In dogs, increased serum ALP activity may reflect necroinflammatory, neoplastic, or cholestatic disorders involving hepatic, biliary, or pancreatic ductal systems; canalicular cholestasis (eg, sepsis impaired organic anion pumps); or simply induction phenomena (ie, endogenous or exogenous glucocorticoids or acute-phase response).1 Glycogen-associated VH is a common syndrome in dogs often associated with endogenous overproduction of steroidogenic hormones or treatment with glucocorticoid medications but may also accompany inflammatory, neoplastic, and infectious hepatic disorders.2 Increased hepatic production of ALP and its release into the systemic circulation accompanies VH, with the corticosteroid-induced ALP isoenzyme predominating.1 Greater than a 3-fold increase in serum ALP activity is common in dogs with VH.3
Histologically, VH in dogs is characterized by the cytosolic accumulation of glycogen within hepatocytes, leading to cell distention, increased fragility, and ballooning degeneration.2,3 Although some clinicians consider VH a benign transformation, it is our observation (SAC and SPM) that progressive VH leads to diffuse hepatic remodeling as vacuolated hepatocytes undergo individual hepatocyte necrosis (degenerative VH), resulting in the formation of parenchymal nodules and intrasinusoidal hypertension secondary to hepatic cord remodeling. In its extreme manifestation, degenerative VH leads to splanchnic hypertension, acquired portosystemic shunts, ascites, and hepatic insufficiency. Although degenerative VH can develop in any dog, we recognized progressive VH as a breed-related disorder in Scottish Terriers.4 Additionally, we identified a high incidence of HCC in this breed.4 Interestingly, a study5 in apparently healthy Scottish Terriers demonstrated a relationship between increased serum ALP activity and increased adrenocorticosteroid hormone concentrations following ACTH stimulation as well as a correlative association between older age and concentration of androstenedione.
The goal of the study reported here was to more fully characterize the phenotype of Scottish Terriers with the syndrome of glycogen-associated VH, including signalment, clinical features, clinicopathologic variables, hepatic ultrasonographic characteristics, hormonal assessments, histologic evaluation of liver specimens, treatment response, and age at death. We sought to determine whether sex, HCC, or pathological hepatic copper accumulation influenced age at death and to confirm the apparent high incidence of HCC.
Materials and Methods
Case selection—Electronic databases (1980 to 2013) of the College of Veterinary Medicine at Cornell University were searched for adult (age > 1 year) Scottish Terriers with histopathologic diagnosis of diffuse glycogen-like VH, HCC, or both. These databases were initially searched for the years 1980 through 2009 for a pilot study, but the study was later expanded to also include information retrieved from 2010 to 2013; criteria for selection remained unchanged throughout.
Histologic evaluation—Available H&E-stained 5-μm sections of neutral-buffered 10% formalin–fixed paraffin-embedded liver specimens underwent independent histologic evaluation by 2 of the authors (SAC and SPM) to select dogs for study inclusion. Glycogen accumulation was identified on the basis of the finding of wispy cytosolic vacuoles with central location of cell nuclei and was differentiated from presumed lipid accumulations appearing as discrete clear vacuoles. A subset of liver specimens (n = 10) was stained for glycogen with periodic acid–Schiff with and without diastase at the time of histopathologic diagnosis; however, specific staining of glycogen was not necessary to confirm hepatocyte vacuolar expansion with a glycogen-like material. Another subset of liver specimens (n = 35) was stained with rhodanine to inspect for possible copper-associated hepatopathy. In these dogs and in 1 additional dog, hepatic copper was quantified by atomic absorption spectroscopy (n = 13) or by digital scanning of rhodanine-stained tissue sections analyzed by means of a proprietary positive pixel algorithm and validated linear regression (23).6,a,b Copper accumulation was considered pathological if it was > 400 μg/g of liver DWB.
Medical record review—Medical records of each dog were reviewed to characterize signalment, clinical signs, clinicopathologic variables (ie, findings on CBC, serum biochemical profile, and urinalysis), coagulation test results (ie, prothrombin time, activated partial thromboplastin time, and serum fibrinogen concentration), abdominal ultrasonographic findings (eg, hepatic echogenicity, hepatic parenchymal nodules, hepatic mass lesions, and adrenal gland dimensions), steroidogenic hormone assessments (results of LDDST and ACTH-stimulation test for cortisol, aldosterone, and sex hormone concentrations), and endogenous plasma ACTH concentrations. Abdominal ultrasonographic examinations were completed by board-certified radiologists or experienced small animal internists. Referring veterinarians or owners of each dog were contacted by telephone to ascertain survival information.
Hormone assessments—Serum cortisol concentration was measured in a subset of dogs before and 1 hour after IV administration of synthetic ACTHc,d (5 μg/kg [2.3 μg/lb]). Plasma for determination of endogenous ACTH concentration was collected into EDTA-anticoagulated tubes in a subset of dogs and then either transported on ice for measurement at the Endocrinology Laboratory, Animal Health Diagnostic Laboratory, College of Veterinary Medicine, Cornell University, or frozen at −20°C before overnight shipping on dry ice for measurement at the Michigan State University Endocrinology Laboratory of the Diagnostic Center for Population and Animal Health, Lansing, Mich. In a subset of dogs, the LDDST was completed by collecting serum for determination of cortisol concentration before and 6 and 8 hours following administration of dexamethasone (0.015 mg/kg [0.007 mg/lb], IV). Serum cortisol and plasma ACTH concentrations were measured by means of either chemiluminescent methodse validated for use in dogs7,8 or radioimmunoassaysf,g validated for use in dogs in each diagnostic laboratory. Sex hormone profiles (androstenedione, progesterone, 17-hydroxyprogesterone, and estradiol) and aldosterone concentrations in a subset of dogs were completed by validated radioimmunologic analyses with serum collected before and after ACTH administration; serum samples were shipped frozen by overnight delivery to the Clinical Endocrinology Service, College of Veterinary Medicine, University of Tennessee.9,h
Statistical analysis—All dogs with diffuse VH were grouped according to sex, presence or absence of HCC without sex stratification, and presence or absence of pathological hepatic copper accumulation. Data were examined for normality by use of box-and-whisker plots and the Kolmogorov-Smirnov test. Dogs with specific clinical features (sex and clinical signs), hepatic ultrasonographic findings (estimated hepatic size [normal, small, or large], hepatic parenchymal nodules, discrete hepatic masses, irregular liver surface contour, ascites, bilateral vs unilateral adrenomegaly, and adrenal nodules), and hormonal assessments (LDDST, baseline and post-ACTH stimulation serum cortisol, aldosterone, and sex hormone [progesterone, 17-hydroxyprogesterone, estradiol, and androstenedione] concentrations and at-home urine cortisol-to-creatinine concentration ratio) were enumerated. Difference in sex distribution was examined in 2-by-2 tables with expected distribution determined on the basis of clinic population data during the course of study. Clinicopathologic variables were compared between groups (presence or absence of HCC and presence or absence of pathological hepatic copper concentration) by means of the Wilcoxon rank sum test, with significance determined on the basis of a 2-sided P value (< 0.05) with Bonferroni correction. Association between age at definitive diagnosis and serum ALP activity was examined by means of Spearman rank correlation. Age at death of all dogs as well as between designated groups was examined by use of Kaplan-Meier survival curves. Comparison of age at death between groups was completed by means of the Gehan-Wilcoxon test to reflect difference in short-term survival rate and the log rank test to reflect difference in long-term survival rate; values of P < 0.05 were considered significant. Median age at death and 95% CI were determined. All statistical computations were completed with the aid of a commercial software program.i
Results
From a population of 168 Scottish Terriers (1980 to 2013) with hepatic histologic sections available for review, 114 dogs met criteria for study inclusion (81 from the initial pilot investigation [1980 to 2009] with 33 additional cases identified from 2010 to 2013); 80 antemortem liver specimens and 34 postmortem liver specimens had been collected for histologic characterization.
