Hy's Law: Identifying Potential Serious Drug-Induced Liver Injury

Hy's Law is one of the most important safety concepts used to recognise a medicine's potential to cause severe idiosyncratic drug-induced liver injury (DILI). It is often reduced to two laboratory thresholds, but the scientific concept is broader: hepatocellular injury accompanied by substantial bilirubin elevation, without a more plausible alternative explanation, may signal a drug capable of causing serious liver injury.

Purpose and Regulatory Context

FDA's final 2009 guidance on premarketing DILI evaluation uses Hy's Law as part of a framework for identifying drugs that may cause severe liver injury during development. The guidance is explicitly non-binding and is intended to support scientific evaluation where actual cases of acute liver failure may be too rare to occur in a typical development programme.

Hy's Law should therefore be understood primarily as a drug-development safety signal concept, not as a stand-alone diagnostic criterion for an individual patient.

The Screening Pattern

Potential Hy's Law cases are commonly identified when a treated subject has:

• ALT or AST greater than about 3× ULN;
• total bilirubin greater than about 2× ULN; and
• no dominant cholestatic pattern or more plausible alternative explanation.

These thresholds identify cases for evaluation. They do not, by themselves, establish DILI or drug causality.

Why the Combination Matters

ALT and AST primarily indicate cellular injury. They are not direct measures of liver function. Total bilirubin provides different information and becomes particularly important when it rises in the setting of hepatocellular injury.

The concern arises from the combination:

hepatocellular injury + impaired bilirubin handling/jaundice + no better explanation.

That pattern has greater implications for severe DILI risk than an isolated transaminase elevation.

Hy's Law Is Not the Same as DILI

DILI is the broader clinical condition. It can be hepatocellular, cholestatic or mixed; mild or severe; and may occur without meeting Hy's Law thresholds.

Conversely, a subject can cross the biochemical screening thresholds and ultimately be found to have viral hepatitis, biliary obstruction, ischaemic injury, another hepatotoxic drug or another disease process.

Therefore:

screening thresholds → medical evaluation → injury-pattern assessment → competing-cause assessment → causality judgement.

Only after this process can the case contribute meaningfully to the drug-level Hy's Law assessment.

Individual Case Versus Drug-Level Signal

Two levels should be kept separate.

At the individual-subject level, the reviewer determines whether the episode is compatible with hepatocellular DILI and whether the investigational medicine is a plausible cause.

At the drug level, the sponsor and regulator assess whether the development programme contains a pattern—potentially including one or more convincing cases—that suggests the medicine itself can cause severe DILI.

Confusing these levels leads to overstatement. A laboratory-triggered case is not automatically a drug-level Hy's Law signal.

Characterising the Liver-Injury Pattern

The classical Hy's Law concept concerns hepatocellular injury. The reviewer should therefore examine ALT, AST, alkaline phosphatase (ALP) and bilirubin together rather than treating the ALT/bilirubin thresholds as sufficient.

ALP is particularly important because substantial cholestasis changes the interpretation of bilirubin elevation. Where useful, the liver-injury pattern can be described using an R-value based on ALT and ALP relative to their ULNs, while recognising that the timing of the measurements used in the calculation matters.

Competing Causes

A potential Hy's Law case requires active investigation of alternative explanations. Depending on the clinical scenario, these may include:

• acute viral hepatitis;
• biliary obstruction;
• ischaemic or hypoxic hepatitis;
• autoimmune hepatitis;
• alcohol-related injury;
• sepsis;
• underlying hepatic disease;
• haemolysis or non-hepatic causes of bilirubin elevation; and
• another medicinal product or supplement.

The absence of a single definitive test for DILI makes this differential diagnosis central to the assessment.

Timing and Longitudinal Review

A laboratory snapshot can obscure the sequence of events. The reviewer should reconstruct:

• baseline liver tests;
• treatment initiation and dose changes;
• onset of aminotransferase elevation;
• onset of bilirubin elevation;
• ALP trajectory;
• symptoms;
• treatment interruption;
• recovery; and
• relevant investigations.

This chronology can distinguish a coherent hepatocellular injury episode from unrelated peaks occurring at different times.

