Exocrine Pancreatic Neoplasia in the Cat: A Case Series
Eight cases of feline pancreatic adenocarcinoma and two cases of pancreatic adenoma were reviewed. The adenomas were incidental findings. Most cats with adenocarcinomas had anorexia (75%) and vomiting (63%), while 38% had abdominal pain, a palpable abdominal mass, and/or jaundice. Diagnostic abnormalities included leukocytosis, hyperglycemia, increased alanine aminotransferase activity, poor serosal detail on abdominal radiography, and an abdominal mass effect on ultrasonography. The majority of cats with carcinomas had metastases (mostly to liver, lung, and small intestine), and all were euthanized or died within 7 days of diagnosis. Clinically, feline pancreatic carcinoma may be difficult to distinguish from feline pancreatitis.
Introduction
Neoplasia of the exocrine pancreas is rare in cats, with estimates of 12.6/100,000 patients per year at risk.1 In one report, pancreatic tumors were found in five of 800 feline necropsies.2 Interest in feline exocrine pancreatic disease, specifically acute and chronic pancreatitis, has dramatically increased in the last 10 years.3–9 Experienced ultrasonographers can now identify the feline pancreas, although the sensitivity of ultrasonography for identifying normal and diseased feline pancreas is debatable.10–14 Cats with pancreatitis may have enlargement of the pancreas, mass effects within the pancreas, fluid accumulation around the pancreas, or increased echogenicity of the pancreas.13 These changes do not distinguish between inflammatory and neoplastic conditions, however.
Pancreatic adenocarcinoma may be found in cats with diabetes mellitus. In a recent study of feline diabetes mellitus that included pancreatic histopathology, eight of 37 cats had pancreatic neoplasia (seven with adenocarcinoma and one with an adenoma).15 In these cats, no correlation was found between the presence of pancreatic neoplasia and glycemic control or survival time, making it difficult to predict which diabetic cats might have pancreatic neoplasia. To further complicate matters, cats have also been identified with concurrent pancreatitis and pancreatic adenocarcinoma.15
Previous studies of exocrine pancreatic adenocarcinomas in cats have described a dermatologic paraneoplastic syndrome associated with the tumors, but they have provided minimal information on clinicopathological and imaging data.16–19 The purpose of this study was to review cases of feline pancreatic exocrine tumors to better characterize the clinical signs, physical examination findings, clinicopathological data, diagnostic imaging results, response to therapy, metastatic pattern, and outcomes associated with these tumors.
Materials and Methods
The medical records of the University of Tennessee College of Veterinary Medicine (UTCVM) were searched from 1980 to 2000 for cats with exocrine pancreatic neoplasia. Cases were included if cytopathological or histopathological confirmation of neoplasia was made. Age, sex, breed, historical information, and physical examination findings were recorded. Complete blood counts, serum biochemical analyses, urinalyses, other clinicopathological tests, and radiographic and ultrasonographic findings were noted, as well as any concurrent diseases, attempted therapy, and outcome. Clinical staging was performed according to the World Health Organization system [Table 1].
Results
Over the 20-year study period, 15,764 cats were evaluated at UTCVM, and 10 cats fit the inclusion criteria listed above. Two cats had pancreatic adenomas (0.013% of admissions), and eight had pancreatic adenocarcinomas (0.05% of admissions). The signalment, clinical signs and duration, physical examination findings, diagnostic findings, and concurrent diseases of these cats are summarized in Table 2. Mean and median ages of the cats were 11 and 12 years, respectively, with a range of 4 to 20 years.
Cats with adenomas presented with signs arising from their concurrent diseases, which included chronic renal failure, polyneuropathy, thromboembolic disease, and bacteremia in case no. 9, and chronic renal failure and gastric mineralization in case no. 10. In both cases, there were no apparent clinical signs associated with the pancreatic adenoma; it was an incidental finding at necropsy. Clinical signs in cats with adenocarcinomas included anorexia (75%), vomiting (63%), abdominal pain (38%), a palpable mass in the cranial abdomen (38%), and jaundice (38%). Duration of clinical signs appeared to be bimodal, with 50% of cats having signs for <7 days and the other 50% having signs for >1 month.
