Abstract
Low-grade alimentary lymphoma (LGAL) was diagnosed by histological and immunohistochemical evaluation of full-thickness biopsies from multiple regions of the gastrointestinal tract collected during exploratory laparotomy in 17 cats. The most common clinical signs were weight loss (n=17) and vomiting and/or diarrhoea (n=15). Clinical signs were chronic in 11 cases. Abdominal palpation was abnormal in 12 cats, including diffuse intestinal thickening (n=8), an abdominal mass due to mesenteric lymph node enlargement (n=5) and a focal mural intestinal mass (n=1). The most common ultrasonographic finding was normal or increased intestinal wall thickness with preservation of layering. Ultrasound-guided fine-needle aspirates of mesenteric lymph nodes (n=9) were incorrectly identified as benign lymphoid hyperplasia in eight cats, in which the histological diagnosis from biopsies was lymphoma. There was neoplastic infiltration of more than one anatomic region of the gastrointestinal tract in 16/17 cats. The jejunum (15/15 cats) and ileum (13/14 cats), followed by the duodenum (10/12 cats), were the most frequently affected sites. Twelve cats were treated with oral prednisolone and high-dose pulse chlorambucil, two with a modified Madison–Wisconsin multiagent protocol and three with a combination of both protocols. Thirteen of the 17 cats (76%) had complete clinical remission with a median remission time of 18.9 months. Cats that achieved complete remission had significantly longer median survival times (19.3 months) than cats that did not achieve complete remission (n=4) (4.1 months; P=0.019). The prognosis for cats with LGAL treated with oral prednisolone in combination with high-dose pulse chlorambucil is good to excellent.
Alimentary lymphoma is the most common anatomic form of lymphoma in the cat. 1–3 It is characterised by infiltration of the gastrointestinal tract with neoplastic lymphocytes, with or without mesenteric lymph node involvement. 4–6 Feline alimentary lymphoma can be histologically classified by the National Cancer Institute Working Formulation (NCIWF) as high-grade, intermediate-grade or low-grade with the latter being common in the diffuse form. 7–11 A less commonly described form of alimentary lymphoma is large granular lymphocytic lymphoma.12,13
Low-grade alimentary lymphoma (LGAL) was first described as a clinical entity in cats in a retrospective study of 29 cases. 10 The diagnosis was based on histology of haematoxylin and eosin (HE)-stained sections from gastroduodenal endoscopic biopsies (EB). 10 Cats with LGAL were reported to have a more favourable outcome when treated with oral prednisolone and high-dose pulse chlorambucil compared to cats with high-grade alimentary lymphoma treated with multiagent chemotherapeutic protocols. 10 The purpose of this study was to review the clinicopathological findings and response to therapy in a cohort of Australian cats diagnosed with LGAL by histological examination of full-thickness biopsies (FTB) taken from multiple regions of the gastrointestinal tract.
Materials and methods
Medical records from The Valentine Charlton Cat Centre (VCCC), University of Sydney and North Shore Veterinary Specialist Centre (NSVSC) were searched for cases of alimentary lymphoma from January 2002 to June 2005. Inclusion criteria were a diagnosis of LGAL including immunophenotype based on histological and immunohistochemical evaluation of FTB taken from the gastrointestinal tract at exploratory laparotomy, complete medical records and use of standard chemotherapeutic protocols. Chemotherapeutic protocols were designated protocol 1; prednisolone (3 mg/kg orally every 24 h PO, tapering to 1–2 mg/kg orally every 24–48 h PO once complete response (CR) was achieved) in combination with high-dose pulse chlorambucil (15 mg/m2 orally every 24 h for 4 consecutive days every 3 weeks PO), 10 protocol 2; a modified Madison–Wisconsin multiagent protocol, 14 or protocol 3; combination of both protocols.
