macrophages and mast cells were notable in the background. The cytological findings were suspicious for lymphoma but hyperplasia could not be excluded. Thoracic and abdominal radiographs were unremarkable. A single skin biopsy from the upper left lip lesion was submitted for histological examination. Methimazole administration was discontinued and a rapidly tapering, 5-day course of prednisolone (Rafter 8 Products, Calgary, Alberta) was initiated for the pruritus (5 mg, PO, q12h, for day 1, 5 mg, PO, q24h for days 2 and 3, 2.5 mg, PO, q24h for days 4 and 5).

Figure 2.

Cytological features of a representative lymph node aspirate from a cat diagnosed with cutaneous lymphoid hyperplasia showing a heterogeneous population of lymphocytes with some medium to large lymphocytes with an atypical appearance. A few mitotic (arrows) figures were seen. Wright’s Giemsa stain, 600 × oil immersion.

Skin biopsies revealed a diffuse, dermal infiltrate of round cells, below an extensively ulcerated surface, that exhibited moderate anisocytosis and anisokaryosis and rare tropism for the follicular epithelium (Figure 3). Mitotic figures were rare. Multicentric lymphoma with cutaneous involvement was considered most likely based on the histological findings and the suspicious cytological findings for the multiple, enlarged lymph nodes. Lomustine (Bristol-Myers Squibb, Montreal) 60 mg/m², PO, q21d, and oral prednisolone (Rafter 8 Products), 5 mg, PO, q24h therapy were initiated on day 28 as lymphadenomegaly persisted and the appearance of the upper left lip lesion was unchanged, despite resolution of the pruritus. The cat remained on atenolol for secondary hypertrophic cardiomyopathy (HCM), 5 mg, PO, q24h. L-asparginase (EUSA Pharma, Limonset, France) 400 IU/kg BW, SC, was added to the chemotherapeutic protocol on day 36.

Figure 3.

Histological features of a skin biopsy of a cat diagnosed with cutaneous lymphoid hyperplasia showing a dense dermal infiltrate of mononuclear cells, below an extensively ulcerated surface, that exhibited moderate anisocytosis and anisokaryosis and tropism for the follicular epithelium (see inset). Hematoxylin and eosin stain, 400 × and 100 × magnification.

The cat received 2 more cycles of lomustine and L-asparaginase on days 75 and 118. The lymph nodes did not appreciably decrease in size until day 89. Since discontinuing methimazole, the cat lost an additional 0.43 kg of body weight that was attributed to unmanaged hyperthyroidism. On day 126, a CBC, serum biochemistry profile, and serum total T4 activity were re-evaluated. Abnormalities were similar to when the cat was initially diagnosed with hyperthyroidism with a mild elevation in ALP (73 U/L) and a mild decrease in total calcium (2.01 mmol/Ll). The serum total T4 concentration was elevated at 186 nmol/L. Carbimazole (Western College of Veterinary Medicine Pharmacy, compounded PCCA Houston, Texas, USA), 2.5 mg, PO, q12h, for 7 d followed by 2.5 mg, PO, q8h thereafter, was prescribed for hyperthyroidism. Two weeks after initiating carbimazole therapy (day 159) the serum total T4 concentration was 70 nmol/L and the cat was clinically well with no recurrence of skin lesions or lymphadenomegaly. The carbimazole dose was subsequently increased (5 mg at 8 am, 2.5 mg at 4 pm and at 10 pm). On day 255 after initial presentation (4 mo after the last dose of chemotherapy) the cat was seen at another clinic for routine dental prophylaxis. A CBC, serum biochemistry profile, and urinalysis were unremarkable and the serum total T4 was 32 nmol/L.

The original diagnosis of multicentric lymphoma with cutaneous involvement was now considered unlikely. The left popliteal lymph node was extirpated while the cat was anesthetized and was submitted for histological examination. The histological diagnosis was normal nodal architecture with evidence of mild follicular hyperplasia. The archived, formalin-fixed, paraffin-embedded skin biopsy from the upper left lip lesion was submitted to the Leukocyte Antigen Biology Laboratory at UC-Davis for immunohistochemistry (IHC) for T- and B-cells and molecular clonality analysis for immunoglobulin heavy chain framework 2 and framework 3 (IgH2, IgH3) and kappa deleting element (KDE) for B-cell and T-cell receptor gene rearrangements. Immunohistochemistry revealed that most of the round cells were T-cells with low numbers of intermingled B-cells that formed follicle-like aggregates. Molecular clonality analysis revealed polyclonal rearrangements. Despite the monomorphic appearance of the lymphocytes, the results of the IHC and molecular clonality analysis supported a reactive lymphoid proliferation.