Relevant features in Scottish Terriers with VH with and without HCC were summarized (Table 1). Age and most clinicopathologic variables had non-Gaussian distribution and thus were reported as median and range. There were 71 male (24 sexually intact and 47 neutered) and 43 female (9 sexually intact and 34 neutered) dogs with no significant (P = 0.08) difference in sex representation. Relevant clinical and clinicopathologic variables for Scottish Terriers that had VH with and without HCC were summarized (Tables 2 and 3). Dogs with VH and HCC were significantly (P = 0.002) older than those without HCC. However, other than discovery of abdominal mass lesions by palpation or well-demarcated hepatic masses by hepatic ultrasonography, no other distinguishing features were found between groups. Although most dogs had increased serum ALT and ALP activities, no significant differences in median serum ALT, ALP, AST, or GGT activities were found between VH-affected dogs with and without HCC. Furthermore, no significant differences were found in proportions of VH-affected dogs with and without HCC having markedly increased (> 10-fold the upper reference range) serum ALT or ALP activities. The serum ALT-to-AST, ALT-to-ALP, and AST-to-ALP activity ratios also were not significantly different between groups (data not shown). In many dogs, progressive increase in serum ALP activity occurred with histologic progression of VH to a degenerative lesion (ie, formation of ultrasonographically visible hypoechoic nodules within a diffusely hyperechoic background parenchyma). Age at onset and rate of change of increased serum ALP activity and progression of VH to visible ultrasonographic hepatic nodules were highly variable among dogs. Although VH was definitively diagnosed in most dogs > 6 years of age, no significant association (P = 0.2; r = 0.13) was found between serum ALP activity and age at definitive diagnosis (Figure 1). Coagulation tests completed in 30 dogs before liver biopsy were within reference intervals except for 1 dog with prolonged activated partial thromboplastin time.
Dot plot of serum ALP activity at the age of definitive diagnosis of VH in 100 Scottish Terriers with and without HCC. Symbols outside the upper limit of the graph represent serum ALP activity > 10,000 U/L.
Citation: Journal of the American Veterinary Medical Association 245, 7; 10.2460/javma.245.7.797
Clinical, clinicopathologic, ultrasonographic, cytologic, and hepatic histologic features of 114 Scottish Terriers that had VH with and without HCC.
| Variable | Tested (No. of dogs) | Confirmed (No. of dogs) | Proportion of dogs (95% CI) |
|---|---|---|---|
| High serum ALP activity | 100 | 98 | 98 (93–100) |
| Hepatomegaly | 101 | 42 | 42 (32–52) |
| Hepatic parenchymal nodules | 101 | 53 | 53 (42–63) |
| VH* | 106 | 106 | 100 (97–100) |
| Copper-associated hepatopathy | 36 | 15 | 42 (26–59) |
| Adrenomegaly | 99 | 26 | 26 (18–36) |
| Signs of hyperadrenocorticism† | 114 | 46 | 40 (31–50) |
| Gallbladder mucocele | 101 | 16 | 16 (9–24) |
| HCC | 114 | 39 | 34 (26–44) |
| TCC of urinary bladder | 114 | 17 | 15 (9–23) |
Eight of 39 dogs with HCC did not have nonneoplastic hepatic tissue submitted for evaluation.
Dogs had ≥ 2 of the following: hepatomegaly, pot-belly appearance, polyphagia, polyuria, polydipsia, truncal alopecia, and high serum ALP activity.
TCC = Transitional cell carcinoma.
Clinical and hematologic features of 114 Scottish Terriers that had VH with or without HCC.
| Dogs without HCC | Dogs with VH and HCC | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Variable | Median (range) | Tested (No.) | Low (No.) | High (No.) | Median (range) | Tested (No.) | Low (No.) | High (No.) | Reference range |
| Age (y) | 8.0 (1–14.0)* | 75 | NA | NA | 10.0 (3.0–16.0)* | 39 | NA | NA | NA |
| Weight (kg) | 11.2 (5.9–15.9) | 75 | NA | NA | 11.7 (6.0–16.8) | 39 | NA | NA | NA |
| PCV (%) | 52 (31–63) | 60 | 6 | 10 | 48 (18–62) | 30 | 7 | 2 | 39–57 |
| Mean corpuscular volume (fL) | 68 (51–74) | 60 | 10 | 1 | 66 (62–73) | 30 | 7 | 0 | 64–73 |
| WBC count (× 103 WBCs/μL) | 10.5 (4.5–39.5) | 60 | 13 | 4 | 11.3 (4.3–38.1) | 30 | 4 | 5 | 7.5–19.9 |
| Neutrophils (× 103 cells/μL) | 8.1 (3.0–34.0) | 60 | 3 | 9 | 8.1 (2.6–30.4) | 30 | 4 | 4 | 3.9–14.7 |
| Lymphocytes (× 103 cells/μL) | 1.4 (0.1–4.0) | 60 | 28 | 0 | 1.3 (0.5–7.5) | 30 | 17 | 1 | 1.5–5.2 |
| Monocytes (× 103 cells/μL) | 0.5 (0–2.9) | 60 | 13 | 2 | 0.5 (0.1–4.4) | 30 | 3 | 1 | 0.3–2.2 |
| Eosinophils (× 103 cells/μL) | 0.3 (0–4.3) | 60 | 11 | 1 | 0.2 (0–1.3) | 30 | 7 | 0 | 0.1–1.6 |
| Basophils (× 103 cells/μL) | 0 (0–0.1) | 60 | NA | 0 | 0 (0–0) | 30 | NA | 0 | 0–0.1 |
| Platelets (× 103 platelets/μL) | 263 (44–898) | 60 | 12 | 4 | 282 (153–692) | 30 | 4 | 1 | 179–510 |
With Bonferroni adjustment, values with P ≤ 0.005 were considered significantly different between dogs with VH without HCC and dogs with VH and HCC.
