Lymphoma, although rare, is the most common haematopoietic neoplasm encountered in horses and can occur at any age, with horses 4–10 years more commonly affected. Lymphoma can be classified into multicentric, alimentary, mediastinal, cutaneous and solitary. Clinical signs are typically nonspecific until the disease has progressed to end-stage at which time clinical signs reflect function of organ(s) involved. Horses with the cutaneous form of lymphoma typically present with multifocal skin lesions and no other clinical signs. Like the nonspecific clinical signs of lymphoma, results of complete blood count (CBC) and serum biochemistries are not often helpful with diagnosis, but lymphoma should be considered if anaemia, hyperfibrinogenaemia, hyperproteinaemia and hypoalbuminaemia are observed without a clear indication of infectious disease. Identification of neoplastic lymphocytes during cytological examination of a body cavity effusion can confirm the presence of lymphoma. Typically, ante mortem confirmation of lymphoma is made through histopathological examination of a biopsy or cytological examination of a fine needle aspirate of a suspected lesion. Observation of compression or destruction of normal tissue architecture by invading neoplastic cells during histological examination is indicative of lymphoma. Additional diagnostics that may improve our knowledge of equine lymphoma include detection of hormone receptors, immunophenotyping, and immunohistochemical analysis for tumour proliferation rates. Prognosis of horses with lymphoma depends on the form of lymphoma and stage at which the horse is presented but death is the common outcome of this disease. Current treatment options are surgical excision, radiation or administration of chemotherapeutic drugs. Treatment of horses with lymphoma can result in palliation and occasional resolution of this disease. Equine oncology is in its infancy, but through further documentation of horses with lymphoma and their response to therapy, a clearer understanding of the disease process and effective treatment of this neoplasm can be developed.
Lymphoma, also known as lymphosarcoma or malignant lymphoma, is a haematopoietic neoplasm arising from lymphoid tissue that includes lymph nodes, spleen and gut-associated lymphoid tissue. Terminology regarding this neoplasm is confusing. Historically the term leukaemia was used to denote a tumour of lymphoid origin. Leukaemia is defined as a cancer arising from the cells of the bone marrow which can result in neoplastic cells in the general circulation. There are leukaemias for every white cell line present in the bone marrow and myelogenous or lymphocytic leukaemias are types of leucocytic leukaemia. Lymphoid leukaemias are the most common type of leucocytic leukaemia. When there is bone marrow affected with lymphoma the terms leukaemic lymphoma or lymphoma with leukaemia are often used to describe this manifestation of the disease.
In human medicine lymphoma is classified as either Hodgkin or non-Hodgkin lymphoma based on a cytological examination of an aspirate or biopsy sample. Although both arise from lymphoid tissue, Hodgkin lymphoma is distinguished from non-Hodgkin lymphoma based on the presence of Reed-Sternberg cells, a giant cell usually derived from B lymphocytes. Non-Hodgkin lymphoma is the most common form of lymphoma identified in domestic animals. Similar to the situation in human medicine following identification of lymphoma, veterinary oncologists attempt to further categorise non-Hodgkin lymphoma by immunophenotyping neoplastic lymphocytes as B cell, T cell or NK cell. This classification allows for diagnosis, disease classification, prognosis prediction, therapeutic selection and monitoring of disease progression.
Incidence and epidemiology
Lymphoma was first reported in a horse in 1858 and is now the most commonly diagnosed haematopoietic neoplasm in horses worldwide (Theilen and Madewell 1979; van den Hoven and Franken 1983). The overall incidence of lymphoma is approximately 1.3–2.8% of all equine tumours and has a prevalence of 0.002–0.5% in the equine population (Savage 1998; Schneider 2003). There is no breed or sex predilection and any age of horse can be affected, but a majority of reported cases are in horses age 4–10 years old. No specific risk factors have been identified and, although there appears to be no genetic predisposition for the development of lymphoma, presence of the neoplasm in an aborted fetus and 2 foals less than 1 year of age have been reported (Tomlinson et al. 1979; Dewes and Blakeley 1980; Haley and Spraker 1983; Seahorn et al. 1988; Savage 1998). Previous reports speculating that malignant transformation of a retrovirus may be a cause of equine lymphoma are questionable because of failure to fulfil Koch's postulates (Meyer et al. 2006).