For all cases signalment, history, physical examination findings, diagnostic investigations, treatment and outcome were recorded. Clinical signs were considered chronic if they were present for >4 weeks. Diagnostic investigations included serum biochemistry, complete blood count (CBC), urinalysis, serum thyroxine concentrations, serology for feline immunodeficiency virus (FIV) antibody and feline leukaemia virus (FeLV) antigen, thoracic and abdominal radiographs, abdominal ultrasonography and exploratory laparotomy. Reference values for small intestinal wall thickness as assessed by ultrasonography were defined as ≤2.8 mm for the duodenum and jejunum and ≤3.2 mm for the ileum. 15 Normal mesenteric lymph node diameter was defined as ≤5 mm. 16 Formalin fixed, paraffin-embedded (FFPE), HE-stained 3–5 μm histological sections were prepared from biopsies collected at exploratory laparotomy. Anatomic sites routinely biopsied included stomach, duodenum, ileum, jejunum, liver, pancreas, mesenteric lymph nodes and, less commonly, colon. FFPE tissues from cases of LGAL were retrieved, cut at 4 μm and labeled with anti-CD3 and anti-CD79a using a previous method. 9 In two cases with eosinophilic infiltrates, additional staining with toluidine blue was performed.
Histological sections of all biopsy specimens retrieved were examined by two pathologists (AC, MK) to confirm the diagnosis of alimentary lymphoma. Tumours were classified according to the NCIWF. Immunophenotype enabled classification according to the Revised European and American Lymphoma/World Health Organisation (REAL/WHO) classification system. 17 Alimentary lymphomas were assigned to high-, intermediate- or low-grade and further classified within these grades on the basis of architecture (follicular or diffuse) and cytological features (nuclear outline and cell size), in accordance with previous applications of NCIWF to feline lymphoma.9,18 LGAL was distinguished from marked lymphoplasmacytic inflammatory bowel disease (IBD) by the relative absence of mixed lymphoid and granulocytic cells and their replacement, sometimes irregularly in early neoplastic disease, with monomorphous sheets of neoplastic lymphoid cells within the lamina propria 17,19–21 ; lack of reactivity of enterocytes in the crypts and villi to the presence of neoplastic lymphoid cells 20 ; effacement of the proprial-epithelial boundary; presence of epitheliotropism17,22; level of extension into the underlying submucosa, tunica muscularis and serosa, 21 and the presence of neoplastic cellular infiltrates in mesenteric lymph nodes.20,21,23
Response to therapy was defined as either CR or no response (NR). CR was defined as a complete resolution of clinical signs, weight gain and resolution of intestinal thickening or abdominal mass on abdominal palpation for greater than or equal to 30 days. NR was defined as any response less than a CR. Response was evaluated each time the cat was presented to a veterinarian for either chemotherapy or laboratory testing. Survival time was defined as the time from first treatment until the animal died or was censored. Cats were censored from statistical evaluation of survival when they were lost to follow-up. Cats that were still alive at the time of analysis had their survival times censored at the last known date of follow-up. Outcomes evaluated were response to therapy and survival time. Categorical variables evaluated for their effect on response and survival times included signalment (sex, breed, age), clinical signs (vomiting, diarrhoea, both vomiting and diarrhoea, inappetence, polyphagia, weight loss, lethargy), focal mass versus diffuse intestinal thickening, mesenteric lymph node enlargement, histological liver involvement, CBC findings (anaemia, neutrophilia) and chemotherapy protocol (1 versus 2 versus 3).
The effect on survival of response to therapy was also examined as a dichotomous variable, yes/no. To determine the statistical influence of the factors listed above on the probability for a cat to enter complete remission, χ2 statistics were used for categorical variables: breed, gender, clinical signs, CBC findings, most of which were dichotomous. An independent T-test was employed to test the influence of age (as a continuous variable) on whether a cat entered complete remission. Variables found to significantly affect remission on univariate analysis (P<0.1) were analysed using binary logistical regression to determine multivariate significance. Kaplan–Meier survival statistics were used to determine overall survival times. Cox regression was used to determine the influences of previously mentioned variables on the duration of survival time. Variables that were found to be significant at the level of P<0.1 in univariate analysis were entered into a forward logistic regression model to determine whether they retained significance under multivariate analysis (P<0.05). A standard statistical program was used for statistical analyses (SPSS 10, Statistical Analytical Software, Chicago, IL).