NA = Not applicable.
Serum biochemical features of 114 Scottish Terriers that had VH with or without HCC.
| Dogs without HCC | Dogs with VH and HCC | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Variable | Median (range) | Tested (No.) | Low (No.) | High (No.) | Median (range) | Tested (No.) | Low (No.) | High (No.) | Reference range |
| Sodium (mEq/L) | 148 (134–158) | 55 | 4 | 8 | 147 (134–153) | 28 | 3 | 5 | 142–151 |
| Potassium (mEq/L) | 4.6 (2.2–6.4) | 54 | 5 | 3 | 4.5 (3.4–7.7) | 28 | 3 | 1 | 3.8–5.6 |
| Chloride (mEq/L) | 111 (81–119) | 53 | 13 | 6 | 109 (88–121) | 28 | 9 | 2 | 108–117 |
| Blood urea nitrogen (mg/dL) | 13 (3–253) | 62 | 10 | 5 | 14 (5–160) | 28 | 3 | 5 | 8–30 |
| Creatinine (mg/dL) | 0.7 (0.3–12.8) | 59 | 4 | 7 | 0.7 (0.3–10.6) | 28 | 3 | 4 | 0.5–1.3 |
| Calcium (mg/dL) | 10.4 (8.9–13.9) | 57 | 4 | 5 | 10.9 (9.1–13.5) | 28 | 1 | 2 | 9.3–11.6 |
| Phosphate (mg/dL) | 4.1 (1.6–22.3) | 57 | 7 | 9 | 4.0 (2.0–9.5) | 28 | 3 | 7 | 2.8–5.3 |
| Glucose (mg/dL) | 98 (65–539) | 60 | 0 | 6 | 97 (70–146) | 28 | 0 | 4 | 58–120 |
| Total protein (g/dL) | 7.0 (4.6–9.4) | 62 | 3 | 29 | 7.0 (4.5–8.2) | 28 | 3 | 15 | 5.6–7.1 |
| Albumin (g/dL) | 3.7 (1.6–5.2) | 61 | 17 | 14 | 3.5 (2.4–4.5) | 28 | 9 | 3 | 3.1–4.1 |
| Globulin (g/dL) | 3.3 (2.2–5.5) | 61 | 0 | 17 | 3.4 (1.7–4.6) | 28 | 1 | 12 | 1.9–3.6 |
| ALT (U/L) | 238 (16–2,700) | 69 | 0 | 49 | 197 (45–10,550) | 29 | 0 | 21 | 12–106 |
| AST (U/L) | 62 (13–924) | 57 | 0 | 35 | 54 (19–6,650) | 27 | 0 | 13 | 13–56 |
| ALP (U/L) | 1,199 (98–22,041) | 69 | 0 | 67 | 1,467 (210–16,278) | 31 | 0 | 31 | 4–122 |
| γ-Glutamyltranspeptidase (U/L) | 10 (0–119) | 45 | NA | 18 | 13 (2–64) | 17 | NA | 9 | < 12 |
| Creatine kinase (U/L) | 128 (59–514) | 36 | 0 | 7 | 170 (55–749) | 17 | 1 | 7 | 58–241 |
| Total bilirubin (mg/dL) | 0.2 (0–7.8) | 63 | NA | 19 | 0.3 (0–12.2) | 29 | NA | 9 | 0–0.3 |
| Cholesterol (mg/dL) | 281 (88–792) | 57 | 5 | 17 | 266 (121–486) | 26 | 3 | 5 | 150–335 |
| Triglycerides (mg/dL) | 153 (35–1,211) | 20 | 0 | 14 | 138 (NA) | 1 | 0 | 1 | 26–108 |
| Amylase (U/dL) | 700 (61–3,203) | 41 | 2 | 8 | 776 (412–1,485) | 17 | 0 | 3 | 286–1,124 |
| Lipase (U/dL) | 280 (36–4,250) | 32 | 0 | 16 | 255 (107–3,434) | 17 | 0 | 6 | 11–275 |
With Bonferroni adjustment, values with P ≤ 0.002 were considered significantly different between dogs with VH without HCC and dogs with VH and HCC.
NA = Not applicable.
Many dogs (101/114 [89%]) had been evaluated for unexplained high serum liver enzyme activity (primarily ALP) before a liver biopsy was pursued. Hyperadrenocorticism was suspected in 46 of 114 (40%) dogs on the basis of unexplained high serum ALP activity and variable clinical features (eg, hepatomegaly, pot-belly appearance, polyuria, polydipsia, polyphagia, and bilaterally symmetric truncal alopecia). Adrenomegaly (adrenal gland width, > 0.7 cm) was detected in 26 of 99 (26%) dogs (22 with adrenal gland dimensions recorded during abdominal ultrasonography, 2 at necropsy, and 2 during exploratory surgery). Unilateral adrenomegaly was identified in 16 of 26 (62%) dogs, involving the left adrenal gland in 10 and right adrenal gland in 6. Hepatomegaly was detected by physical examination, hepatic ultrasonography, or necropsy in 42 of 101 (42%) dogs. A small nodular liver was described for 4 of 114 (4%) dogs; these dogs also had acquired portosystemic shunts and abdominal effusion characterized as pure or modified transudate. One dog with a small liver had chronic hepatic injury attributed to severe copper-associated hepatopathy. The remainder had small nodular livers secondary to progressive degenerative VH. Hepatic ultrasonography revealed hyperechoic hepatic parenchyma with hypoechoic nodules in 53 of 101 (52%) dogs; otherwise, hepatic parenchyma was typically characterized as mottled or coarse. In a few dogs (3/101 [3%]), no abnormalities were identified on ultrasonographic evaluation.
Of 46 dogs with suspected hyperadrenocorticism (made on the basis of clinical signs, routine laboratory testing, or adrenomegaly on ultrasonographic imaging), all underwent ≥ 1 assessment of adrenal function. Thirty-seven of the 46 (80%) dogs underwent an ACTH response test; of these 37 dogs, 17 (46%) had high basal serum cortisol concentration and 25 (68%) had high ACTH-stimulated serum cortisol concentration. Twenty-one of 46 (46%) dogs underwent an LDDST; 5 of 21 (24%) dogs tested lacked adequate cortisol suppression. Of 7 dogs suspected of hyperadrenocorticism for which urine cortisol-to-creatinine concentration ratios were determined, 5 had ratios inconsistent with cortisol excess. An ACTH-stimulated sex hormone profile was completed in 27 of 46 (59%) dogs with suspected adrenal gland hyperactivity; for each analyte, the fold increase exceeding the upper limit of the reference range (95% CI used as the reference range) for sex and reproductive status was determined (Figure 2). Dogs with fold increase > 1.0 relative to the upper limit of the reference range specific for sex and reproductive statush were considered to have high serum hormone concentrations.
