Patient characteristics, histopathological findings and outcome in 97 cats with extranodal subcutaneous lymphoma (2007-2011).

Original Article

DOI: 10.1111/vco.12081

Patient characteristics, histopathological findings and outcome in 97 cats with extranodal subcutaneous lymphoma (2007–2011) K. Meichner1 and W. von Bomhard2 1

Clinic of Small Animal Medicine, Center of Clinical Veterinary Medicine, Ludwig-Maximilian-University of Munich, Munich, Germany 2 Specialty Practice for Veterinary Pathology, Munich, Germany

Abstract

Keywords feline, histopathology, lymphosarcoma, skin

This study describes epidemiologic, clinical, macro- and microscopic tumour characteristics and outcome in 97 cats with subcutaneous lymphoma, an uncommon variant of feline extranodal lymphoma. Middle-aged (median 11 years), male (60.8%), Domestic Shorthair cats (89.7%) were commonly affected. Most tumours presented as a painless, firm, subcutaneous nodule or mass, with predilection to the lateral thoracic or abdominal wall, and the interscapular region. Deep subcutaneous invasion with extension into superficial or underlying tissues, extensive central areas of necrosis and peripheral inflammation were characteristic histopathological findings. Prevalence of retroviral infection was low. Local relapses after therapy were common (43.5%), and 32.2% had distant involvement later in course. Median overall survival was 148 days. Subcutaneous lymphoma should be considered a rare but important differential diagnosis for a subcutaneous mass in cats. Tumours show an aggressive biological behaviour. Treatment options including prognosis should be investigated in further studies.

Introduction

Correspondence address: K. Meichner Department of Population Health and Pathobiology, College of Veterinary Medicine, North Carolina State University, 1052 William Moore Drive Raleigh, NC 27607, USA e-mail: [email protected]

Lymphoma is the most common haematopoietic tumour in cats.1 – 6 From the 1970s to the early 1990s, the multicentric and mediastinal forms were the most common presentations for malignant lymphoma in cats.6 – 10 During that period, up to 80% of cats with lymphoid malignancies were tested positive for feline leukaemia virus (FeLV) antigen.7,11 – 14 Prevalence of progressive FeLV infection decreased in the general cat population due to routine screening of cats, separation of progressively infected cats and preventative FeLV vaccination programs since the 1990s. Consequently the number of FeLV-associated lymphomas declined as well.15 – 20

© 2014 John Wiley & Sons Ltd

Following the gastrointestinal location, extranodal lymphomas are presently the second most common diagnosed feline lymphoid tumours. The majority of cats with this presentation is FeLV antigen-negative.10,16,17,20,21 The term ‘extranodal’ describes lymphomas with solitary, non-lymphoid organ involvement, like the nasal cavity, kidneys, central nervous system (CNS) or eyes.22 Lymphoma affecting the skin is generally rare in cats,6 reported in 1.7% of all lymphoma cases23 and in 3.36% of cats with extranodal lymphomas.24 There are a few reports about epidermal and dermal forms,25 – 29 and even less is known about subcutaneous involvement. To the authors’ knowledge only few cases of lymphoma affecting the subcutaneous tissue have been described in the

1

2 K. Meichner and W. von Bomhard

literature so far.21,23,30 – 33 Generally, those cases were included in studies together with different lymphoma subtypes.21,23,31 – 33 In the past few years, the authors repeatedly diagnosed and managed cases of solitary lymphoma in cats located in the subcutaneous tissue. Detailed information about this distinct subtype is missing so far. Thus, the aim of this study was to collect data about epidemiologic, clinical, macro- and microscopic tumour characteristics, and to evaluate the biologic behaviour of subcutaneous lymphoma in cats.

Patients

Material and methods

Histopathology

Design of the study

Specimens of all selected cases were reviewed using light microscopy on hematoxylin and eosin (H&E) stained sections, and re-evaluated by a boardcertified veterinary pathologist (WvB). Tumour resection margins of full tumour samples were assessed depending on eligibility. Margins were described as ‘incomplete’ if tumour cells extended to the surgical edge, and as ‘complete’ if 5 mm of normal tissue was observed around the tumour. Necrosis interferes with immunohistochemistry.35 Considering the necrotic character of the neoplasms, immunohistochemical analysis was therefore conducted on 22 selected cases that contained sufficient neoplastic cells and only small necrotic zones. Consecutive 5 μm sections were mounted on positively charged slides (Superfrost plus, Menzel, Braunschweig, Germany). Following heat antigen retrieval CD3 (DAKO, Glostrup, Denmark) and CD20 (Thermo Scientific, Waltham, MA, USA) were used at dilutions of 1:400 and 1:1300, respectively. Antibody binding was detected by a horseradish peroxidase labelled detection system (DCS, Hamburg, Germany). Hyperplastic canine lymph nodes were used for positive controls. CD20 labelled follicular lymphocytes positive, whereas CD3 labelled paracortical lymphocytes. For negative controls buffer replaced the primary antibody.

This study represents a retrospective review of client-owned cats diagnosed with subcutaneous lymphoma at a specialized veterinary histopathologic diagnostic laboratory in Munich, Germany. The laboratory serves Germany and parts of Austria, Italy and Denmark. Data was derived from cases submitted for histopathology (full tumour samples and/or biopsies) by reviewing laboratory records (submitting forms) and/or the electronic database between January 2007 and February 2011. Additional informations, particularly regarding history, clinical staging tests, treatment and outcome, were derived from a survey which was sent as a questionnaire to each submitting veterinarian with an eligible case.

Selection of cases The cases labelled in the computer database with a diagnosis of ‘lymphoma’ or ‘round cell neoplasm’ confined to the subcutaneous tissue were re-evaluated by a board-certified veterinary pathologist (WvB). Cats in which the subcutaneous tumour was identified to be a lymph node, cats in which the clinical course was consistent with secondary subcutaneous involvement (e.g. multicentric lymphoma) and cats without a definitive diagnosis of lymphoma (e.g. undifferentiated round cell neoplasia) were excluded from analysis. Patients were considered only once to prevent bias to potential follow-up histopathological examinations.