Abnormal clinicopathological results are summarized in Table 2. Among cats with adenocarcinomas, 4/8 (50%) had neutrophilia, 2/6 (33%) had elevations in serum alkaline phosphatase activity, and alanine aminotransferase (ALT) activity was increased in 4/7 (57%) cats. Only two of these cats (case nos. 7, 8) had concurrent diseases (i.e., mucopurulent cholangitis and diabetes mellitus, respectively) that are commonly associated with increased liver enzymes. Mild hyperglycemia was identified in 5/8 (63%) cats. Case no. 7 had a twofold increase in amylase, and case no. 8 had a twofold increase in lipase. These enzymes were not evaluated in the other cats. Pre- and postprandial bile acids were increased in the only cat evaluated, and partial thromboplastin time (PTT) was significantly prolonged in the one case in which it was measured. Three cats were azotemic; in one, the urine specific gravity supported a renal origin of the azotemia.
Radiographic and ultrasonographic findings, tumor type and distribution, clinical stage, and concurrent diseases are summarized in Table 2. Seven cats with adenocarcinomas had abdominal radiographs taken, and all radiographs were abnormal. In four cats, a mass was visible, but it could only be localized to the pancreas in one cat (case no. 1). In five cats, there was loss of serosal detail consistent with peritoneal inflammation [Figures 1A, 1B]. Thoracic radiographs were taken of four cats; of these, two cats were normal and two had pleural effusion. Abdominal ultrasonography was done on four cats. A pancreatic mass was identified in two of the cats, and a nonlocalized mass was seen in one [Figures 2A, 2B].
Diagnosis of pancreatic neoplasia was made at necropsy in eight cats, including both cases of pancreatic adenoma. Four of the necropsied cats had ultrasound-guided fine-needle aspirates of an abdominal mass (n=3) or peritoneal effusion (n=1) prior to necropsy. In two of these cases, the aspirate was diagnostic for malignancy, and in two cases it was nondiagnostic. Of the six cats with adenocarcinomas that were necropsied, two were stage T1N0M0, two were T1N1M1, one was T1N2M1, and one was T1N0M1. Four cats had metastases to a variety of sites, including the liver (n=3), lung (n=3), small intestine (n=3), heart, diaphragm, and mesenteric, tracheobronchial, sublumbar, cranial mediastinal, and hepatic lymph nodes (n=1 each). Six of the eight adenocarcinomas involved the entire pancreas. In one cat, the tumor was confined to the left lobe, and in another cat the tumor was in the body of the pancreas. The adenocarcinomas appeared as one large mass (6/8) or as scattered nodules (2/8). Both cats with adenomas had multiple nodules scattered throughout the pancreas.
Case no. 8 had the most significant concurrent diseases, which included pancreatic adenocarcinoma, hyperadrenocorticism, and diabetes mellitus with insulin resistance. Two adrenocorticotropin (ACTH) stimulation tests were performed in this cat. On the first test, baseline cortisol was slightly elevated (3.9 μg/dL; reference range, 1.8 to 3.7 μg/dL), and the post-ACTH (12 μg/dL; reference range, <13 μg/dL) showed normal stimulation. Results of a second ACTH-response test performed the next week were similar (pre-ACTH cortisol of 5.9 μg/dL and post-ACTH cortisol of 13 μg/dL). Results of both tests were suggestive of, but not conclusive, for hyperadrenocorticism. On a low-dose (0.01 mg/kg intravenously) dexamethasone suppression test, baseline cortisol was 13 μg/dL (reference range, 1.8 to 3.7 μg/dL), cortisol at 30 minutes was 11 μg/dL (reference range, <1.5 μg/dL), and cortisol at 60 minutes was 13 μg/dL (reference range, <1.5 μg/dL). These results were consistent with feline hyperadrenocorticism.20 These adrenal function tests in combination with insulin-resistant diabetes mellitus, alopecia, and bilateral adrenomegaly on ultrasonography supported the diagnosis of feline hyperadrenocorticism. The alopecia in this cat was nonpruritic, bilaterally symmetrical, and affected both forelimbs and the head immediately caudal to the ears.
Treatment was attempted with supportive care in seven cats with adenocarcinomas. Four were given intravenous fluids, five were given antibiotics, and one each was given fentanyl, vitamin K, prednisone, and glipizide (given to the cat with diabetes mellitus). None of these treatments improved the clinical condition of the cats.
