Introduction

A 14-year-old 5.6-kg (12.3-lb) castrated male Toy Poodle was referred for evaluation of a hepatic mass. Results of routine clinicopathologic analyses performed a month earlier indicated that the dog had high serum activities of ALT (427 U/L; reference range, 22 to 74 U/L) and ALP (1,379 U/L; reference range, 12 to 116 U/L). At that time, abdominal ultrasonography revealed a mass (approx 5 cm in diameter) in the left medial liver lobe. An ultrasound-guided fine-needle aspirate specimen was obtained for cytologic evaluation, the results of which were not diagnostic.

At the referral examination, the dog was bright and alert, and its heart rate, respiratory rate, and rectal temperature were within reference limits. There was no detectable lymphadenopathy. On palpation, the dog's abdomen was soft and apparently nonpainful with no palpable masses or fluid wave. Presurgical 64-slice CT of the dog's abdomen to locate and delineate the mass was scheduled for the following week. Pre- and postcontrast (immediate and delayed) CT images of the abdomen were obtained. Along the midline of the liver, there was a soft tissue mass (approx 6.8 X 7.1 X 6.5 cm) with mild to moderate contrast enhancement. Cranially, the mass extended to just caudal of the left divisional branch of the hepatic vein. Caudally, the mass abutted the left divisional branch of the portal vein. The most caudal aspect of the mass contained a 1.7-cm-long (the longest dimension in transverse images) fluid component. In the ventral aspect of the caudate process of the caudate lobe, there was a 1.8-cm-long (the longest dimension in transverse images), non–contrast-enhancing cyst. Other CT findings included a few small contrast-enhancing splenic nodules and small cortical renal cysts; there were a few small mineralized bodies in both the left caudal portion of the abdomen and right inguinal region.

After completion of CT, a partial liver lobectomy was performed to remove as much of the solitary mass as possible. The dog was administered methadone and propofol, and then anesthesia was maintained at an appropriate surgical level by inhalation of isoflurane with oxygen. A midline laparotomy incision was created to gain access to the liver. An electrosurgical device^a was used to divide omental adhesions to the hepatic mass. Two ligating loops^b were used to ligate and resect the largest accessible portion of the mass. Oxidized regenerated cellulose^c was applied to the exposed surface of the tumor to control local capillary hemorrhage. Approximately a third of the mass was successfully removed. The abdomen was lavaged with sterile saline (0.9% NaCl) solution; after the fluid was suctioned from the abdomen, the linea alba was closed with 2-0 polydioxanone suture^d in a continuous suture pattern. The skin was apposed with 3-0 polydioxanone suture^e in a continuous intradermal suture pattern followed by application of skin staples.^f

Histologic examination of sections of the mass revealed well-differentiated hepatocytes arranged in cords (Figure 1). Most of the hepatocytes were swollen with poorly defined to occasionally distinct, round, unstained cytoplasmic vacuoles. Other microscopic findings included mild anisocytosis and mild anisokaryosis; there were 2 mitotic figures/10 hpf (400X). Adjacent unaffected hepatic parenchyma was not observed in the examined tissue sections. Given the location of the dog's tumor, the top 2 differential diagnoses were hepatocellular adenoma and well-differentiated hepatocellular carcinoma. On the basis of the histopathologic findings, the reporting patholo-gist favored a diagnosis of adenoma. The mass had arrangements of cords composed of relatively bland hepatocytes (ie, the proliferating cells had minimal atypia with few cytologic criteria of malignancy), although the clinical behavior of this mass was not representative of an adenoma. Histologically, hepato-cellular adenomas typically have well-differentiated hepatocytes of bland appearance that are arranged in cords with absence of portal triad and interlobular bile ducts, which aids in the distinction from normal adjacent liver tissue. However, because this mass was incompletely excised, tumorous and nontumorous hepatic tissue could not be compared to achieve a definitive diagnosis. Consultation with the oncology service after 1 month to determine whether radiation therapy could be used to treat the dog's residual disease was advised.

View Full Size

Photomicrograph of a section of the liver mass in a 14-year-old Toy Poodle that was removed during partial liver lobectomy. Notice the attempted formation of cords by relatively bland hepatocytes with low criteria of malignancy. The reporting pathologist favored a diagnosis of adenoma on the basis of this morphological feature of the mass but could not provide a definitive diagnosis without evaluation of the tumor junction (tumorous vs nontumorous hepatic tissue) because the mass was partially resected. Evidence of invasion into adjacent parenchyma is often used as a distinguishing factor between hepatocellular adenoma and carcinoma. H&E stain; bar = 50 μm.

