Glatiramer acetate–induced acute hepatotoxicity in an adolescent with MS

Abstract

Glatiramer acetate (GA), a synthetic copolymer, is a frequently used first-line treatment for relapsing-remitting multiple sclerosis (RRMS). Probable autoimmune hepatotoxicity during GA treatment has been reported,^1–4 but GA hepatotoxicity in the absence of positive autoimmune markers has not previously been described. Here, we report GA-induced hepatotoxicity in a pediatric patient with multiple sclerosis (MS).

Glatiramer acetate (GA), a synthetic copolymer, is a frequently used first-line treatment for relapsing-remitting multiple sclerosis (RRMS). Probable autoimmune hepatotoxicity during GA treatment has been reported,^1–4 but GA hepatotoxicity in the absence of positive autoimmune markers has not previously been described. Here, we report GA-induced hepatotoxicity in a pediatric patient with multiple sclerosis (MS).

Case report.

We diagnosed a 15-year-old girl with RRMS following 2 attacks of brainstem signs and symptoms (October and December 2011) with fluid-attenuated inversion recovery signal abnormalities on MRI characteristic for MS and spinal fluid analysis that revealed the presence of oligoclonal bands. We initiated immunomodulatory therapy with subcutaneous (SC) interferon-β-1a in February 2012 and titrated the dose gradually to 44 µg 3 times a week. Our patient had no other medical conditions and was taking no other prescribed or herbal medications. There was a remote history of transient unexplained jaundice in her father. Aspartate aminotransferase (AST = 370, normal range 0–36) and alanine aminotransferase (ALT 145, normal range 0–40) were elevated in May 2012 and we discontinued interferon therapy. Liver transaminases normalized by August 2012 (AST 20 and ALT 35) and we initiated GA at a dose of 20 mg SC once daily (figure 1A).

Figure 1. Liver transaminases over time and liver pathology in a 15-year-old patient with multiple sclerosis.

(A) Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels in our patient. Arrows indicate time of initiation of interferon-β-1a (IFN) and glatiramer acetate (GA). Pathology shows (B) isolated single hepatocyte cell death at earliest phase (blue arrow) and (C) late phase hepatocyte death triggering an inflammatory response. (D) Immunostain for CD8+ cytotoxic T cells shows that these are the principal cells of the inflammatory response.

In October 2012, routine blood tests showed elevated ALT (253) and serum ferritin (119, normal range 5.5–67.4). Repeat testing 1 week later revealed further elevations in ALT, AST, and serum ferritin (figure 1A). Physical examination revealed no jaundice, hepatomegaly, or other stigmata of liver disease. Our patient's only symptom was profound fatigue. Hepatic ultrasound showed nonspecific echogenicities suggestive of fatty infiltration without hepatomegaly. Liver function tests including serum international normalized ratio, partial thromboplastin time, bilirubin, and albumin were normal. Total white blood cell count, hemoglobin, and platelet levels were normal. Inflammatory markers including erythrocyte sedimentation rate and C-reactive protein were normal as was serum iron. Serologic tests for recent Epstein-Barr virus infection and for hepatitis A, B, and C were negative. Tests for autoimmune hepatitis including antinuclear antibody (ANA), anti–smooth muscle antibody, antiliver/kidney antibody, and immunoglobulin G (IgG) levels were all normal.

Liver biopsy revealed a mild parenchymal, mainly CD8+, lymphocytic inflammatory infiltrate within the portal tracts with mild portal fibrosis, no plasma cells, and no signs of chronic liver disease (figure 1, B–D). There was hepatocyte necrosis. Cytomegalovirus and adenovirus immunostains were negative. Electron microscopy revealed microvesicular steatosis and increased intracellular lipofuscin deposition with mild smooth endoplasmic reticulum and mitochondrial proliferation (figure e-1 on the Neurology® Web site at www.neurology.org). Some mitochondria showed nonspecific reactive changes, were enlarged, and contained crystalline inclusions. Necrotic cells did not show activation or cytoplasmic expression of the apoptotic pathway protein Caspase-3 by immunohistochemistry, suggesting a nonapoptotic cellular death mechanism. The liver biopsy was consistent with drug-induced hepatocellular injury with mitochondrial damage.

We discontinued GA therapy on November 7, 2012. Liver transaminases and serum ferritin normalized by December 31, 2012.

Discussion.

GA has been linked to the development or exacerbation of autoimmune conditions including myasthenia gravis,⁵ autoimmune hyperthyroidism,⁶ and autoimmune hepatitis.^1,2 There was no evidence for autoimmune injury in our patient as ANA, anti–smooth muscle, and antiliver/kidney antibodies were absent, serum IgG level was normal, and there was no plasma cell infiltrate on liver histology. Instead the microvesicular steatosis, hepatocyte necrosis, and structural mitochondrial changes suggest cell death mediated by mitochondrial metabolic stress activation, a novel pattern of drug-induced injury associated with GA. This pattern is well-recognized with other drugs and the presumed mechanism is inhibition of mitochondrial function with accumulation of reactive oxygen species, lipid peroxidation with fat accumulation, and cell death.⁷

Although 2 prior case reports described possible GA hepatotoxicity, the patient in the first report had a concomitant elevated ANA titer (1:320),⁴ and the patient in the second report displayed anti–smooth muscle antibodies acutely (1:320),³ suggesting underlying autoimmune mechanisms as opposed to direct toxic injury.

Importantly, our patient previously displayed elevated liver transaminases with interferon-β-1a treatment, resulting in interferon discontinuation and a switch to GA. While interferon-induced hepatotoxicity has successfully been managed by changing to GA,² this case suggests that, for some patients, transaminitis with interferon treatment may signal a predisposition to GA-induced hepatotoxicity.

This report should alert practitioners to the possibility of acute hepatitis induced by GA. While regulatory agencies currently do not require blood tests for safety monitoring during GA treatment, we suggest that individuals who commence GA following transaminitis with interferons should be regularly monitored for signs of hepatotoxicity. Hepatotoxicity occurred within 3 months of therapy initiation, suggesting early monitoring of transaminases is necessary. Increased vigilance may be warranted in children.