Feline cutaneous lymphoma

Introduction

Lymphoma is the most diagnosed neoplasm in the cat, representing around 50% of all tumors seen in the species. From this overall percentage cutaneous lymphoma accounts for 0.2-3.0% of cases, highlighting the fact that it is a rare but malignant neoplasm 1. To date only a handful of reports on feline cutaneous lymphoma cases have been published, whereas in dogs and people it is a well-recognized and documented disease 23.

As per the classification system used in human medicine, cutaneous lymphoma is separated into two main types: epitheliotropic and non-epitheliotropic. The classification indicates if the epidermis, dermis and adnexa have been infiltrated with neoplastic lymphocytes (epitheliotropic) versus just the dermis and subcutis (non-epitheliotropic) 4 – although to be more exact, feline cutaneous epitheliotropic lymphoma (CETL) does not involve the adnexal glands 2. 

Cutaneous epitheliotropic lymphoma is a subset of cutaneous T-cell lymphoma (CTCL) and is further divided into mycosis fungoides, Sézary syndrome (a leukemic variant) and pagetoid reticulosis 234; in cats this latter condition has yet to be described 2. Cutaneous non-epitheliotropic lymphoma (CNEL) is the more common condition in cats and includes indolent T-cell lymphoma (or cutaneous lymphocytosis), diffuse T-cell lymphoma, T-cell-rich large B-cell lymphoma, and lymphoplasmacytic lymphoma 1. In more recent years there has been acknowledgement of feline CNEL subgroups: cutaneous lymphoma at injection sites, tarsal lymphoma, and CNEL associated with a fracture site. They all lack epitheliotropism, hence their subgrouping, but clinical and pathological features discern them as unique disease entities 156. In dogs and humans CETL is the more common condition and has consequently influenced research in the cat 1.

Etiology and predisposing factors

In general, cutaneous lymphoma affects older cats (with a median age of onset at 10 years) and there are no apparent breed or sex predispositions 37. As a corollary to the rarity of feline cutaneous lymphoma, the etiology is still not well understood 37. For more common feline lymphomas (e.g., mediastinal and multicentric) there is an evident association between viruses (e.g., feline leukemia virus [FeLV] and feline immunodeficiency virus [FIV]) and lymphoma, but such a relationship has not been established for feline cutaneous lymphoma 1238. However, one study 9 employed a PCR technique to highlight integrated FeLV provirus in tumor DNA from a cat with CNEL. This finding, along with other suggestions in the field, does not conclusively exclude FeLV and/or other feline viruses as possible causes 3, and advanced molecular diagnostics may be required to definitively identify any link 9.

Other contributory factors have also been proposed. Whilst there is limited information available for the extremely rare dermatopathy follicular mucinosis (a presumed immune-mediated disease), two cats have been described with the disease which both subsequently developed CETL 710. Although it is not possible to confirm the evolution of follicular mucinosis to CETL, follicular mucinosis in cats may be a predisposing factor or an indicator of more severe disease to come. Furthermore, pre-existing chronic dermatitis in dogs and people with CTCL has been researched as a possible predisposing factor to skin cancer, although (in dogs, people and cats) transformation of inflammatory skin disease (e.g., atopic dermatitis) to neoplasia of the integument has not been proven 3.

Predisposing factors for feline CNEL have also been evaluated: a study of 17 cats with CNEL at injection sites was linked to a case report of a cat with fracture site associated CNEL. The study and case report were linked as both authors proposed chronic inflammation (post injection and fracture respectively) to be a possible cause for malignant transformation 15. Chronic inflammation is well documented in humans as a potential nidus for B-cell lymphoma, and other applicable nidus examples include surgery, trauma, metallic implants and viral infections 1.

Clinical presentation and diagnosis

Cutaneous epitheliotropic lymphoma

Most reported cases of feline CETL are the mycosis fungoides form 3. When the neoplasm occurs, it has a slow progression and at the time of presentation there may be no significant systemic signs and/or abnormalities on routine hematology and biochemistry 311. Moreover, cats with Sézary syndrome typically present with intense pruritus and lymphadenomegaly, and on cytology have circulating neoplastic lymphocytes (leukemia) 211.