Cases were categorized according to: year of diagnosis, breed, age at diagnosis, sex, clinical signs, duration of disease, location and characterization of the tumour, stage,34 substage, feline immunodeficiency virus (FIV) and FeLV status, date of onset and treatment regime, outcome including response to treatment, date and location of local or distant relapse, and follow-up information including cause of death. If necessary, additional information was obtained from submitting veterinarians via telephone interviews.

Outcome analysis For outcome analysis, overall remission rate (ORR), progression-free survival (PFS) and overall survival (OS) were evaluated for cats receiving treatment.

© 2014 John Wiley & Sons Ltd, Veterinary and Comparative Oncology, doi: 10.1111/vco.12081

Extranodal subcutaneous lymphoma in cats

The design of the study prevented the exact determination of complete and partial remission (PR). Therefore, response to treatment is reported as an ORR, which was defined as the percentage of treated patients that achieved complete or PR within the first 4 weeks after initiation of therapy. Complete remission (CR) was considered as resolution of all clinically detectable disease and PR as a decrease in tumour size more than 50% but less than 100%. Stable disease (SD) was the definition of less than 50% tumour reduction, no reduction or less than 25% increase in size of all measurable tumours. Progressive disease (PD) was considered as over 25% increase in size of all measurable tumours, or the appearance of new lesions. New lesions were confirmed by cytology and/or histopathology. PFS was defined as the time in days from initiation of treatment until the first date when criteria for PD were met, or death from any cause occurred. OS was defined as the time in days from initiation of treatment until death from any cause occurred.

Statistical analysis Descriptive statistical analyses were performed in terms of patient characteristics, and macro- and microscopic tumour characteristics. Kaplan–Meier product limit analysis was used for PFS and OS analysis. Differences in outcome were assessed by the log-rank test. Cats were censored from PFS analysis if PD had not occurred, if they were still alive at the time of analysis or if they were lost to follow-up. Cats were censored from OS analysis if they were still alive at the time of analysis, or if they were lost to follow-up. Statistical significance was set at a P value ≤0.05. All calculations were performed using PASW statistics (SPSS), version 18.0.

Results Selected cases Of 261 796 cases submitted to the laboratory from the years 2007 to 2011, 23 810 submissions were feline cutaneous or subcutaneous masses. A total of 111 cats were identified from that database. Fourteen cases were excluded after reviewing

3

histopathology and additional clinical information, and 97 cases clearly fulfilled the criteria of a subcutaneous lymphoma and were included in this study.

Patient characteristics History and signalment Among 97 cats with subcutaneous lesions, in 3 cats an injection-site reaction, and in 24 cats a fibrosarcoma was the presumptive diagnosis by the submitting veterinarian. Duration of the mass before diagnosis was recorded in 61 cats. According to owners’ observation, the mass had been present for a few days in 5 cats (8.2%), for a few weeks in most cats (41/61, 67.2%) and several months in 15 cats (24.6%). Patient characteristics are summarized in Table 1. The age of affected cats at diagnosis ranged from 1 to 17 years, with a mean of 10.8 years and a median of 11 years, respectively. Age was not recorded for five cats. The population of 97 cats was composed of 46 (47.4%) neutered males, 32 (33.0%) neutered females, 13 (13.4%) intact males and 6 (6.2%) intact females. The male to female ratio was 1.6:1. Represented breeds included 87 Domestic Shorthairs, 5 Maine Coon, 2 Carthusian and Persian each and 1 Norwegian Forest cat. Clinical data In 70 cases, clinical data was completed by return of the questionnaires. Besides a physical examination performed in all 70 cats, one or more additional clinical staging tests (blood work, determination of FeLV and FIV status, thoracic and/or abdominal radiographs and abdominal ultrasound) were performed in 46 cats. Stage, substage and retroviral status are depicted in Table 1. Substage b patients (12 cats, 17.1%) presented with one or more of the following clinical signs: lethargy (10 cats), anorexia (2 cats) and weight loss (5 cats); four cats with tumours affecting the extremities presented with swollen legs and lameness.

Tumour characteristics Macroscopic findings Distribution of tumour sizes (5 cm), measured for the largest diameter, was nearly



Table 1. Characteristics and outcome of cats with subcutaneous lymphoma

Variable (n) Median age in years (range) Sex Male Female

59/97 (60.8%) 38/97 (39.2%)

Breed Domestic Shorthair Maine Coon Carthusian Persian Norwegian Forest Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Substage a Substage b

87/97 (89.7%) 5/97 (5.1%) 2/97 (2.1%) 2/97 (2.1%) 1/97 (1.0%) 52/70 9/70 6/70 1/70 2/70 58/70 (82.9%) 12/70 (17.1%)

Median PFS in days (95% CI)

Median OS in days (95% CI)

83 (29–137) 101 (43–158) P = 0.493

168 (124–211) 137 (7–266) P = 0.923

108 (65–150) 34 (0–72) P = 0.047

173 (131–214) 39 (34–43) P = 0.002

11 (1–17)

Retroviral status FIV positive FeLV positive Tumour size 5 cm

33/89 (37.1%) 29/89 (32.6%) 27/89 (30.3%)

115 (43–187) 110 (44–175) 61 (2–119) P = 0.628

234 (125–342) 168 (108–227) 135 (32–237) P = 0.166

Immunophenotype B-cell T-cell

15/22 (68.2%) 7/22 (31.8%)

77 (67–86) 108 (34–181) P = 0.190

138 (128–147) 214 (0–571) P = 0.560

83 (18–147)

137 (80–194)

Tumour margins Infiltrated Clean

1/46 (2.2%) 3/46 (6.5%)