Equine lymphoma is classified into the following clinical syndromes: multicentric or generalised, alimentary, mediastinal, cutaneous and solitary tumours of extranodal sites. Although clinical signs reflect the function of the organ involved and the degree and duration of involvement, common clinical signs for all forms of equine lymphoma include weight loss, depression, lethargy, oedema of the ventral portion of the body wall or distal limb, recurrent fever and lymphadenopathy if peripheral lymph nodes are involved (Neufeld 1973; van den Hoven and Franken 1983; Sweeney and Gillette 1989; Carlson 1995; Savage 1998; McClure 2000; Knottenbelt 2003; Schneider 2003; Meyer et al. 2006; Munoz et al. 2009). Clinical signs of lymphoma typically develop insidiously, but can occur acutely depending on organ involvement. Unfortunately, the diagnosis of lymphoma for most affected horses is late in the course of the disease because of the insidious nature of lymphoma and the lack of pathognomonic clinical signs.
Since IgM deficiency has been associated with equine lymphoma, measurement of serum IgM concentration was claimed to aid diagnosis (Dopson et al. 1983; Furr et al. 1992; Ansar Ahmed et al. 1993; Perkins et al. 2003). People with B cell non-Hodgkin lymphoma typically have decreased production of the majority of immunoglobulin subclasses and this is suspected as the cause for decreased IgM in horses with lymphoma but is unproven (Biggar et al. 2009). One study of horses with lymphoma indicated that, for horses with IgM concentration ≤60 mg/dl, the test had a sensitivity of 50% and specificity of 35% for diagnosis of lymphoma (Ansar Ahmed et al. 1993). However, Perkins et al. (2003) found that when IgM of normal, fit horses was compared to that of horses with lymphoma, an IgM ≤23 mg/dl was more indicative of IgM deficiency. This new value for IgM decreased the sensitivity of the test to 23% but the specificity improved dramatically to 88% (Perkins et al. 2003). Although this new value for normal concentration of serum IgM increases the likelihood of an erroneous ruling out of lymphoma, it is still not a reliable screening test for the diagnosis of equine lymphoma.
Fine-needle aspiration or biopsy (incisional or excisional) of suspected lesions is the preferred method for diagnosis of lymphoma. Tissue samples allow for not only histological diagnosis but also for categorising the lymphoma into B or T cell origin, determining the proliferation rate and presence of hormone receptors. Histological characteristics of lymphoma that distinguish it from lymphoid hyperplasia include compression or destruction of normal tissue architecture, a single population of cells with unorganised chromatin pattern, and variably sized and shaped nucleoli and parafollicular atrophy (Fig 3) (Meyer et al. 2006; de Bruijn et al. 2007; Tournquist 2008). Currently, a histological diagnosis of equine lymphoma is usually the end of an investigation, but further classification, as is done in human oncology, might allow for a more refined prognosis, strategic therapy and therapeutic monitoring. Immunophenotyping is a technique in which lymphoma cells that bare a specific antigen particular to a cell lineage are exposed to marked antibodies for that lineage resulting in a reaction that identifies the lymphoma type (Kelley and Mahaffey 1998; Gudgin and Erber 2005). Immunophenotyping of neoplastic lymphocytes in horses with lymphoma indicates that multicentric and alimentary lymphomas are predominately T cell origin; mediastinal are almost always T cell origin and cutaneous are of either T cell origin or T cell-rich B cell lymphoma (Asahina et al. 1994; Kelley and Mahaffey 1998; Gavazza et al. 2003; Taylor et al. 2006; de Bruijn et al. 2007; Munoz et al. 2009). This may be unfortunate as T cell origin lymphomas are a more aggressive type of lymphoma and carry a poorer prognosis when compared to B cell lymphomas; however, further investigation into the disease in the horse is warranted (Vail and Young 2002). Another diagnostic tool used by human and small animal oncologists to aid in establishing a prognosis and to monitor response to therapy is determination of tumour proliferation rate by immunohistochemical analysis of antigens associated with the cell cycle (Kelley and Mahaffey 1998). To date this diagnostic tool indicates that both T and B cell lymphomas have high proliferation rates, but until more information is gathered the correlation between the recorded values and prognosis and response to treatment is unknown in horses.
In mares with cutaneous lesions reported to regress during pregnancy, an additional immunohistochemical analysis for progesterone receptors on neoplastic cells may be indicated to offer a possible therapeutic option and to document the prevalence of oestrogen or progesterone receptors in various forms of equine lymphoma. Normal equine lymphoid tissue was determined to contain 1.9% progesterone-positive receptors while lymphomas, especially those rich in B cells, had a high percentage of progesterone-positive receptors (55% overall; 64% in B cell lymphomas, 58% in T cell-rich B cell lymphomas and 33% in T cell lymphomas) and neither normal nor lymphoma tissue was positive for oestrogen receptors (Henson et al. 2000). This study also found that the anatomical location of lymphoma correlated with the percentage of neoplastic cells that had progesterone-positive receptors (100% of splenic lymphomas, 67% of cutaneous, 60% of thoracic, 40% of multicentric and 25% of alimentary of equine lymphomas had progesterone-positive receptors) (Henson et al. 2000). Based on the positive response of women with breast cancer to hormonal management of tumours that has oestrogen or progesterone positive receptors, information needs to be obtained to determine if therapy of horses with either progesterone or anti-progesterone drugs is possible in the management of lymphoma with progesterone-positive receptors (Muss 1992).