Citation: Journal of the American Veterinary Medical Association 259, 4; 10.2460/javma.259.4.392

• Download Figure
• Download figure as PowerPoint slide

One month following surgery, the dog underwent repeated CT; the findings indicated tumor regrowth. At this time, serum activities of ALT (675 U/L), ALP (1,570 U/L), and γ-glutamyltransferase (171 U/L; reference range, 0 to 10 U/L) were high. Compared with the initial CT measurements, the hepatic mass was only slightly smaller (6.7 X 6.3 X 5.6 cm) at this time. No other hepatic nodules or masses were identified. A diagnosis of hepatocellular carcinoma best fit the disease characteristics. Owing to the inability to remove the entire mass during the initial surgery and evidence of rapid recurrence, radiation therapy was recommended.

On 5 consecutive days, the dog received treatment with IMRT-based SBRT with a simultaneous integrated radiation boost to the center of the tumor. The dog received a base dose of 20 Gy that was then boosted to 28 Gy. The 2-mm postcontrast CT images and a computerized 3-D treatment planning system^g were used to guide the radiation therapy. Contrast-enhancing gross tumor delineated with a CT scan was defined as the gross tumor volume. There was no further expansion for sub-clinical disease, and the planning target volume was defined as the gross tumor volume. Seven-field static beam IMRT was delivered with a linear accelerator.^h The prescription dose for the planning target volume was 4 Gy on each of the 5 consecutive days (Monday through Friday) dosed to the 95th isodose line. The 95th isodose line was used to normalize the dose and thus increase the dose by 5%, allowing the operator to improve target volume coverage and facilitate a sharper dose drop-off. There were 3 simultaneously integrated boost zones defined as 5-, 10-, and 20-mm inner margins from tumor delineation. Concentric subvolumes were used to maneuver the radiation dose into the center of the tumor and away from the neighboring normal tissues. The center of the tumor was the region boosted in doses. The mean percentages of the radiation dose received by the 5-, 10-, and 20-mm inner margins were 116.9%, 120.4%, and 134.1%, respectively. The prescription dose to the 5- and 10-mm inner margins was 25 Gy; the prescription dose to the 20-mm inner margin was 28 Gy. The intent was not only to push the radiation dose toward the center of the tumor and away from normal adjacent tissues but also to boost the radiation dose to center of the tumor. The same prescribed dose was applied to the 5- and 10-mm inner margins in an effort to provide conservative radiation therapy because this large liver mass was the first the authors had treated in this manner, and internal volumes had been used to calculate the radiation dose. The liver contour was defined as the liver-planning target volume; the maximum, minimum, and mean radiation doses to the liver were 24.40, 0.48, and 10.18 Gy, respectively. The stomach, kidneys, and pancreas radiation doses were all within previously defined dose tolerance limits.¹ Each treatment day, positioning of the patient was verified by matching the cone-beam CT imaging with the pretreatment radiation planning CT scan.

The dog had no complications during radiation therapy and did well clinically thereafter. Repeated CT and clinicopathologic analyses were performed 3 months after completion of radiation therapy. Recheck imaging revealed no evidence of distant or local meta-static disease; the previously detected focal soft tissue opacification on the ventral tip of the left caudal lung lobe was no longer present. The liver mass measured 4.5 X 4.0 X 5.0 cm. The dog's serum activities of ALP (874 U/L) and ALT (304 U/L) had improved.

At 6 months after completion of radiation therapy, thoracic and abdominal CT and clinicopathologic analyses were performed. Computed tomography revealed that the liver mass had reduced further in size (approx 3.8 X 4.7 X 5.1 cm). The remainder of the liver, including the small cyst in the caudate lobe, was unchanged. However, there was another soft tissue nodule (maximum transverse dimension, 0.6 cm) in the caudodorsal aspect of the right caudal lung lobe adjacent to a tertiary bronchus. No additional pulmonary lesions were detected. Repeated serum biochemical analysis revealed continued decreases in the dog's ALP activity (185 U/L) and ALT activity (425 U/L).

At 10 months after completion of radiation therapy (ie, 11 months after surgery), CT revealed further reduction in the size of the liver mass (3.2 X 3.9 X 4.7 cm); at this time, the mass was approximately 75% smaller than its size prior to radiation therapy and 36% smaller than its size 4 months earlier. On the basis of published criteria for solid tumors,² the response of the mass to radiation therapy was partial, characterized by a > 30% reduction in overall tumor size. However, the dog's serum activities of ALP (529 U/L) and ALT (222 U/L) were slightly elevated, compared with findings at 6 months after completion of radiation therapy. The owner reported that the dog had been doing well. A recheck examination and repeated clinicopathologic analyses in 1 month were recommended; however, the dog was lost to follow-up.

