ELSEVIER
The Veterinary Journal
journal homepage: www.elsevier.com/locate/tvjl
The Veterinary Journal
Epidemiology of massive hepatocellular carcinoma in dogs: A 4-year retrospective study
Check for updates
R. Leela-arporna, H. Ohtaª, N. Nagataª, K. Sasaokaª, M. Tamuraª, A. Dermlima, K. Nisaª, K. Morishitab, N. Sasakiª, K. Nakamurab,1, S. Takagib,2, K. Hosoyab, M. Takiguchia,*
a Laboratory of Veterinary Internal Medicine, Department of Veterinary Clinical Sciences, Graduate School of Veterinary Medicine, Hokkaido University, Sapporo, Hokkaido 060-0818, Japan
b Veterinary Teaching Hospital, Graduate School of Veterinary Medicine, Hokkaido University, Sapporo, Hokkaido 060-0818, Japan
ARTICLE INFO
Article history: Accepted 22 April 2019
Keywords:
Canine Hepatocellular carcinoma Hyperadrenocorticism Risk factors
ABSTRACT
Hepatocellular carcinoma (HCC) is the most common primary liver tumour in dogs. However, the clinical features and risk factors of HCC have not been confirmed. The objective of this study was to investigate the clinical features and risk factors for canine HCC. Medical records of 44 dogs diagnosed with HCC at Hokkaido University Veterinary Teaching Hospital between 2013 and 2017 were retrospectively reviewed. All dogs evaluated at the teaching hospital during the study period were used as the reference population for breed, age, sex predispositions or possible related factors for HCC, including concurrent disorders. Clinical characteristics of HCC were determined using propensity score matching analysis.
The prevalence of HCC diagnosis was 0.96%. Multivariate analysis revealed that dogs diagnosed with HCC were significantly older (odds ratio [OR], 1.20; 95% confidence intervals [CI], 1.07-1.33) than the reference population. Welsh Corgis (OR, 3.68; 95% CI, 1.56-8.67) and Beagles (OR, 4.33; 95% CI, 1.58-11.90) were significantly predisposed to HCC. Twenty-seven of 44 dogs with HCC had at least one concurrent disorder. The most common concurrent disorder was hyperadrenocorticism (n = 10), and the adjusted odds of hyperadrenocorticism in dogs with HCC were 4.13 higher than those of the reference population (95% CI, 1.95-8.76). Propensity score matching analysis revealed that thrombocytosis (n=30/43), increased alanine aminotransferase (n=41/44), increased alkaline phosphatase (n=42/44), and hypercalcemia (n= 13/32) were significantly associated with HCC diagnosis. The results of this study suggest that Welsh Corgis and Beagles are breeds with a predisposition for HCC and that hyperadrenocorticism might be a potential risk factor.
@ 2019 Elsevier Ltd. All rights reserved.
Introduction
Hepatocellular carcinoma (HCC) is one of the most common primary liver tumours in humans (Mittal and El-Serag, 2013) and dogs (Patnaik et al., 1980, 1981; Liptak, 2013). In humans, the development of HCC is associated with major risk factors, including cirrhosis, chronic infection with hepatitis B and C viruses, alcoholic and non-alcoholic fatty liver disease. However, similar risk factors have not been identified in dogs because a viral etiology has not been detected in dogs, and an association between
* Corresponding author. E-mail address: mtaki@vetmed.hokudai.ac.jp (M. Takiguchi).
1 Dr. K. Nakamura’s present address is Organization for Promotion of Tenure Track, University of Miyazaki, Miyazaki 889-2192, Japan.
2 Dr. S. Takagi’s present address is Department of Veterinary Medicine, Azabu University, Kanagawa, 252-5201, Japan.
cirrhosis and HCC in dogs is rare, representing only 7% of dogs with HCC (Patnaik et al., 1981; Gumerlock et al., 1992).
A few studies have explored the risk factors for HCC in dogs and have revealed that certain breeds of dogs, particularly Miniature Schnauzers and Shih Tzus, and male dogs are overrepresented (Patnaik et al., 1980,1981; Liptak et al., 2004; Hirose et al., 2014). However, the risk factors for HCC in dogs have not yet been confirmed. Previous studies have reported that vacuolar hepat- opathy (VH) in Scottish terriers may be associated with HCC development, suggesting that VH might be a risk factor for HCC (Cortright et al., 2014; Peyron et al., 2014, 2015). In humans, recent studies have reported that hypothyroidism and diabetes mellitus are related to HCC (Hassan et al., 2009; Wang et al., 2014; Banal et al., 2017) due to the association with non-alcoholic steatohe- patitis (NASH; Liangpunsakul and Chalasani, 2003; El-Serag et al., 2006), which is considered to be a predisposing condition for HCC development (Fingas et al., 2016; Cholankeril et al., 2017).
