Lomustine (1-[2-chloroethyl]-3-cyclohexyl-1-nitrosourea) is an alkylating agent used to treat dogs with various neoplastic diseases, such as mast cell tumors, lymphomas, CNS tumors, and histiocytic sarcomas.1–5 It is currently considered the first-line treatment for several aggressive tumor types, including canine histiocytic sarcoma and epitheliotropic lymphoma. Lomustine can be administered by the PO route and has high lipophilicity, which facilitates its penetration across the blood-brain barrier.6,7 Thus, lomustine is a popular treatment alternative for veterinary patients.8 Despite the positive attributes of lomustine, it has been associated with hepatopathy in dogs (as evidenced by increases in ALT activity) and, rarely, liver failure.9
The hepatotoxic effects of lomustine were first noted in dogs administered high doses of the drug in preclinical studies10,11 and were characterized by an increase in ALT activity and histopathologic changes that persisted after liver enzyme activities returned to normal levels. It is important to note that the dogs of those studies10,11 received intentionally high doses of lomustine or were administered the drug over long periods, neither of which reflects clinical use of lomustine in dogs with neoplastic diseases. However, a lomustine-induced increase in ALT activity is common in dogs, occurring in 51% to 86% of cancer-bearing dogs treated with the drug.5,9,12,13 When dogs receiving lomustine develop high ALT activities, most veterinarians discontinue or extend the dosing interval of the drug, although no definitive guidelines have been established. Dogs treated with lomustine commonly receive glucocorticoids concurrently for the treatment of hematopoietic malignancies, and corticosteroid-induced enzymopathies confound interpretation of increased liver enzyme activities.14
In dogs, lomustine administration is most commonly discontinued owing to signs of hepatopathy, despite its apparent efficacy against the neoplastic process.1,2,12 Previous reports9,13,15 indicate that lomustine administration is permanently discontinued in 24% to 33% of treated dogs because of high ALT activity, even in the absence of clinical signs of liver disease. Treatment delays or premature cessation of lomustine may negatively affect the likelihood of remission and patient survival. Although lomustine-treated dogs commonly develop abnormally increased ALT activity, clinical liver failure is rare, affecting only 1% to 6% of patients.16,17 The discrepancy between the high incidence of liver enzymopathy and low incidence of liver failure suggests that lomustine may cause an adaptive change in many treated dogs such that transient increases in liver enzyme activities occur in the absence of substantial histologic structural changes in the liver.18
The mechanism responsible for lomustine-induced hepatopathy has not been well studied. Multiple reports9,11,16,19 describe a delayed, chronic hepatopathy with acute increases in liver enzyme activity after administration of several doses of lomustine, without concurrent clinical signs. Evidence is conflicting on whether this phenomenon is related to the cumulative dose of lomustine administered.9,16 In lomustine-treated rats and humans, an isocyanate metabolite has been shown to inhibit glutathione reductase, suggesting decreases in reduced glutathione levels.20 Decreases in concentrations of endogenous antioxidants, such as glutathione, have been implicated in the development of drug-induced liver injury in dogs, cats, and mice.21,22 Although it is unclear whether a decrease in endogenous antioxidant concentrations has a role in lomustine-induced hepatopathy, the use of supplemental antioxidants that increase hepatic glutathione concentration mitigates the increase in liver enzyme activities in dogs with cancer that are treated with lomustine.13
The objective of the experimental study reported here was to characterize the biochemical, functional, and histopathologic changes associated with lomustine-induced liver injury in healthy dogs administered the drug in accordance with a commonly used chemotherapy protocol with and without concurrent administration of prednisone. We hypothesized that liver enzyme activities, results of liver function tests, histologic scores, and glutathione depletion would consistently increase for all dogs. We also hypothesized that indices of liver function, such as BA concentration, might be better predictors of histologic injury to the liver than ALT activity. We anticipated that increases in ALT activity would supersede changes in liver function and that prednisone-induced changes in liver enzyme activities would confound interpretation of the effects of lomustine on those activities in dogs administered both drugs concurrently.
Materials and Methods
Animals
All study procedures were reviewed and approved by the Louisiana State University Institutional Animal Care and Use Committee (IACUC No. 15–097). Ten purpose-bred hounds with a mean age of 3.8 years (range, 3 to 5 years) and body weight of 27 kg (range, 24 to 34 kg) were selected for study enrollment from a research colony. Only sexually intact female hounds were enrolled in the study on the basis of their availability in coordination with the laboratory animal medicine faculty and their cooperative temperaments. All dogs were determined to be healthy on the basis of results of a physical examination, CBC, serum biochemical analysis, and urinalysis.
To assess liver function prior to study initiation (baseline; day 0), food was withheld from dogs for 12 hours and a blood sample (approx 3 to 5 mL) was collected via jugular venipuncture to determine serum ammonia (fasting ammonia) and preprandial serum total BA concentrations. The dogs were then fed, and another blood sample was collected for determination of postprandial serum total BA concentration. Fasting ammonia and pre- and postprandial serum total BA concentrations were quantified as described.23 Additionally, focused abdominal ultrasonographic evaluation of the liver and ultrasound-guided collection of percutaneous liver biopsy specimens were performed at baseline.