Results
Forty-two cats were identified from medical records with alimentary lymphoma during the study period, including 21 cats with high-grade disease and 21 with LGAL. On review of histological sections from the 21 cats initially diagnosed with LGAL, 19 were classified as LGAL of the small lymphocytic type and two were reclassified as intermediate-grade alimentary lymphoma (large cleaved cell lymphoma) in accordance with previous applications of NCIWF to feline lymphoma. Of the 19 LGAL tumours, immunophenotyping could not be performed in one case because additional FFPE tissues were not available. The remaining 18 cases of LGAL and both intermediate-grade tumours were of T-cell immunophenotype and were classified according to the REAL/WHO system as intestinal T-cell lymphomas. Seventeen of the 19 cases of LGAL met the inclusion criteria. One was excluded because of use of a non-standardised chemotherapy protocol. The clinicopathological findings, treatment and outcomes of the 17 cats included in the study are summarised in Table 1.
Clinicopathological findings and outcome from 17 cats with LGAL
Focal=focal abdominal mass, Diffuse=diffuse intestinal thickening, SI=small intestine, LN=lymph node, mes LN=mesenteric lymph node, Euth=euthanasia.
The median age of affected cats was 13 years (range 9–18 years). There were 16 domestic crossbred cats and one Burmese. All cats were desexed, including eight males and nine females. Historical findings included weight loss (n=17), inappetence (n=7), vomiting (n=6), diarrhoea (n=4), vomiting and diarrhoea (n=5), lethargy (n=4) and polyphagia (n=2). Vomiting and/or diarrhoea were chronic in 11/15 cases. Abdominal palpation was abnormal in 12 cats, including six cats with diffuse intestinal thickening, four cats with a focal abdominal mass and two cats with diffuse intestinal thickening and a focal abdominal mass. On abdominal ultrasonography, the abdominal mass was identified as mesenteric lymph node enlargement in five cats and a focal intestinal mass in one cat.
Results of serum biochemistry and CBC were available for 15 cats. Mature neutrophilia (n=4, range 12.3–37.1×106/l, reference interval (RI) 3.8–10.8×106/l) and anaemia (n=3, haematocrit (Hct) 0.16–0.25 l/l, RI 0.30–0.45 l/l) were the most common haematological abnormalities. The anaemia was regenerative in two cats (Hct<0.30 l/l, absolute aggregate reticulocyte count>50×109/l) and non-regenerative in one cat with concurrent chronic kidney disease. The anaemia improved with treatment, but failed to resolve in all three cats. Serum albumin was normal in all cats tested. Four cats had mild to moderate elevations in urea (range 11.6–33.3 mmol/l, RI 5.87–10.6 mmol/l), which were considered pre-renal in two (urine specific gravity (USG)>1.040). One cat had a moderate elevation in alanine aminotransferase (ALT) (318 U/l, RI 1–80 U/l), due to concurrent neutrophilic cholangitis. Serum thyroxine level was elevated in 1/9 cats tested (54 nmol/l, RI 5–39 nmol/l). Serological testing for FIV antibody was performed in nine cats, of which one was positive. All eight cats tested for FeLV antigen were negative. Urinalysis results were available for 13 cats, of which eight had adequate urine concentrating ability (USG>1.040).
Abdominal ultrasonography was performed in 16 cats, including measurement of small intestinal wall thickness in 15 cats. Ultrasonographic findings included mesenteric lymph node enlargement (n=12), diffuse small intestinal wall thickening (n=9), and a focal mural intestinal mass with loss of normal layering (n=1). Mesenteric lymph node diameter was recorded in 9/12 cats with lymph node enlargement. The mean lymph node diameter was 15.9 mm (median 10 mm, range 6.5–30 mm). In the nine cats with diffuse intestinal thickening, the mean wall thickness was 4.3 mm (median 4.5 mm, range 3.4–5 mm). A focal mural thickening of 2 cm was detected in one case. Ultrasound-guided fine-needle aspiration of mesenteric lymph nodes was performed in nine cats. Cytological analysis of all samples suggested benign lymphoid hyperplasia. One cat, with concurrent bacterial cholangitis had a diffuse increase in echogenicity of the liver and abnormal biliary tract architecture. Renal cortical echogenicity was increased in six cats. Two cats had loss of renal corticomedullary distinction. The spleen and pancreas were considered normal in all cats.
