Primary hepatic neoplasms are less common than metastatic neoplasms in the liver and are either carcinomas, carcinoids, sarcomas, or of hemolymphatic origin. Metastatic neoplasia of the liver can originate from multiple visceral organs and can include lymphosarcoma. There may be no clinicopathologic features indicating the presence of metastatic neoplasia within the liver.
Primary tumors are most often found in older animals (>9 yr) and can be malignant or benign. The most common primary hepatic neoplasm in dogs is hepatocellular carcinoma, whereas the most common primary hepatic neoplasms in cats include biliary adenomas and carcinomas and cystadenomas (not considered a neoplasm) most common in older cats. Additional tumor types include hemangiosarcomas, carcinoids, and sarcomas in dogs; lymphomas and myeloproliferative disease in cats; and less frequently, leiomyosarcomas, GI stromal tumors, and myelolipomas.
Hepatocellular carcinoma is most common in dogs but also may occur rarely in cats. Diagnosis may be initially pursued because of palpation of an abdominal mass or recognition of serially increasing ALT, ALP, or GGT activities. Less commonly, mass lesions are discovered as the cause of critical abdominal hemorrhage. Radiography may disclose a large mass lesion or emphysematous abscess within a necrotic tumor core. Ultrasonography is more sensitive for detection of mass lesions and can discriminate single from multiple lobe involvement. Small hepatocellular carcinomas may appear hypoechoic, hyperechoic, or heteroechoic. However, the large size of some masses precludes clear differentiation of tumor impingement on or invasion into adjacent viscera and vasculature. Hepatocellular carcinomas can occur as a single large mass in one liver lobe with or without smaller masses in other lobes (massive), as discrete nodules located in multiple lobes (nodular), or as infiltrative disease throughout the liver without obvious discrete nodularity (diffuse). Nodular and diffuse hepatocellular carcinomas, which account for 29% and 10%, respectively, of all hepatocellular carcinomas, involve multiple liver lobes and are not usually amenable to surgical removal. Single massive hepatocellular carcinomas represent 61% of all canine hepatocellular carcinomas and are potentially resectable with good outcome. Tumors involving the left liver lobes carry the best prognosis. Hepatocellular carcinomas are locally invasive and may spread to local lymph nodes but are slow to metastasize elsewhere. Multiple histologic patterns have been characterized for canine hepatocellular carcinoma, including (but not limited to) solid, clear cell, trabecular, peliod, pseudoglandular, scirrhous, and anaplastic.
Common clinical signs in dogs include weight loss, inappetence, fever, and lethargy; less common signs include vomiting, PU/PD, and seizures (hypoglycemia). However, dogs may be asymptomatic until the tumor reaches massive size or develops a necrotic core leading to critical abdominal hemorrhage. On abdominal palpation, a mass may be detected, and pain notable. Abdominal effusion is uncommon. Laboratory tests may indicate nonregenerative anemia, RBC microcytosis, thrombocytosis, increased serum activity of ALP and AST, and hypercholesterolemia. High ALT and AST concentrations may reflect invasion of adjacent normal tissue or central tumor necrosis and may indicate a poor prognosis. Increased ALP may reflect association with increased steroidogenesis (particularly androgens and progestins). Hypoglycemia may develop either due to large tumor mass or a paraneoplastic effect. Pulmonary metastases are uncommon. Tumor margins should be demarcated on submitted specimens to judge adequacy of mass resection (tumor-free margin). Mass excision with wide margins can be curative in both dogs and cats. Liver biopsies from regions distal to the tumor are recommended in dogs, because antecedent dysplastic foci may portend possible tumor recurrence. Finding dysplastic foci should increase ultrasonographic and clinicopathologic (a sudden peak in ALP or transaminase activity) surveillance for tumor recurrence.
Differentiation of hepatocellular carcinoma from hepatocellular adenoma can be difficult, involving judgement of the degree of dysplasia and characterization of the histologic features. Like hepatocellular carcinomas, adenomas also may be accompanied by microscopic dysplastic foci in unaffected liver lobes. Hepatocellular adenomas and dysplastic foci are more common in dogs with atypical adrenal hyperplasia associated with increased androgen or progesterone concentrations. Both lesions are often associated with increased liver enzyme activity (ALP primarily, with more modest change in transaminases). Rather than single mass lesions, multiple dysplastic foci are commonly identified within a single biopsy sample and among multiple biopsies from the same animal. Neoplastic transformation of dysplastic foci is suspected.
Similar to hepatocellular carcinomas, hepatocellular adenomas may outgrow their central blood supply, developing a necrotic core that may serve as a nidus for abscess formation. Mass lesions also may rupture and cause critical abdominal hemorrhage. Hepatocellular adenoma is curable by wide resection, although recurrence is possible in dogs with dysplastic foci. Tumor margins should be demarcated on submitted specimens to enable microscopic assessment of the adequacy of mass resection (tumor-free margin).
Variably classified as cholangiocellular carcinoma or adenocarcinomas and hepatocellular adenocarcinomas, these tumors are the most common primary malignant hepatic neoplasm in cats and may derive from intrahepatic or extrahepatic bile ducts, pancreas, or gallbladder. Pancreatic adenocarcinomas, invasive into hepatic structures, are also common in cats. Immunohistochemical staining may be necessary to definitively characterize the tissue of origin (biliary epithelium vs hepatocellular). Biliary cysts can be mistaken on gross inspection for primary biliary adenocarcinomas.