Experimental design
Dogs were randomly assigned by means of pulling numbers from a container to receive lomustine alone (n = 5) or in combination with prednisone (5). For both treatment groups, the target dose of lomustine to be administered was 75 mg/m2. The drug was provided in 10-mg capsulesa; therefore, the actual dose administered was rounded up to the next multiple of 10. Lomustine was administered PO, every 21 days for a total of 5 doses. The first dose of lomustine was administered on day 0 after baseline data were collected. The dose of lomustine was reduced for individual dogs owing to signs of toxicosis on the basis of clinical evaluation, but the lomustine dose remained within the standard clinical dose range (60 to 80 mg/m2) for all dogs throughout the experimental period.
Dogs assigned to the lomustine-prednisone group also received prednisoneb (1.5 to 1.7 mg/kg, PO, q 24 h for 12 weeks). Following completion of lomustine administration (day 84), the prednisone dose was gradually tapered over 2 weeks and discontinued on day 98. All dogs were administered cefpodoximec (6.0 to 8.4 mg/kg, PO, q 24 h for 7 days) beginning the fifth day after administration of each lomustine dose (ie, days 5, 26, 47, 68, and 89) as prophylaxis to minimize the risk for potential infections subsequent to lomustine-induced neutropenia. Cefpodoxime was chosen because of its broad spectrum of activity against gram-positive and gram-negative organisms and the fact that it required only once-daily administration. All dogs also received a flea and heartworm preventatived on a monthly basis for the duration of the experimental period.
Sample and data collection
For each dog, blood (approx 8 to 10 mL) was collected as previously described for a CBC, serum biochemical analysis, and determination of pre- and postprandial serum total BA concentrations on days 0, 21, 42, 63, 84, 104, 161, and 220 and for determination of fasting ammonia concentration on days 21, 42, 63, 84, and 104. A urine sample (5 mL) was collected by cystocentesis for a urinalysis on days 0 and 104. One week after lomustine administration, a CBC was performed for each dog to monitor for neutrophil and platelet nadirs.
Ultrasonography was used to evaluate the liver and guide collection of liver biopsy specimens on days 0, 63, 84, 104, and 220 for all dogs and within 3 days after lomustine administration for any dog that was assigned a VCOG adverse event24 of grade 3 or greater or that developed an ALT activity > 240 U/L (ie, a 4-fold increase from the upper limit of the reference range). All ultrasonographic examinations were performed by the same ultrasonographer (NR) with a 10.5-MHz convex transducer probe.e
For the liver biopsy procedure, each dog was sedated with butorphanol tartratef (0.4 mg/kg, IM) and dexmedetomidine hydrochlorideg (5 μg/kg, IM). Maropitanth (1 mg/kg, SC) was administered to dogs that developed signs of nausea. Dogs were positioned in right lateral recumbency for ultrasonographic evaluation of the liver and identification of the optimum window for biopsy specimen collection. A No. 10 scalpel blade was used to create a stab incision through the skin at the optimum biopsy window (either caudal to the last rib on the left side or on midline at the level of the last rib). A spring-loaded biopsy guni with a 14-gauge biopsy needle was used to obtain 4 core liver biopsy specimens with ultrasound guidance. One biopsy specimen was snap frozen with liquid nitrogen for measurement of hepatic reduced glutathione concentration. The 3 remaining specimens were fixed in neutral-buffered 10% formalin for histologic examination. Sedation was reversed with atipamezolej (50 μg/kg, IM). Tramadolk (2 to 5 mg/kg, PO) was administered as needed for analgesia for 24 hours after each biopsy procedure.
Histologic examination
Formalin-fixed liver biopsy specimens were processed and embedded in paraffin in a routine manner. Specimens were cut into 5-μm-thick sections, placed on a microscope slide, and stained with H&E stain. All biopsy specimens were histologically evaluated by 1 board-certified veterinary anatomic pathologist (IML) who was unaware of (blinded to) the treatment group assignment of each dog. Histopathologic changes and the severity of those changes (in particular hepatocellular vacuolar degeneration, inflammation, and biliary duct damage) were assessed in accordance with the World Small Animal Veterinary Association Liver Standardization Group guidelines.25 Briefly, the histologic score for a given biopsy specimen could range from 0 to 50, with higher scores indicative of greater lesion severity. When warranted, additional sections of each biopsy specimen were processed and stained to assess the presence of copper (rhodanine stain), iron (Perl Prussian blue stain), lipofuscin (Schmorl stain), bile (Hall stain), and fibrosis (Masson trichrome stain).
Hepatic reduced glutathione analysis
The hepatic reduced glutathione concentration in snap-frozen biopsy specimens was determined within 1 week after specimen acquisition by use of high-performance liquid chromatography with bromobimane derivatization, as described.26 The limit of quantitation for reduced glutathione concentration was 1μM. The assay had an intra-assay coefficient of variation of 1.4% to 2.8% and an interassay coefficient of variation of 7.4% to 13.9% within the glutathione concentration range relevant to this study.