On exploratory laparotomy and subsequent FTB examination, lymphoma was detected in the stomach of 4/12 cats (33%), duodenum of 10/12 cats (83%), jejunum 15/15 cats (100%), ileum 13/14 cats (93%), mesenteric lymph nodes 10/17 cats (59%), liver 4/15 cats (27%), colon 1/5 cats (20%) and pancreas 1/14 cats (7%) (Table 1). All but one cat had histological evidence of diffuse intestinal disease with neoplastic infiltration of more than one anatomic region of the gastrointestinal tract (excluding lymph nodes) (Table 1). In one cat, lymphoma was restricted to the jejunum, whereas IBD was present in other intestinal regions. Three other cats with lymphoma had histological evidence of concurrent lymphoplasmacytic IBD affecting other regions of the alimentary tract. In two cats, large numbers of eosinophils accompanied the neoplastic infiltrate in the intestine and mesenteric lymph nodes. In both cases the neoplastic cells were CD3 positive and did not stain with toluidine blue. Eight cats with histological evidence of lymphoma of the mesenteric lymph nodes had been incorrectly identified as having benign lymphoid hyperplasia on cytological evaluation. Lymphocytic infiltrates present in the liver of six cats and in the pancreas of four cats were considered inflammatory. Other histopathological findings included mild chronic pancreatitis (n=1), neutrophilic cholangitis (n=1) and hepatic lipidosis (n=1).
Twelve cats received oral chemotherapy (protocol 1) and two were treated using multiagent chemotherapy (protocol 2). The remaining three cats received both oral and multiagent chemotherapy protocols sequentially (protocol 3) (Table 1). One cat was changed to oral chemotherapy 3 months from diagnosis after failing to achieve or maintain remission on multiagent chemotherapy. The other cat was changed to oral chemotherapy 11 months from diagnosis following an anaphylactic reaction to vincristine. The remaining cat was changed from oral chemotherapy to multiagent chemotherapy at 5 months from diagnosis, due to the development of high-grade (lymphoblastic type) renal lymphoma. Adverse reactions to oral chemotherapy were uncommon, including mild self-limiting gastrointestinal signs in two cats and myelosuppression in two cats. Of the latter, one cat developed transient moderate neutropenia (2.57×106/l), whilst the other cat developed severe thrombocytopenia (platelet count 7×109/l, reference values 300–700×109/l) after 10 months of protocol 1. The majority of platelets present were macroplatelets. No clinical signs of thrombocytopenia were apparent. Thrombocytopenia resolved 6 months after discontinuation of chlorambucil.
Thirteen cats (76%) went into complete remission with a median remission time of 18.9 months (range 3.5–73 months). The only factor that appeared to predict whether a cat achieved remission or not, was the reporting of lethargy by the owner. Of 13 cats that were not lethargic, 12 (92%) achieved a complete remission. Of four cats that were lethargic, only one achieved a complete remission (P=0.022). Survival data were available for all 17 cats. Two cats were censored, as they were still alive at the completion of the study period. Median survival time was 14.9 months (range 0.5–76 months) ( Fig 1 ). Cats that achieved complete remission (n=13) had significantly longer median survival times (19.3 months) than cats that did not achieve complete remission (n=4) (4.1 months; P=0.019) ( Fig 2 ). Of the two cats still alive at the completion of the study, both remained in clinical remission. One cat was being treated with prednisolone at a reduced dose and frequency (2 mg/kg orally q 48 h) and chlorambucil at reduced frequency (15 mg/m2 orally for 4 consecutive days every 4 weeks). In the other cat, therapy was ceased for 10 months following development of severe thrombocytopenia, and relapse occurred after a period of 25 months from diagnosis. Protocol 1 was recommenced. There was no recurrence of thrombocytopenia. Remission was re-established and the cat was in clinical remission at the time of writing.