In dogs, one study showed a disruption in mitochondrial ultrastructure and metabolism and modification of keratin filaments in VH livers (Peyron et al., 2015). Similar ultrastructural and metabolic changes in the liver have also been observed in humans with NASH (Takaki et al., 2013). Therefore, it is possible that VH-related disorders can increase the risk of HCC develop- ment, as 9/55 dogs with VH developed HCC (Peyron et al., 2015). However, a search for concurrent disorders in dogs with HCC has not been performed.
Due to limited information regarding the epidemiological features of HCC in dogs, the aims of this study were to estimate the prevalence of HCC and to identify potential risk factors associated with HCC, including clinicopathologic factors and concurrent disorders.
Materials and methods
Study population
A retrospective study was carried out in the Hokkaido University Veterinary Teaching Hospital (HUVTH) from 1 May 2013 to 31 May 2017. Informed consent was obtained from all owners of dogs involved in this study. HCC cases were identified by abdominal ultrasound and histopathologic examination following surgery. All histopathologic examinations were performed by a board-certified pathologist. The pathologic diagnosis of HCC was defined according to the guidelines of the World Small Animal Veterinary Association (WSAVA) Liver Standardization Group (Cullen, 2009). To estimate prevalence, to examine age, sex and breed predispositions, and to investigate risk factors for HCC including concurrent disorders, all dog cases presented to HUVTH during the study period were used as the reference population.
To characterize the clinical features of HCC, one-to-one propensity score matching combined with covariate adjustment was used to select a pair of dogs with and without HCC in the same conditions, resulting in no differences in age, sex, breed, and comorbidities for the case-control analysis.
Data collection
For both HCC cases and controls, data extracted from the medical records included signalment (age, sex, breed, and bodyweight); history of long-term steroid use in anti-inflammatory or immunosuppressive doses (0.5-2.0 mg/kg/day; >2 weeks; Reusch et al., 2015); clinicopathologic findings, including hematologic and serum biochemical analyses, endocrine test results, imaging results and concurrent diseases.
Hematological abnormalities were defined as follows: leucocytosis, white blood cell (WBC) count >17 x 103 cells/uL (reference range, 6-17 x 103 cells/u.L); anaemia, hematocrit (HCT) <37% (reference range, 37-55%); and thrombocytosis, platelet (PLT) count >500 × 103 cells/p.L (reference range, 200-500 x 103 cells/p.L). Serum biochemical abnormalities were defined as follows: hypoproteinemia, total protein (TP) content <50 g/L (reference range 50-72 g/L); hypoalbuminemia, albumin (Alb) content <26 g/L (reference range 26-40 g/L); and hypoglycemia, serum glucose (Glu) <4.2 mmol/L (reference range 4.2-7.1 mmol/L). The upper limits of the reference ranges for liver enzymes, including serum alanine aminotransferase (ALT), alkaline phosphatase (ALP), aspartate aminotransferase (AST) and gamma-glutamyl transferase (GGT), were 78 IU/L (reference range, 17- 78 IU/L), 254 IU/L (reference range, 47-254 IU/L), 44 IU/L (reference range, 17- 44 IU/L) and 14IU/L (reference range, 5-14IU/L), respectively. In addition, hyperbilirubinemia was defined as a total bilirubin (T-bil) concentration >8.6 mmol/L (reference range 1.7-8.6 mmol/L). Other serum biochemical abnor- malities were defined as hypercalcemia, hypertriglyceridemia and hypercholes- terolemia if the total calcium (tCa), triglyceride (TG) and total cholesterol (TCho) concentrations were >3.0 mmol/L (reference range, 2.3-3.0 mmol/L), >1.5 mmol/L (reference range, 0.3-1.5 mmol/L) and >8.1 mmol/L (reference range, 2.9- 8.1 mmol/L), respectively.
For endocrine testing, endocrine disorders including hyperadrenocorticism, hypothyroidism and diabetes mellitus that were diagnosed at a private animal hospital or our teaching hospital were considered in this study. Hyper- adrenocorticism was determined if the dogs had a historical diagnosis within 6 months of HCC presentation (Mesich et al., 2009), on the basis of a positive result with either a low-dose dexamethasone suppression test or an adrenocorticotrophic hormone (ACTH) stimulation test, in combination with one or more common clinical signs other than abdominal distension and hepatomegaly, as described in the ACVIM consensus statement (Behrend et al., 2013). Diagnosis of hypothyroidism was based on a historical diagnosis of a thyroid panel and low total or free thyroxine levels with elevated thyroid- stimulating hormone levels within 6 months of HCC presentation (Mesich et al., 2009). Diabetes mellitus was considered present if the dogs had a historical diagnosis prior to or within 3 months after HCC presentation (Mesich et al., 2009), based on persistent fasting hyperglycemia with clinical signs.