Statistical analysis
Dependent variables of interest included serum ALT and ALP activities, total bilirubin and BA concentrations, liver histologic score, and hepatic reduced glutathione concentration. The data distribution for each dependent variable was assessed for normality by visual analysis of residual plots, which exhibited skewness or asymmetry. Results indicated that none of the dependent variables were normally distributed; therefore, the Box-Cox method was used to determine the optimal monotonic transformation to induce normality for each dependent variable prior to analysis (Supplementary Appendix S1, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.81.10.810). Repeated-measures ANOVA was used to determine whether each dependent variable differed significantly between the 2 treatment groups over time. Each model included fixed effects for sample acquisition time (time), group (lomustine-only or lomustine-prednisone), and the interaction between time and group and a random effect to control for repeated measures within subjects. A compound symmetry correlation structure was assumed, and model residuals were checked by use of smooth density estimation and normal probability plotting. We expected the fixed effect of time to be statistically significant. However, we determined that the interaction between time and group was the most important independent variable, which indicated that there was a difference between the 2 treatment groups when time was held constant. Post hoc contrast estimation was used to compare the means between the 2 groups at various times. To assess the respective relationships between peak biochemical measures and the histologic score, simple linear regression plots were created and the Pearson correlation coefficient (r) was calculated. Values of P < 0.05 were considered significant for all analyses. All analyses were performed with a commercially available statistical software program.l
Results
Dogs
Results of all baseline bloodwork and ultrasonographic liver evaluations were within the expected reference ranges for all dogs except 1. That dog had mild hypoalbuminemia (albumin, 2.1 g/dL; reference range, 2.6 to 4.2 g/dL); however, its liver function was deemed normal on the basis of serum fasting ammonia and preprandial and postprandial BA concentrations, and proteinuria was not detected during urinalysis. Therefore, the dog was enrolled in the study.
For all 10 study dogs, the mean lomustine dose initially administered was 74 mg/m2 (range, 71 to 80 mg/m2). Two dogs (1 in each group) developed VCOG grade 4 neutropenia but no clinical signs of disease; therefore, the subsequent dose of lomustine administered to each of those dogs was reduced by 15%.
One dog in the lomustine-only group developed signs of acute hepatopathy following administration of the second dose of lomustine. On day 41 (prior to administration of the third dose of lomustine), that dog was icteric and had high serum ALT (6,383 U/L; reference range, 0 to 60 U/L), ALP (3,556 U/L; reference range, 0 to 100 U/L), and GGT (179 U/L; reference range, 0 to 8 U/L) activities and high preprandial (1,040 μmol/L; reference range, < 15.0 μmol/L) and postprandial (930 μmol/L; reference range, < 30.0 μmol/L) BA concentrations. On day 20, the dog had not been icteric, and all biochemical variables had been within the respective reference ranges. Ultrasonographic evaluation on day 41 revealed a subjectively small liver that was subjectively hyperechoic but hypoechoic relative to the spleen. The portal vessels appeared thickened and irregular. Ascites was not observed. The dog was withdrawn from the study on day 41, treated with a nutritional supplementm intended to promote liver health, and monitored. Thus, biochemical and histologic data from only 9 dogs were included in analyses after day 41. The remaining 9 dogs received all 5 planned doses of lomustine and completed the study. The mean cumulative dose of lomustine administered to those 9 dogs was 370 mg/m2.
Six of the 9 dogs that completed the lomustine protocol developed clinical signs of liver failure beginning approximately 11 weeks after administration of the final dose of lomustine (ie, beginning approx 160 days after initiation of the lomustine protocol). Seven of the 10 dogs (4 dogs in the lomustine-prednisone group and 3 dogs in the lomustine-only group) initially enrolled in the study were euthanized by means of a barbiturate overdose because of clinical liver failure and underwent necropsy. During necropsy, the most common gross findings were microhepatia (n = 7), diffuse hepatopathy with or without ascites (5), and acquired extrahepatic portosystemic shunts (3). One dog had ammonium urate urolithiasis and another had pleural effusion.
Ultrasonographic findings
Nonspecific ultrasonographic changes to the liver (subjective hepatomegaly and hyperechogenicity) were noted during the experimental period when lomustine was administered. No major adverse events associated with the acquisition of liver biopsy specimens were observed. The most notable ultrasonographic changes were observed following completion of the lomustine protocol when the dogs developed clinical signs of liver failure. Commonly observed ultrasonographic changes in dogs with clinical signs of liver failure were a homogenous appearance (n = 9), subjective microhepatia (6), irregular liver margins (6), variable anechoic fluid free within the peritoneum (5), and subjective hepatomegaly (2). The liver appeared hypoechoic in 2 dogs and hyperechoic in 3 dogs. One dog had ultrasonographic evidence of pleural effusion.