Dermatological examination may identify lesions which can be localized or generalized, arising anywhere on the body. Skin lesions include exfoliative erythroderma, patches, plaques, erosions and ulcers (Figure 1), and lesions at mucocutaneous junctions (Figure 2) and in the oral cavity (Figure 3) 312. As a result, skin lesions are comparable in appearance to those caused by parasites (e.g., Demodex spp.), infection (e.g., dermatophyte species) and allergic skin disease (e.g., eosinophilic granuloma complex) 712. However, one case report noted a cat with feline CETL where the clinical summary was remarkably different, with acute onset of nodular lesions starting on the tail 13. This further complicates the list of differential diagnoses, as clinically CETL seems to have an unpredictable manifestation. 

Diagnosis is based on histopathology. For a definitive diagnosis, small-to-medium or medium-to-large lymphocytes with tropism for the epidermis must be identified 12, but the pathologist may also variably describe Pautrier’s microabscessation, spongiosis and apoptosis of keratinocytes, a mixed inflammatory reaction and orthokeratosis and parakeratosis of the epidermis 3. Histopathology alone does not reveal any specific features clearly distinguishing mycosis fungoides from Sézary syndrome 711.

It is widely accepted that CETL is nearly always T-cell in origin, but beyond that little research exists for the more precise immunophenotype and immunopathology of feline CETL. Some pathologists believe that the immunophenotype of feline CETL most closely resembles mycosis fungoides in people, as the T-cells involved are thought to be helper T-cells (CD4 cells) 3. However, this has been challenged by the identification of perforin in neoplastic T-cells from a cat with CETL. Perforin is a pore-forming protein stored in the cytoplasmic granules of cytotoxic T-cells (CD8 cells) and has a role in targeted cellular killing. Unfortunately, in this case CD8 expression was not reviewed, but by comparison to relevant human studies it was suggested that the T-cells were most likely CD8 cells 13. Essentially more work is required to elucidate the immunophenotype of feline CETL.

Cutaneous non-epitheliotropic lymphoma

Clinically, CNEL presents as non-pruritic solitary or multifocal plaques and/or nodules that are often ulcerated (Figure 4 and 5); erythema, crusting and scaling occur sporadically (Figure 6) 214 There are no listed predilection sites but a literature search suggests lesions can develop on peripheral areas, at common injection sites (cutaneous lymphoma at injection sites), on the tarsus (cutaneous lymphoma of the tarsus) and where there was a previous fracture (CNEL associated with a fracture site) 15614. These can manifest as edematous swelling, a nodule, or a subcutaneous mass 156. Cats with these more atypical lesions represent distinct forms of cutaneous lymphoma, reinforcing the point that it should always be a differential diagnosis for any swelling, nodule or mass within the dermis or subcutis, irrespective of bodily location. Moreover, it is unclear how long lesions are observed before diagnosis; this can range from possibly months to years versus rapid progression, where lymphatic and systemic dissemination can occur within days 125614.

CNEL has been defined as a diffuse, poorly differentiated lymphocytic lymphoma 2, thus making it difficult to separate from other cutaneous round cell tumors, histiocytic proliferative disorders and advanced mycosis fungoides. More recent case reports that include histopathology describe an infiltrative growth of well-differentiated lymphocytic tumor cells, medium-to-large in size, between the corium and subcutis 514. Additionally, for the CNEL subgroups, shared histopathological findings include neoplastic cells arranged in sheets with high mitotic indices and areas of necrosis 156.

CNEL is a disease of either T- or B-cell immunophenotype and this has been neatly demonstrated by two separate studies. In one study more cats were classified as having B-cell lymphoma than T-cell 6, whereas the opposite occurred in the other report 1. Immunophenotyping was achieved via immunohistochemistry and the specific identification of surface antigens expressed by the neoplastic lymphocytes: CD3 by T-cells and CD79 by B-cells 16. However, CNEL predominantly has a T-cell phenotype, and diagnosis of a B-cell tumor is generally considered extremely rare 167.

Treatment – CETL

Evidence-based medicine for the treatment of CETL in cats is scarce and as a result has been adapted from canine or human medicine. First-line treatment in people with localized, superficial or early lesions is skin-directed, namely topical therapy, phototherapy, and photodynamic and radiation therapy. Systemic chemotherapy is considered second-line treatment and reserved for advanced cases and/or palliation 15. The following is an outline of current and evolving treatment modalities for humans and dogs and, with the caveat that most drugs discussed are not licensed for use, could theoretically be applied to cats.