Statistical analysis
Period prevalence was evaluated for dogs diagnosed with HCC. Continuous variables, including age and bodyweight, were assessed using the Mann-Whitney U-test and were expressed as the median and range. Categorical variables, including breed, sex, clinicopathologic findings and concurrent disorders, were analysed using Pearson’s chi-square test or Fischer’s exact test. Factors possibly associated with HCC, including age, breed, sex and concurrent disorders, were assessed using univariate and multivariate logistic regression analysis. Odds ratios (ORs) and 95% confidence intervals (CIs) for univariate and multivariate associations between HCC and possible risk factors were also estimated. Statistical power analysis was conducted to determine the effect of a significant breed predisposition to HCC. Propensity score matching (1:1 match) was performed to minimize the effect of potential confounders on selection bias for case-control analysis, using multiple logistic regressions to estimate the probability of having specific clinical features for HCC. The covariates used in the propensity score were age, breed, and comorbidities. A Bonferroni correction was applied to account for the multiplicity of breeds. Statistical analyses were performed using commercial statistical software packages (JMP Pro, version 14.0.0, SAS Institute, and R 3.4.1, The R Project for Statistical Computing). P <0.05 was considered statistically significant (P<0.0036 after Bonferroni correction).
Results
Prevalence estimates
The study population consisted of 4607 client-owned dogs that were presented during the study period. Forty-one dogs were diagnosed with massive-type HCC, giving a prevalence of 0.96%.
Risk factors for HCC
The ages of the dogs diagnosed with HCC (median, 11 years; range, 8-15 years) were significantly higher (P <0.001) than those of the reference population (median, 9 years; range 0-20 years). The median bodyweight of dogs with HCC was 7 kg (range, 1.7- 32.5 kg). The HCC group included 18 females and 26 males. Compared with each sex category in the reference population (n=2107 females, n=2456 males, there was no significant difference with the HCC group (P=0.3186).
Details regarding the dog breeds are shown in Table 1. The HCC group included seven Welsh Corgis (15.9%), five Beagles (11.4%), five Shih Tzus (11.4%), five Chihuahuas (11.4%), four Miniature Dachshunds (9.1%), four Yorkshire terriers (9.1%), four Toy Poodles (9.1%), four Mixed breeds (9.1%) and one each of the following: Pug (2.3%), Shiba Inu (2.3%), Boston terrier (2.3%), Golden retriever (2.3%), Miniature Schnauzer (2.3%) and Pomeranian (2.3%). The total number of cases without HCC during the study period was 4563. Of these, 3293 dogs belonged to one of the dog breeds in which HCC was described. The number and proportion of each breed among the cases without HCC were as follows: 217 Welsh
| Breed | n | Total n | Odds ratio | 95% CI | P |
|---|---|---|---|---|---|
| Welsh Corgi | 7 | 224 | 3.79 | 1.67-8.60 | 0.0014* |
| Beagle | 5 | 112 | 5.34 | 2.06-13.81 | 0.0006 |
| Shih Tzu | 5 | 211 | 2.71 | 1.06-6.95 | 0.0378 |
| Chihuahua | 5 | 373 | 1.46 | 0.57-3.73 | 0.4274 |
| Miniature Dachshund | 4 | 852 | 0.44 | 0.16-1.23 | 0.1165 |
| Yorkshire terrier | 4 | 163 | 2.77 | 0.98-7.84 | 0.0548 |
| Toy Poodle | 4 | 338 | 1.27 | 0.45-3.56 | 0.6546 |
| Mixed breed | 4 | 342 | 1.25 | 0.44-3.51 | 0.6722 |
| Pug | 1 | 76 | 1.39 | 0.19-10.24 | 0.7455 |
| Shiba Inu | 1 | 179 | 0.57 | 0.08-4.18 | 0.5829 |
| Boston terrier | 1 | 37 | 2.92 | 0.39-21.82 | 0.2953 |
| Golden retriever | 1 | 94 | 1.12 | 0.15-8.20 | 0.9128 |
| Miniature Schnauzer | 1 | 167 | 0.62 | 0.08-4.50 | 0.6330 |
| Pomeranian | 1 | 125 | 0.83 | 0.11-6.09 | 0.8568 |
95% CI, 95% confidence interval.
* P < 0.0036 were statistically significant by Bonferroni correction.
Corgis (4.8%), 107 Beagles (2.8%), 206 Shih Tzus (4.5%), 368 Chihuahuas (8.1%), 848 Miniature Dachshunds (18.6%), 159 Yorkshire terriers (3.5%), 334 Toy Poodles (7.3%), 338 Mixed breeds (7.4%), 75 Pugs (1.6%), 178 Shiba Inus (3.9%), 36 Boston terriers (0.8%), 93 Golden retrievers (2%), 166 Miniature Schnauz- ers (3.6%) and 124 Pomeranians (2.7%). A significant breed predisposition to HCC was observed in Welsh Corgis (OR, 3.79; 95% CI, 1.67-8.60; P=0.0014) and Beagles (OR, 5.34; 95% CI, 2.06- 13.81; P=0.0006).