Biochemical changes
Serum ALT and ALP activities and total bilirubin and BA concentrations over time were plotted (Figure 1) For all dogs, serum ALT activity increased significantly (P < 0.001) over time and generally peaked after the third dose of lomustine was administered on day 41. Eight of 10 dogs had a VCOG grade 4 increase in ALT activity (> 600 U/L; > 10 times the upper limit of the reference range) at least once during the experimental period. Serum ALP activity also increased significantly (P < 0.001) over time and generally peaked after the third dose of lomustine was administered. Four of the 5 dogs in the lomustine-only group had a VCOG grade 4 increase in ALP activity (> 2,000 U/L; > 20 times the upper limit of the reference range). Only 1 dog (lomustine-only group) had an elevated serum total bilirubin concentration (0.7 mg/dL; reference range, 0.0 to 0.4 mg/dL), which was detected after administration of the fourth dose of lomustine. However, the median serum bilirubin concentration did not differ significantly from baseline at any time for either treatment group. Serum BA concentrations increased significantly (P < 0.001) over time. All dogs in the lomustine-prednisone group and 3 of 5 dogs in the lomustine-only group developed elevated postprandial BA concentrations (range, 67 to 249 μmol/L). Although the peak BA concentration occurred after administration of the fourth dose of lomustine in 1 dog, the peak BA concentration occurred after cessation of lomustine administration in all other dogs. Most dogs did not have persistently elevated BA concentrations until they developed clinical liver failure. Serum fasting ammonia concentrations were erratic, and the median fasting ammonia concentration was not significantly increased from baseline at any time for either group.
Plots of serum ALT activity (A), ALP activity (B), total bilirubin concentration (C), BA concentration (D), liver histologic score (E), and hepatic reduced glutathione concentration (F) over time for adult sexually intact female hounds that received lomustine (approx 75 mg/m2, PO, q 2l d for 5 doses) only (white circles; n = 5) or concurrently with prednisone (approx 1.5 mg/kg, PO, q 24 h for 12 weeks; black circles; 5). All measurements reported for day 0 were acquired immediately before administration of the first dose of lomustine. Lines represent the medians for dogs treated with lomustine only (dashed line) or with lomustine and prednisone (solid line). One dog in the lomustine-only group was removed from the scatterplot owing to early clinical signs of liver failure. Variations in data points acquired for histology (E) and glutathione (F) are dependent on individual dogs' peak serum ALT activity, which triggered the acquisition of liver biopsy specimens.
Citation: American Journal of Veterinary Research 81, 10; 10.2460/ajvr.81.10.810
Histologic scores
Histologic examination of baseline liver biopsy specimens revealed mild deposition of lipofuscin and scattered iron in Kupffer cells and portal macrophages but no evidence of fibrosis, inflammation, or structural changes. Following completion of the lomustine protocol, the most common histologic changes observed in liver biopsy specimens were loss of portal vein profiles, mild to moderate atypia of the biliary epithelium, and mild to moderate fibrosis. Histologic examination of liver tissue specimens obtained from the 7 dogs that were euthanized because of clinical signs of liver failure revealed histiocytic inflammation within the portal tract and adjacent parenchyma. The portal veins were diminished and persistent biliary hyperplasia was evident. Representative photomicrographs of sections of liver tissue for a dog in the lomustine-only group that was euthanized because of clinical liver failure were provided (Figure 2) to demonstrate the progression of histologic changes associated with lomustine administration.
Photomicrographs of sections of liver tissue that depict the histologic progression of lomustine-induced hepatopathy in a dog that received lomustine only as described in Figure 1. A—Photomicrograph of a section of a liver biopsy specimen obtained before initiation of lomustine administration (baseline; day 0). Notice that portal tracts are within reference limits and hepatocytes are uniform in size and mildly vacuolated. B—Photomicrograph of a section of a liver biopsy specimen obtained on day 104 (3 weeks after completion of the lomustine protocol). Notice that there is prominent fibrosis circumferentially around the bile ducts. Mild biliary hyperplasia and accumulation of pigmented macrophages are also evident. C—Photomicrograph of a section of liver tissue obtained during necropsy performed immediately after euthanasia owing to clinical signs of liver failure 12 weeks after completion of the lomustine protocol. Notice that portal vein profiles are diminished, prominent pigmented macrophages are present in the portal tract and adjacent parenchyma, and persistent biliary hyperplasia is evident. H&E stain; bar = 300 μm.
Citation: American Journal of Veterinary Research 81, 10; 10.2460/ajvr.81.10.810
For both treatment groups, the median histologic score was increased from baseline at 3 weeks after administration of the final dose of lomustine (ie, day 104; P < 0.001) and at the time the final liver biopsy specimens were obtained (ie, final histologic score [time relative to completion of lomustine protocol was variable among dogs and was dependent on date of euthanasia]; P < 0.001; Figure 1). The median histologic score did not differ significantly between the lomustine-only group and lomustine-prednisone group at any time. The histologic score was positively correlated with peak ALT (r = 0.79; P = 0.012) and ALP (r = 0.90; P < 0.001) activities and peak BA concentration (r = 0.68; P = 0.045).
Hepatic glutathione concentration
The hepatic reduced glutathione concentration over time was plotted for all dogs (Figure 1). The median hepatic reduced glutathione concentration decreased significantly (P = 0.01) over time for both groups. The nadir of hepatic glutathione concentrations occurred after administration of the last dose of lomustine, and both nadirs were significantly lower (P < 0.001) than baseline. There was no correlation between the magnitude of glutathione concentration change and final histologic score (r = 0.03; P = 0.93).