The owner elected to proceed with radioactive iodine (RAI) therapy as a permanent cure for the hyperthyroidism. Repeat laboratory work prior to RAI treatment was unremarkable. Thoracic radiographs were unchanged from previous plain-film radiographs. A left anterior vesicular block was detected on a 6-lead electrocardiogram so atenolol was continued. Fundoscopic examination and systemic arterial blood pressure measurement were both normal. On day 313, 3.3 mCi of Iodine-131 was administered SC and the cat was discharged 10 d later. Eighteen hours following discharge the cat died unexpectedly. A complete postmortem examination revealed subacute chylothorax, restrictive cardiomyopathy with endocardial fibrosis, a right thyroid adenoma, and lymphocytic dermatitis of the left upper lip.

Discussion

Peripheral eosinophilia, decreased creatinine and total calcium, and increased serum ALP activity have been previously documented in hyperthyroid cats (2–5). The eosinophilic lymphadenitis and the cutaneous lymphoid hyperplasia (CLH) that mimicked cutaneous lymphoma, induced or exacerbated by treatment with methimazole, were unusual aspects of this case.

Hyperthyroidism or a hypersensitivity disorder could explain the peripheral eosinophilia and mild eosinophilic lymphadenitis. Peripheral eosinophilia has been reported in a subset of hyperthyroid cats (3,6). Peripheral eosinophilia, lymphadenitis and pruritic, scaling, erythematous and edematous facial lesions have also been reported in people with autoimmune Graves’ hyperthyroidism (7,8). Resolution of the peripheral eosinophilia and the eosinophilic component of the lymphadenitis shortly after initiating treatment for hyperthyroidism suggests that hyperthyroidism was responsible for these abnormalities. However, given the chronic nature of the lip lesion in this cat, a concurrent underlying hypersensitivity to another allergen was also suspected.

A reaction to methimazole was suspected as the cause for the pruritis and self-induced excoriations of the face and neck seen shortly after initiating treatment. This is an uncommon, yet, well-known side-effect of methimazole in cats. Such signs resolve quickly when methimazole is discontinued, as was the case here (9,10).

The marked generalized lymphadenomegaly observed 2 wk after initiating treatment with methimazole was initially considered most suspicious for multicentric lymphoma with cutaneous involvement based on the cytological and histological findings. However, ultimately IHC and the polyclonal nature of the infiltrate confirmed by molecular clonality analysis on the archived, formalin-fixed, paraffin-embedded upper left lip lesion biopsies, along with the indolent clinical course, supported a benign reactive, non-neoplastic, lymphoid proliferation. Methimazole appeared to exacerbate this condition. Lymphadenomegaly as a side-effect of methimazole treatment in a cat has been reported, and there is an anecodotal report of lymphadenomegaly secondary to carbimazole treatment in a cat (11,12). Lymphadenomegaly, with or without concurrent cutaneous signs consistent with CLH, has also been seen in humans with drug-induced pseudolymphoma including cases following administration of methimazole and carbimazole (13–16). It was surprising that this cat did not react to carbimazole since carbimazole is a prodrug that is converted in vivo to methimazole. Other terms for CLH used in human medicine include cutaneous pseudolymphoma (CPL), lymphocytoma cutis, and lymphadenosis benigna cutis (14,17). Cutaneous lymphoid hyperplasia and CPL are the most commonly used terms but all refer to a heterogeneous group of benign, reactive lymphoid proliferations in the skin of humans (18). Histologically, a monomorphic lymphocyte infiltrate may be seen that can mimic cutaneous lymphoma or there may be variable numbers of medium- to large-sized lymphocytes, which often appear atypical, and may exist alongside many ordinary small lymphocytes. The presence of other inflammatory cells (e.g., histiocytes, mast cells, plasma cells, and eosinophils) is also common (13,17–18). Lymph node findings in cases with concurrent lymphadenopathy can manifest as either benign lymphoid hyperplasia or a pseudolymphoma with obliteration of normal nodal architecture (13).

Major advances have been made in human medicine with respect to immunohistochemistry and clonality studies to facilitate differentiation of CPL/CLH from cutaneous lymphoma. Immunophenotypic studies have shown that most human cases of CLH are similar to the case described and consist of a mixture of polytypic B- and T-cells, macrophages, and dendritic cells (14,17). Occasionally, the lesions are composed predominantly of either T- or B-cells (17–18). Clonality, as assessed by amplification of the T-cell receptor-gamma (TCRG) gene locus for T-cells or the immunoglobulin heavy chain (IgH) gene locus for B-cells, using either polymerase chain reaction (PCR) or a southern blot, is commonly employed in human cases of suspected CLH to rule out cutaneous lymphoma (17). With CLH, an indolent clinical course is expected and clonality analysis reveals a polyclonal infiltrate of lymphocytes; this is in contrast to “true” cutaneous lymphoma where a monoclonal infiltrate is seen and metastasis and progression within a year are expected (17–20). However, this distinction is not absolute. Cases of monoclonal CLH that do not progress and some apparent cases of polyclonal CLH that evolve into “true” cutaneous lymphoma have been reported (14,17). In the latter cases of CLH, the dominant polyclonal T- or B-cell infiltrate may harbor an occult monoclonal B- or T-cell population. These cases are referred to as “clonal CLH” (21).