As human oncology has progressed and become more sophisticated, the classification for lymphoma has evolved from the Kiel system and National Cancer Institute-Working Formulation to the revised European-American lymphoma system (REAL) to the most recent classification system developed by the World Health Organization (WHO) in 2008 (Armitage 2005; WHO 2008). Unfortunately, lymphoma of domestic animals does not fit neatly into these human classification systems and although the human non-Hodgkin lymphoma working formulation and WHO classification systems have been modified to allow classification of lymphoma for domestic animals, at this time the modified classification systems do not allow for the same predictions of prognosis and therapeutic response as it allows for people (Vail and Young 2002).
Treatment responses for equine lymphoma are unfortunately unknown since the majority of affected horses are not diagnosed as having the disease until it has reached end-stage. The therapeutic approach to equine lymphoma is similar to companion animals in that the patient's overall physiological status, extent and location of disease and client's financial and personal commitment to treatment needs to be evaluated before therapy is begun (Vail and Young 2002; Burns and Couto 2009). Staging the disease may aid in determining what treatment options, if any, may be available (Vail and Young 2002). WHO has developed a clinical staging system for lymphoma in domestic animals based on anatomic site and extent of organ involvement and clinical signs (Table 1) (Vail and Young 2002). Based on clinical stage, options available include surgical excision of solitary tumours, radiotherapy and chemotherapy. However, it is important to remind the owner that therapy is unlikely to be curative but only palliative.
Table 1. World Health Organisation clinical staging system for lymphoma in domestic animals
Limited to single node or lymphoid tissue within single organ
Without systemic signs
With systemic signs
Involvement of several lymph nodes within a regional area
Without systemic signs
With systemic signs
Generalised lymph node involvement
Without systemic signs
With systemic signs
Liver or spleen involvement with or without stage 3 disease
Without systemic signs
With systemic signs
Manifestation in the blood and involvement of bone marrow and/or other organ systems with or without stage 4 disease
Without systemic signs
With systemic signs
There are several reports describing outcome of horses undergoing surgical excision or reduction of solitary masses of lymphoma involving the large colon, eye and upper airway in the horse (Burba et al. 1991; Dabareiner et al. 1996; Rebhun and Del Piero 1998). There are reports of surgical resection of the large colon for removal of solitary masses involving the right ventral colon in one horse and left dorsal colon in the other (Dabareiner et al. 1996). The horse that underwent resection of the right dorsal colon was reported to be healthy 18 months following surgery but the other horse was subjected to euthanasia 14 months after surgery because it developed multicentric lymphoma (Dabareiner et al. 1996). Regression of cutaneous lymphoma nodules in a mare was reported following removal of a granulosa theca cell tumour (Henson et al. 1998). For this horse and anecdotally for others, there have been reports of regression of cutaneous form of lymphoma nodules after parenteral administration of synthetic progestin (Henson et al. 1998). For horses with lymphoma of the upper airway, surgical reduction of the mass was only palliative but surgical reduction could potentially be used in conjunction with radiotherapy (Meschter and Allen 1984).
Radiation therapy in the horse is accomplished using one of 2 techniques: brachytherapy or teletherapy. Brachytherapy is the use of a sealed radioactive source (Iridium-92, iodine-125, or strontium-90) through implantation or surface treatment of the targeted tissue. Teletherapy is the deployment of radiation from an external beam such as a linear accelerator or cobalt-60 machine. Teletherapy is usually limited to treatment of neoplasia within a distance of 80–100 cm from the radiation source (Henson and Dobson 2004). Whether radiation therapy is possible depends on the volume and location of the lymphoma (Weaver et al. 1996; Vail and Young 2002; Henson and Dobson 2004; Gerard et al. 2010). Typically, teletherapy is employed in treatment of stage 1 lymphoma or in management of problematic lesions, such as cutaneous lymphoma over an articular surface or perineum (Henson and Dobson 2004). Radiation dose is based on tumour location, size and depth. The total dose of radiation is divided into as few treatments as possible. A decrease in radiation treatments increases normal tissue tolerance and decreases the number of times the patient is placed under general anaesthesia, required for accurate positioning and immobilisation (Henson and Dobson 2004). Side effects of radiation therapy in the horse are localised to the area being treated and include inflammation, ulceration and necrosis which in the case of skin can lead to growth of white hair and temporary regional lymphadenopathy (Weaver et al. 1996; Henson et al. 2004; Henson and Dobson 2004). In a recent report, radiation therapy was used to treat 3 horses with lymphoma, 2 with solitary skin masses and one with a mass in a nasal passage (Henson et al. 2004). These horses were tumour-free for up to 9 years at the time of the report (Henson et al. 2004).