Fold increases (determined on the basis of sex- and reproductive status-specific reference ranges) in serum steroidogenic hormone concentrations before (baseline; panel A) and after (post; panel B) ACTH stimulation in 27 Scottish Terriers with histologically confirmed VH. Hormone values represent fold increase above the upper limit of the reference range. After 2002, androstenedione was only measured up to 10 ng/mL, with higher concentrations represented as > 10 ng/mL, limiting maximal fold increase of androstenedione to 2.52. Solid black quadrangle represents a dog that developed spontaneous hypoadrenocorticism.
Citation: Journal of the American Veterinary Medical Association 245, 7; 10.2460/javma.245.7.797
Increased pre-ACTH and post-ACTH stimulation serum sex hormone concentrations (exceeding the upper limit of the reference range for the sex and reproductive status for that individual) occurred for aldosterone in 21% (4/19) and 15% (3/20), for progesterone in 65% (17/26) and 72% (18/25), for 17-hydroxyprogesterone in 27% (7/26) and 31% (8/26), for estradiol in 62% (16/26) and 42% (11/26), and for androstenedione in 73% (19/26) and 62% (16/26) of dogs tested, respectively. All dogs tested had a high serum concentration of at least 1 steroidogenic hormone at baseline or following ACTH stimulation; 25 of 28 (89%) dogs had high serum concentrations of ≥ 2 steroidogenic hormones at baseline or following ACTH stimulation. Endogenous plasma ACTH concentrations were measured in 4 of 46 (9%) dogs, and no consistent pattern relative to adrenal hormones was found (data not shown). Fifteen dogs had discrete unilateral adrenal masses visualized by ultrasonographic imaging. Histologic evaluation of the adrenal gland in 6 of these dogs revealed nodular hyperplasia (n = 2), adenoma (3), and carcinoma (1).
Photomicrographs of representative histologic sections of liver lesions characteristic of hepatopathy in Scottish Terriers. A—Typical wispy cytosolic hepatocyte vacuolation. H&E stain; bar = 70 μm. B—Glycogen-like cytosolic expansion. Magenta color, periodic acid–Schiff without amylase; bar = 300 μm. C—Digestion of glycogen-like storage material. Periodic acid–Schiff stain with amylase; bar = 300 μm. D—Thin bands of fibrillar connective tissue dissecting between nodules of vacuolated hepatocytes with greater vacuolation at nodule margin. Blue staining, Masson trichrome; bar = 2 mm. E—Advanced fibrosis as a result of chronic progressive degenerative VH. Masson trichrome; bar = 200 μm. F—Dysplastic focus (arrows). H&E stain; bar = 2 mm. G—Same dysplastic focus (arrows) as shown in panel F. Notice the loss of normal reticulin support but retention in nondysplastic hepatocytes. Reticulin stain; bar = 2 mm. H and I—Morphology of dysplastic hepatocytes (left field [H] and central field [I]), compared with nondysplastic surrounding vacuolated hepatocytes, and loss of organized sinusoidal cords. H&E stain; bar = 200 μm.
Citation: Journal of the American Veterinary Medical Association 245, 7; 10.2460/javma.245.7.797
On the basis of histologic features of dogs with nonneoplastic liver specimens available for evaluation, all had diffuse glycogen-like VH as the predominant histologic feature (Figure 3). Abundant cytosolic glycogen was confirmed in liver biopsy specimens from 10 dogs that had been stained with Periodic acid–Schiff (with and without diastase). Glycogen accumulation, easily detected on routine H&E staining, reconciled with finding diffuse hyperechoic hepatic parenchymal echogenicity on hepatic ultrasonography. Ballooning degeneration of severely vacuolated hepatocytes appeared to lead to apoptotic cell loss and formation of distinct nodules composed of less vacuolated cells that were marginated by a thin rim of fibrillar collagen confirmed by reticulin and Masson trichrome staining. Such parenchymal remodeling seemingly reconciled with the ultrasonographic observation that hypoechoic nodules contrasted against a diffusely hyperechoic parenchymal background. As expected with a chronic hepatopathy, fibrosis was more extensive in dogs that died of liver failure with microhepatia and acquired portosystemic shunts. Thirty-nine of 114 (34%) dogs had HCC detected at surgery or necropsy or by abdominal ultrasonography. Mass lesions ranged in largest dimension from 0.2 to 9.0 cm with no significant difference in hepatic lobe involvement (10 tumors located in right liver lobes, 10 located in left liver lobes, 18 with multifocal locations, and 1 with unspecified location). When sufficient nonneoplastic tissue was available for evaluation (31/39 dogs), diffuse VH also was commonly identified. In 8 dogs, many small dysplastic foci of hepatocytes (small proliferating hepatocytes forming hepatic cords ranging in width from 2 to 5 cells that also lacked dense glycogen vacuolation) were observed that had irregular borders interdigitating with more normal-appearing vacuolated hepatocytes in liver specimens collected distant (ie, separate liver lobe) to HCC. Multiple dysplastic foci also were identified in liver specimens collected from 11 dogs lacking a diagnosis of HCC. Dysplastic foci were densely eosinophilic when stained with Masson trichrome and lacked expected hepatocyte cord definition with reticulin staining. Serum sex hormone concentrations assessed before and following ACTH stimulation in 1 dog with HCC and 4 dogs with dysplastic hepatic foci confirmed progesterone concentrations (5/5) and androstenedione concentrations (4/5) that exceeded the upper limit of the reference range.
In addition to diffuse VH, 11 dogs without HCC had nonhepatocellular neoplasia within biopsied liver specimens (lymphosarcoma, 4; metastatic insulinoma, 1; uncharacterized neuroendocrine neoplasia, 1; biliary adenocarcinoma, 1; carcinoma of undetermined origin, 1; hemangiosarcoma, 1; angiomyelolipoma, 1; and melanoma, 1), and 6 dogs had necroinflammatory lymphoplasmacytic hepatitis. Gallbladder mucocele was identified in 16 of 101 (16%) dogs, which led to cholecystectomy and liver biopsy.
