Drug-induced liver injury and Hy's Law
Drug-induced liver injury is a safety concern for regulators, and it is hard to detect during drug development. A case that meets the Hy's Law criteria signals a drug's potential to cause serious liver injury, and the liver tests behind those criteria are monitored during trials and plotted to find such cases.
Why the regulator watches the liver
Drug-induced liver injury (DILI) is, in the US Food and Drug Administration's words, the "most frequent single cause of safety-related drug marketing withdrawals for the past 50 years". Hepatotoxicity continues to be a concern for patients. Examples of drugs withdrawn because of DILI include Oraflex, Duract, Marsilid, Serzone, Diflurex and Rezulin. Mild to moderate cases are treatable, but severe ones may result in irreversible liver failure that is fatal or requires liver transplantation.
Routine liver function test results are used to gauge a drug's potential to cause DILI, and the FDA has issued guidance on the premarketing clinical evaluation of potential DILI cases. Regulatory writers and other medical communicators need particular information to interpret the DILI-related safety data obtained before a drug is marketed.
Why it is hard to detect
DILI is challenging to detect because no particular class of drugs or mechanism of action is specifically associated with it, and because many liver diseases are known to raise liver function levels, among them viral hepatitis, acute cholestasis, biliary tract disorders and liver cirrhosis. For these reasons, a firm diagnosis of DILI can be difficult. DILI is also hard to detect because it has no clear identifying signs or symptoms, resembles many known liver diseases, is rarely seen during drug development, is idiosyncratic and unpredictable, and is not clearly dose-related.
Hy's Law
Almost four decades before 2015, Hyman Zimmerman observed that patients with drug-induced hepatocellular injury in combination with jaundice had a poor prognosis, with 10% to 50% mortality from acute liver failure.
What is now informally called Hy's Law is defined by the FDA with three criteria:
- The drug causes hepatocellular injury, shown as increases in alanine aminotransferase (ALT) and/or aspartate aminotransferase (AST) of at least three times the upper limit of normal (ULN).
- Among patients with such increases, some also show impaired liver function, with total bilirubin at least twice the ULN, but no initial signs of cholestasis, that is, no elevated alkaline phosphatase (ALP).
- No other reason can be found to explain the liver injury.
The FDA's guidance on Hy's Law, as it stood in 2015, calls pure hepatocellular injury sufficient to cause hyperbilirubinaemia "an ominous indicator of the potential for a drug to cause serious liver injury". It states that finding one Hy's Law case in the clinical trial database is worrisome, while finding two is considered highly predictive of a drug's potential to cause severe DILI when given to a larger population.
The liver panel
Several liver panel tests are used to identify Hy's Law cases:
| Liver panel test | Function | Normal range | Hy's Law range |
|---|---|---|---|
| ALT | Enzyme for energy metabolism, involved in gluconeogenesis and amino acid degradation | 10 to 40 IU/L | At least three times the ULN |
| AST | Enzyme for energy metabolism, involved in gluconeogenesis and amino acid degradation | 10 to 34 IU/L | At least three times the ULN |
| Total bilirubin | 0.3 to 1.9 mg/dL | At least twice the ULN | |
| ALP | Enzyme for protein regulation, involved in dephosphorylation | 20 to 140 IU/L | Above 140 IU/L, not likely a Hy's Law case |
Monitoring during clinical trials
The FDA's recommendations for monitoring DILI during clinical trials include:
- Test ALT, AST, ALP and total bilirubin every two to four weeks for several months.
- Follow up an ALT increase of at least three times the ULN by repeat testing within 48 to 72 hours.
- Repeat these four tests often, to confirm and determine whether their levels are increasing or decreasing.
- Ask about symptoms.
In longer clinical trials, liver panel monitoring, sometimes called serum chemistry monitoring, can be spaced to once every two to three months if there are no signs of liver injury after about three months. In clinical development programmes whose early trials identified no signs of hepatotoxicity, less frequent liver panel monitoring can be used in later trials as well.
Plotting cases with eDISH
The FDA created the Evaluation of Drug-Induced Serious Hepatotoxicity (eDISH) tool to help its reviewers identify potential cases of DILI, and it can be used to review laboratory data from large clinical trials. With eDISH, users can generate scatter plots of total bilirubin over its ULN against ALT over its ULN, rapidly identify individuals with laboratory findings that meet the Hy's Law criteria, and find cases that might be DILI. An example the FDA published to illustrate the tool, drawn from real clinical study data, plots peak total bilirubin against peak ALT for 1,960 patients on drug X and 1,962 on drug C. Both axes run from 0.1 to 100 times the upper limit of the reference range, with lines at three times for ALT and twice for bilirubin, and the plot's regions are labelled the normal range, the cholestasis range, the Temple's Corollary range and the Hy's Law range.
Keeping up with the field
As of 2015, the FDA and the American Association for the Study of Liver Diseases sponsor an annual DILI conference, and the association's website archives or links the proceedings of all previous DILI conferences. The website is a resource regulatory writers can use to stay up to date on DILI.