The principles of chemotherapy are to maximise neoplastic cell death while minimising local and systemic effects. Although chemotherapy is often only palliative, curative chemotherapy is possible if a solitary mass is involved. Due to the potential for drug resistance, a multi-drug protocol is often used (Vail and Young 2002). Selection of chemotherapeutic drugs for use in a multi-drug protocol is based on the ability to administer the drugs at the maximum recommended dose based on toxicity, drugs that avoid overlapping toxicities and those that have a known effect against the neoplasm being treated (Vail and Young 2002; Burns and Couto 2009). Several multi-drug protocols have been reported for treatment of equine lymphoma with varying success (Tables 2 and 3) (Gollagher et al. 1993; Saulez et al. 2004; Theon et al. 2007; Burns and Couto 2009). Drug doses for horses can be calculated by using the following formula: body surface area (m2) = wt (g 2/3) × 10.5/104 (Burns and Couto 2009). After beginning chemotherapy, remission of equine lymphoma can be seen within 2–4 weeks and treatment is continued for an additional 2–3 months (Burns and Couto 2009). If remission is still evident after this time and side effects minimal, chemotherapy can be continued. While the lymphoma is in remission, therapy is continued but the intervals between doses of chemotherapeutic drugs is increased by 1 week for 2–3 months then increased by an additional week. This continual therapy has been reported to allow for patients to remain in remission for an additional 6–8 months, but once therapy was discontinued, the lymphoma returned (Burns and Couto 2009). Side effects of chemotherapy are not often encountered but include bone marrow suppression, gastrointestinal disturbances, alopecia and laminitis (Vail and Young 2002; Burns and Couto 2009). Thus regular (every 2 weeks) physical examinations and CBC are recommended. If significant abnormalities are observed during physical examination or on CBC, or serum biochemistry evaluation, it may be necessary to delay chemotherapy until those parameters return to within an acceptable range. Options for chemotherapy other than multi-drug protocols include administration of a single chemotherapeutic drug either systemically or by local injection of a solitary lesion. Doxorubicin has been administered intravenously at a dose of 30–60 mg/m2 every 2–3 weeks; however, this drug should be used with caution because myocardial damage and pulmonary fibrosis are potential side effects (Burns and Couto 2009). Intralesional administration of cisplatin combined with sesame oil at a dose of 1 mg of cisplatin/cm3 of tumour injected in planes spaced 1 cm apart repeated every 2 weeks resulted in a success rate of 96.2% for horses with cutaneous lymphoma lesions (Theon et al. 2007).
Table 2. Chemotherapeutic drugs used in treatment of equine lymphoma
Mechanism of action
Inhibits DNA, RNA and protein synthesis
Bone marrow suppression, bladder irritation (in canine)
Bone marrow suppression
Cytosine arabinoside (Cytarabine)
Kills cell in S-phase and blocks progression from G1 to S phase of DNA synthesis
Inhibition of protein synthesis, free radical formation
Cardiotoxic, nephrotoxic, potent vesicant
Inhibition of intracellular microtubules thus disrupting the cell cycle
Perivascular tissue reaction, hepatotoxic
Inhibits DNA synthesis
Deprive cells of amino acid thus inhibit protein synthesis
Binds DNA strands preventing protein synthesis
Can be nephrotoxic in dogs
Table 3. Multi-drug protocols
Every 2 weeks alternating basis
i.m. or subcut
1 mg/kg bwt
q. 7–14 days
Chlorambucil or cyclophosphamide
20 mg/m2 200 mg/m2
per os i.v. (catheter)
q. 14 days q. 14–21 days
1.1–2.2 mg/kg bwt
q. 48 h
Vincristine (can be added if no response initially)
q. 7 days
Single agent therapy
q. 2–3 weeks
Cyclophosphamide or vincristine
200 mg/m2 0.5 mg/m2
i.v. (catheter) i.v. (catheter)
q. 2–3 weeks
Combo chemotherapeutic with autologous vaccine
Given Days 1 and 36
Autologous tumour vaccine
2 ml injected at 4 sites
Days 4, 21 and 39
q. 3 weeks
Cisplatin (1 ml of 10 mg/ml cisplatin and 2 ml sesame oil)