Of the 44 HCC cases, 27 (61.4%) had at least one concurrent disease (Table 2). The most frequent concurrent disease with HCC was hyperadrenocorticism (total n=10; n=3 Beagles, n=2 Chihuahuas and one each of the following: Welsh Corgi, Mixed breed, Pomeranian, Boston terrier and Toy Poodle). The association of HCC with concurrent disorders, especially endocrinopathies, is shown in Table 3. Chi-square testing revealed that the odds of hyperadrenocorticism in dogs diagnosed with HCC were 6.92 times those of the controls (95% CI, 3.37-14.22; P<0.0001). However, there was no significant association between HCC and hypothy- roidism or diabetes mellitus (P>0.05). In addition, only two HCC cases had a history of long-term steroid use (4.5%).
Multivariate logistic regression analysis confirmed that age was significantly associated with HCC, with increased risk in older dogs (OR, 1.20; 95% CI, 1.07-1.33; P=0.0005). Welsh Corgis (OR, 3.68; 95% CI, 1.56-8.67; P=0.0029) and Beagles (OR, 4.33; 95% CI, 1.58- 11.90; P=0.0044) were the only breeds with a statistically significant predisposition to HCC (statistical power =75.3% and 76.9%, respectively). Although Shih Tzus were a predisposed breed in univariate analysis, the statistical power was only 46%. Hyperadrenocorticism was significantly associated with HCC as a concurrent disorder (OR, 4.13; 95% CI, 1.95-8.76; P=0.0002). However, sex and hypothyroidism or diabetes mellitus were not associated with HCC. Variables associated with HCC in univariate and multivariate analysis are summarized in Table 3.
Clinical characteristics of HCC
According to the one-to-one propensity score matching, 44 dogs without HCC from the reference population were matched with 44 HCC cases. The clinicopathologic findings for the HCC cases were compared with those for dogs without HCC as the control
| Category | n |
|---|---|
| Endocrinopathy/metabolic | |
| Hypothyroidism | 2 |
| Hyperadrenocorticism | 10 |
| Diabetes mellitus | 1 |
| Thyroid carcinoma | 1 |
| Hepatic/pancreatic | |
| Nodular hyperplasia | 3 |
| Gallbladder mucocele | 1 |
| Cardiovascular | |
| Myxomatous mitral valve degeneration | 2 |
| Heart-base tumour | 1 |
| Gastrointestinal | |
| Tooth root abscess | 1 |
| Leiomyoma of the ileum | 1 |
| Urinary | |
| Membranous glomerulonephritis | 1 |
| Chronic kidney disease | 1 |
| Bladder calculi | 1 |
| Neurological | |
| Idiopathic epilepsy | 1 |
| Cauda equina syndrome | 1 |
| Meningioma | 1 |
| Others | 4 |
| Total | 33 |
group. The results of the clinicopathologic findings are summa- rized in Table 4. Hematology was performed in 44 dogs, and data were available for 43 dogs for WBC count, HCT and PLT count. Serum biochemical analysis was performed in 44 dogs, and ALT, ALP and TP were evaluated in all dogs. Alb and T-bil concentrations were evaluated in 42 dogs; Glu was evaluated in 41 dogs; and tCa concentrations were evaluated in 32 dogs. Serum AST and GGT activities and TCho and TG concentrations were assessed in 30, 28, 27 and 15 dogs, respectively. Thrombocytosis (n=30/43; 69.8%; P=0.0002), elevated ALT (n=41/44; 93.2%; P<0.0001), elevated ALP (n=42/44; 95.5%; P=0.0034), and hypercalcemia (n=13/32; 40.6%; P=0.0042) were significantly associated with HCC.
Discussion
This study investigated the prevalence, risk factors and clinical characteristics associated with HCC in dogs. Our results revealed a higher prevalence of HCC than that observed in a previous study (Patnaik et al., 1981) and confirmed the risk of HCC development in older dogs, as reported in previous studies (Patnaik et al., 1980,1981). In addition, we report for the first time a breed predisposition for HCC in Welsh Corgis and Beagles and an association between HCC and hyperadrenocorticism. This study also found a significant association between canine HCC and thrombocytosis, elevated ALT and ALP and hypercalcemia. However, in contrast to the results of previous studies (Patnaik et al., 1980, 1981), there was no sex predisposition for HCC in this study.
In this study, the prevalence of HCC was higher than in a previous report, in which HCC was observed in 0.46% of dogs at necropsy (Patnaik et al., 1981). This discrepancy might be due to recent advances in diagnostic technology and/or to differences in denominator populations. However, our results supported the results of previous studies that reported increased risk of HCC in dogs >10 years old (Patnaik et al., 1980, 1981).
