Feline low-grade alimentary lymphoma: how common is it?

Abstract

Low-grade alimentary lymphoma (LGAL) requires histological assessment of biopsies for diagnosis whereas intermediate- (IGAL) and high-grade (HGAL) alimentary lymphoma (AL) can be diagnosed by cytology of intestinal or mesenteric lymph node aspirates. Assessment of the relative frequency of subtypes of AL using histology alone may be skewed towards an increased frequency of LGAL as cases of IGAL or HGAL diagnosed cytologically may not progress to biopsy. We investigated the relative prevalence of AL subtypes diagnosed by both histopathology and cytology among primary accession cases across Australia during a 5-year period. Clinicopathological features of LGAL were compared with those of IGAL/HGAL. Fifty-three cases of AL were identified, including 30 diagnosed by histology (15 LGAL, 13 HGAL, two IGAL) and 23 IGAL/HGAL diagnosed by cytology. LGAL cases comprised 50% of histological diagnoses, but only 28% of all AL. A palpable abdominal mass was more common in IGAL/HGAL (43%) than in LGAL (7%) [odds ratio (OR) 7.6, P = 0.01]. Anaemia was more common in IGAL/HGAL (41%) compared with LGAL (7%) (OR 9.6, P = 0.02). On abdominal ultrasound, a gastrointestinal mural mass was visualised in 41% of IGAL/HGAL and 0% of LGAL (P = 0.01). Where a detailed abdominal ultrasound report was provided, gastric/intestinal wall thickening was the most commonly reported abnormality (82%). In cats with intestinal thickening, a loss of normal layering was more common (P = 0.02) in cats with IGAL/HGAL (71%) compared with those with LGAL (20%). The relative prevalence of LGAL was lower when cases diagnosed by cytology were included in addition to those diagnosed by histology in the study population. The relative frequency with which LGAL is diagnosed has increased since initial reports from this region. A number of significant clinicopathological findings are useful to distinguish LGAL from IGAL/HGAL.

Short Communication

Alimentary lymphoma (AL) is the most common anatomical form of lymphoma in cats.^1-5 It is characterised by an infiltration of the gastrointestinal tract and/or associated lymph nodes with neoplastic lymphocytes.^6-8 Feline AL can be classified histologically according to the National Cancer Institute Working Formulation (NCI WF), based on mitotic rate and cell type, as low-grade (LGAL), intermediate-grade (IGAL) or high-grade (HGAL). ⁹ Most commonly, LGAL occurs as a mucosal infiltrate of small-to-medium sized T-lymphocytes and presumably represents a neoplastic transformation of the existing high population of CD3+ T cells present in the diffuse mucosal-associated lymphoid tissue in the small intestine. ¹⁰ IGAL and HGAL are usually characterised by a higher proportion of large lymphoblastic cells and have an increased tendency for transmural spread.

A diagnosis of feline AL can be achieved by histopathological examination of full-thickness gastrointestinal biopsy specimens obtained laparoscopically or at exploratory laparotomy, or endoscopic gastrointestinal biopsies, which sample the mucosa but, often, only scant submucosa. Although less invasive, endoscopy would be considered inferior to exploratory surgery for biopsy procurement as partial-thickness specimens provide little information about transmural spread. Furthermore, the use of endoscopy precludes sampling of the mid-to-distal jejunum, which has been reported as the most common site for LGAL. ¹⁰ In many cases of IGAL and HGAL, a diagnosis can be achieved by cytological examination of fine needle aspirate (FNA) biopsies of gastrointestinal mass lesions or associated lymphoid structures. However, it is usually not possible to distinguish LGAL from inflammatory bowel disease or benign lymphoid hyperplasia on cytology alone.^6,11

Over the last decade, LGAL has become increasingly recognised as a clinical entity. Several retrospective studies have identified LGAL as a disease that is clinically distinguishable from HGAL, with a favourable prognosis when treated with oral prednisolone and chlorambucil.^6,12-14 However, the relative prevalence of the different NCI WF grades of feline AL in the general feline population has not been well established. An Australian study published in 1999 found that LGAL accounted for 11% of histologically-diagnosed feline AL. ¹ A more recent study from the same institution found that LGAL accounted for 45% of histological diagnoses. ⁶ This suggests that feline AL is characterised by a higher proportion of low-grade disease than identified previously. ⁶ Studies from North America have also demonstrated a high proportion of histologically low-grade disease, ranging from 37–75%.^10,12,15 Cases of IGAL and HGAL diagnosed by FNA cytology of an intestinal- or mesenteric lymph node-associated mass, do not necessarily proceed to tissue biopsy and a histological diagnosis. The exclusion of cases of AL diagnosed by cytology could result in the overestimation of the prevalence of LGAL in histopathological studies of feline AL. The primary aim of this study was to determine the relative prevalence of LGAL diagnosed at three primary accession feline veterinary hospitals in Australia over a 5-year period. The secondary aim was to compare clinicopathological findings in cats with LGAL to those with IGAL and HGAL.