1/31 30/31

similar for different categories at diagnosis (Table 1). Size was not recorded for eight cats. Figure 1 illustrates distribution among different subcutaneous locations. The most common site was the lateral aspect of the thoracic wall (31/95, 32.6%), with the right side (16/95, 16.8%) almost equally affected compared to the left side (15/95, 15.8%). In 29 cats (30.5%) the tumour was located on the abdominal wall (including the flank region), with an even distribution for the left and right side (10/95 each, 10.5% each), followed in frequency by the ventral and inguinal abdominal region (9/95, 9.5%). Twenty-one cats (22.1%) had the tumour located in the caudal dorsal neck and interscapular region, respectively. The extremities were affected in nine cats, with predominance for the hind limbs

(8/95, 8.1%). Other, less common locations were the fore limbs (1/95, 1.1%) and the ventral neck region (1/95, 1.1%). Four of 95 cats (4.2%) presented with more than one tumour at different subcutaneous sites. Location was not recorded in two cats. Two cats had concurrent diagnoses of a subcutaneous lymphoma and a fibrosarcoma at different subcutaneous locations. The most common clinical presentation was a painless (23/28, 82.1%), firm (25/28, 89.3%), subcutaneous nodule or mass (87/89, 97.8%) (Fig. 2). Multi-nodular presentation was observed in 18 of 89 (20.2%) cases, and 4 of 89 cats (4.5%) showed a diffuse infiltration of the skin. Upon palpation, description of the lesions was variable and included: moveable (17/32, 53.1%), fixed (15/32, 46.9%),



5

Figure 3. Feline subcutaneous lymphoma, H&E, ×1.25.

The subcutaneous tissue is effaced by sheets of neoplastic round cells. Note central tumour necrosis.

Figure 1. Pie chart diagram depicting distribution in percent (%) of subcutaneous lymphoma in 95 cats. Four cats had more than one tumour. ‘Others’ included the tail base and the ventral neck.

Figure 2. A 10-year-old, male-neutered Domestic Shorthair cat with a subcutaneous lymphoma (→ arrow) located on the ventral abdominal skin (area is shaved): subcutaneous mass with focal alopecia, erythema and central crusting (→ arrow). Subcutaneous swelling and erythema involving the inguinal area (left) are also observed.

well-demarcated (21/29, 72.4%) and invasive (27.6%, 8/29) tumours. One of the authors (K. M.) also observed ulceration, alopecia, erythema and peritumoural swelling in several cases (Fig. 2). Microscopic findings In 25 of 97 patients, diagnosis was obtained by true cut or incisional biopsy, whereas 72 cats

had histopathology performed on excisional biopsy specimens. All tumours had similar histopathological features (Figs 3–6). A poorly demarcated sheet of atypical round cells effaced the deep dermis and the panniculus (Figs 3 and 4). The cells had a small rim of deeply basophilic cytoplasm, and large centrally located round to oval nuclei with one or multiple nucleoli (Fig. 6). Anisocytosis and anisokaryosis were mild to moderate. The mitotic index ranged from 2 to 8 mitoses per high power field (×40). There were scattered macrophages with phagocytized nuclear debris, creating a starry sky phenomenon (Fig. 4). All masses had extensive central areas of coagulative necrosis (Figs 3–5). Within the necrotic centre, few blood vessels remained intact, surrounded by a slim zone of neoplastic round cells (Fig. 5). Peripheral to the neoplastic population there was necrotic adipose tissue and there were perivascular aggregates of mature lymphocytes, few plasma cell and macrophages. Lymphocytes occasionally form lymphoid aggregates and follicles. The neoplastic population frequently extended into the superficial dermis or the underlying skeletal muscle. According to the World Health Organization (WHO) classification, the diagnosis of an extranodal diffuse large cell lymphoma was established.36 Among specimens with eligible margins following tumour excision, margins were infiltrated by neoplastic cells in 1 cat, whereas 30 cats had clean margins (Table 1). The immunophenotype could be determined for all 22 lymphomas (Table 1), and in all cases more than 80% of the atypical lymphoid cells reacted



Figure 4. Feline subcutaneous lymphoma, H&E, ×4.


Sheets of deeply basophilic neoplastic round cells efface the subcutis. Scattered are macrophages with phagocytized nuclear debris. The sheets of neoplastic cells are flanked by aggregates of mature small lymphocytes.

Sheets of atypical large lymphoblasts with one prominent nucleolus. Note scattered mitotic figures.


Extensive necrosis and collar lesions. Two intact vessels are surrounded by cuffs of atypical lymphoblasts.

with either CD3 or CD20 (Figs 7 and 8). Fifteen (68.2%) lymphomas were of B-cell origin and seven (31.8%) were T-cell origin.

Treatment and outcome Information in treatment and outcome was available for 69 cats. The majority of cats had surgery alone (53/69, 76.8%), or surgery combined with other modalities (5/69, 7.3%) consistent of: prednisolone (2 cats), prednisolone plus chlorambucil (1 cat), L-asparaginase, cyclophosphamide, vincristine, prednisolone (L-COP)-based chemotherapy (1 cat) and COP-based chemotherapy combined with definitive radiation therapy (1 cat). Response rates after therapy for surgically

Figure 7. Feline subcutaneous B-cell lymphoma. Immunostaining for CD20, ×40.

removed tumours were considered meaningless and thus not reported. Four cats (5.8%) had only chemotherapy: prednisolone (2 cats), Lasparaginase, cyclophosphamide, doxorubicin, vincristine, prednisolone (L-CHOP) and lomustine (1 cat), L-asparaginase, dexamethasone, chlorambucil and cyclophosphamide (1 cat). One of these four cats showed a partial response, two experienced SD and one cat developed PD. Seven cats (10.1%) recsived no specific anti-tumour treatment. Median PFS for all 62 cats receiving specific treatment was 101 days [95% confidence interval (CI) 58–144 days] (Fig. 9). Seven cats died between 7 and 362 days after start of treatment. Twelve cats got censored from PFS analysis because they were still alive and in their first remission (6 cats, 9.7%, censored between day 202 and 1186), or lost to follow-up (6 cats, 9.7%,



7

Figure 8. Feline subcutaneous T-cell lymphoma. Immunostaining for CD3, ×40.

Figure 10. Kaplan–Meier OS estimation for cats with

subcutaneous lymphoma (n = 62). MST = 148 days. Bars indicate censored cases.