Interestingly, this study found an increased risk of HCC in Welsh Corgis and Beagles with a power of 75.3% and 76.9%, respectively. This result is inconsistent with previous studies reporting an overrepresentation of HCC in Miniature Schnauzers (Liptak et al., 2004) and Shih Tzus (Hirose et al., 2014). Although Shih Tzus were predisposed to HCC in univariate analysis, the power statistic for Shih Tzus was only 46.6%. In addition, multivariate analysis confirmed that Shih Tzus were not predisposed to HCC. According to this analysis, there is low possibility that Shih Tzus are a predisposed breed in this study. However, we cannot exclude a predisposition of Shih Tzus to HCC based on underpowered statistics. Thus, further studies with a large number of Shih Tzus are needed to confirm the possibility that Shih Tzus are predisposed to HCC. Differences in breed predisposition among studies might also occur due to regional differences. Moreover, it is possible that there are genetic differences in Welsh Corgis and Beagles from the area where this study was performed compared to those in other studies.
In humans, HCC is associated with chronic liver diseases, such as NASH. Previous studies have indicated an association between NASH and a metabolic syndrome characterized by lipid accumula- tion in hepatocytes (Fingas et al., 2016; Cholankeril et al., 2017). Lipid accumulation leads to mitochondrial dysfunction, which results in oxidative stress in hepatocytes and can lead to the development of HCC (Paschos and Paletas, 2009; Vanni et al., 2010; Eshraghian and Jahromi, 2014). In dogs, VH is a common hepatic disorder that has histopathologic characteristics similar to NASH in humans, although the pathophysiology of both disorders is different because VH is mostly associated with glycogen accumu- lation secondary to endogenous or exogenous glucocorticoid excess. However, NASH may be a form of VH, since a previous
| Variable | Unadjusted OR (95% CI) | P | Adjusted OR (95% CI)" | P |
|---|---|---|---|---|
| Age | 1.25 (1.13-1.38) | < 0.0001 | 1.20 (1.07-1.33) | 0.0005 |
| Breed | ||||
| Welsh Corgis | 3.79 (1.67-8.60) | 0.0014* | 3.68 (1.56-8.67) | 0.0029* |
| Beagles | 5.34 (2.06-13.81) | 0.0006 | 4.33 (1.58-11.90) | 0.0044 |
| Shih Tzus | 2.71 (1.06-6.95) | 0.0378 | 2.61 (0.98-6.99) | 0.0556 |
| Male sex | 1.36 (0.74-2.51) | 0.3186 | 1.47 (0.79-2.73) | 0.2189 |
| Concurrent disorder | ||||
| Hyperadrenocorticism | 6.92 (3.37-14.22) | <0.0001 | 4.13 (1.95-8.76) | 0.0002 |
| Hypothyroidism | 1.29 (0.31-5.36) | 0.7302 | 0.82 (0.19-3.51) | 0.7854 |
| Diabetes mellitus | 2.06 (0.28-15.23) | 0.4799 | 1.69 (0.21-13.43) | 0.6209 |
OR, Odds ratio; 95% CI, 95% confidence interval.
* P < 0.05 were statistically significant.
| Parameter | Reference range | HCC cases (n=44) | Controls (n=44) | P | ||||
|---|---|---|---|---|---|---|---|---|
| n | Median (range) | Abnormal (%) | n | Median (range) | Abnormal (%) | |||
| Hematologic findings | ||||||||
| Leukocytes (x103 cells/uL) | 6.0-17.0 | 43 | 9.8 (5.4-23.5) | 9.3 | 43 | 11.1 (4.8-47.9) | 18.6 | 0.2072 |
| Hematocrit (%) | 37.0-55.0 | 43 | 41.4 (20.6-57.7) | 27.9 | 44 | 42.9 (14.8-54.5) | 20.5 | 0.4607 |
| Platelets (×103 cells/p.L) | 200-500 | 43 | 575 (165-1160) | 69.8 | 43 | 403 (77-756) | 25.6 | 0.0002 |
| Serum biochemical findings | ||||||||
| Total protein (g/L) | 50-72 | 44 | 73 (56-92) | 0 | 44 | 69 (40-98) | 4.6 | 0.1055 |
| Albumin (g/L) | 26-40 | 42 | 33 (21-48) | 4.8 | 43 | 32 (12-39) | 16.3 | 0.1561 |
| Glucose (mmol/L) | 4.2-7.1 | 41 | 5.7 (2.1-9.3) | 2.4 | 43 | 6.1 (4.0-12.5) | 2.3 | 0.9717 |
| ALT (IU/L) | 17-78 | 44 | 314.5 (64-1001) | 93.2 | 44 | 73 (17-1001) | 45.5 | <0.0001 |
| ALP (IU/L) | 47-254 | 44 | 2551 (179-3591) | 95.5 | 44 | 477 (72-3501) | 70.5 | 0.0034 |
| AST (IU/L) | 17-44 | 30 | 38.5 (17-369) | 43.3 | 20 | 33.5 (12-848) | 25.0 | 0.2370 |
| GGT (IU/L) | 5-14 | 28 | 13 (1-1076) | 42.9 | 18 | 11 (0-1201) | 38.9 | 1.0000 |
| Total bilirubin (µmol/L) | 1.7-8.6 | 42 | 3.4 (1.7-8.6) | 0 | 41 | 1.7 (1.7-172.8) | 7.3 | 0.1160 |
| Total calcium (mmol/L) | 2.3-3.0 | 32 | 2.9 (2.2-3.4) | 40.6 | 39 | 2.7 (1.7-3.3) | 7.7 | 0.0042 |
| Total cholesterol (mmol/L) | 2.9-8.1 | 27 | 7.2 (2.7-11.7) | 44.4 | 21 | 6.7 (2.8-11.7) | 33.3 | 0.5553 |
| Triglycerides (mmol/L) | 0.3-1.5 | 15 | 1.1 (0.7-4.0) | 26.7 | 9 | 0.6 (0.5-5.7) | 22.2 | 1.0000 |
ALT, alanine aminotransferase; ALP, alkaline phosphatase; AST, aspartate aminotransferase; GGT, gamma-glutamyl transferase. P<0.05 were statistically significant.