Twenty-six of 36 (72%) dogs with diffuse VH that had hepatic copper concentration evaluated had hepatic copper concentrations exceeding the upper limit of the reference range (ie, > 400 μg/g DWB).6,10 Of dogs with high hepatic copper concentrations, 15 of 26 (58%) had histologic features consistent with copper-associated hepatopathy.10 Although copper-associated hepatopathy seemingly complicated illness in these dogs, none had HCC. The subset of dogs with histologic features consistent with copper-associated hepatopathy had significantly (P = 0.04) higher serum ALT activity (median, 639 U/L [range, 66 to 2,003 U/L]) than dogs with measured copper lacking this lesion (276 U/L [range, 30 to 1,798 U/L]). However, the overlapping ranges of serum ALT activities precluded exclusive use of this variable in discriminating involvement of copper in an individual dog's disease process. Considering dogs with measured hepatic copper concentrations, those with copper-associated hepatopathy had significantly (P < 0.001) higher median hepatic copper concentration (1,255 μg/g DWB; range, 536 to 3,079 μg/g DWB), compared with dogs lacking this lesion (337 μg/g DWB; range, 61 to 832 μg/g DWB). Most dogs with measured liver copper concentrations had liver biopsy specimens collected after 1997 (32/36 [89%]).
Control of adrenocortical function was attempted in 13 dogs with clinical signs and endocrinologic test results suggestive of adrenal gland hyperactivity. Mitotane was administered to 4 dogs, trilostane to 4 dogs, and ketoconazole to 5 dogs. In an attempt to control hyperadrenocorticism, 3 dogs received treatment with > 1 drug after washout intervals. Treatment with mitotane or ketoconazole led to acute death, hypoadrenocorticism, or generalized illness. Although treatment with trilostane was better tolerated in the small number of treated dogs, it failed to alleviate clinical signs, decrease serum ALP activity, or improve VH after months of administration, and it provoked signs of toxicity in 1 dog.
Scottish Terriers with progressive VH (Figure 4) had a median age at death of 11 years (95% CI, 10 to 12 years), with some dogs living as long as 16 years. Median age at death was significantly (Gehan-Wilcoxon test, P = 0.04; log rank test, P = 0.015) greater for males (11.0 years; 95% CI, 9.5 to 12.5 years) than for females (10 years; 95% CI, 8 to 11 years). There was no significant (Gehan-Wilcoxon test, P = 0.37; log rank test, P = 0.50) difference in age at death between dogs with (11 years; 95% CI, 10 to 12 years) and without HCC (10 years; 95% CI, 9 to 11.2 years). Of 39 dogs with HCC, HCC was diagnosed at necropsy in 19 dogs that were euthanized because of progressive or severe clinical illness. No significant (Gehan-Wilcoxon test, P = 0.44; log rank test, P = 0.66) difference was found in age at death between dogs with (10.6 years; 95% CI, 9 to 11.2 years) and without (11.5 years; 95% CI, 9.0 to 13.0 years) copper-associated hepatopathy. In most of these dogs, antemortem diagnosis was made from findings on histologic evaluation of liver biopsy specimens.
Kaplan-Meier survival curves for 114 Scottish Terriers that had VH with and without HCC (A) and 106 male and female Scottish Terriers with VH (B). Eight dogs with HCC did not have nonneoplastic tissue submitted for histologic evaluation. Male dogs survived significantly (P < 0.05) longer than female dogs. Survivorship S(t) = Survival probability.
Citation: Journal of the American Veterinary Medical Association 245, 7; 10.2460/javma.245.7.797
Discussion
This retrospective study of 114 Scottish Terriers confirmed that glycogen-like VH accompanied by a progressive increase in serum ALP activity is found in this breed. Surprisingly, VH was also associated with development of HCC in these Scottish Terriers. Sequential monitoring over years in individual dogs confirmed the progressive nature of this syndrome, with increases in serum ALP activity coinciding with development of diffuse hepatic parenchymal hyperechogenicity and numerous small hypoechoic parenchymal nodules evident on ultrasonographic examination. Although ultrasonographic detection of features such as hyperechoic or coarse-appearing hepatic parenchyma, hypoechoic nodules contrasting against hyperechoic parenchyma, and discrete mass lesions is common, this diagnostic modality cannot definitively confirm VH in Scottish Terriers without histologic evaluation of hepatic tissue. Ultrasonographic features reflect hepatic parenchymal remodeling and progression of nondegenerative to degenerative VH with formation of regenerative nodules sometimes associated with dysplastic hepatocellular foci. It is possible that these foci are antecedent to development of HCC as occurs in humans and experimental models used to study HCC in animals. Identification of large and progressive hepatic mass lesions, development of gallbladder mucocele, and detection of adrenomegaly impacted clinical recommendations.
In this study, Scottish Terriers with VH had a broad range of historical, physical, and clinicopathologic features with a phenotype consistent with typical hyperadrenocorticism in 40% (46/114) of dogs. Findings confirm that the VH is not always benign, but rather may rapidly progress to liver failure in some dogs with development of small nodular liver, portal hypertension, and ascites. Yet, in other dogs with VH, chronically increased serum ALP activity (3- to 20-fold) persisted for > 10 years without apparent clinical deterioration. Clinicopathologic features, including magnitude of increases in serum ALP and ALT activities, failed to differentiate between dogs with VH with and without HCC or with and without copper-associated hepatopathy and could not predict early death from hepatic failure. Consequently, considering the variable age of syndrome onset and rate of progression, it is impossible to predict relative risk for hepatic failure or development of HCC without sequential patient monitoring. Although we did not find a significant difference in age at death of affected dogs with and without HCC, it is prudent to consider the retrospective nature of this study. Indeed, some dogs with HCC were taken directly to surgery for tumor resection, which likely increased their survival chances. Sequential biochemical monitoring of Scottish Terriers with this syndrome has demonstrated that abrupt increases in serum ALP activity may function as a paraneoplastic sentinel of emerging HCC or development of gallbladder mucocele. Monitoring with abdominal ultrasonography has permitted detection of enlarging hepatic mass lesions, which is supportive of recommending earlier surgical extirpation of small HCC. We therefore recommend that Scottish Terriers with increased serum ALP activity undergo semiannual monitoring of serum liver enzyme activities and synthetic markers of liver function as well as yearly hepatic ultrasonographic examination. Increased surveillance frequency is advised for dogs with sudden marked increases in serum ALP activity or transitioning to a nodular hepatopathy on ultrasonographic evaluation. This strategy is aimed at early detection of hepatic mass lesions likely to represent surgically excisable HCC and recognition of subclinical gallbladder mucoceles that may allow uncomplicated cholecystectomy.
