Trial and Error Investigational Drug Induced Liver Injury, A Case Series Report

Abstract

This is a case report series of four patients who exhibited signs and symptoms of acute liver dysfunction during participation in a Phase I trial of a novel non-steroidal anti-inflammatory drug (NSAID) designed to inhibit microsomal prostaglandin synthase 1 (MPGES1). Within one month of trial initiation, all four patients presented with epigastric pain, fatigue, nausea, and increasing liver function tests (LFTs). Two out of four patients required hospitalization, underwent liver biopsies, and were treated with N-acetylcysteine. The remaining two patients were managed as outpatients. Liver biopsies were consistent with drug induced liver injury (DILI). Within three months of stopping the investigational drug, symptoms subsided and LFTs normalized in all patients. This case report series signifies the importance of NSAIDs and novel drug agents in general as potentially hepatotoxic substances, the need for a high level of suspicion of DILI when considering possible etiologies of acute liver failure, and the need for prompt withdrawal of the causative agent in management of patients presenting with DILI.

Introduction

Drug induced liver injury (DILI) refers to liver diseases caused by drugs and toxic substances. Not only is DILI the leading cause of acute liver failure (ALF) in the United States, it is the most common reason for drug non-approval and withdrawal by the US Food and Drug Administration (FDA).¹ Older women (> 50–60 years old) appear to be more likely to develop DILI.² Aside from acetaminophen, NSAIDs, along with antibiotics, top the list for causes of DILI.³ Approximately 10% of total drug-induced hepatotoxicity is related to NSAIDs.³ DILI can mimic all forms of liver disease, making diagnosis a challenging one. Furthermore, histology indicates the type and degree of liver injury and not the etiology of it.⁴

A high level of suspicion is necessary to make a diagnosis. The Roussel Uclaf Causality Assessment Method (RUCAM-CIOMS) is a scoring system that can aid diagnosis of DILI (Figure 1). It includes temporal relationship with drug use, exclusion of other causes, and drug's hepatotoxic potential.⁴ Hy's Law is also valuable in identifying patients at heightened risk for complications of DILI. According to Hy's Law simultaneous presence of elevated serum bilirubin (>2x ULN) and elevated aminotransferases (>3x ULN) identifies patients at risk for complications.²

Figure 1.

RUCAM CIOMS Scoring System for Hepatocellular DILI

Once a diagnosis of DILI has been established, prompt removal of the causative agent, symptomatic treatment, and monitoring for ALF are paramount.⁴ DILI is associated with a high level of morbidity and mortality, and thus, prompt withdrawal of the drug is necessary to prevent permanent liver damage, reduce the need for liver transplantation, and reduce mortality.⁴ There are no firmly established treatment protocols for many of the causes of DILI outside of treatment for acetaminophen toxicity. Hence, the Drug Induced Liver Injury Network (DILIN), sponsored by the National Institutes of Health (NIH), is considering conducting treatment trials for severe DILI.⁵

Prostaglandin E2 (PGE2) is a principal mediator of inflammation and the most abundant prostaglandin in the human body, making it a promising target of novel anti-inflammatory therapy.⁶ While relatively safe, NSAIDs are associated with gastric ulcers, pro-thrombotic events, and other adverse effects. Novel NSAIDs targeting various pathways of prostaglandin synthesis are the topic of current research in hopes of avoiding these adverse effects. Microsomal prostaglandin synthase 1 (MPGESI) is one such enzyme that selectively targets the formation of PGE2.⁷ Thus, it was theorized that MPGES1 inhibition would target inflammation while minimizing gastric and pro- thrombotic side effects associated with other NSAIDs. Studies have shown that MPGES1 generated PGE2 plays a key role in inflammation, pain, fever, anorexia, atherosclerosis, stroke, and tumorgenesis.⁸ It has also been shown to be up regulated in synovial tissue.⁶ Inhibition of MPGES1 could theoretically reduce pro-inflammatory PGE2 while sparing other prostanoids, thus minimizing adverse side effects typical of other NSAIDs.^6,7

Case Report

Four female patients over fifty years of age presented with signs and symptoms of acute liver dysfunction during participation in a Phase I trial of a novel NSAID inhibiting MPGES1. At the first sign of liver dysfunction, the trial's local principle investigators referred them to the Queen's Medical Center (QMC) for evaluation and notified the QMC Liver Center. Two patients were admitted, while two were managed as outpatients. The first patient was referred due to progressively worsening liver function tests (LFTs) in addition to experiencing epigastric discomfort, nausea without vomiting, and loss of appetite. The second patient presented with elevated LFTs and nausea without vomiting. These patients were admitted to QMC for further evaluation and treatment. At the time of admission, ALT and total bilirubin were 1005 and 1.5, and 1400 and 0.9, respectively. Other labs including CBC, INR, and renal function, were within normal limits. Both patients exhibited negative Hepatitis A, B, and C serologies. Temporal presentation of illness to drug exposure, exclusion of viral hepatitis, and the hepatotoxic potential of NSAIDs documented in animals supported DILI as a leading diagnosis.