study reported that VH in dogs also leads to mitochondrial dysfunction in hepatocytes, which is similar to the effects of NASH in humans (Peyron et al., 2015). Therefore, the association of HCC with NASH in humans may be similar to the association with VH in dogs. Thus, VH may contribute to the development of HCC in dogs.
A previous study in Scottish terriers suggested that VH can cause hepatic remodelling and may progress to degenerative VH with the formation of regenerative foci. This transition may exhibit dysplastic characteristics and precede the development of HCC, as reported in human and experimental animal models (Cortright et al., 2014). However, other dog breeds can also develop degenerative VH, as reported in a previous study where an association between VH and neoplasia was suggested (Sepesy et al., 2006). This indicates that it is possible that VH secondary to hyperadrenocorticism might play a role in the pathogenesis of HCC. Therefore, HCC should be considered when liver pathology is diagnosed in dogs with hyperadrenocorticism. However, the association between HCC and hyperadrenocorticism is in our study is inconsistent with a recent report of disease associations in dogs with hyperadrenocorticism (Hoffman et al., 2018). Differ- ences in associated comorbidities might be due to difference between study designs. The present study used the same period of disease occurrence as a condition for both HCC cases and the reference population, in contrast to the previous report. Thus, it is possible that the association between HCC and hyperadrenocor- ticism could be present within the same period rather than at the
same time point (i.e., death), since massive HCC can be treated by surgical resection before death.
This study did not find an association between HCC and hypothyroidism or diabetes mellitus, although these two diseases are chronic disorders and can cause VH (Watson, 2017). However, we cannot be certain that there were no such associations due to the small number of HCC cases with those two diseases. Thus, further studies are needed to investigate a large-scale HCC population to confirm the results of this study and determine whether there are any differences in the pathophysiology of VH in dogs with lipid and glycogen accumulation.
Although clinicopathologic features are usually nonspecific (Bexfield, 2017; Selmic, 2017), thrombocytosis and hypercalcemia were overrepresented in the dogs with HCC examined here, which is similar to the results of previous reports (Patnaik et al., 1980; Liptak et al., 2004). The causes of HCC-related thrombocytosis and hypercalcemia in dogs are still unclear. These conditions may result from paraneoplastic syndrome, as observed in human HCC (Luo et al., 1999; Hwang et al., 2004; Chang et al., 2013). However, for hypercalcemia, the present study only evaluated the tCa concentration. Therefore, further investigation is needed to evaluate the ionized calcium concentration to confirm the presentation of hypercalcemia in dogs with HCC and determine whether these two conditions are paraneoplastic phenomena. Moreover, ALT and ALP levels were frequently increased in this study, supporting the results of previous studies, which reported
that dogs with HCC typically present with high serum liver enzyme (Patnaik et al., 1980; Liptak et al., 2004). However, this observation is not specific for liver tumours.
This study has several limitations. Firstly, there was a small number of dogs with HCC, which may limited our ability to demonstrate an association in some breeds and with hypothyroidism or diabetes mellitus. Secondly, we did not investigate an association between HCC and long-term steroid use, or the physiological effects of exogenous glucocorticoids on HCC development in this study, due to the small number of HCC cases with long-term glucocorticoid administration and the difficulty of collecting the history of long-term steroid use in the reference population because of the retrospective nature of this study. Thus, the possibility of HCC development associated with excess exogenous glucocorticoids remains unknown. In addition, due to the retrospective study design, clinicopathologic findings were not established for all dogs. Missing data may also have affected our results. Concurrent disorders occurring within 6 months of HCC presentation may not necessarily have been related to HCC, although this period provided adequate time for examining diseases suspected at the time of HCC diagnosis. There is also a possibility of false-positive diagnosis of hyperadrenocorticism in HCC cases. To minimize this limitation, we used stricter diagnostic criteria for hyperadrenocorticism, including only HCC cases presenting with common clinical signs of hyper- adrenocorticism other than abdominal distension and hepatomegaly in combination with positive endocrine tests. Finally, this retrospective study cannot confirm the role of hyperadrenocorticism in the pathogenesis of HCC development. Therefore, a prospective study with a large-scale population should be conducted to define any associations between HCC and hyperadrenocorticism or other comorbidities.