Figure 9. Kaplan–Meier PFS estimation for cats with subcutaneous lymphoma (n = 62). Median PFS = 101 days. Bars indicate censored cases.

censored between day 0 and 11). Among different treatment subgroups, cats with only surgery (53/62) experienced a median PFS of 73 days (95% CI 49–97 days), cats with surgery and additional therapy (chemo-/radiotherapy) had a mean PFS of 756 days (range 101–1186 days, median was not reached) and cats with only chemotherapy had a median PFS of 108 days (95% CI 13–138 days). Median OS for all 62 cats receiving specific treatment was 148 days (95% CI 107–189 days) (Fig. 10). In 39 cats (62.9%) death or euthanasia was related to the neoplastic disease, whereas 3 cats died due to non-tumour-related reasons (heart and chronic renal disease). Twenty cats

got censored from analysis because they were still alive (9 cats, 14.5%, censored between day 202 and 1569) or lost to follow-up (11 cats, 17.7%, censored between day 0 and 301). Among different treatment subgroups, cats with only surgery (53/62) experienced a median OS of 135 days (95% CI 121 to 148 days), cats with surgery and additional therapy (chemo-/radiotherapy) had a mean OS of 786 days (range 326–1186 days, median was not reached) and cats with only chemotherapy had a median OS of 148 days (95% CI 138–180 days). Seven cats without specific anti-tumour treatment lived between 0 and 218 days. Patients presented in substage a showed a significantly longer PFS and OS (108 and 173 days, respectively) compared to substage b patients (34 and 39 days, respectively) (P = 0.047 and 0.002, respectively). No additional variables with impact on outcome could be identified (Table 1).

Follow-up Twenty-seven of 62 cats (43.5%) had a relapse at the original location, and 10 cats (16.1%) developed new tumours at a different subcutaneous site without evidence of reoccurrence of the primary tumour. Twenty cats (32.2%) were diagnosed with new lesions at distant sites (lymphoid and nonlymphoid tissue) later in course.



Nine cats (14.5%) experienced long-term survival: seven of them lived more than 2 years, and six of nine cats lived without evidence of tumour recurrence after initial therapy.

Discussion Tumours of the lymphoid systems and tumours affecting the skin and subcutaneous tissue are the most common neoplasms in cats.1 – 6,37 – 39 However, lymphoid tumours affecting the feline skin are rare.6,23,24,37 In this study, subcutaneous lymphoma presented 0.41% of all submitted feline skin masses, which is comparable with a previous large survey where lymphoid tumours presented less than 1% of all feline skin tumours.37 In contrast to other subcutaneous tumours, like feline injection-site sarcoma (FISS) which presents 17.88% of all feline skin tumours,40 subcutaneous lymphoma can thus be generally regarded rare in cats. Patient characteristics in this study regarding gender, age at diagnosis and affected breeds are comparable with previous feline lymphoma reports,3,16 – 18,20 – 24,41,42 with older cats, male cats and domestic shorthaired cats being more commonly affected. Interestingly, important clinical features of subcutaneous lymphomas are also characteristics of FISS. First, the clinical presentation of a firm, painless, subcutaneous mass may resemble FISS. Second of all, the localizations of the lymphomas on the lateral aspect of the thoracic or abdominal wall, and in the interscapular region are also predilection sites for FISS,43 – 45 and present typical injection sites in cats. Thirdly, not only grossly, even subgrossly subcutaneous lymphomas mimic FISS due to its poorly demarcated character, the large central necrosis leading to cavitation, and the location in the subcutis with extension into the upper dermis or skeletal muscle.43,44 However, on higher magnification a monotonous round cell population consistent with neoplastic lymphoblasts became apparent. The diagnosis of a malignant lymphoma could thus be established and a FISS was ruled out. Subcutaneous lymphomas in cats may therefore suggest clinically a FISS or an injection reaction, and cytology or histopathology is crucial for the exact distinction.

Though lymphomas are considered to exfoliate well upon fine-needle aspiration in general, to the authors’ experience cytological examination is more problematic in subcutaneous lymphomas. Specimens frequently revealed low-cellularity or non-diagnostic samples, most likely due to the large necrotic centre of this neoplasm. Tumour necrosis is a common feature of many different neoplasms, and is considered as a result of fast growth and insufficient angiogenesis.46 Within necrotic centres of lymphoid tumours intact neoplastic cells may form a rim of about 20 cells around functional vessels. This phenomenon is called ‘collar lesion’.47 The authors consider this an important histopathological feature for the diagnosis of subcutaneous lymphomas in cats. Collar lesions were observed in all masses examined in this study. In various human lymphoid neoplasms, foci of necrosis or apoptosis were supposed to be associated with cytotoxicity exhibited by infiltration of reactive48,49 or neoplastic50 – 52 cytolytic lymphocytes. If this might be attributable for necrosis in feline subcutaneous lymphoma needs further investigation. Inflammatory infiltrates, lymph follicles and necrotic adipose tissue consistently flanked the tumour. While mild inflammation is a common feature of neoplasms,53 the peripheral inflammatory component examined here was quite pronounced. Interestingly, this is yet another diagnostic feature that FISS and feline subcutaneous lymphoma do share,44,54,55 and could be suggestive of an antitumour host immunity response, or of a preceding local trauma. The latter is supported by the fact that most tumour localizations present common subcutaneous injection sites. Chronic inflammation, e.g. triggered by injection of certain drugs or vaccines, is supposed to act as a tumour promoter in FISS.43,56 Cats seem to be predisposed to tumour development at sites of chronic inflammation. Interestingly, two cats in this study had subcutaneous lymphoma and FISS concurrently at different locations. This finding might be coincidental, because injection-site sarcomas are relative common feline tumours. However, one case series reported different lymphoma types in cats with a history of FISS and suspected similar risk factors for the development of both neoplasms.31 Another report of a cat with a subcutaneous