Discussion
The most remarkable aspect of the study reported here was the variability of signalment, presentation, and diagnostic findings. Few generalities emerged. The age range (4 to 20 years) was large. Among 20 previously reported cases, only three cats were <10 years of age.16–24 In this study, 3/8 cats were <10 years old. There was no sex predisposition in this series of cats, which was consistent with previous reports.16–24 The study population was too small to make generalizations regarding breeds affected; although in previous reports, no clear breed predisposition emerged.16–192425
Clinical signs included anorexia, vomiting, abdominal pain, a palpable abdominal mass, and/or jaundice in the cats with adenocarcinomas. Many previously reported cases presented with severe weight loss despite a normal appetite, which was not evident in this series of cats.216181924 Previous studies occasionally identified a palpable abdominal mass, but in other reports a mass was not palpable.216–1924 Clinicopathological changes were inconsistent, although leukocytosis and elevations in liver enzyme activity were common. Leukocytosis and liver enzyme elevations have been reported in some cases of feline pancreatic carcinoma but not in others.16181925 Abdominal radiography was able to identify abdominal disease in all cases of pancreatic adenocarcinoma (e.g., a mass or poor serosal detail). Some prior reports found abnormalities in abdominal radiographs in cats with pancreatic neoplasia, while others did not.16–1924 Abdominal ultrasonography was more helpful than plain radiographs for localizing a mass to the pancreas (50%). This was consistent with previous studies, in which some ultrasonographic examinations were unremarkable and others identified abdominal disease.1624
Two of four cats with pancreatic adenocarcinomas, for which thoracic radiographs were available, had pleural effusion. Both of these cats also had peritoneal effusion. One of these cats was necropsied and had metastases to the lung and diaphragm. The pleural effusion may have arisen as a direct result of tumor interfering with venous or lymphatic drainage from the chest, or as a result of release of active digestive enzymes from the abnormal pancreas, causing vasculitis, as can occur with pancreatitis.11 Half of the cytopathological examinations performed in the cats of this study were diagnostic for malignancy. This was lower than prior results in dogs, where fine-needle aspirates of pancreatic masses were diagnostic for carcinoma in 7/8 cases.25
In the study reported here, the incidence of metastasis seemed to be unrelated to the duration of clinical signs. Among the eight cats for which necropsy results were available, four cats had metastases. Their durations of clinical signs were 4 days (n=2), 7 days, and 7 months. The two cats without detectable metastasis had durations of clinical signs of 7 days and 5 weeks. Interestingly, the liver, lung, and small intestine were equally affected with metastases. The high occurrence of small intestinal metastases has not been described previously. In this series of cases, the tumor diffusely affected most of the pancreas in 75% of the cats, unlike a previous report that found the head of the pancreas to be involved more than the tail.23
None of the cats in this study received any therapies specifically designed to treat their cancer. Not surprisingly, supportive care alone was ineffective. Few attempts have been made to treat pancreatic cancer in small animals. One cat with pancreatic adenocarcinoma treated with surgical resection had 18 weeks of clinical remission, but signs recurred when metastases developed in the liver, spleen, and peritoneum.24
In humans, pancreatic carcinoma remains the fourth most common cause of cancer-related death in the United States; it occurs most commonly in people who smoke or have chronic pancreatitis or diabetes mellitus.26 Despite advances in multimodality therapy combining surgery, radiation, and chemotherapy, 5-year survival rates are consistently ≤18% in people.27 In the future, the use of gemcitabine as a radiosensitizer or manipulation of cyclooxygenase or lipoxygenase may improve survival times in humans.2829 The major cause of therapeutic failure in humans with pancreatic cancer is the advanced stage of disease at the time of diagnosis; therefore, current efforts focus on early detection. Conventional diagnostic imaging may have difficulty distinguishing chronic pancreatitis and pancreatic carcinoma; therefore, screening of pancreatic secretions for mutations associated with malignancy is under development as a potentially more sensitive and specific diagnostic tool.30 The etiology, environmental risks, and underlying mutations associated with feline pancreatic cancer are unknown.