Adrenal gland hyperactivity associated with VH in Scottish Terriers predominantly involves sex hormones (most notably progesterone and androstenedione) and may be accompanied by overt adrenomegaly (unilateral or bilateral). Despite adrenomegaly, data from this study and a previous report5 confirm that cortisol assessments (baseline and post-ACTH stimulation testing, LDDST, or at-home collected urine cortisol-to-creatinine concentration ratios) inconsistently verify typical hyperadrenocorticism (Cushing's syndrome). In this study, assessment of pre- and post-ACTH stimulation serum sex hormone concentrations supported adrenal gland hyperactivity in up to 80% (20/25) to 88% (22/25) of dogs tested. It remains possible that adrenal hyperplasia secondary to chronic stress or illness in these dogs accounted for the increased serum concentrations of cortisol and sex hormones.11 However, it is also plausible that the adrenal gland hyperactivity and associated VH reflect a breed-related genetic disorder affecting adrenal steroidogenesis that leads to ALP induction, hepatocellular glycogen accumulation, progressive VH, and ultimately formation of HCC. Interestingly, in this and the previous study5 of apparently healthy Scottish Terriers with an increased serum ALP activity, the most common sex hormone abnormalities were increased concentrations of progesterone and androstenedione.
Our finding that age at death was significantly greater for male dogs with VH remains unexplained. Seemingly, an initial glycogen-like VH evolves gradually over a dog's lifespan at different rates among individuals. The lesion may transition to degenerative VH with formation of regenerative foci that may have dysplastic characteristics. Finding dysplastic foci in dogs with severe diffuse VH and in liver lobes distant to recognized HCC suggests that dysplastic foci may precede neoplastic transformation. Our finding of a predisposition for HCC in Scottish Terriers is unprecedented.
Worldwide, HCC is the sixth most common cancer and ranks as the third highest cause of cancer-related death in human beings.12–14 This neoplasm is associated with chronically diseased liver in populations affected with chronic aflatoxin exposure; excessive alcohol intake; infections with hepatitis B, C, or E; congenital hemochromatosis; treatment with androgens; or nonalcoholic steatohepatitis associated with the metabolic syndrome.12–14 In humans and rodents, HCC is known to undergo multistaged progression characterized by histopathologic, molecular biologic, and diagnostic imaging studies.15–29 Small nodular hypercellular lesions referred to as foci of adenomatous hyperplasia or atypical adenomatous hyperplasia in chronically diseased liver have molecular signatures consistent with preneoplastic lesions.15,16 In humans, these preneoplastic lesions are classified as either low-grade or high-grade dysplasia depending on the degree of atypia (ie, nuclear hyperchromasia, nuclear contour irregularities, cytoplasmic basophilia or clear cell [vacuolated] change, and increased nuclear-cytoplasmic ratio), presence of occasional mitotic figures, and cytoplasmic features suggestive of a clonal population.16 Dysplastic foci are most often discovered in cirrhotic livers and precede or accompany development of HCC.16–22 Finding HCC within dysplastic foci (nodules within nodules) strongly supports the concept that dysplastic atypia is the precursor to neoplastic transformation in human beings.16 Follow-up studies23 in humans with high-grade dysplastic foci show transformation to HCC within a few years (estimated 4-fold relative risk for HCC in patients with dysplastic foci). The association between dysplastic foci and development of HCC is consistent with observations in dogs in the present study. This is the first report of this phenomenon in canine patients.
It is widely recognized that normal livers are functionally and morphologically influenced by sex hormones and that HCC has a male sex predilection in humans (male-to-female ratio of up to 3:1 reported).13,14,30 Epidemiological studies confirm increased risk for HCC with long-term use of oral contraceptives or anabolic androgens, with each drug category also associated with formation of dysplastic foci.30,31 One retrospective study31 of HCC confirmed nearly 100 human patients after chronic androgen (eg, testosterone, methyl testosterone, oxymetholone, danazol, and testosterone combined with estrogen) exposure for illness or for anabolic effects (body building). Increased risk for HCC has also been directly linked with plasma testosterone concentrations in men, corroborated in several animal models of HCC.32,33 Indeed, studies34–40 in rodents confirm a link between androgen exposure and heightened risk for chemically induced and naturally occurring HCC in female or castrated male mice, whereas castration provides a relative protective influence in sexually intact males. Both short- and long-term toxicity studies of oral androstenedione in rats and mice confirm its association with vacuolar (glycogen and lipid) hepatocellular transformation and enhanced risk for preneoplastic dysplastic foci or HCC.41 Although exact mechanisms remain unclear, proteins involved with androgen transport and metabolism are implicated as carcinogenic risk factors. Estrogen exposure also increases risk for HCC.33 Thus, a body of data involving naturally occurring HCC in humans and experimental models of HCC in animals clearly implicates a role for sex hormones in tumor initiation, with most evidence incriminating androgens.33 Findings in this group of Scottish Terriers with a high proportion of HCC and dysplastic hepatic foci included high serum sex hormone (androgens, progesterone, and estradiol) concentrations in many dogs. It appears that there was an increased incidence of gallbladder mucocele in our study population considering that 16 of 114 dogs had this condition. An association between gallbladder mucocele and hyperadrenocorticism has previously been proposed.42 Further evidence implicating a role of steroidogenic hormones in mucocele formation is that cystic mucosal hyperplasia of the gallbladder wall, typically associated with gallbladder mucoceles, can be induced in dogs by exposure to progestational compounds.43 Thus, we believe that there is an association between mucocele formation and abnormal steroidogenesis.
Among all dog breeds, a 0.46% frequency of HCC was reported in a large retrospective case series44 (12,245 canine necropsies at a single institution). There were 446 dogs of numerous breeds with a definitive diagnosis of HCC in our clinical teaching hospital database during the 33-year study interval. Of this population, 5% were Scottish Terriers (a breed comprising < 0.5% of dogs evaluated at our hospital in a year). This equates to a 10-fold relative risk for HCC, compared with other breeds. In the previous study44 of HCC in dogs, mean age at diagnosis was > 11 years, with a male predilection (male-to-female ratio, 1.7:1). In that study,44 although predilection for larger dogs was noted, no breed predominance was recognized. Increased serum activities of ALP in 89% (28 tested), ALT in 80% (30 tested), and AST in 56% (25 tested) of dogs were reported, with hyperbilirubinemia detected in 28% (25 tested) and increased total protein concentration found in 26% (31 tested).44 Findings in our studied population of Scottish Terriers with HCC were similar, with median age at diagnosis of 10 years, male sex predilection (male-to-female ratio, 1.7:1), and increased serum activities of ALP (98/100 [98%]), ALT (70/98 [71%]), and AST (48/84 [57%]) as well as hyperbilirubinemia (28/92 [30%]) and increased total protein concentration (44/90 [49%]). However, clinicopathologic features could not discriminate between Scottish Terriers with VH with or without HCC. There also were no hepatic enzyme ratios that could differentiate between dogs with VH with or without HCC, as has been previously suggested.45 Distribution of HCC in the previous study44 showed 30 of 49 (61%) massive lesions within a single liver lobe (lateralizing to left liver lobes in 20/30 [67%] and right liver lobes in 6/30 [20%]); however, mass lesions were also commonly identified in other liver lobes (24/30 [80%]). In the present study, an even distribution of HCC was observed between right and left liver lobes with a similar common recognition of multifocal neoplasms. No difference was found in age at death between Scottish Terriers with VH with or without HCC. However, this finding must be interpreted cautiously because it may be biased by the fact that the study population included 19 of 39 (49%) dogs with HCC diagnosed at necropsy (euthanasia because an hepatic mass was diagnosed), with the remainder having HCC removal, which likely extended their lifespan.