angioinvasive lymphoma also hypothesized a possible causality between the tumour and a rabies vaccine the cat received 6 months prior at the same location.30 Probably a reactive, longstanding, polyclonal lymphocytic inflammation might progress to a monoclonal, neoplastic population later in course, which was shown so far in humans.57 In contrast to FISS,44,55,58 none of the lymphomas in this study contained macrophages with phagocytized, vaccine-like product. It is nevertheless tempting to speculate that the same pathogenesis underlies those two tumour entities, and whether both share similarities in oncogenesis needs further investigation. Owing to the lymphoblastic character of the cells and the mild to moderate anisocytosis, the diagnosis of a malignant lymphoma could already be established on H&E examination. However, in a diagnostic setting it may not always be possible to reach a final diagnosis based on H&E alone. In those cases, it will be necessary to differentiate subcutaneous lymphomas from other round cell neoplasms that occur in the subcutis of cats, such as histiocytic sarcomas, mast cell tumours or plasma cell tumours. Immunohistochemistry proved useful to verify the histogenesis of the neoplasms in this study and established readily the immunophenotype. The majority of tumours was large B-cell lymphomas, with less large lymphoblastic T-cell lymphomas. In contrast to dogs, where T-cell lymphomas are associated with a worse outcome,59,60 immunophenotype failed to reveal a prognostic significance in our study. This correlates with previous studies on feline lymphoma.10,18,32 However, immunohistochemistry was not performed on all specimens in this study. Thus, additional research on the prognostic role of the immunophenotype in subcutaneous lymphoma is necessary. A low percentage of cats tested positive for feline retroviruses. None of feline skin lymphomas in other reports was associated with progressive FeLV infection,20,21,24 which implies a negligible role of FeLV in subcutaneous lymphomagenesis, and most likely reflects the decreasing prevalence of FeLV among cats in most countries and among cats with lymphomas in particular.16 – 21,61 In contrast to other studies, the majority of cats (82.9%) in this study appeared healthy at the

9

time of diagnosis.10,18,21,24,62 This is not surprising: compared to other types of lymphomas involving the gastrointestinal tract, CNS or nasal cavity, a subcutaneous mass can be easily recognized by the owners, and cats presumably presented in an early, local stage. In addition, general health is not expected to be compromised due to the subcutaneous location of the mass. Nonetheless, those patients that presented with clinical signs showed a worse outcome. This finding is supported by some previous studies,10,18,63 whereas other studies failed to show a correlation.21,62 However, due to the retrospective nature of this study a potential bias cannot be excluded, as some owners might have declined therapy due to the cat’s poorer general condition. Although the majority of cats initially presented with local disease and with substage a, more than one third of the patients developed distant involvement later in course. Staging was definitely heterogeneous and not-standardized among the present studied population, and for some patients stage migration can be expected if a more sensitive test would have been used. Thus, more advanced stages might have been potentially underestimated at diagnosis, leading to a false increase of the systemic spread rate later in course. However, our findings are comparable with other extranodal locations, like nasal lymphoma, where patients also develop distant tumour spread despite the initial local character of the disease.64,65 Patients in this study showed a high local relapse rate (43.3%), and the outcome regarding PFS (101 days) and OS (148 days) was worse compared to other studies, which reported remission and survival durations between 164 and 264 days, and 266 and 571 days, respectively.18,21,24 These cited studies included different anatomical types, and the majority of cats were treated with a combination chemotherapy protocol. Most cats in this study had surgery alone, the minority in combination with other modalities. Surgery might be a good treatment option for a local restricted, low-grade, indolent lymphoma, but may not be the single therapy of choice for a high-grade lymphoid neoplasm like subcutaneous lymphoma in this study. Even patients that underwent surgery with clean tumour margins


10

K. Meichner and W. von Bomhard

on histopathological examination showed a fast disease progression (median 83 days). Furthermore, relapses were not only locally restricted, but also a high proportion of cats had distant relapses involving other subcutaneous (16.1%) or systemic (32.2%) sites. These findings suggest not only a locally, but also a systemic aggressive disease, which also might account for the poor outcome. However, outcome results of this study should be interpreted with caution as data were obtained retrospectively and outcome information was provided by means of questionnaires. Prospective studies including standardized treatment protocols, recheck and follow-up regimes are necessary to better evaluate outcome and prognosis of cats with subcutaneous lymphoma. External beam radiotherapy induced rapid responses of macroscopic tumours, but the tumour progressed in some patients beyond the radiation field (data not shown). This was also reported in the past for one cat with subcutaneous lymphoma treated with definitive radiation therapy.33 Likewise, response of patients initially receiving only chemotherapy was poor, and also for patients treated with a combination chemotherapy protocol following relapse after surgery (data not shown). This might implicate that chemotherapy alone is insufficient in gross disease control. In contrast, a previous study reported CR with a multi-agent, continuous protocol, lasting up to 9 months in two cats with lymphoma involving the subcutaneous tissue.21 However, due to lack of standardized protocols and the small number of patients treated only with chemotherapy in this study, efficacy of chemotherapy in controlling feline subcutaneous lymphoma should be interpreted with caution. Further studies including larger numbers of cats and using standardized chemotherapy protocols are needed to draw definite conclusions. At this point, it seems that local therapy alone and chemotherapy alone are insufficient treatment options. Thus, a multimodality approach may be necessary to control subcutaneous lymphoma in cats. This is supported by the small subgroup of patients (5 cats) receiving combined modality therapy. They showed the best outcome, with three of them experienced long-term survival in their first