The association of diabetes mellitus, hyperadrenocorticism, and exocrine pancreatic neoplasia as seen in case no. 8 has been reported previously in two cats.21 It is unknown whether diabetes mellitus develops secondary to compression of islet cells by tumor, secondary to cortisol-induced beta-cell degeneration, or secondary to decreased carbohydrate metabolism (which is common in humans with pancreatic carcinoma).26 In humans, long-standing diabetes mellitus is a risk factor for the development of pancreatic cancer.26 It is not yet clear if this association is also true in the cat. There is circumstantial evidence linking the diseases, as a recent study found that 8/37 cats with diabetes mellitus had pancreatic neoplasia.15
Previous reports of pancreatic adenocarcinomas in cats have identified a paraneoplastic dermatological condition consisting of nonpruritic, symmetrical alopecia affecting the face, ventral body, and medial aspect of the limbs of cats. The skin was usually glistening but not fragile, and there were some crusty lesions on the footpads.16181924 Only the diabetic cat in this series developed alopecia consistent with this dermatological condition. A similar paraneoplastic alopecia has also been reported in cats with bile-duct carcinomas.1631 The pathogenesis of this dermatological disease is unknown. Histopathologically, it is characterized by loss of the stratum corneum and severe follicular atrophy with miniaturized hair bulbs.161824
Interpretation of the significance of the data presented in the study reported here was compromised by the high incidence of concurrent disease in these cats and by the similarity of their clinical signs with those of acute pancreatitis. Most cats with acute pancreatitis have anorexia, lethargy, vomiting, weight loss, and potentially abdominal pain or icterus as the predominant clinical signs.4–681112 Cats with experimentally induced pancreatitis often have fever and tachycardia, with or without abdominal pain, and rare vomiting.32
The clinicopathological changes often found in cats with pancreatitis are also similar to the changes seen in the cats with pancreatic carcinoma in this study. In a prior study of cats with histopathologically confirmed pancreatitis, 62% had leukocytosis, 38% were hyperbilirubinemic, and 24% had increased ALT activity.12 This same study reported that 6/21 cats had decreased serosal detail in the cranial abdomen on radiographs, and 5/21 had an enlarged hypoechoic pancreas on ultrasonography. All these findings are similar to those reported here for cats with pancreatic carcinoma.
Few diagnostic tests are capable of distinguishing pancreatitis from pancreatic carcinoma in cats antemortem, yet it is important to differentiate the two conditions, because pancreatic adenocarcinoma is associated with a grave prognosis. The presence of lesions consistent with metastases on radiography or ultrasonography may suggest malignancy, but cytopathological or histopathological confirmation is required for more diagnostic certainty. Computed tomography and magnetic resonance imaging have been used to attempt visualization of the feline pancreas, but results have been conflicting.123334 Unfortunately, no data is available regarding the results of imaging pancreatic adenocarcinoma versus pancreatitis using either modality in the cat.
Serum trypsin-like immunoreactivity has been reported to be the most sensitive test for feline pancreatitis, but other studies have found it to be a poor predictor of pancreatic inflammation.123437 Trypsin-like immunoreactivity has only been evaluated in one cat with pancreatic cancer, and it was within the reference range.24 Amylase and lipase, which are useful in the diagnosis of pancreatitis in dogs, are not very helpful in cats.3538 Dogs with pancreatic or hepatic neoplasia may have significant elevations in serum lipase with only minimal elevations of serum amylase, implying tumor production of lipase.3639 Amylase and lipase were mildly elevated in two cats in this study, but these enzymes were evaluated in too few cats to allow reliable conclusions to be drawn. One other report found significant elevation of only lipase in a cat.25 Because these abnormalities are not very useful diagnostically, cytopathology and histopathology are currently the best methods to differentiate between inflammation and neoplasia of the pancreas.
Conclusion
In this study, it was found that pancreatic adenomas in the cat were not associated with clinical signs, but that pancreatic adenocarcinomas caused anorexia, vomiting, and abdominal pain. Physical examination identified a palpable abdominal mass and/or jaundice in cats with pancreatic adenocarcinomas. Clinicopathological changes were variable. Radiographic loss of serosal detail was common, and a mass was sometimes visible on abdominal ultrasonography. These clinical and imaging findings were similar to those associated with feline pancreatitis. Most pancreatic adenocarcinomas had metastasized at the time of diagnosis, and most cats died or were euthanized within 7 days of diagnosis.
Citation: Journal of the American Animal Hospital Association 40, 3; 10.5326/0400238
Citation: Journal of the American Animal Hospital Association 40, 3; 10.5326/0400238
Lateral and ventrodorsal abdominal radiographs of an 11-year-old domestic shorthair cat with pancreatic adenocarcinoma (case no. 1). Note the loss of serosal detail in the abdomen.