Prednisone effect
When time was controlled (held constant) during the analyses, the median serum ALT (P < 0.001), ALP (P = 0.002), and GGT (P < 0.001) activities and albumin (P = 0.002), cholesterol (P < 0.001), and BA (P = 0.005) concentrations for the lomustine-prednisone group were significantly increased, compared with those for the lomustine-only group. However, the median serum total bilirubin (P = 0.052) and fasting ammonia (P = 0.45) concentrations, hepatic reduced glutathione concentration (P = 0.25), and histologic score (P = 0.47) did not differ significantly between the 2 groups. This suggested that for dogs treated with lomustine, prednisone had a significant effect on some biochemical variables, but its effect on other indices of liver function and structure was less certain.
Discussion
Results of the present study indicated that administration of lomustine to healthy dogs at a dosage of approximately 75 mg/m2, PO, every 21 days for 5 doses commonly caused hepatopathy characterized by significant increases in liver enzyme activities, other biochemical indices of liver function, and gross and histologic changes in liver structure. Most dogs of the present study developed an acute elevation in liver enzyme activity following administration of the second or third dose of lomustine but did not manifest clinical signs of liver disease. Liver enzyme activities then returned to normal levels. However, histologic changes in the liver tissue continued to progress for weeks to months after completion of the lomustine protocol. Abnormalities in indices of liver function (serum BA concentration) became evident at approximately the same time that clinical signs of liver disease, such as ascites and icterus, began to manifest. Concurrent administration of prednisone did not appear to be protective against lomustine-induced hepatopathy for the dogs of the present study and in fact appeared to exacerbate increases in liver enzyme activities and other indices of liver function. Seven of the 10 dogs of the present study were euthanized because of clinical signs of liver failure beginning 11 weeks after discontinuation of lomustine administration. Necropsies were performed on all 7 dogs, and the most common abnormalities observed included a decrease in portal vein profiles, mild histiocytic inflammation, and portal fibrosis with multiple acquired portosystemic shunts. Abnormalities were observed in dogs that did and did not receive prednisone in conjunction with lomustine.
In studies16,17 involving dogs with neoplastic disease, ≤ 6% of those treated with lomustine developed clinical signs of liver failure following completion of the treatment protocol. The 10 dogs of the present study were healthy at study enrollment, yet 7 of them subsequently developed clinical signs of liver failure and were euthanized several months after completion of the lomustine protocol. The high morbidity and mortality rates observed in this study were likely the result of continued administration of lomustine despite increases in liver enzyme activity as well as exclusion of concurrent hepatoprotectant medications. Serum activities of aminotransferases have been used as biomarkers for hepatic damage in preclinical studies for decades. Alanine aminotransferase is considered the gold standard for monitoring hepatocyte damage; however, its specificity in biomarker validation is questioned owing to the lack of correlation between high ALT activity and histopathologic changes.27 Veterinarians routinely monitor liver enzyme activities in patients, particularly those receiving drugs that may impair liver function, and commonly discontinue drug administration when liver enzyme activities become abnormally increased. Discontinuing administration of hepatotoxic drugs as soon as liver enzyme activities become abnormally increased generally minimizes liver damage and allows the organ to heal. The dogs of the present study continued to receive lomustine despite developing VCOG grade 3 or greater increases in ALT activity as long as they did not have clinical signs of liver disease (eg, icterus or ascites). This likely exhausted the liver's reparative mechanisms (eg, the antioxidant response to oxidative stress) and contributed to the advanced hepatopathy observed in most dogs.
Dogs that relapse with lymphoma or macroscopic mast cell tumors generally receive a median of 1 dose of lomustine before resistance to chemotherapy develops.1,2 Hence, some lomustine-treated dogs may not receive a sufficient cumulative dose of the drug to cause clinical liver disease, or they die or are euthanized because of cancer progression before lomustine-induced hepatopathy becomes clinically evident. Consideration of the hepatotoxic effects of lomustine is most important for patients for which extended survival times are expected, such as dogs with lymphoma and mast cell tumors.17,28
Most dogs of the present study had an acute increase in ALT activity following administration of the third dose of lomustine. That finding was consistent with results of other studies9,13 in which dogs received a median of 3 doses of lomustine before significant increases in serum ALT activity were observed. In the present study, 2 of the 3 dogs that did not develop sufficient liver disease to warrant euthanasia did not have a VCOG grade 3 or greater increase in ALT activity during the lomustine protocol. In human medicine, although serum ALT activity is not significantly correlated with the extent of liver damage, increases in ALT activity greater than twice the upper limit of the reference range generally cause concern about irreversible hepatic changes.18
In the present study, the dogs administered prednisone concurrently with lomustine had significantly greater median peak serum ALT, ALP, and GGT activities and BA concentrations than did the dogs administered lomustine only. Glucocorticoids can induce an increase in ALP and GGT activities,29 which may have contributed to the significant differences in the median peak activities observed for those 2 enzymes between the lomustine-only group and lomustine-prednisone group. Additionally, steroid hepatopathy is characterized by abnormally increased serum GGT activity and BA concentrations in dogs.29,30 In clinical practice, most dogs treated with lomustine receive corticosteroids concurrently owing to their anti-inflammatory and antineoplastic properties. Therefore, monitoring liver enzyme activities and other biochemical indices of liver function before and during treatment is critical so that the drug regimen can be adjusted as necessary in a timely manner. The median serum total bilirubin and fasting ammonia concentrations, hepatic reduced glutathione concentration, and histologic score did not differ significantly between the lomustine-only and lomustine-prednisone groups of the present study. However, the study population was small and there may have been insufficient power to detect such changes. Further research involving a larger population of dogs is necessary to better elucidate the effects of concurrent administration of prednisone and lomustine.