remission (data not shown). However, the number of cats was very small. Further studies including larger numbers of cats are needed to clearly prove a benefit of a multimodality treatment. This study has several limitations. There could be inaccuracies in data retrieval especially for parameters like staging results, treatment and outcome information, given the retrospective nature of the study and the collection of these parameters by means of questionnaires. Patients were evaluated and treated by multiple different institutions; clinical evaluation, staging, treatment and follow-up were non-standardized and possibly inconsistent. In addition, the small numbers of cats in some subgroups limited the power of the study. Randomized studies will be helpful to exclude the potential bias of owners’ and clinicians’ selection of treatment types influenced by costs, duration and the cat’s stage and condition. This might help to guide treatment protocols and to establish a standard of care for cats with subcutaneous lymphoma. Immunohistochemistry was only applied to a subset of cases and clonality was not assessed. Further studies are therefore necessary to subclassify and fully characterize subcutaneous lymphoma in cats. Additional research in this field is currently conducted by a separate group (P. Roccabianca and V. Affolter, personal communication). If this lymphoma subtype has a distinct geographic distribution needs further investigation by obtaining data from other countries. Findings of this study conclude that subcutaneous lymphoma in cats may mimic feline injection-site sarcoma by clinical and macroscopic presentation, and should be considered a rare but important differential diagnosis for a subcutaneous mass in a cat. There is evidence that there is a high risk of relapse, both locally and distantly. The role of chemotherapy is unclear, but it should be considered for control of systemic or microscopic disease. Future studies will be helpful to investigate whether multimodality treatment is superior to surgery, chemotherapy or radiotherapy alone.

Acknowledgements The authors thank the referring veterinarians who responded to the survey, Fredie Elbert Davis III



for assistance with language editing, and Prof. Dr Johannes Hirschberger for his intellectual input. 12.

Conflict of interest The authors did not use any source of funding and indicate that they have no affiliations or involvement with any organization with a financial interest in the subjects discussed in this paper.

13.

14.

References 1. Jarrett WF, Crawford EM, Martin WB and Davie F. A virus-like particle associated with leukemia (Lymphosarcoma). Nature 1964; 202: 567–569. 2. Dorn CR. The epidemiology of cancer in animals. California Medicine 1967; 107: 481–489. 3. Dorn CR and Taylor DO. Survey of animal neoplasms in Alameda and Contra Costa countries, California. I. Methodology and description of cases. Journal of the National Cancer Institute 1968; 40: 295–305. 4. Dorn CR and Taylor DO. Survey of animal neoplasms in Alameda and Contra Costa countries, Carlifornia. II. Cancer morbidity in dogs and cats from Alameda County. Journal of the National Cancer Institute 1968; 40: 307–318. 5. Hardy WD Jr. Feline oncoretroviruses. In: The Retroviridae 2. JA Levy Ed., New York, Plenum Press, 1993: 109–180. 6. Hardy WD Jr. Hematopoietic tumors of cats. Journal of the American Animal Hospital Association 1981; 17: 921–940. 7. Francis DP, Cotter SM, Hardy WD Jr and Essex M. Comparison of virus-positive and virus-negative cases of feline leukemia and lymphoma. Cancer Research 1979; 39: 3866–3870. 8. Jackson ML, Wood SL, Misra V and Haines DM. Immunohistochemical identification of B and T lymphocytes in formalin-fixed, paraffin-embedded feline lymphosarcomas: relation to feline leukemia virus status, tumor site, and patient age. Canadian Journal of Veterinary Research 1996; 60: 199–204. 9. Day MJ. Review of thymic pathology in 30 cats and 36 dogs. Journal of Small Animal Practice 1997; 38: 393–403. 10. Vail DM, Moore AS and Volk LM. Feline lymphoma (145 cases): proliferation indices, cluster of differentiation, immunoreactivity, and their association with prognosis in 90 cats. Journal of Veterinary Internal Medicine 1998; 12: 349–354. 11. Cotter SM, Hardy WD Jr and Essex M. Association of feline leukemia virus with lymphosarcoma and

15.

16.

17.

18.

19.

20.

21.

22.

11

other disorders in the cat. Journal of the American Veterinary Medical Association 1975; 166: 449–454. Francis DP, Essex M, Cotter SM, Gutensohn N, Jakowski R and Hardy WD Jr. Epidemiologic association between virus-negative feline leukemia and the horizontally transmitted feline leukemia virus. Cancer Letters 1981; 12: 37–42. Reinacher M. Infektionen mit dem felinen Leuk¨amie-Virus (FeLV) bei sezierten Katzen. Kleintierpraxis 1987; 32: 65–72. Shelton GH, Grant CK, Cotter SM, Gardner MB, Hardy WD Jr and Digiacomo RF. Feline immunodeficiency virus and feline leukemia virus infections and their relationships to lymphoid malignancies in cats: a retrospective study (1968–1988). Journal of Acquired Immune Deficiency Syndromes 1990; 3: 623–630. Cotter SM. Changing epidemiology of FeLV. In: Proceedings of the 15th Annual ACVIM Forum, Lake Buena Vista, FL, USA, 1997: 22–25. Teske E, Van Straten G, Van Noort R and Rutteman GR. Chemotherapy with cyclophosphamide, vincristine, and prednisolone (COP) in cats with malignant lymphoma: new results with an old protocol. Journal of Veterinary Internal Medicine 2002; 16: 179–186. Louwerens M, London CA, Pedersen NC and Lyons LA. Feline lymphoma in the post-feline leukemia virus era. Journal of Veterinary Internal Medicine 2005; 19: 329–335. Brenn SH, Couto SS, Craft DM, Leung C and Bergman PJ. Evaluation of P-glycoprotein expression in feline lymphoma and correlation with clinical outcome. Veterinary and Comparative Oncology 2008; 6: 201–211. Stuetzer B, Simon K, Lutz H, Majzoub M, Hermanns W, Hirschberger J, Sauter-Louis C and Hartmann K. Incidence of persistent viraemia and latent feline leukaemia virus infection in cats with lymphoma. Journal of Feline Medicine and Surgery 2010; 13: 81–87. Meichner K, Kruse DB, Hirschberger J and Hartmann K. Changes in prevalence of progressive feline leukaemia virus infection in cats with lymphoma in Germany. Veterinary Record 2012; 171: 348. doi:10.1136/vr.100813. Simon D, Eberle N, Laacke-Singer L and Nolte I. Combination chemotherapy in feline lymphoma: treatment outcome, tolerability, and duration in 23 cats. Journal of Veterinary Internal Medicine 2008; 22: 394–400. Vail DM. Feline lymphoma and leukemia. In: Withrow and MacEwen’s Small Animal Clinical Oncology. 5th edn., SJ Withrow, DM Vail and RL