Kaplan–Meier survival analysis for 17 cats with LGAL. Two cats were censored (cross marks), as they were still alive at the completion of the study period. Median survival time was 14.9 months (range 0.5–76.1 months).
Thirteen cats that achieved complete remission had significantly longer median survival time (19.3 months, solid line) than four cats that did not achieve complete remission (4.1 months, dashed line; P=0.019). Crosses indicate two cats censored still alive at the end of the study period.
Discussion
Alimentary lymphoma is now identified as the most common anatomic form of lymphoma in cats. 1 Our study provides a clinicopathological assessment of LGAL. Most previous studies of feline alimentary lymphoma have not reported histological grades or have included mainly cases of high-grade alimentary lymphoma, as LGAL has only recently been described.10,22,24 In a previous Australian case series, 90% of alimentary lymphomas were intermediate- or high-grade. 9 In contrast, Fondacaro et al found 75% of alimentary lymphomas were low-grade. 10 This raises the issue of the relative incidence of high-, intermediate- and low-grade alimentary lymphoma. In our study, procurement of cases led to roughly equal numbers of high-grade and low-grade alimentary lymphomas. The observed variability in prevalence of the three histological grades may be due to differences in geographic distribution, method used to procure biopsy samples, grading system used by pathologists and increased awareness of low-grade disease. Prospective studies to determine the relative prevalence of the three histological grades of alimentary lymphoma and to elucidate the reasons for their relative frequencies are warranted.
Cats diagnosed with LGAL were middle-aged or older. No apparent breed or sex predisposition was identified. In the majority of cases, clinical signs were chronic and indistinguishable from those associated with inflammatory intestinal disease. Similarly, a recent study comparing cats with IBD and alimentary lymphoma, found no correlation between the clinical findings and the final diagnosis. 25 Abdominal palpation in the majority of cats revealed either generalised intestinal thickening or a focal abdominal mass usually due to mesenteric lymph node enlargement. Previously, the finding of a palpable abdominal mass in this setting has been considered suggestive of high-grade alimentary lymphoma, in which there is segmental, often eccentric mural thickening and/or mesenteric lymph node enlargement.4,10 Our study demonstrates that a palpable mass is also consistent with LGAL.
We found no association between retroviral infection and LGAL in agreement with other studies of alimentary lymphoma.1,10,26 No cats were hypoalbuminaemic in our study. By contrast, hypoalbuminaemia has been reported in 50–75% of cats with predominantly high-grade alimentary lymphoma10,26 and in 49% of cats with LGAL. 10 Hypoalbuminaemia is thought to occur primarily due to loss of low molecular weight proteins across a compromised intestinal wall. It is likely that in LGAL, the integrity of the intestinal wall can be maintained until late in the disease process. Anaemia was present in 20% of cats tested here and was attributed to concurrent chronic kidney disease and suspected gastrointestinal haemorrhage. Mild anaemia has been reported in 43% of cats with high-grade alimentary lymphoma. 4
The reported ultrasonographic features of alimentary lymphoma in cats include transmural intestinal thickening with disruption of normal wall layering, reduced wall echogenicity, localised hypomotility and abdominal lymphadenopathy. Loss of intestinal wall layering on sonography occurs due to infiltration of the intestinal wall with neoplastic or inflammatory cells, as well as secondary necrosis, oedema and/or haemorrhage.27,28 Although histological grade of lymphoma was not specified in these studies, the reported changes likely pertain to high-grade alimentary lymphoma. In the present study, ultrasonographic findings of diffuse intestinal wall thickening with preservation of the intestinal wall layering, was common. Histological evaluation of intestinal biopsies is warranted when thickened intestinal wall or mesenteric lymphadenomegaly is detected by abdominal ultrasonography; generalised intestinal wall thickening and mesenteric lymph node enlargement are common in both LGAL and IBD. 25 However, the finding of normal intestinal thickness does not rule out a diagnosis of LGAL and should not preclude histological evaluation of intestinal biopsies. There was a disparity between ultrasonographical and histological assessment of the liver in our study; the liver was ultrasonographically normal in all but one cat, but 4/15 cases where the liver was biopsied had concurrent hepatic involvement. We also identified that cytological evaluation of enlarged mesenteric lymph nodes was misleading in establishing a diagnosis of LGAL. Histological assessment of lymph node architecture is required to distinguish between benign hyperplasia and neoplasia.