Given the evidence of significant liver dysfunction provided by severely elevated LFTs, a decision was made to proceed with liver biopsy. Liver biopsies of the two patients were consistent with DILI (Figures 2, 3, 4). Figure 2 demonstrates numerous portal and lobular eosinophils. Figure 3 shows micro-granulomas. Lastly, Figure 4 highlights zone 3 cholestasis and marked lobular hepatocyte necrosis consistent with DILI. Furthermore, biopsies were negative for alpha-1 inclusions, fibrosis, and steatosis.

Figure 2.

Intermediate power image showing eosinophils consistent with drug injury

Figure 3.

Intermediate power image showing granuloma consistent with drug injury

Figure 4.

High power image of Zone 3 cholestasis and marked lobular hepatocyte necrosis consistent with drug injury

Although this was not a case of acetaminophen toxicity, N-acetylcysteine was administered given the critical nature of their illness and lack of alternative treatment options. LFTs began to improve immediately following cessation of investigational NSAID and normalized between six and twelve weeks of terminating the investigational drug (Figure 5). They were closely followed at the QMC Liver Center after discharge. Two additional patients were asymptomatic and were referred to the QMC Liver Center after trial investigators noticed increasing LFTs. They also recovered without complication following timely removal of investigational agent.

Figure 5.

ALT values over time from start of treatment to normalization. Patients 1 and 2 stopped experimental treatment, were admitted, and treated with N-acetylcysteine on Day 19. Patients 3 and 4 stopped treatment on Day 12.

Given the temporal relationship of presentation to trial participation and improvement upon cessation, age over 55, exclusion of other causes of potential ALF including viral hepatitis, alcoholic hepatitis, and other infections, and knowledge of the hepatotoxic potential of NSAIDs, a diagnosis of DILI was made. The RUCAM-CIOMS scale was used to determine the likelihood of DILI as a diagnosis in these four women. All four women received a score between 6 and 8, corresponding to a “probable” diagnosis of DILI. Patients were not re-challenged with the drug in question to avoid the possibility of further liver damage. Thus, they each had the potential to earn 3 additional points on RUCAM-CIOMS scale if they had experienced worsening of symptoms and LFTs with re-introduction of drug in question. This would have made a diagnosis of DILI “highly probable.”

Discussion

Though FDA approved NSAIDs are widely used and report of DILI is rare, it should always be in the differential diagnosis in the clinical setting of acute liver injury. Older women may be at increased risk for NSAID induced liver injury as shown previously and again illustrated in our case series.

While targeting MPGES1 sought to reduce gastric and pro- thrombotic side effects associated with other NSAIDs, the high incidence of acute liver injury nationwide prompted the sponsor to immediately terminate the trial. High clinical suspicion played a key role in prompt diagnoses and treatment of the patients who were referred to the QMC Liver Center. Nonetheless, it took up to three months for their LFTs to normalize. DILI can quickly deteriorate into fulminate liver failure necessitating liver transplantation if not diagnosed and treated early in the course of disease. RUCAM-CIOMS scoring system and Hy's Law are helpful when considering a diagnosis of DILI and determining the risk of complications. Clinicians should not hesitate to obtain liver biopsy early in the course of the disease as it will aid in establishing diagnosis and will be more technically difficult to obtain when disease progresses and coagulopathy develops. One must keep in mind that biopsy indicates type and degree of liver injury, not etiology (not a specific drug that caused the injury). Prompt removal of the offending agent and supportive treatment remain the corner stone of DILI treatment. There is some data to indicate that acetylcysteine is helpful in other types of DILI and should be considered given relatively benign nature of this therapy.

Future research is being directed at possibly identifying individuals susceptible to DILI by specific drugs. NSAIDs are metabolized by the liver's cytochrome P450 system. Various genes encode for members of the cytochrome P450 superfamily of enzymes. It has been hypothesized that allelic variation in one or more members of the P450 superfamily could explain why certain people are more susceptible to NSAID induced liver injury.² It is conceivable that the women in our case series were slow metabolizers of the investigational drug, making them susceptible to drug induced liver injury. Pharmacogenetics based investigation of allelic variation of various P450 enzyme polymorphisms coupled with knowledge regarding how various drugs are metabolized could serve as a possible screening method to reduce the number of cases of DILI in the future and represents a topic of current DILI research.⁹

The search remains for the development of a novel and safe target of inflammation. While MPGES1 appears to be a novel target of inflammation, the most recent Phase I trial has failed to prove its safety. As research pursues other enzymatic targets of inflammation, this case demonstrates the need for DILI to remain high on the differential when presented with severe liver dysfunction of unknown etiology.

In conclusion, we have presented a case series of four patients with DILI due to a novel investigational NSAID. Our case report series highlights the potential for NSAID induced hepatic toxicity, the importance of recognition of drug induced liver injury, use of the RUCAM-CIOMS scoring system in aiding diagnosis, and prompt withdrawal of offending agent. Additionally, it calls to attention the need for awareness that early phase clinical trials are being conducted here in Hawai‘i and that patients presenting with acute liver failure may have been exposed to an investigational agent with unknown toxicities.

Conflict of Interest

The authors report no conflict of interest.