The high morbidity and mortality rates for the dogs of the present study were unexpected given the low incidence of lomustine-induced liver failure in clinical patients.16,17 Continued administration of lomustine and the lack of treatment with hepatoprotective agents likely contributed to the high morbidity rate. It is also possible that the breed of the study dogs might have played a role in the development of liver disease. The biotransformation of lomustine is mediated by cytochrome P450s.31 The purpose-bred hounds used in this study might have been predisposed to lomustine-induced hepatopathy owing to unidentified pharmacogenetic factors.
The mechanism by which lomustine causes hepatic injury is not fully understood. Dogs and humans appear to metabolize lomustine in a similar manner, and the main metabolites of lomustine are trans-4-hydroxylomustine and cis-4-hydroxylomustine.32 Interestingly, hepatopathy secondary to administration of lomustine or other nitrosoureas is rare in human patients.33 Although the alkylating ability of lomustine is responsible for its cytotoxic effects on target (neoplastic) tissues, the hepatotoxic effects of lomustine are believed to be the result of carbamoylation. Presumably, lomustine leads to the formation of the reactive metabolite diethyl ethylphosphonate isocyanate, which leads to carbamoylation of proteins and makes them nonfunctional.34 Isocyanates decreased hepatic reduced glutathione concentration in rats treated with fotemustine, a nitrosourea similar to lomustine.35 For the dogs of the present study, hepatic reduced glutathione concentration decreased significantly following initiation of the lomustine protocol. Glutathione is an antioxidant present in hepatocytes that protects the liver from damage caused by oxidative stress. It is speculated that oxidative stress is involved in the hepatotoxic mechanism of lomustine, but it has yet to be determined whether the decrease in hepatic reduced glutathione concentration is a cause or effect of lomustine-induced hepatopathy.
Glutathione precursors and antioxidants minimize the extent of glutathione depletion in rat hepatocytes exposed to fotemustine in vitro.36 In dogs with neoplasia that are treated with lomustine, concurrent administration of a nutritional supplement containing S-adenosyl methionine and silybin results in less severe increases in liver enzyme activities, mitigates signs of hepatopathy, and increases the likelihood of the patient completing the lomustine treatment regimen relative to similar dogs that do not receive the nutritional supplement.13 Although administration of supplemental glutathione may decrease the risk of hepatopathy, it may also alter the efficacy of alkylating agents such as nitrosoureas.34 Tumor resistance to alkylating agents is most often the result of altered DNA repair pathways; however, it has also been associated with intracellular inactivation of alkylating agents.7 Abnormally increased intracellular glutathione concentration and glutathione S-transferase activities have been detected in cells resistant to alkylating agents.37 Further evaluation of lomustine metabolite concentrations and the mechanisms by which hepatoprotective agents mitigate the hepatotoxic effects of lomustine will facilitate the development of guidelines for administration of the drug to clinical patients.
The present study had several limitations. The study population was small and homogeneous, consisting of only 10 sexually intact female purpose-bred hounds. Thus, we likely had inadequate power to fully evaluate some outcomes. Additionally, the high incidence of liver failure subsequent to lomustine administration observed in the dogs of this study may not be applicable to other breeds of dogs owing to breed-specific pharmacogenetic factors. Ideally, liver biopsy specimens would have been obtained from each dog at each sample acquisition time so that histologic changes could have been compared between dogs with and without severe increases in serum ALT activity. However, liver biopsy is an invasive procedure, with reported minor and major complication rates of 5.8% and 1.2%, respectively.38 Therefore, we opted to perform liver biopsies only for dogs with suspected hepatopathy at each sample acquisition time. Finally, ultrasonographic examination of the entire abdomen was not performed for any of the study subjects at any time. It is possible that acquired portosystemic shunts that developed near the kidneys or midabdominal region remained undetected.
Findings of the present study suggested that in dogs, lomustine-induced hepatopathy was characterized by an acute increase in liver enzyme activities followed by a spontaneous return to baseline activities despite the absence of clinical signs of liver disease. Abnormal liver function test results were not observed until weeks to months after lomustine administration was discontinued and coincided with the manifestation of clinical signs of liver disease. Those signs generally progressed and led to clinical deterioration sufficient to warrant euthanasia. On the basis of these findings, we recommend that dogs being treated with lomustine undergo frequent monitoring of liver enzyme activities and discontinuation of lomustine administration when an acute increase in ALT or ALP activity is detected. Ultrasonographic examination of the liver yielded nonspecific results and was not useful for detection of disease until dogs were in advanced stages of liver failure. The decrease in hepatic reduced glutathione concentration during lomustine administration for the dogs of this study suggested that oxidative stress is involved in lomustine-induced hepatic injury and validated observations that concurrent administration of antioxidants, such as S-adenosyl methionine and silybin, might be beneficial for lomustine-treated dogs. Further research is necessary to elucidate the efficacy of hepatoprotective agents and glutathione supplementation for mitigation or prevention of the hepatotoxic effects of lomustine in dogs.