12

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

K. Meichner and W. von Bomhard

Page Eds., St. Louis, W.B. Saunders, 2013: 638–653. Gabor LJ, Malik R and Canfield PJ. Clinical and anatomical features of lymphosarcoma in 118 cats. Australian Veterinary Journal 1998; 76: 725–732. Taylor SS, Goodfellow MR, Browne WJ, Walding B, Murphy S, Tzannes S, Gerou-Ferriani M, Schwartz A and Dobson JM. Feline extranodal lymphoma: response to chemotherapy and survival in 110 cats. Journal of Small Animal Practice 2009; 50: 584–592. Day MJ. Immunophenotypic characterization of cutaneous lymphoid neoplasia in the dog and cat. Journal of Comparative Pathology 1995; 112: 79–96. Fox LE. Feline cutaneous and subcutaneous neoplasms. Veterinary Clinics of North America: Small Animal Practice 1995; 25: 961–979. Fontaine J, Heimann M and Day MJ. Cutaneous epitheliotropic T-cell lymphoma in the cat: a review of the literature and five new cases. Veterinary Dermatology 2011; 22: 454–461. Schick RO, Murphy GF and Goldschmidt MH. Cutaneous lymphosarcoma and leukemia in a cat. Journal of the American Veterinary Medical Association 1993; 203: 1155–1158. Tobey JC, Houston DM, Breur GJ, Jackson ML and Stubbington DA. Cutaneous T-cell lymphoma in a cat. Journal of the American Veterinary Medical Association 1994; 204: 606–609. Rogers DG and Aliano VA. Metastatic angioinvasive lymphoma (lymphomatoid granulomatosis) in a cat. Journal of Veterinary Diagnostic Investigation 2009; 21: 390–394. Madewell BR, Gieger TL, Pesavento PA and Kent MS. Vaccine site-associated sarcoma and malignant lymphoma in cats: a report of six cases (1997–2002). Journal of the American Animal Hospital Association 2004; 40: 47–50. Gabor LJ, Canfield PJ and Malik R. Immunophenotypic and histological characterisation of 109 cases of feline lymphosarcoma. Australian Veterinary Journal 1999; 77: 436–441. Elmslie RE, Ogilvie GK, Gillette EL and McChesney-Gillette S. Radiotherapy with and without chemotherapy for localized lymphoma in 10 cats. Veterinary Radiology & Ultrasound 1991; 32: 277–280. Mooney SC and Hayes AA. Lymphoma in the cat: an approach to diagnosis and management. Seminars in Veterinary Medicine and Surgery (small animal) 1986; 1: 51–57. Bain BJ, Clark DM and Wilkins BS. Infection and reactive changes. In: Bone Marrow Pathology. 4th

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

edn., BJ Bain, DM Clark and BS Wilkins Eds., West Sussex, Wiley-Blackwell, 2010: 100–165. Valli VE, Jacobs RM, Parodi AL, Vernau W and Moore PR. Histological classification of hematopoietic tumors of domestic animals. In: World Health Organization International Histological Classification of Tumors of Domestic Animals. 2nd edn., FY Schulman Ed., Washington, DC, Armed Forces Institute of Pathology, American Registry of Pathology, 2002: 35–38. Miller MA, Nelson SL, Turk JR, Pace LW, Brown TP, Shaw DP, Fischer JR and Gosser HS. Cutaneous neoplasia in 340 cats. Veterinary Pathology 1991; 28: 389–395. Brodey RS. Canine and feline neoplasia. Advances in Veterinary Science and Comparative Medicine 1970; 14: 309–354. Nielsen SW. Classification of tumors in dogs and cats. Journal of the American Animal Hospital Association 1983; 19: 13–60. Hauck ML. Tumors of the skin and subcutaneous tissues. In: Withrow and MacEwen’s Small Animal Clinical Oncology. 5th edn., SJ Withrow, DM Vail and RL Page Eds., St. Louis, W.B. Saunders, 2013: 305–320. Kristal O, Lana SE, Ogilvie GK, Rand WM, Cotter SM and Moore AS. Single agent chemotherapy with doxorubicin for feline lymphoma: a retrospective study of 19 cases (1994–1997). Journal of Veterinary Internal Medicine 2001; 15: 125–130. Milner RJ, Peyton J, Cooke K, Fox LE, Gallagher A, Gordon P and Hester J. Response rates and survival times for cats with lymphoma treated with the University of Wisconsin-Madison chemotherapy protocol: 38 cases (1996–2003). Journal of the American Veterinary Medical Association 2005; 227: 1118–1122. Hendrick MJ and Brooks JJ. Postvaccinal sarcomas in the cat: histology and immunohistochemistry. Veterinary Pathology 1994; 31: 126–129. Doddy FD, Glickman LT, Glickman NW and Janovitz EB. Feline fibrosarcomas at vaccination sites and non-vaccination sites. Journal of Comparative Pathology 1996; 114: 165–174. Shaw SC, Kent MS, Gordon IK, Collins CJ, Greasby TA, Beckett LA, Hammond GM and Skorupski KA. Temporal changes in characteristics of injection-site sarcomas in cats: 392 cases (1990–2006). Journal of the American Veterinary Medical Association 2009; 234: 376–380. Wicki A and Christofori G. The angiogenic switch in tumorigenesis. In: Tumor angiogenesis. Basic Mechanisms and Cancer Therapy. 1st edn., D



47.

48.

49.

50.

51.

52.

53. 54. 55.

56.

