Imatinib and Panax Ginseng : A Potential Interaction Resulting in Liver Toxicity

Case Report

A 26-year-old man diagnosed with CML had been receiving 400 mg of imatinib daily since January 2002. He had been seen in the clinic every 3-6 months since his initial visit in 2004, and his response to imatinib was monitored via polymerase chain reaction for the bcr-abl gene. He has maintained a major molecular response since his initial presentation.

In July 2009, 7 years after starting imatinib, he presented with a 2-day history of right upper quadrant pain. The pain was described as episodes of sharp, stabbing sensations that were exacerbated with movement. He had no history of heavy lifting, alcohol intake, or acetaminophen ingestion. Liver enzyme tests at that time showed alanine aminotransferase (ALT) 1069 U/L, aspartate aminotransferase (AST) 481 U/L, alkaline phosphatase 124 IU/L, total bilirubin 1.4 mg/dL, albumin 4.0 g/dL, and international normalized ratio 1.08. After a detailed history, it was found that the only lifestyle modification prior to the identification of abnormal results of liver function tests was daily ingestion of Panax ginseng via Full Throttle energy drinks for the past 3 months. Results of abdominal and pelvic computed tomography (CT) scans were unremarkable. In light of no obvious abnormalities seen on CT, we discontinued imatinib and counseled the patient to avoid all ginseng-containing products while we evaluated for other causes of hepatitis. Results of tests for hepatitis B DNA, hepatitis B core immunoglobulin (Ig) M antibody, hepatitis B surface antigen, hepatitis A antibody, hepatitis C RNA, Epstein-Barr virus capsid IgM and IgG antibody, Epstein-Barr virus nuclear antigen antibody, cytomegalovirus DNA, and herpes simplex virus IgM antibody were all negative or nonreactive, ruling out a viral etiology. All findings from an autoimmune workup including anti-nuclear, anti-smooth muscle, anti-parietal cell, and anti-mitochondrial antibody studies were negative. Investigation of potential enzymatic deficiencies also ruled out alpha-1-antitrypsin deficiency and Wilson's disease, and Lyme polyvalent enzyme-linked immunosorbent assay was negative.

An ultrasound-guided liver biopsy subsequently revealed spotty hepatocellular necrosis, apoptosis, inflammation (predominantly centrilobular), Kupffer cell hyperplasia, mild steatosis, and mild portal inflammation with no bridging necrosis. The portal tracts showed a mild infiltration of lymphocytes, macrophages, and plasma cells, as well as rare eosinophils. Staining with trichrome showed mild pericellular fibrosis but no portal fibrosis or cirrhosis. Results of acid-fast bacteria and Gomori-Grocott methenamine silver stains were negative. In sum, the patient's liver histopathology was consistent with resolving or late-stage acute lobular hepatitis, most likely from a drug-induced etiology.

The patient was treated with prednisone 60 mg daily for 1 week, followed by 40 mg daily for 3 days and then a taper of 10 mg every 3 days. After 19 days of prednisone therapy, his elevated liver enzyme levels resolved and remained normalized for an additional 4 weeks with no corticosteroids. Given his good clinical response to imatinib, imatinib was restarted at the same dose to treat his CML. He has been on imatinib for more than 3 months and has not demonstrated any repeat elevations in his liver enzyme levels.

Discussion

This case illustrates a common dilemma for clinicians: determining whether a toxic reaction is related to a previously tolerated drug, a newly added medication, or some interaction between the two. The patient's history and subsequent tolerance to reintroduction of imatinib strongly argue in favor of hepatotoxicity being caused by ginsengmediated inhibition of imatinib metabolism. Review of the available literature supports this conclusion.

Studies have shown imatinib to be hepatotoxic; however, this usually manifests within the first 1-2 years of therapy.^2–4 Therefore, while the patient's presentation fits known toxic reactions to imatinib, the timeline is well outside that expected from the literature reports. To our knowledge, no case report has shown ginseng alone to cause liver damage, making this possibility equally unlikely. The remaining possibility, an interaction between ginseng and imatinib, is far more plausible. We found no information specifically assessing the risk of imatinib hepatotoxicity with concurrent ginseng ingestion, but pharmacologic studies do provide support for this hypothesis.

Imatinib is metabolized primarily via CYP3A4. In vitro, this has been widely confirmed with the use of CYP3A4 inducers and inhibitors. Phenytoin (inducer) was found to decrease the plasma concentration of imatinib by one fifth when coadministered, ² while ketoconazole (inhibitor) was found to increase the plasma concentration of imatinib by 40% after just a single dose. ² Other cytochrome P450 isoenzymes that play a minor role in the metabolism of imatinib are CYP2D6, 1A2, 2C9, and 2C19.^2,11

Several in vitro and in vivo studies have also evaluated the effects of ginseng on cytochrome P450 isoenzymes. In vitro^2,6–8,11 and in vivo,^2,6,7,9,12 these studies have found significant interactions between ginseng and the isoenzymes. In vivo, Smith et al. found ginseng to increase the plasma concentration of nifedipine, a drug metabolized via CYP3A4, when coadministered. This study concluded that ginseng has an inhibitory effect on CYP3A4. ¹² Conversely, some in vivo studies have not found a significant interaction between ginseng and CYP3A4.^8,10,13–15 Ginseng has not been shown to alter the activity of other cytochrome P450 isoenzymes that play a minor role in the metabolism of imatinib. ¹⁵

Our patient's liver toxicity can be explained given the above information. His daily ingestion of ginseng led to reduced CYP3A4 activity, which subsequently decreased the metabolism of imatinib. Compromised imatinib metabolism increased its plasma concentration, thus potentiating its hepatotoxic effect. Based on the Naranjo probability scale, ¹⁶ this indicates a probable likelihood that imatinib caused this patient's hepatotoxicity. Application of the Horn drug interaction probability scale ¹⁷ also indicates a probable interaction between imatinib and ginseng.

In light of the potential effect on CYP3A4, unlikely timeframe for imatinib hepatotoxicity, and the lack of reports of isolated ginseng-induced hepatitis, we propose that the simultaneous administration of ginseng and imatinib led to the late onset of imatinib-induced hepatotoxicity in this patient.

Although we focus on the above-described interaction to explain this patient's hepatotoxicity, a second potential interaction mechanism involving drug transporters may play a role. P-glycoprotein (P-GP), an intestinal enterocyte drug transporter, is responsible for pumping drugs out of the body, thus serving to decrease drug bioavailability. Imatinib is a P-GP substrate ¹⁸ ; therefore, its plasma concentration may be dependent on P-GP activity. P. ginseng has been shown to inhibit P-GP ¹⁹ ; as a result, the intestinal absorption and bioavailability of imatinib could theoretically increase with concurrent ginseng ingestion. This drug-herb interaction involving drug transporters may act in synergy with the interaction on CYP3A4, further potentiating imatinib's hepatotoxic effect.

CML is a rare disease, with about 4500 patients diagnosed every year. With imatinib, greater than 80% of patients will be alive with no evidence of disease progression after at least 7 years. Thus, the prevalence of this disease will slowly increase and more patients will be on long-term therapy with this drug. This case emphasizes 2 points: the importance of continuous monitoring of liver function tests in patients on imatinib therapy, even after several years, and the importance of counseling patients to avoid ginseng and any other over-the-counter herbal supplements that may interact with imatinib.