Topical therapies

Topical corticosteroids are used initially in people with CTCL and work by inhibiting lymphocyte endothelial binding and inducing apoptosis; they have an excellent response rate (82 to 94%) 15. Cats presenting with early lesions could be treated in the same way, the primary aim being to achieve remission or, at the very least, provide symptomatic relief 7. Topical chemotherapy in the form of mechlorethamine has also been used successfully in both people and dogs: in a human study 75% of participants (155 in total) with early lesions completely responded, and for dogs with patch and plaque-stage lesions a good response has been described 715. Application of tretinoin gel has been anecdotally useful in a few dogs with CTCL. This treatment was inspired by the synthetic retinoid bexarotene used in humans 715, as topical retinoids selectively bind and activate RX receptors, regulating cellular differentiation, proliferation and apoptosis 15. Another potential topical requiring further research in both human and veterinary medicine is imiquimod, an immunomodulatory agent that has shown antitumor activity in basal cell carcinoma 15. 

Phototherapy, photodynamic therapy and radiation 

In human medicine phototherapy (ultraviolet light radiation) is a more researched technique than photodynamic therapy for treatment of CTCL, and various studies support its efficacy in early cases 15. Whereas in veterinary medicine, despite remaining in its infancy, photodynamic therapy has produced exciting results for the remission of cutaneous squamous cell carcinoma and small, non-invasive tumors in cats. The technique involves administration of a tumor-localizing photosensitizer (topical, oral or intravenous) followed by light activation, resulting in tissue damage 16. 

Electron beam radiation therapy uses low-energy electrons requiring specialized technology. In human medicine, when the entire skin surface is treated, it is known as total skin electron beam (TSEB) radiation, and is actively used in people with superficial lesions, plaques and tumors that have not responded to topical therapy 15. Initial response to TSEB in selected cases is good; however, relapses are common and there appears to be a correlation between CTCL stage and relapse rate: the more advanced the CTCL, the higher the rate of relapse 1517. In veterinary medicine TSEB therapy has been investigated for several years and applied to chosen patients; one case report described its use in a dog with chemoresistant CTCL, which induced remission for 19 months 18, but there are no reports of its use in cats. 

Disease-modifying agents and systemic chemotherapy

In human CTCL, as the neoplasm progresses, malignant T-cells proliferate and produce a cytokine imbalance, which is the target of disease-modifying agents. Such agents are used as second-line treatment when skin-directed therapy has failed and/or the disease is more advanced 15. In veterinary medicine the use of synthetic retinoids has been explored and proven favorable due to a discovered non-overlapping toxicity when combined with standard cytotoxic therapy. Systemic retinoids are natural or synthetic analogues of vitamin A that exert profound effects on cellular growth, maturation and differentiation 19. In one study dogs with CTCL were treated with a mixture of synthetic retinoids (isotretinoin and etretinate) with a 42% response rate recorded 20. Unfortunately, their use has not accelerated due to the lag phase between initiation of therapy and clinical effect, and also cost 20.

The most researched treatment option for CETL in veterinary medicine is systemic chemotherapy, and there are various single agent versus multiagent protocols. This is partly because the disease is often already at an advanced stage by the time of diagnosis, requiring a more aggressive approach. Neoplastic lymphocytes remain sensitive to oral corticosteroids, which are reported to have clinical and palliative effects 219. Unfortunately, corticosteroid monotherapy rarely provides long-term effective treatment; they are more efficacious when incorporated into multiagent chemotherapy protocols 19.

Chemotherapy protocols for feline CETL have not yet been determined; in dogs, single agent chemotherapy protocols are most commonly reported. An initial pilot study monitoring the response of 7 dogs, including 5 dogs with epitheliotropic lymphoma, to lomustine revealed that all dogs achieved complete remission (2 months to over a year) 21. Two retrospective studies published in 2006 evaluated the response of dogs with CTCL to lomustine, and in both approximately 80% of dogs achieved a measurable response. Inevitably the high response rates were accompanied by side effects (myelosuppression and increased hepatic enzymes), and in addition some of the dogs in these studies had previously received other chemotherapy agents and/or were being treated with corticosteroids 192223. Therefore, despite these studies giving veterinary oncologists and dermatologists the confidence to use lomustine for canine CTCL, more research is required to establish protocols for patients showing chemoresistance versus those that are chemonaive 19. Due to its efficacy in dogs, lomustine has been recommended for use in cats with CETL 3.