Conclusions
There was increased risk of HCC development with age, and Welsh Corgis and Beagles were predisposed to HCC. In addition, a significant association between HCC and hyperadrenocorticism was observed, suggesting that hyperadrenocorticism might be a predisposing factor for HCC development.
Conflict of interest statement
None of the authors have any financial or personal relationship that could inappropriately influence or bias the content of the paper.
Acknowledgements
No specific grant was received for this research from funding agencies in the public, commercial, or not-for-profit sectors. Preliminary results were presented as an abstract and poster presentation at the 2018 ACVIM Forum in Seattle, Washington, USA (14-16 June 2018). The authors wish to thank Dr. Yumiko Kagawa (YK), an American College of Veterinary Pathologists board- certified pathologist, for her help with interpretation of all the histopathologic findings.
References
Banal, K.A., Paz-Pacheco, E., de Villa, V., 2017. Diabetes mellitus and prediabetes in patients with hepatocellular carcinoma in a tertiary Philippine hospital. Journal of the ASEAN Federation of Endocrine Societies 32, 32-37.
Behrend, E.N., Kooistra, H.S., Nelson, R., Reusch, C.E., Scott-Moncrieff, J.C., 2013. Diagnosis of spontaneous canine hyperadrenocorticism: 2012 ACVIM consensus statement (small animal). Journal of Veterinary Internal Medicine 27, 1292-1304.
Bexfield, N., 2017. Neoplasms of the liver, In: Ettinger, S.J., Feldman, E.C., Cote, E. (Eds.), Textbook of Veterinary Internal Medicine. 8th ed. Elsevier Saunders, St. Louis, MO, USA, pp. 4065-4074.
Chang, P.E., Ong, W.C., Lui, H.F., Tan, C.K., 2013. Epidemiology and prognosis of paraneoplastic syndromes in hepatocellular carcinoma. ISRN Oncology 2013, 684026.
Cholankeril, G., Patel, R., Khurana, S., Satapathy, S.K., 2017. Hepatocellular carcinoma in non-alcoholic steatohepatitis: current knowledge and implications for management. World Journal of Hepatology 9, 533-543.
Cortright, C.C., Center, S.A., Randolph, J.F., McDonough, S.P., Fecteau, K.A., Warner, K. L., Chiapella, A.M., Pierce, R.L., Graham, A.H., Wall, L.J., et al., 2014. Clinical features of progressive vacuolar hepatopathy in Scottish terriers with and without hepatocellular carcinoma: 114 cases (1980-2013). Journal of the American Veterinary Medical Association 245, 797-808.
Cullen, J.M., 2009. Summary of the World Small Animal Veterinary Association standardization committee guide to classification of liver disease in dogs and cats. Veterinary Clinics of North America: Small Animal Practice 39, 395-418.
El-Serag, H.B., Hampel, H., Javadi, F., 2006. The association between diabetes and hepatocellular carcinoma: a systematic review of epidemiologic evidence. Clinical Gastroenterology and Hepatology 4, 369-380.
Eshraghian, A., Jahromi, A.H., 2014. Non-alcoholic fatty liver disease and thyroid dysfunction: a systematic review. World Journal of Gastroenterology 20, 8102-8109.
Fingas, C.D., Best, J., Sowa, J.P., Canbay, A., 2016. Epidemiology of nonalcoholic steatohepatitis and hepatocellular carcinoma. Clinical Liver Disease 8, 119-122.
Gumerlock, P.H., Kraegel, S.A., Madewell, B.R., 1992. Detection of mammalian and avian hepadenovirus by the polymerase chain reaction. Veterinary Microbiology 32, 273-280.
Hassan, M.M., Kaseb, A., Li, D., Patt, Y.Z., Vauthey, J.N., Thomas, M.B., Curley, S.A., Spitz, M.R., Sherman, S.I., Abdalla, E.K., et al., 2009. Association between hypothyroidism and hepatocellular carcinoma: a case-control study in the United States. Hepatology 49, 1563-1570.
Hirose, N., Uchida, K., Kanemoto, H., Ohno, K., Chambers, J.K., Nakayama, H., 2014. A retrospective histopathological survey on canine and feline liver diseases at the University of Tokyo between 2006 and 2012. Journal of Veterinary Medical Science 76, 1015-1020.
Hoffman, J.M., Lourenço, B.N., Promislow, D.E.L., Creevy, K.E., 2018. Canine hyperadrenocorticism associations with significant, selected comorbidities and mortality within North American veterinary teaching hospitals. Journal of Small Animal Practice 59, 681-690.