Clinical signs usually include anorexia, lethargy, and vomiting, and some cats are jaundiced. Many cats have a history of antecedent liver disease based on historical biochemical profiles; histologically, the chronic liver disease is a nonsuppurative cholangiohepatitis. A mass or large liver may be palpable. Increased ALT, AST, ALP, and GGT activities and increased cholesterol and bilirubin concentrations are common. However, some cats with biliary adenocarcinomas have no clinical signs or laboratory abnormalities. Biliary tree obstruction is identified in some but not all cats with neoplasia associated with the common bile duct and gallbladder. Abdominal radiographs may disclose mass lesions disrupting the hepatic silhouette. Ultrasonography usually delineates mass lesions, their dimensions, and lobe location and documents the presence or absence of biliary tree and/or gallbladder obstruction. Some cats develop abdominal effusion and carcinomatosis.
Surgical resection of neoplastic lesions associated with biliary structures distal to the porta hepatis and lesions associated with the gallbladder is possible. Neoplasia involving the common duct may be palliated with stent placement through the sphincter of Oddi into the duodenum (poorly tolerated by cats) or surgical creation of a biliary diversion. Some cats survive for months with palliative supportive care (without surgery) despite total bile duct obstruction. However, the longterm prognosis is poor. Metastatic lesions are often found in the local lymph nodes, peritoneum, and lungs.
The most common hemolymphatic tumor found in the liver in both dogs and cats is lymphoma; this may be primary or metastatic (from primary enteric or multifocal disease). Other myeloproliferative diseases and mast cell neoplasia also can involve the liver, especially in cats. Animals with infiltrative hepatic lymphoma may remain asymptomatic in regard to clinicopathologic features except for hyperbilirubinemia and hepatomegaly. Ultrasonographic imaging may not recognize architectural changes or discover nodules or mass lesions. Abdominal effusion may develop secondary to compressive presinusoidal and sinusoidal intrahepatic portal hypertension. Sampling of the effusion and hepatic aspiration may allow cytologic diagnosis of lymphoma. Circulating blood should first be reviewed for presence of neoplastic lymphocytes.
These benign tumors are composed of adipose cells and hematopoietic elements. The cellular composition closely resembles cellular elements found in bone marrow. These tumors are usually serendipitously discovered during abdominal ultrasonography, appear densely hyperechoic, have a clearly demarcated border, and are usually small. Aspiration cytology can easily characterize the cellular features. Unless large vessels and biliary structures are compressed, these lesions do not require surgical removal.
The most common tumors metastasizing to the liver in dogs include lymphoma, pancreatic carcinoma, mammary carcinoma, pheochromocytoma, intestinal carcinoma, thyroid carcinoma, fibrosarcoma, osteosarcoma, hemangiosarcoma, mast cell tumors, and transitional cell carcinoma. Metastatic tumors of the liver are less common in cats but include pancreatic, intestinal, and renal cell carcinomas; mast cell tumors; and lymphoma. Metastatic tumors are often multifocal but may initially exist only within vascular and lymphatic elements.
Clinical signs can be nonspecific or specific to the liver and resemble features associated with primary hepatobiliary neoplasia, including anorexia, weight loss, vomiting, PU/PD, and variable hyperbilirubinemia. Metastatic hepatic neoplasia is more likely to be associated with a malignant abdominal effusion. Neurologic signs may indicate metastatic lesions within the brain, with associated clinical signs mistaken for HE. Abnormalities on baseline hematologic or chemistry profiles may be minimal. Although a nonregenerative anemia may develop, there are no consistent changes in WBC count or distribution. Schistocytes may be seen when neoplasia invades hepatic sinusoids, creating a shearing effect on RBCs. Eosinophilia can develop with mast cell tumors and lymphoma, especially in cats. Liver enzymes may be normal or variably increased. Hypoglycemia due to large tumor mass (enhanced glucose utilization) or a paraneoplastic effect (insulin-like effect) is sometimes identified. Hyperbilirubinemia and increased AST are more frequent in canine metastatic disease than in primary hepatic neoplasia. Although radiographic findings are variable, ultrasonography can discriminate single vs multiple lobe involvement from diffuse infiltrative disease. However, aspirates of mass lesions or biopsy is needed for definitive diagnosis. If one liver lobe is involved, surgical removal is recommended. If lymphoma or mastocytosis is diagnosed, appropriate chemotherapy may prolong life.
Hepatic hemangiosarcoma is treated by surgical resection as primary therapy if a defined mass lesion is characterized, with follow-up chemotherapy. There is no specific study of survival expectations of dogs with hepatic hemangiosarcoma, with or without surgical debulking or chemotherapy. Surgery is palliative to arrest active hemorrhage but does not provide longterm survival benefit for most dogs with liver involvement. Survival in dogs with grossly evident metastatic disease have the worst prognosis, with median survival ranging from 68–136 days. A number of combination chemotherapy protocols have been investigated in dogs with hemangiosarcoma, providing median survival ranging from 145 to 250 days in dogs with incisional or excisional biopsies (all sites of tumor not limited to the liver). In a chemotherapy study with doxorubin (free or pegylated liposome-encapsulated drug) compared with a random prospective clinical trial (n=17 for each group), there was no survival difference between treatments in dogs with splenic hemangiosarcoma. Pegylated liposome-encapsulated doxorubicin was associated with more adverse effects. Another combination protocol for dogs with advanced-stage noncutaneous hemangiosarcoma (doxorubicin, dacarbazine, and vincristine) with nonresectable stage II and stage III hemangiosarcoma (doxorubicin and dacarbazine day 1, vincristine days 8 and 15, protocol repeated every 21 days for a maximum of six cycles or until disease progression) was studied. In this study of 24 dogs, a 47.4% response rate (five complete responses, four partial responses) was realized; median time to tumor progression was 101 days with a median survival of 125 days. Significant toxicities included hematologic and GI adverse effects.