Single agent chemotherapy with L-asparaginase or doxorubicin has also been investigated, but so far only in dogs. A study observing pegylated (polyethylene glycol capsule) L-asparaginase included seven dogs with CTCL, and clinically all initially appeared to improve, but long-term the responses were partial and short-lived 24. Another prospective study assessed the toxicosis and response of dogs with various neoplasms to pegylated doxorubicin; nine of the cases had CTCL, and three of them went into complete remission (with a median of 90 days). The report noted that pegylated doxorubicin had a markedly reduced cardiotoxic and myelosuppressive effect compared to free doxorubicin, but remains impractical due to cost 25.

In human medicine multiagent chemotherapy protocols for CTCL have been described, but are not considered superior based on survival rates 15. Combination chemotherapy has been trialed for dogs with CTCL (with various permutations of prednisolone, vincristine, cyclophosphamide and doxorubicin) and a moderate response noted, with survival times varying from 2 to 6 months 719. COP (cyclophosphamide, vincristine and prednisolone) and CHOP (cyclophosphamide, doxorubicin, vincristine and prednisolone) protocols have also been used in a small case series of canine CTCL and one case report of feline CETL, with moderate success 471319.

Other options

Other treatment options for feline cutaneous lymphoma include placental lysate, surgical excision of solitary lesions, and fibronectin (local and intravenous) 36. Fibronectin is a glycoprotein with the antineoplastic ability to opsonize target cells for macrophagic and monocytic destruction 3. Despite publication of different treatment modalities, none have been supported by an adequate case series with reliable treatment response and survival rates 6, and all require further research in the field. 

Multimodal and symptomatic treatment

A multimodal treatment approach does not seem to improve survival rates in people with CTCL, but it does successfully increase complete response rates. Nonetheless, combining treatment modalities with non-overlapping toxicities (e.g., excision of a mass with adjunctive chemotherapy) is a rational protocol and should be adopted for chosen veterinary patients. Importantly, symptomatic treatment (e.g., analgesia and antibiotics) should be initiated parallel to the conventional treatment plan to optimize quality of life, especially in palliative cases 19.

Treatment – CNEL

As CNEL is less common in dogs, insufficient research has been performed to create a foundation for treatment recommendations, and consequently there is no standard of care for treating feline CNEL. Despite this, relevant publications exist which help make an evidence-based decision for therapy options. Two case reports of cats with CNEL detail the therapeutic approach elected and response experienced: one cat treated with lomustine went into complete remission for approximately 4 months, and the other received a modified CHOP protocol, achieving maintenance for a total of 4 weeks 514. 

A retrospective study of 23 cats with tarsal CNEL reported that treating affected cases with combination therapy (radiation and chemotherapy) or surgery (with or without chemotherapy) led to a significantly longer mean survival time compared to a monotherapy approach (corticosteroids or chemotherapy), at 316 days versus 155 days 6. Cats being treated for CNEL should of course also receive multimodal and symptomatic treatment (as above) when appropriate.

Prognosis

With little information available for feline cutaneous lymphoma it is difficult to comment on prognosis. In humans, survival rates are determined by the type of CTCL diagnosed; for example, Sézary syndrome has a worse prognosis than mycosis fungoides (33% versus 89-93% 5 year survival rate) 3 although in general the prognosis of CTCL in people is considered good 15. Dogs diagnosed with early CTCL and treated appropriately may survive beyond 12 months, whereas dogs diagnosed with an advanced version may only live for another 6 months, regardless of response to treatment 19. In cats the median survival time reported for CETL is approximately 10 months, but for CNEL the prognosis is graver, with a median survival time of 4 to 8 months 37. 

Conclusion

Cutaneous lymphoma is a rare neoplasm in cats, and little has been published about the manifestation, treatment and prognosis of the two forms. However, when managing a case of suspected feline cutaneous lymphoma early diagnosis is key to an improved prognosis; whenever possible skin biopsy should be performed near the start of the diagnostic investigation. For treatment, an outline of standard of care has not been defined and undeniably requires further research, but on review cats have generally been treated with systemic chemotherapy, and where appropriate a multimodal approach should be adopted to optimize treatment response. Additionally, alongside the main therapeutic, symptomatic treatment should be initiated to enhance quality of life, particularly for cats being treated palliatively.