Lymphoma was diagnosed in more than one region of the gastrointestinal tract in all but one case, confirming that LGAL is typically a diffuse or multifocal disease. Neoplastic infiltrates were found in at least one region of the distal small intestine in all cats biopsied. Duodenal involvement was not present in all cases, and only 4/12 cats (33%) had gastric involvement. These findings serve to highlight potential limitations when upper gastrointestinal endoscopy is used to procure intestinal biopsies. It is possible that endoscopic diagnosis of LGAL could be improved if endoscopic gastrointestinal studies routinely include ileal biopsies obtained via the ileocaecocolic valve. EB, which only sample the intestinal epithelium and submucosa, are usually adequate for the histological diagnosis of gastric lymphoma, but duodenal biopsies may be insufficient for accurate differentiation between IBD and LGAL. In a study of 10 cats with LGAL, gastroduodenal EB specimens were compared with FTB from multiple gastrointestinal regions from the same cats. On evaluation of FTB, all 10 cats had lymphoma in the jejunum and ileum, and nine had duodenal involvement. However, an incorrect diagnosis of IBD was made in 5/9 cats on evaluation of duodenal EB specimens alone. Technical difficulties may have hampered the quality of EB specimens, as two cats had only ‘partial’ duodenal assessment and in three cats duodenal biopsy was performed blindly. 25 Further studies are required to compare the results of ileal EB and FTB specimens in the diagnosis of LGAL in cats.
In the present study lymphoma was demonstrated in only 1/5 colonic biopsies taken. Colonic biopsies were taken from the mid to distal colon. It is not known whether a similar predilection for ileocaecocolic junction involvement exists for LGAL, as it does for high-grade alimentary lymphoma. 29
Concurrent involvement of the mesenteric lymph nodes and liver was common, in agreement with previous studies.22,25 Sonographic changes in kidneys were common, but as renal biopsies were not taken the proportion of cats with renal involvement could not be determined. Paraneoplastic eosinophilic infiltrates, previously reported in a cat with intestinal T-lymphoma were identified in two cats in this study. 30 Immunophenotyping and toluidine blue staining were performed to differentiate lymphoma from mast cell neoplasia, as the latter can also be accompanied by eosinophilic infiltrates. In lymphoma, eosinophil chemotaxis is thought to result from the production of interleukin-5 by neoplastic lymphocytes. 31 T-cell intestinal lymphoma with eosinophilic infiltrates has also been reported in dogs and humans. 32
Concurrent lymphoplasmacytic IBD was diagnosed in four cases here, which has been reported by others. 22 It has been proposed that IBD may be a precursor to lymphoid malignancy of the intestinal tract in cats.17,19,20 Alternatively, some cases of IBD could have been erroneously diagnosed with lymphoma. In one study of 32 cats diagnosed with alimentary lymphoma from HE-stained sections, adjunctive immunohistochemical stains revealed that in five cases (15%) the ‘neoplastic’ infiltrate was composed of a mixed population of small B and T lymphocytes and plasma cells. On this basis, all five cases were considered to have IBD. 11 Also, severe lymphocytic intestinal inflammation secondary to food intolerance, for which the initial histological diagnosis from HE-stained intestinal biopsies was alimentary lymphoma, has been described.33,34 All cases of LGAL in our study were of T-cell immunophenotype. LGAL of cats is widely considered to be a disorder of lymphocytes of T-cell lineage and is also known as enteropathy-associated T-cell lymphoma.17,19,22,35 However, it should be noted that expansion of T-cell populations in intestinal mucosal-associated lymphoid tissue can occur in both inflammatory and neoplastic intestinal disease in cats. 36 Determination of clonality of T-cell populations in the lymphocytic intestinal infiltrate by molecular techniques shows promise as an adjunct to histology and immunohistological phenotyping for diagnosing LGAL. Clonality testing was 89% sensitive in detection of T-cell lymphoma, based on clonal or oligoclonal T-cell populations that were considered to be neoplastic. 36 Using polymerase chain reaction (PCR), 22/28 cats were found to have clonal rearrangements of the genes encoding the gamma variable region of the T-cell receptor, while three had oligoclonal rearrangements. In comparison, polyclonal rearrangements were detected in 3/3 cats with normal intestinal histology and in 9/9 cats with lymphocytic–plasmacytic intestinal inflammation. 36