Discussion

In dogs, approximately 50% of primary liver neoplasms are hepatocellular carcinomas; such tumors most commonly affect dogs ≥ 10 years old.³ Conversely, hepatocellular adenomas in cats and dogs seldom cause clinical signs. Hepatocellular adenomas are often incidental findings and are more prevalent in cats.⁴ In addition, differentiation of adenomas from carcinomas on the basis of morphological characteristics alone can be challenging, and histologic evidence of invasion into adjacent parenchyma is often used as the distinguishing factor. Thus, it is important to be able to compare tumorous to nontumorous hepatic tissue. The prognosis for surgically treated dogs with well-differentiated hepatocellular tumors is excellent. Resection of hepatocellular tumors is recommended because even liver lobectomy with incomplete margins or a partial liver lobectomy is associated with low tumor recurrence and distant metastatic rate.⁵ High circulating liver enzyme activities indicate hepatocellular injury caused by either tumor-induced hepatic tissue compression or aggressive tumor behavior and are considered to be poor prognostic indicators. Hepatocellular adenomas are not typically associated with a fast rate of postoperative tumor regrowth or severe postoperative increases in liver enzyme activities. Because the dog of the present report had high serum liver enzyme activities even after a debulking procedure, the concern was that the carcinoma's behavior was aggressive. Radiation therapy was elected to stabilize or decrease tumor size.

Stereotactic radiation is defined as the precise delivery of highly conformal ablative doses of radiation outside of the brain; doses are generally delivered in 1 to 5 treatments to well-defined targets. To the authors' knowledge, there are no published data in the veterinary medical literature regarding the use of SBRT to control residual liver cancers; however, this treatment is becoming more common in human medicine. In human medicine, various SBRT fractionation regimens have been shown to be well tolerated by patients, but such treatments have a variable success rate with regard to local tumor control. In 41 humans with either hepatocellular carcinoma or intrahepatic cholangiocarcinoma, 6-fraction SBRT with a median dose of 36 Gy was administered. Of the 41 patients with hepatocellular carcinoma and preexisting large vessel thrombosis, there were a wide range of responses (6% had a complete response, 19% had a partial response, and 38% had stable disease), and the median survival rate was 11.6 months.⁶ Additionally, there is a report⁷ of a patient with early-stage hepato-cellular carcinoma who received a total dose of 60 Gy administered in 3 fractions; this regimen was considered successful for that patient. The mass decreased in size in 2 months, and the patient's circulating liver enzyme activities normalized in 8 months after completion of SBRT. However, liver enzyme activities should not be solely used to determine treatment success because eventual increases in liver enzyme activities may be an adverse effect of radiation therapy.

In veterinary medicine, there has been 1 study⁸ investigating the use of 3DCRT on hepatocellular carcinomas in dogs. With 3DCRT, a conformal radiation dose is delivered to tumors while surrounding normal structures are spared. However, 3DCRT lacks the precision of IMRT-based SBRT, which greatly spares adjacent normal tissues. Theoretically, 3DCRT would be associated with more radiation-induced adverse effects than IMRT-based SBRT. In a study⁸ of 6 dogs with massive hepatocellular carcinomas that were ineligible for resection, 3DCRT with a cumulative dose of 18 to 42 Gy was administered in 3 to 7 fractions. A partial response was reported for 5 dogs. Radiation-induced adverse effects were reported for only 1 dog; those effects were not associated with clinical signs, were reversible, and occurred within 4 months after completion of 3DCRT. The median survival time for the 6 dogs was 567 days.⁸

In the case described in the present report, the combination of IMRT-based SBRT with a simultaneous integrated radiation boost to the center of the tumor allowed delivery of different doses of radiation to different target areas in a single treatment fraction, which in theory would spare the healthy surrounding tissue. In the human medical literature, the radiation doses used to treat hepatocellular carcinomas vary and are dependent on size and the number of lesions. In 1 study,⁹ 47 patients with inoperable hepatocellular carcinoma were treated with SBRT as a local salvage treatment after incomplete transarterial chemoembolization, and 42 to 60 Gy was provided in 3 fractions. Among the 47 patients, the overall survival rate was 68.7% with a reported 2-year local control rate of 94.6%; 18 (38.3%) patients had a complete response. In another study,¹⁰ 102 patients with hepatocellular carcinoma were treated with 24 to 54 Gy in 6 fractions. Among the 102 patients, the reported 1-year local control rate was 87% with a complete response in 11% of patients. For the dog of the present report, the same prescribed dose was applied to the 5- and 10-mm inner margins in an effort to provide conservative radiation therapy because this large liver mass was the first the authors had treated in this manner, and internal volumes had been used to calculate the radiation dose. However, the authors have subsequently increased each subvolume when treating similar patients.

In the case described in the present report, the use of IMRT-SBRT on a well-differentiated hepatocellular carcinoma in a dog after partial liver lobectomy was successful in reducing the size of the remaining mass and normalizing circulating liver enzyme activities. This dog did not develop acute adverse effects; at 10 months after treatment, the procedure was not associated with late-term adverse effects and the dog had been doing well clinically. Although this particular SBRT plan was effective, more studies should be conducted to assess different radiation dosing and fractionation regimens while minimizing radiation-included adverse effects.