Drug-induced Liver Injury Associated with the Glucagon-like Peptide 1 (GLP-1) Agonist Liraglutide

Purpose: Data on the adverse event profiles of the incretin-based hypoglycemic agents are limited. We present the first case of drug-induced liver injury (DILI) secondary to the glucagon-like peptide 1 (GLP-1) agonist liraglutide. Case: A 29-year-old woman with type 2 diabetes and vitiligo presented with 10 days of nausea, emesis, and acute hepatitis. Other than starting liraglutide 1.2 mg daily four months prior, the patient reported no medication changes, supplements, or acetaminophen use. Her social history was unremarkable. Admission labs: AST 991 U/L, ALT 1123 U/L, bilirubin (total/direct) 9.5/6.2 mg/dL, alkaline phosphatase 90 U/L, platelets 224 K/uL, and INR 1.3. Physical exam revealed a nontender abdomen and intact mentation. Liver ultrasound showed increased periportal echogenicity and no biliary dilation. Extensive evaluation for viral (A-E, CMV, EBV, adenovirus), autoimmune, and metabolic causes was unrevealing. Liver biopsy showed acute hepatitis with mixed infiltrate, significant eosinophils, rare plasma cells, and no interface hepatitis (Figure 1a-b). The patient was discharged, but, nine days later, developed worsening symptoms and rising transaminases. She was admitted for repeat biopsy, which showed hepatic necrosis and an extensive eosinophilic infiltrate (Figure 1c-d). Steroids were started for presumed marker-negative druginduced autoimmune hepatitis, and continued given clinical improvement. Six months after stopping liraglutide, AST/ALT are 143/156 U/L, total bilirubin is 3.0 mg/dL, and INR is 1.0.Figure 1: Liver histology, hematoxylin and eosin stain. A. Infiltration of a portal tract with predominantly eosinophils. B. High power view demonstrating ballooning change and eosinophilic infiltrate. C. Submassive hepatic necrosis. D. High power view of adjacent preserved liver with eosinophilic infiltrate.Discussion: This is the first reported case of liraglutide-induced hepatitis. Liraglutide is an incretinbased drug, along with the dipeptidyl peptidase-4 (DPP-4) inhibitors that have been implicated in cases of DILI. Initial GLP-1 studies did not report hepatotoxicity, although they were likely underpowered. Hepatotoxicity may be an incretin analogue class effect with a long latency period. Post-marketing studies are needed to define the hepatotoxic potential of GLP-1 agonists.