In companion animals, the term cutaneous lymphocytosis (CL) has been proposed for cases analogous to pseudolymphoma in humans (22,23). The clinical, morphological, and immunohistochemical features of CL in 23 cats and in 8 dogs have been described (22,23). However, the current use of the term CL is problematic since most cases of CL in dogs and cats, on subsequent clonality testing, have turned out to be monoclonal (23–24). Also, in many cases, there is evolution to widespread systemic involvement, often with cytological features typical of either low-grade or high-grade lymphoma, after years of stable or very slowly progressive disease (22–24). The apparent transformation from a reactive to a neoplastic population may be due to initial misdiagnosis (i.e., failure to detect an occult clonal population) or genuine transformation to or replacement by lymphoma (17). The case described may represent a true case of feline CLH due to the benign clinical course and the polyclonal nature of the infiltrate. It is also possible that an occult clonal population was not detected and that with more time, progression to lymphoma would have occurred.

In veterinary medicine, there are fewer available immunophenotypic markers and the state of knowledge regarding immunophenotyping and clonality testing is less advanced than in human medicine. With more widespread commercial availability of clonality testing and increased testing of infiltrative cutaneous lymphoid lesions more cases of CLH may be diagnosed in cats and dogs. However, while clonality studies were helpful in our case they should never be the sole criterion used to differentiate CLH from cutaneous lymphoma, as recent studies in cats, dogs, and humans suggest monoclonality does not always indicate an aggressive clinical course (17,22–23).

Recently, it has been proposed that the cutaneous lymphoproliferative diseases, including CLH and cutaneous lymphoma, represent points on a continuum of lymphoproliferative disorders with monoclonal cutaneous lymphoma representing the extreme malignant end of the spectrum and polyclonal CLH the benign end of the spectrum (19–20). Cases of monoclonal CL in dogs and cats or monoclonal CPL in humans may lie somewhere between polyclonal CLH and monoclonal cutaneous lymphoma (17). It is also possible that cases of polyclonal CPL consist of 2 groups: those with and those without an occult clonal T- or B-cell population at the time of diagnosis with the former representing a subset more likely to develop lymphoma (17). It is also possible that cases of CL and CPL do indeed start as a benign reactive lymphoid infiltrate and evolve slowly in a stepwise process to lymphoma as a result of immune dysregulation.

Cutaneous lymphoid hyperplasia in humans has been attributed to a diverse array of causes; however, in the majority of cases the cause is unknown (idiopathic form). Drug-induced forms, including cases secondary to methimazole and to various anticonvulsants; infection-induced forms (herpes virus, Borrelia burgdorferi, Bartonella henselae); and cases secondary to exposure to contact allergens or from introduction of foreign antigens into the skin (e.g., arthropod bites, vaccinations, tattoo ink) are also reported (14,15,25). These associations have lead to the suggestion that CLH represents an exaggerated immune response to a newly encountered antigen. A single case of pseudolymphoma induced by phenobarbital administration in a cat has been been reported and methimazole was implicated in another case of CL in a cat (22,26). In our case, CLH in the upper left lip lesion was likely induced by a chronic hypersensitivity reaction to an unknown allergen and administration of methimazole is suspected to have exacerbated this condition.

Treatment of CLH relies on elimination of the offending allergen or drug and/or immunosuppression (15,18–21,23,25,26). Prednisone given in conjunction with chlorambucil or lomustine provided good to complete resolution in some cats initially diagnosed with CL and this combination was effective in this case (18). No response was seen in any of the canine cases of suspected CL to long-term steroid therapy or other therapies but despite this, the clinical course remained indolent (19). Other treatments including cyclosporine, doxycycline or tetracycline, cryosurgery, laser ablation, surgical excision, and radiotherapy have been reported to be effective in some human idiopathic CLH cases (15,21,23).

This case illustrates that reactive cutaneous lymphoid proliferations in cats can mimic lymphoma. In this case, the indolent clinical course and the clonality studies provided support for the diagnosis of CLH. The sudden unexpected death of this cat was not related to the diagnosis of CLH but was likely secondary to a fatal arrhythmia due to cardiomyopathy.

Cutaneous lymphoid hyperplasia needs to be kept in mind by clinicians and pathologists in order to avoid a misdiagnosis of cutaneous lymphoma that could lead to premature euthanasia or overly aggressive chemotherapy. Immunohistochemistry and clonality studies should be part of the diagnostic work-up for cases of suspected cutaneous lymphoma in cats and dogs to help rule in possible reactive lymphoid proliferations. However, immunohistochemistry and clonality studies should not be completely relied upon for predicting the clinical course of the disease and close follow-up is advised in all cases of possible CLH because cases can progress to lymphoma (23). CVJ