Diagnosis of copper-associated hepatopathy has increased in dogs of all breeds since 1997 when the biochemical form of copper complexes in commercial dog foods was changed.10 Most liver biopsy specimens with copper measurements in this study were collected after 1997 (32/36 [89%]). Hepatic copper concentrations exceeding the upper limit of the reference range (400 μg/g DWB) were found in 72% (26/36) of analyzed biopsy specimens, with copper > 800 μg/g DWB in 31% (11/36) of analyzed biopsy specimens. Copper-associated hepatopathy complicated VH in 15 of 114 (13%) dogs. Recognition of this complicating comorbidity could not be deduced from clinicopathologic or hepatic ultrasonographic findings without histologic evaluation of liver biopsy specimens. It is important to identify copper-associated hepatopathy because chelation and antioxidant treatment along with dietary restriction of copper is curable. Finding increased hepatic copper in some dogs without evidence of copper-mediated liver injury was not unexpected on the basis of our extensive experience with hepatic copper accumulation in dogs (SPM and SAC). Unlike humans and cats, dogs do not appear to accumulate large quantities of copper secondary to cholestatic or other forms of liver injury, and thus, we do not believe that copper accumulated secondary to VH. Considering the transition metal status of copper and its ability to augment oxidative injury of any origin, its presence warrants concern as a factor that might escalate liver injury.
Finding high serum ALP activity or VH in a Scottish Terrier may lead to adrenal function assessments that show exaggerated adrenal hormone production, which may be interpreted to represent typical or atypical hyperadrenocorticism. Among 13 Scottish Terriers undergoing conventional adrenomodulatory treatment, there were no positive outcomes. All treatments were either ineffective or resulted in adverse reactions (eg, glucocorticoid insufficiency, hepatotoxicity, and death). The apparent frailty of Scottish Terriers with VH treated with conventional adrenomodulatory agents remains unexplained. If the underlying metabolic abnormality causing VH in Scottish Terriers is associated with sex hormone–related adrenal gland hyperactivity, treatment with trilostane may be ill-advised, considering that studies46–48 in dogs have shown that trilostane fosters accumulation of sex hormone precursors, fails to control overproduction of androstenedione, and leads to adrenomegaly. Further research aimed at characterizing the possible genetic basis of this syndrome might allow development of rational and effective treatment strategies.
ABBREVIATIONS
| ALP | Alkaline phosphatase |
| ALT | Alanine aminotransferase |
| AST | Aspartate aminotransferase |
| CI | Confidence interval |
| DWB | Dry weight basis |
| HCC | Hepatocellular carcinoma |
| LDDST | Low-dose dexamethasone suppression test |
| VH | Vacuolar hepatopathy |
Aperio ScanScope CS, Aperio Technologies Inc, Vista, Calif.
Aperio 2004–08–11, version 8.1, Aperio Technologies Inc, Vista, Calif.
Cortrosyn, Amphastar Pharmaceuticals, Rancho Cucamonga, Calif.
Cortrosyn, Organon Inc, West Orange, NJ.
Immulite 1000, Siemens Medical Solutions Diagnostics, Siemens Healthcare Diagnostics, Deerfield, Ill.
Cortisol Kit, Siemens Healthcare Diagnostics, Deerfield, Ill.
ACTH IRMA Kit, Diagnostic Products Corp, Deerfield, Ill.
Clinical Endocrinology Service, College of Veterinary Medicine, University of Tennessee, Knoxville, Tenn.
Statistix, version 9, Analytical Software, Tallahassee, Fla.
References
1. Center SA. Interpretation of liver enzymes. Vet Clin North Am Small Anim Pract 2007; 37: 297–333.
2. Sepesy LM, Center SA, Randolph JF, et al. Vacuolar hepatopathy in dogs: 336 cases (1993–2005). J Am Vet Med Assoc 2006; 229: 246–252.
3. Badylak SF, Van Vleet JF. Sequential morphologic and clinicopathologic alterations in dogs with experimentally induced glucocorticoid hepatopathy. Am J Vet Res 1981; 42: 1310–1318.
4. Center SA. Breed specific hepatopathies in Scottish Terrier and Maltese dogs, in Proceedings. Am College Vet Intern Med Forum 2012; 694.
5. Zimmerman KL, Panciera DL, Panciera RJ, et al. Hyperphosphatasemia and concurrent adrenal gland dysfunction in apparently healthy Scottish Terriers. J Am Vet Med Assoc 2010; 237: 178–186.
6. Center SA, McDonough SP, Bogdonavic L. Digital image analysis of rhodanine stained canine liver biopsies for calculation of hepatic copper concentrations. Am J Vet Res 2013; 74: 1474–1480.
7. Reimers TJ, Salerno VJ, Lamb SV. Validation and application of solid-phase chemiluminescent immunoassays for diagnosis of endocrine diseases in animals. Comp Haematol Int 1996; 6: 170–175.
8. Scott-Moncrieff JC, Koshko MA, Brown JA, et al. Validation of a chemiluminescent enzyme immunometric assay for plasma adrenocorticotropic hormone in the dog. Vet Clin Pathol 2003; 32: 180–187.
9. Frank LA, Rohrbach BW, Bailey EM, et al. Steroid hormone concentration profiles in healthy intact and neutered dogs before and after cosyntropin administration. Domest Anim Endocrinol 2003; 24: 43–57.
10. Johnston AN, Center SA, McDonough SP, et al. Hepatic copper concentrations in Labrador Retrievers with and without chronic hepatitis: 72 cases (1980–2010). J Am Vet Med Assoc 2013; 242: 372–380.
11. Behrend EN, Kennis R. Atypical Cushing's syndrome in dogs: arguments for and against. Vet Clin North Am Small Anim Pract 2010; 40: 285–296.
12. Parkin DM, Bray F, Ferlay J, et al. Global cancer statistics, 2002. CA Cancer J Clin 2005; 55: 74–108.
13. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999; 340: 745–750.
14. Ince N, Wands JR. The increasing incidence of hepatocellular carcinoma (edit). N Engl J Med 1999; 340: 798–799.
15. Effendi K, Sakamoto M. Molecular pathology in early hepatocarcinogenesis. Oncology 2010; 78: 157–160.
16. Roskams T, Kojiro M. Pathology of early hepatocellular carcinoma: conventional and molecular diagnosis. Semin Liver Dis 2010; 30: 17–25.