Acknowledgments
The authors thank Kristen Ballard for technical assistance; Drs. Rhett Stout, Carmen Arsuaga, and Raphael Malbrue for assistance with animal care and procedures; and Dr. John Cullen for evaluation of histologic specimens.
ABBREVIATIONS
| ALP | Alkaline phosphatase |
| ALT | Alanine aminotransferase |
| BA | Bile acid |
| GGT | γ-Glutamyltransferase |
| VCOG | Veterinary Cooperative Oncology Group |
Footnotes
Gleostine, NextSource Biotechnology, Miami, Fla.
West-Ward Pharmaceuticals, Eatontown, NJ.
Simplicef, Zoetis, Parsippany, NJ.
Triflexis, Elanco, Greenfield, Ind.
Noblus, Hitachi Aloka Medical America Inc, Wallingford, Conn.
Torbugesic, Zoetis, Parsippany, NJ.
Dexdormitor, Zoetis, Parsippany, NJ.
Cerenia, Zoetis, Parsippany, NJ.
Magnum Biopsy Instrument, Bard Peripheral Vascular Inc, Tempe, Ariz.
Antisedan, Zoetis, Parsippany, NJ.
Virtus Pharmaceuticals, Newtown, Pa.
SAS, version 9.4, SAS Institute Inc, Cary, NC.
Marin Plus, Nutramax Laboratories, Lancaster, SC.
References
1. Moore AS, London CA, Wood CA, et al. Lomustine (CCNU) for the treatment of resistant lymphoma in dogs. J Vet Intern Med 1999;13:395–398.
2. Rassnick KM, Moore AS, Williams LE, et al. Treatment of canine mast cell tumors with CCNU (lomustine). J Vet Intern Med 1999;13:601–605.
3. Skorupski KA, Clifford CA, Paoloni MC, et al. CCNU for the treatment of dogs with histiocytic sarcoma. J Vet Intern Med 2007;21:121–126.
4. Van Meervenne S, Verhoeven PS, de Vos J, et al. Comparison between symptomatic treatment and lomustine supplementation in 71 dogs with intracranial, space-occupying lesions. Vet Comp Oncol 2014;12:67–77.
5. Williams LE, Rassnick KM, Power HT, et al. CCNU in the treatment of canine epitheliotropic lymphoma. J Vet Intern Med 2006;20:136–143.
6. Schabel FM Jr, Johnston TP, McCaleb GS, et al. Experimental evaluation of potential anticancer agents VIII. Effects of certain nitrosoureas on intracerebral L1210 leukemia. Cancer Res 1963;23:725–733.
7. Leukemias and plasma cell tumors. In: DeVita VT Jr, Lawrence TS, Rosenberg SA, eds. DeVita, Hellman, and Rosenberg's cancer: principles and practice of oncology. 10th ed. Philadelphia: Wolters Kluwer Health, 2015;1602–1719.
8. Gustafson DL, Bailey DB. Cancer chemotherapy. In: Withrow SJ, Vail DM, Page RL, eds. Withrow and MacEwen's small animal clinical oncology. 6th ed. St Louis: Elsevier Saunders, 2019;182–208.
9. Hosoya K, Lord LK, Lara-Garcia A, et al. Prevalence of elevated alanine transaminase activity in dogs treated with CCNU (lomustine). Vet Comp Oncol 2009;7:244–255.
10. Carter SK, Newman JW. Nitrosoureas: 1,3-bis(2-chloroethyl)-1-nitrosourea (NSC-409962;BCNU) and 1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea (NSC-79037; CCNU)—clinical brochure. Cancer Chemother Rep 3 1968;1:115–151.
11. Henry MC, Davis RD, Schein PS. Hepatotoxicity of 1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea. (CCNU) in dogs: the use of serial percutaneous liver biopsies. Toxicol Appl Pharmacol 1973;25:410–417.
12. Heading KL, Brockley LK, Bennett PF. CCNU (lomustine) toxicity in dogs: a retrospective study (2002–07). Aust Vet J 2011;89:109–116.
13. Skorupski KA, Hammond GM, Irish AM, et al. Prospective randomized clinical trial assessing the efficacy of Denamarin for prevention of CCNU-induced hepatopathy in tumor-bearing dogs. J Vet Intern Med 2011;25:838–845.
14. Center SA, Slater MR, Manwarren T, et al. Diagnostic efficacy of serum alkaline phosphatase and γ-glutamyltransferase in dogs with histologically confirmed hepatobiliary disease: 270 cases (1980–1990). J Am Vet Med Assoc 1992;201:1258–1264.
15. Vail DM, von Euler H, Rusk AW, et al. A randomized trial investigating the efficacy and safety of water soluble micellar paclitaxel (Paccal Vet) for treatment of nonresectable grade 2 or 3 mast cell tumors in dogs. J Vet Intern Med 2012;26:598–607.