Hwang, S.J., Luo, J.C., Li, C.P., Chu, C.W., Wu, J.C., Lai, C.R., Chiang, J.H., Chau, G.Y., Lui, W.Y., Lee, C.C., et al., 2004. Thrombocytosis: a paraneoplastic syndrome in patients with hepatocellular carcinoma. World Journal of Gastroenterology 10, 2472-2477. Liangpunsakul, S., Chalasani, N., 2003. Is hypothyroidism a risk factor for nonalcoholic steatohepatitis? Journal of Clinical Gastroenterology 37, 340-343. Liptak, J.M., 2013. Hepatobillary tumors, In: Withrow, S.J., Vail, D.M., Page, R.L. (Eds.), Withrow and MacEwen’s Small Animal Clinical Oncology. 5th ed. Elsevier Saunders, St. Louis, MO, USA, pp. 405-412.
Liptak, J.M., Dernell, W.S., Monnet, E., Powers, B.E., Bachand, A.M., Kenney, J.G., Withrow, S.J., 2004. Massive hepatocellular carcinoma in dogs: 48 cases (1992- 2002). Journal of the American Veterinary Medical Association 225, 1225-1230.
Luo, J.C., Hwang, S.J., Wu, J.C., Li, C.P., Hsiao, L.T., Lai, C.R., Chiang, J.H., Lui, W.Y., Chang, F.Y., Lee, S.D., 1999. Paraneoplastic syndromes in patients with hepatocellular carcinoma in Taiwan. Cancer 86, 799-804.
Mesich, M.L.L., Mayhew, P.D., Paek, M., Holt, D.E., Brown, D.C., 2009. Gall bladder mucoceles and their association with endocrinopathies in dogs: a retrospective case-control study. Journal of Small Animal Practice 50, 630-635.
Mittal, S., El-Serag, H.B., 2013. Epidemiology of HCC: consider the population. Journal of Clinical Gastroenterology 47, 1-10.
Paschos, P., Paletas, K., 2009. Non alcoholic fatty liver disease and metabolic syndrome. Hippokratia 13, 9-19.
Patnaik, A.K., Hurvitz, A.I., Lieberman, P.H., 1980. Canine hepatic neoplasms: a clinicopathologic study. Veterinary Pathology 17, 553-564.
Patnaik, A.K., Hurvitz, A.I., Lieberman, P.H., Johnson, G.F., 1981. Canine hepatocellular carcinoma. Veterinary Pathology 18, 427-438.
Peyron, C., Chevallier, M., Lecoindre, P., Guerret, S., Pagnon, A., 2014. Clinical, blood biochemical and hepatic histological data in 49 French Scottish terrier dogs according to their plasma ALP activity, hepatic vacuolation and the presence or absence of hepatocellular carcinoma. Revue de Médecine Vétérinaire 165, 245-251.
Peyron, C., Lecoindre, P., Chevallier, M., Guerret, S., Pagnon, A., 2015. Vacuolar hepatopathy in 43 French Scottish terriers: a morphological study. Revue de Médecine Vétérinaire 166, 176-184.
Reusch, C.E., 2015. Glucocorticoid therapy, In: Feldman, E.C., Nelson, R.W., Reusch, C., Scott-Moncrieff, J.C. (Eds.), Canine and Feline Endocrinology. 4th ed. Elsevier Saunders, St. Louis, MO, USA, pp. 555-577.
Selmic, L.E., 2017. Hepatobiliary neoplasia. Veterinary Clinics of North America: Small Animal Practice 47, 725-735.
Sepesy, L.M., Center, S.A., Randolph, J.F., Warner, K.L., Erb, H.N., 2006. Vacuolar hepatopathy in dogs: 336 cases (1993-2005). Journal of the American Veterinary Medical Association 229, 246-252.
Takaki, A., Kawai, D., Yamamoto, K., 2013. Multiple hits, including oxidative stress, as pathogenesis and treatment target in non-alcoholic steatohepatitis (NASH). International Journal of Molecular Sciences 14, 20704-20728.
Vanni, E., Bugianesi, E., Kotronen, A., De Minicis, S., Yki-Järvinen, H., Svegliati- Baroni, G., 2010. From the metabolic syndrome to NAFLD or vice versa? Digestive and Liver Disease 42, 320-330.
Wang, Y.G., Wang, P., Wang, B., Fu, Z.J., Zhao, W.J., Yan, S.L., 2014. Diabetes mellitus and poorer prognosis in hepatocellular carcinoma: a systematic review and meta-analysis. PLoS One 9, e95485.
Watson, P.J., 2017. Metabolic diseases of the liver, In: Ettinger, S.J., Feldman, E.C., Cote, E. (Eds.), Textbook of Veterinary Internal Medicine. 8th ed. Elsevier Saunders, St. Louis, MO, USA, pp. 4037-4051.