17. Arakawa M, Kage M, Sugihara S, et al. Emergence of malignant lesions within an adenomatous hyperplastic nodule in a cirrhotic liver. Observations in five cases. Gastroenterology 1986; 91: 198–208.
18. Mion F, Grozel L, Boillot O, et al. Adult cirrhotic liver explants: precancerous lesions and undetected small hepatocellular carcinomas. Gastroenterology 1996; 111: 1587–1592.
19. Terada T, Terasaki S, Nakanuma Y. A clinicopathologic study of adenomatous hyperplasia of the liver in 209 consecutive cirrhotic livers examined by autopsy. Cancer 1993; 72: 1551–1556.
20. Hytiroglou P, Theise ND, Schwartz M, et al. Macroregenerative nodules in a series of adult cirrhotic liver explants: issues of classification and nomenclature. Hepatology 1995; 21: 703–708.
21. Libbrecht L, Bielen D, Verslype C, et al. Focal lesions in cirrhotic explant livers: pathological evaluation and accuracy of pretransplantation imaging examinations. Liver Transpl 2002; 8: 749–761.
22. Libbrecht L, Desmet V, Roskams T. Preneoplastic lesions in human hepatocarcinogenesis. Liver Int 2005; 25: 16–27.
23. Borzio M, Fargion S, Borzio F, et al. Impact of large regenerative, low grade and high grade dysplastic nodules in hepatocellular carcinoma development. J Hepatol 2003; 39: 208–214.
24. Tsuda H, Hirohashi S, Shimamoto Y, et al. Clonal origin of atypical adenomatous hyperplasia of the liver and clonal identity with hepatocellular carcinoma. Gastroenterology 1988; 95: 1664–1666.
25. Takayama T, Makuuchi M, Hirohashi S, et al. Malignant transformation of adenomatous hyperplasia to hepatocellular carcinoma. Lancet 1990; 336: 1150–1153.
26. Sakamoto M, Hirohashi S, Shimamoto Y. Early stages of multi-step hepatocarcinogenesis: adenomatous hyperplasia and early hepatocellular carcinoma. Hum Pathol 1991; 22: 172–178.
27. Hirohashi S, Ishak KG, Kojiro M, et al. Hepatocellular carcinoma. In: Hamilton SR, Aaltonen LA, eds. Pathology and genetics of tumors of the digestive system. Lyon, France: IARC Press, 2000; 159–172.
28. Matsui O, Kadoya M, Kameyama T, et al. Benign and malignant nodules in cirrhotic livers: distinction based on blood supply. Radiology 1991;178: 493–497.
29. Kudo M. Morphological diagnosis of hepatocellular carcinoma: special emphasis on intranodular hemodynamic imaging. Hepatogastroenterology 1998; 45: 1226–1231.
30. Di Maio M, Daniele B, Pignata S, et al. Is human hepatocellular carcinoma a hormone-responsive tumor? World J Gastroenterol 2008; 14: 1682–1689.
31. Velazquez I, Alter BP. Androgens and liver tumors: Fanconi's anemia and non-Fanconi's conditions. Am J Hematol 2004; 77: 257–267.
32. Yu M-W, Change Y-C, Yang S-Y, et al. Hormonal markers and hepatitis B virus-related hepatocellular carcinoma risk: a nested case-control study among men. J Natl Cancer Inst 2001; 93: 1644–1651.
33. De Maria N, Manno M, Villa E. Sex hormones and liver cancer. Mol Cell Endocrinol 2002; 193: 59–63.
34. Agnew LR, Gardner WU. The incidence of spontaneous hepatomas in C3H, C3H (low milk factor), and CBA mice and the effect of estrogen and androgen on the occurrence of these tumors in C3H mice. Cancer Res 1952; 12: 757–761.
35. Vesselinovitch SD, Mihailovich N. The effect of gonadectomy on the development of hepatomas induced by urethan. Cancer Res 1967; 27: 1788–1791.
36. Toh YC. Effect of neonatal castration on liver tumor induction by N-2-fluorenylacetamide in suckling BALB/c mice. Carcinogenesis 1981; 2: 1219–1221.
37. Firminger HI, Reuber MD. Influence of adrenocortical, androgenic, and anabolic hormones on the development of carcinoma and cirrhosis of the liver in A×C rats fed N-2-fluorenyldiacetamide. J Natl Cancer Inst 1961; 27: 559–595.
38. Vesselinovitch SD, Itze L, Mihailovich N, et al. Modifying role of partial hepatectomy and gonadectomy in ethylnitrosourea-induced hepatocarcinogenesis. Cancer Res 1980; 40: 1538–1542.
39. Kemp CJ, Leary CN, Drinkwater NR. Promotion of murine hepatocarcinogenesis by testosterone is androgen receptor-dependent but not cell autonomous. Proc Natl Acad Sci U S A 1989; 86: 7505–7509.
40. Yu L, Nagasue N, Yamaguchi M, et al. Effects of castration and androgen replacement on tumour growth of human hepatocellular carcinoma in nude mice. J Hepatol 1996; 25: 362–369.
41. Toxicology and carcinogenesis studies of androstenedione (CAS No. 63–05–8) in F344/ N rats and B6c3f1 mice (gavage studies). Natl Toxicol Program Tech Rep Ser 2010; 510: 1–190.
42. Mawdesley-Thomas LE, Noel PR. Cystic hyperplasia of the gall bladder in the Beagle, associated with the administration of progestational compounds. Vet Rec 1967; 80: 658–659.
43. Mesich MLL, Mayhew PD, Paek M, et al. Gallbladder mucoceles and their association with endocrinopathies in dogs: a retrospective case-control study. J Small Anim Pract 2009; 50: 630–635.
44. Patnaik AK, Hurvitz AI, Lieberman PH, et al. Canine hepatocellular carcinoma. Vet Pathol 1981; 18: 427–438.
45. Patnaik AK, Hurvitz AI, Lieberman PH. Canine hepatic neoplasms: a clinicopathologic study. Vet Pathol 1980; 17: 553–564.
46. Sieber-Ruckstuhl NS, Boretti FS, Wenger M, et al. Cortisol, aldosterone, cortisol precursors, androgen and endogenous ACTH concentrations in dogs with pituitary-dependent hyperadrenocorticism treated with trilostane. Domest Anim Endocrinol 2006; 31: 63–75.
47. Ruckstuhl NS, Nett CS, Reusch CE. Results of clinical examinations, laboratory tests, and ultrasonography in dogs with pituitary-dependent hyperadrenocorticism treated with trilostane. Am J Vet Res 2002; 63: 506–512.
48. Mantis P, Lamb CR, Witt AL, et al. Changes in ultrasonographic appearance of adrenal glands in dogs with pituitary-dependent hyperadrenocorticism treated with trilostane. Vet Radiol Ultrasound 2003; 44: 682–685.