Response to chemotherapy was good in most cases. Complete remission was achieved in 76% of cats with an overall median remission time of 18.9 months. Those cats that achieved complete remission were found to have significantly longer survival times when compared to cats that failed to achieve complete remission. This finding is consistent with other studies of feline lymphoma.2,3 Cats that were lethargic were less likely to achieve complete remission. Cats with clinical illness at diagnosis have been reported to have shorter survivals in other studies.3,4 Interestingly, one cat in this study showed no clinical response to multiagent chemotherapy, but had long-term remission (73 months) with oral prednisolone and chlorambucil. This suggests that LGAL has a different biological behaviour and may respond better to slow alkylating agents, such as chlorambucil. There are few reports documenting the treatment and outcome of cats with LGAL. Results of this and other studies suggest that cats with LGAL generally show an excellent response to treatment with oral prednisolone and high-dose pulse chlorambucil therapy.10,22,35
High-dose chlorambucil therapy is associated with an increased potential for gastrointestinal toxicity and myelosuppression. 37 Mild self-limiting gastrointestinal signs, such as vomiting, diarrhoea and inappetence, as noted in two cats, can be difficult to distinguish from progression of primary gastrointestinal disease. Myelosuppression was documented in two cats, of which one developed severe thrombocytopenia after 10 months of chlorambucil therapy. Thrombocytopenia has been documented previously as a major adverse effect of chlorambucil therapy. 38
There are limitations in this clinical study. Because of the retrospective nature of the study, complete data were not available for all cases. Ultrasonographic measures of small intestinal wall thickness and mesenteric lymph node size were not recorded in all cases. The frequency of thickening of the small intestinal wall may have been underestimated because of subtle cases where measurements were not recorded. In addition, the small number of cases in the study limited statistical analysis. Finally, some authors have reported a third response category in the treatment of lymphoma, partial response (PR) being >50% but less than 100% response. As the assignation and differentiation of anything less than a CR in this retrospective study requires access to data that the authors do not have, we have taken the more conservative approach of assigning any PR to the NR category.
Conclusion
LGAL is a chronic, indolent disease of middle to old age cats with clinical signs indistinguishable from IBD. In the majority of cats disease is diffuse and involves more than one region of the gastrointestinal tract. The jejunum and ileum are the most commonly affected sites. Generalised intestinal thickening and mesenteric lymph node enlargement are commonly detected on physical and abdominal ultrasonographic examination. Focal intestinal masses can also occur. Whilst abdominal ultrasound and cytological examination may assist in the evaluation of gastrointestinal disease, histological examination is necessary to establish a definitive diagnosis of LGAL. CR to chemotherapy is a positive prognostic indicator. Cats with LGAL may have durable remissions when treated with oral prednisolone and high-dose pulse chlorambucil.
Footnotes
Acknowledgments
The authors would like to thank staff at the Valentine Charlton Cat Centre, University of Sydney and North Shore Veterinary Specialist Centre and referring veterinarians for their assistance with cases. This study was funded in part by the Australian Companion Animal Health Foundation and Payten cancer research fund.