The medical records from three primary accession, feline-only, veterinary hospitals in Sydney, Melbourne and Brisbane, Australia, were searched for cases of feline AL diagnosed between December 2004 and December 2009. A total of 58 cases were identified. A definitive diagnosis was obtained in 53 cases from biopsy samples sent to an external veterinary pathology laboratory, comprising smears prepared from FNAs of alimentary structures in 23 cats and tissue biopsies submitted for histopathology in 30 cats. The remaining five cases were diagnosed based on in-house cytology results alone and were excluded from the study. Cases diagnosed histologically were assigned an NCI WF grade based on the pathologist’s interpretation. Those diagnosed by FNA cytology were assigned arbitrarily as IGAL/HGAL as it is not possible to differentiate between cases of intermediate- and high-grade disease on cytological features alone. Cases of IGAL and HGAL diagnosed histologically, and all cases diagnosed cytologically, were combined as a single group (IGAL/HGAL) for statistical analyses. The age, breed, sex, body weight, body condition score, abdominal palpation findings, packed cell volume (PCV), serum albumin concentration, abdominal ultrasound findings, and results for feline immunodeficiency virus (FIV) and feline leukaemia virus (FeLV) tests were recorded, when available. A maximum likelihood χ² test was used to test for an association between categorical variables and the grade of AL (LGAL versus IGAL/HGAL). However, the Fisher’s exact test was used when the assumptions of χ² test were invalid, ie, either expected frequencies were 0 in any cell of the contingency table or were <5 in more than 20% of the cells. Continuous variables (age and body weight) were compared between grades using two-sample t-test if the assumptions of normality and equal variances were valid or using Mann–Whitney U test if they were not. The level of significance for this study was P <0.05. All the statistical analyses were conducted using GenStat ver 13 (VSN International).

Of the 30 cases diagnosed by histopathology, 15 (50%) were LGAL, two (7%) were IGAL and 13 (43%) were HGAL. A further 23 cases of IGAL/HGAL were diagnosed by cytology. Overall, 28% (15/53) of the AL were LGAL. Sites of AL sampled by FNA (n = 23), excisional biopsy (n = 24) or endoscopic biopsy (n = 6), were mesenteric lymph node (n = 6), stomach (n = 7), small intestine (n = 26), small intestine and mesenteric lymph node (n = 10) and colon (n = 4). There was no statistical association between anatomical site and the grade of lymphoma (P = 0.97). Overall, domestic shorthair cats were most commonly affected (n = 24, 45%), while Burmese were the most common purebred (n = 6, 11%). This is likely to reflect the popularity of domestic shorthair and Burmese cats in Australia, ¹⁶ rather than a true association between breed and the development of AL. There was no significant association between pedigree status and the grade of AL (P = 0.21). Female cats (n = 31, 58%) predominated slightly over male cats (n = 22, 42%). There was no statistical association between sex and the grade of lymphoma (P = 0.89). The median age of cats from the combined population was 12.6 years, and the mean age of both subpopulations was 12.16 years.

The most common clinical signs in all grades of AL were weight loss (n = 30, 58%), inappetence (n = 28, 54%), vomiting (n = 18, 35%) and diarrhoea (n = 13, 25%). There was no statistical difference in the prevalence of these clinical signs between cats with LGAL or IGAL/HGAL. Abdominal palpation was abnormal in 23/52 (44%) cats. Interestingly, cases with IGAL/HGAL were no more likely to present with palpable abnormalities compared with cats with LGAL (P = 0.70). However, when abnormalities were detected, the described changes varied between the two groups of cats. A palpable abdominal mass was detected in 16/37 (43%) cats with IGAL/HGAL compared with just 1/15 (7%) cats with LGAL. This association was statistically significant (P = 0.01). Palpable diffuse intestinal thickening was significantly (P = 0.012) more common in cats with LGAL (n = 6, 40%) compared with those with IGAL/HGAL (n = 3, 8%).

Basic details of an abdominal ultrasound examination were recorded in 45 cases, comprising 34 IGAL/HGAL and 11 LGAL. The presence of a mass associated with the gastric or intestinal wall on abdominal ultrasound (14/45, 31%) was correlated with IGAL/HGAL. All 14 cats with a gastric or intestinal mass were diagnosed with IGAL/HGAL, and this association was significant (P = 0.01). A descriptive ultrasound report detailing the appearance of the gastrointestinal tract was provided for 33 cats. Segmental gastric or intestinal wall thickening was the most common abnormality in cats with LGAL (10/11, 91%) or IGAL/HGAL (17/22, 77%). The remaining six cats had evidence of a mesenteric lymphadenopathy only. Of the 27 cats with intestinal wall thickening, loss of normal wall layering was significantly (P = 0.02) more common in cats with IGAL/HGAL (12/17, 71%) compared with those with LGAL (2/10, 20%).

Fourteen of 47 (30%) cats where PCV values were recorded were anaemic (PCV<30%). Anaemia was significantly more common in cats with IGAL/HGAL (13/32, 41%) compared with LGAL (1/15, 7%) (P = 0.02). Six of 24 (25%) cats with IGAL/HGAL that had serum albumin measurements recorded were hypoalbuminaemic (<24 g/l). In comparison, none of the 15 cats with LGAL were hypoalbuminaemic, although the association was not significant (P = 0.07). Ten cats with IGAL/HGAL and five cats with LGAL were tested for FIV antibodies and FeLV antigen using serology. Three of the 15 (20%) cats were seropositive for FIV, two of which were diagnosed with IGAL/HGAL and 1 with LGAL (P = 1.00). None of the 15 cats tested positive for circulating FeLV antigen.

Overall, approximately one quarter of AL in this study was low-grade (28%, 15/53). This is lower than previous findings from referral institutions, where LGAL accounted for 37%, 45%, 67% and 75% of AL, respectively.^6,10,12,15 It is likely that the lower prevalence of LGAL reported in our study is at least partially attributable to the inclusion of cases diagnosed by FNA cytology. In fact, the prevalence of LGAL in our population would have been even lower if cases diagnosed by in-house cytology had not been excluded. Furthermore, exclusion of all cases diagnosed by cytology would have nearly doubled the reported prevalence of LGAL in our study. Studies documenting the relative prevalence of LGAL that include only cases diagnosed histologically,^6,10,12,15 are therefore likely to provide a significant underestimation of the true prevalence of intermediate- and high-grade disease in both the general and referral feline population.