16. Kristal O, Rassnick KM, Gliatto JM, et al. Hepatotoxicity associated with CCNU (lomustine) chemotherapy in dogs. J Vet Intern Med 2004;18:75–80.
17. Hosoya K, Kisseberth WC, Alvarez FJ, et al. Adjuvant CCNU (lomustine) and prednisone chemotherapy for dogs with incompletely excised grade 2 mast cell tumors. J Am Anim Hosp Assoc 2009;45:14–18.
18. Hall AP, Elcombe CR, Foster JR, et al. Liver hypertrophy: a review of adaptive (adverse and non-adverse) changes—conclusions from the 3rd International ESTP Expert Workshop. Toxicol Pathol 2012;40:971–994.
19. Henry MC, Marlow M. Serum enzymes in hepatotoxicity induced by chloroethylcyclohexyl nitrosourea and arabinofuranosyl-6-mercaptopurine. Toxicol Appl Pharmacol 1973;24:250–255.
20. Borel AG, Abbott FS. Identification of carbamoylated thiol conjugates as metabolites of the antineoplastic 1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea, in rats and humans. Drug Metab Dispos 1993;21:889–901.
21. Center SA, Warner KL, Erb HN. Liver glutathione concentrations in dogs and cats with naturally occurring liver disease. Am J Vet Res 2002;63:1187–1197.
22. Mitchell JR, Jollow DJ, Potter WZ, et al. Acetaminophen-induced hepatic necrosis. IV. Protective role of glutathione. J Pharmacol Exp Ther 1973;187:211–217.
23. Chapman SE, Hostutler RA. A laboratory diagnostic approach to hepatobiliary disease in small animals. Vet Clin North Am Small Anim Pract 2013;43:1209–1225.
24. VCOG. Veterinary Cooperative Oncology Group—common terminology criteria for adverse events (VCOG-CTCAE) following chemotherapy or biological antineoplastic therapy in dogs and cats v1.1. Vet Comp Oncol 2016;14:417–446.
25. World Small Animal Veterinary Association Liver Standardization Group. WSAVA standards for clinical and histological diagnosis of canine and feline liver diseases. Philadelphia: Saunders Elsevier, 2006.
26. Barry-Heffernan C, Ekena J, Dowling S, et al. Biomarkers of oxidative stress as an assessment of the redox status of the liver in dogs. J Vet Intern Med 2019;33:611–617.
27. Anadón A, Martínez-Larrañaga MR, Ares I, et al. Biomarkers of drug toxicity and safety evaluation. In: Gupta RC, ed. Biomarkers in toxicology. 2nd ed. San Diego: Academic Press, 2019;655–691.
28. Brown PM, Tzannes S, Nguyen S, et al. LOPP chemotherapy as a first-line treatment for dogs with T-cell lymphoma. Vet Comp Oncol 2018;16:108–113.
29. Tennant BC, Center SA. Hepatic function. In: Kaneko JJ, Harvey JW, Bruss ML, eds. Clinical biochemistry of domestic animals. 6th ed. San Diego: Academic Press, 2008;379–412.
30. Center SA, Baldwin BH, Erb HN, et al. Bile acid concentrations in the diagnosis of hepatobiliary disease in the dog. J Am Vet Med Assoc 1985;187:935–940.
31. Kramer RA. Cytochrome P-450-dependent formation of alkylating metabolites of the 2-chloroethylnitrosoureas MeCCNU and CCNU. Biochem Pharmacol 1989;38:3185–3192.
32. Chakkath T, Lavergne SN, Fan TM, et al. Preliminary metabolism of lomustine in dogs and comparative cytotoxicity of lomustine and its major metabolites in canine cells. Vet Sci 2014;1:159–173.
33. Musser ML, Quinn HT, Chretin JD. Low apparent risk of CCNU (lomustine)-associated clinical hepatotoxicity in cats. J Feline Med Surg 2012;14:871–875.
34. Gerson SL, Bulgar AD, Weeks LD, et al. Alkylating agents. Part A: classical alkylating agents. In: Chabner BA, Longo DL, eds. Cancer chemotherapy and biotherapy: principles and practice. 5th ed. Philadelphia: Wolters Kluwer-Lippincott Williams and Wilkins Health, 2011;267–292.
35. Brakenhoff JP, Commandeur JN, Wormhoudt LW, et al. Molecular mechanisms of toxic effects of fotemustine in rat hepatocytes and subcellular rat liver fractions. Carcinogenesis 1996;17:715–724.
36. Vermeulen NP, Commandeur JN, Groot EJ, et al. Toxicity of fotemustine in rat hepatocytes and mechanism-based protection against it. Chem Biol Interact 1998;110:139–158.
37. Robson CN, Lewis AD, Wolf CR, et al. Reduced levels of drug-induced DNA cross-linking in nitrogen mustard-resistant Chinese hamster ovary cells expressing elevated glutathione S-transferase activity. Cancer Res 1987;47:6022–6027.
38. Léveillé R, Partington BP, Biller DS, et al. Complications after ultrasound-guided biopsy of abdominal structures in dogs and cats: 246 cases (1984–1991). J Am Vet Med Assoc 1993;203:413–415.