Baseline Abnormalities and Special Populations

Patients with chronic liver disease may enter trials with elevated ALT, AST, ALP or bilirubin. Applying simple multiples of ULN without considering baseline can therefore misclassify treatment-emergent injury.

In such populations, the reviewer may need to consider change from baseline, prior variability and disease-specific context in addition to conventional thresholds.

The principle remains the same: the thresholds are screening tools, while the medical question is whether treatment is associated with a new clinically meaningful hepatocellular injury pattern.

eDISH as a Case-Finding Tool

eDISH can display peak aminotransferase and bilirubin values across a study population and make potential Hy's Law-range subjects easier to identify.

A point in the upper-right quadrant should trigger medical review, not automatic classification. The underlying case still needs longitudinal analysis, ALP assessment, competing-cause evaluation and causal judgement.

For detailed interpretation see [[edish]].

Comparator and Population Context

Drug-level interpretation should consider more than the individual case. Relevant questions include:

• Are aminotransferase elevations more frequent in the active arm than in controls?
• Are potential cases confined to the investigational treatment?
• Is there a dose or exposure pattern?
• Do cases share a similar phenotype or latency?
• Are there supportive nonclinical or mechanistic findings?
• Is the observed pattern plausible in the underlying disease population?

A convincing case can be highly informative, but the strength of the overall signal depends on the quality and coherence of the full evidence base.

Practical Medical Review Sequence

A practical assessment can proceed as follows:

1. confirm that the biochemical trigger is genuine;
2. reconstruct baseline and longitudinal liver tests;
3. determine whether the injury is hepatocellular, cholestatic or mixed;
4. assess the timing relative to treatment;
5. investigate alternative causes actively;
6. review concomitant hepatotoxic exposures;
7. assess dechallenge and any clinically relevant rechallenge information;
8. determine the individual-case causal interpretation; and
9. integrate the adjudicated case with the rest of the development or post-marketing evidence.

This sequence is recommended scientific practice rather than a legally prescribed workflow.

Potential Failure Modes

The following are illustrative failure modes, not reported regulatory findings.

Pharmacovigilance After Approval

Hy's Law is most closely associated with clinical development because systematic laboratory monitoring provides complete longitudinal data. After approval, spontaneous reports often contain less complete laboratory information.

Post-marketing DILI assessment therefore requires strong follow-up for laboratory values, timing, alternative causes, concomitant therapies and outcomes. A clinically convincing spontaneous report may still be highly informative, particularly when supported by recurrence, epidemiology, literature, class effects or a coherent series of cases.

Relationship With Other QPPV.com Articles

For the broader disease entity see [[drug-induced-liver-injury-dili]]. For graphical population review see [[edish]]. Together these articles distinguish three levels of reasoning:

• DILI — the clinical injury;
• Hy's Law — a serious hepatotoxicity signal concept;
• eDISH — a visual method for finding and contextualising relevant trial subjects.

Key Takeaways

Hy's Law is not simply a pair of laboratory thresholds and is not a diagnostic test for an individual patient.

The biochemical pattern identifies subjects who require structured evaluation for hepatocellular DILI and alternative causes.

The most important scientific distinction is between an individual potential case and the drug-level conclusion that a medicine may have the capacity to cause severe liver injury.

Longitudinal review, ALP, competing causes, baseline status, comparator experience and the totality of evidence determine how much weight a potential case should carry.

References

1. U.S. Food and Drug Administration. Drug-Induced Liver Injury: Premarketing Clinical Evaluation. Guidance for Industry. July 2009.
2. Watkins PB, Desai M, Berkowitz SD, et al. Evaluation of drug-induced serious hepatotoxicity (eDISH): application of this data organization approach to phase III clinical trials of rivaroxaban after total hip or knee replacement surgery. Drug Safety. 2011;34(3):243-252.
3. International Council for Harmonisation. Relevant nonclinical and clinical safety guidance as applicable to the development programme.

Regulatory Note

FDA's July 2009 DILI guidance remains final, non-binding guidance. The commonly used biochemical thresholds are part of a scientific framework for identifying potential cases; they do not create an automatic legal classification of DILI or drug causality. Current product- and programme-specific regulatory advice should be followed where additional requirements apply.