Marme and N Fusenig Eds., Heidelberg, Springer, 2007: 68–86. Valli VE and Ward JM. T-cell and NK-cell neoplasms. In: Veterinary Comparative Hemotopathology. 1st edn., VE Valli Ed., Ames, Blackwell Publishing, 2007: 275–365. ten Berge RL, Oudejans JJ, Dukers DF, Meijer JW, Ossenkoppele GJ and Meijer CJ. Percentage of activated cytotoxic T-lymphocytes in anaplastic large cell lymphoma and Hodgkin’s disease: an independent biological prognostic marker. Leukemia 2001; 15: 458–464. Morice WG, Kurtin PJ and Myers JL. Expression of cytolytic lymphocyte-associated antigens in pulmonary lymphomatoid granulomatosis. American Journal of Clinical Pathology 2002; 18: 391–398. Jaffe ES, Chan JK, Su IJ, Frizzera G, Mori S, Feller AC and Ho FC. Report of the Workshop on Nasal and Related Extranodal Angiocentric T/Natural Killer Cell Lymphomas: definitions, differential diagnosis, and epidemiology. The American Journal of Surgical Pathology 1996; 20: 103–111. Chan AC, Ho JW, Chiang AK and Srivastava G. Phenotypic and cytotoxic characteristics of peripheral T-cell and NK-cell lymphomas in relation to Epstein-Barr virus association. Histopathology 1999; 34: 16–24. Guitart J. Subcutaneous lymphoma and related conditions. Dermatologic Therapy 2010; 23: 350–355. Coussens LM and Werb Z. Inflammation and Cancer. Nature 2002; 420: 860–867. Hendrick MJ. Feline vaccine-associated sarcomas. Cancer Investigation 1999; 17: 273–277. Couto SS, Griffey SM, Duarte PC and Madewell BR. Feline vaccine-associated fibrosarcoma: morphologic distinctions. Veterinary Pathology 2002; 39: 33–41. Kass PH, Barnes WG Jr, Spangler WL, Chomel BB and Culbertson MR. Epidemiologic evidence for a causal relation between vaccination and fibrosarcoma tumorigenesis in cats. Journal of the American Veterinary Medical Association 1993; 203: 396–405.

13

57. Kapanci Y and Toccanier MF. Lymphomatoid granulomatosis of the lung. An immunohistochemical study. Applied Pathology 1983; 1: 97–114. 58. Hendrick MJ, Goldschmidt MH, Shofer FS, Wang YY and Somlyo AP. Postvaccinal sarcomas in the cat: epidemiology and electron probe microanalytical identification of aluminum. Cancer Research 1992; 52: 5391–5394. 59. Teske E, van Heerde P, Rutteman GR, Kurzman ID, Moore PF and MacEwen EG. Prognostic factors for treatment of malignant lymphoma in dogs. Journal of the American Veterinary Medical Association 1994; 205: 1722–1728. 60. Dobson JM, Blackwood LB, McInnes EF, Bostock DE, Nicholas P, Hoather TM and Tom BD. Prognostic variables in canine multicentric lymphosarcoma. Journal of Small Animal Practice 2001; 42: 377–384. 61. Gabor LJ, Jackson ML, Trask B, Malik R and Canfield PJ. Feline leukaemia virus status of Australian cats with lymphosarcoma. Australian Veterinary Journal 2001; 79: 476–481. 62. Hadden AG, Cotter SM, Rand W, Moore AS, Davis RM and Morrissey P. Efficacy and toxicosis of VELCAP-C treatment of lymphoma in cats. Journal of Veterinary Internal Medicine 2008; 22: 153–157. 63. Mooney SC, Hayes AA, MacEwen EG, Matus RE, Geary A and Shurgot BA. Treatment and prognostic factors in lymphoma in cats: 103 cases (1977 1981). Journal of the American Veterinary Medical Association 1989; 194: 696–702. 64. Haney SM, Beaver L, Turrel J, Clifford CA, Klein MK, Crawford S, Poulson JM and Azuma C. Survival analysis of 97 cats with nasal lymphoma: a multi-institutional retrospective study (1986–2006). Journal of Veterinary Internal Medicine 2009; 23: 287–294. 65. Sfiligoi G, Th´eon AP and Kent MS. Response of nineteen cats with nasal lymphoma to radiation therapy and chemotherapy. Veterinary Radiology & Ultrasound 2007; 48: 388–393.


Clinical characteristics of 26 patients with primary extranodal Hodgkin lymphoma.

A patient with pneumopericardium secondary to extranodal T-cell lymphoma.

Histopathological difficulties in an adolescent lymphoma patient.

Patient characteristics, prognostic factors and outcome of dogs with high-grade primary mediastinal lymphoma.

Clinical characteristics and treatment of 69 patients with extranodal natural killer T-cell lymphoma.

Rapidly progressive renal failure in a patient with extranodal non-Hodgkin's lymphoma.

Clinical findings, diagnostic test results, and treatment outcome in cats with spontaneous hyperadrenocorticism: 30 cases.

Extranodal lymphoma with peripheral nervous system involvement in a dog.

Clinical characteristics of 95 patients with ocular adnexal and uveal lymphoma: treatment outcomes in extranodal marginal zone subtype.

Echocardiographic findings in 11 cats with acromegaly.

Endoscopic Findings of Gastric Extranodal Marginal Zone B-Cell Mucosa-Associated Lymphoid Tissue Lymphoma.

Vanishing bile duct syndrome in a Hodgkin's lymphoma patient with fatal outcome despite lymphoma remission.

Histopathological findings in a critically ill patient with avian influenza A (H7N9).

Subcutaneous sarcoidal granuloma underlying porokeratosis in a patient with sarcoid-lymphoma syndrome.

Extranodal lymphoma with an unusual constellation of features.

Extranodal pulmonary marginal zone B-cell lymphoma.

Multifocal extranodal lymphoma: A case report.

Extranodal lymphoma of the posterior auricle.

Primary extranodal variant of non-Hodgkin's lymphoma.

Extranodal T-cell lymphoma of nasal cavity.

Tumour angiogenic factor associated with subcutaneous lymphoma.

Comparison of ultrasonographic findings in cats with and without azotaemia.

Can clinical signs, clinicopathological findings and abdominal ultrasonography predict the site of histopathological abnormalities of the alimentary tract in cats?

Guillain-Barre Syndrome in a Patient with Primary Extranodal Intestinal Non-Hodgkin's Lymphoma: Paraneoplastic, Drug Induced or Coincidental?