Drug-Induced Cardiac Toxicity: Cardiovascular Safety Assessment in Clinical Development and Pharmacovigilance

Drug-induced cardiac toxicity encompasses adverse effects on cardiac electrophysiology, myocardium, conduction, heart rate, vascular tone and haemodynamics. It is broader than QT prolongation and cannot be reduced to a single ECG, biomarker or imaging threshold.

Purpose and Scientific Framework

The core question is what cardiovascular phenotype occurred, how clinically important it is, and whether the medicinal product is the most plausible explanation.

Relevant evidence may include:

• symptoms and clinical events;
• ECG and rhythm monitoring;
• heart rate and blood pressure;
• cardiac troponin and natriuretic peptides;
• echocardiography or other imaging;
• exposure and pharmacology;
• baseline cardiovascular disease;
• electrolyte and organ-function changes;
• concomitant medicines; and
• alternative causes.

ICH S7A addresses nonclinical safety pharmacology of vital functions, including cardiovascular function. ICH E14 and S7B provide the specialised framework for QT/QTc prolongation and proarrhythmic potential; they should not be treated as a complete framework for every form of cardiotoxicity.

Major Cardiovascular Phenotypes

Electrophysiological Effects

Medicines can alter ventricular repolarisation, conduction, automaticity or heart rate. QT/QTc prolongation is one important phenotype, but bradyarrhythmia, atrioventricular block, QRS widening and atrial or ventricular arrhythmias may arise through different mechanisms.

The dedicated [[drug-induced-qt-prolongation-cardiac-safety]] article addresses QT and proarrhythmic risk in detail.

Myocardial Injury

Cardiac troponin is a sensitive marker of myocardial injury, but elevation does not by itself establish myocardial infarction or drug causality. Myocarditis, heart failure, tachyarrhythmia, pulmonary embolism, sepsis, renal dysfunction and other acute illnesses can also elevate troponin.

Interpretation therefore requires symptoms, ECG findings, imaging, kinetics and clinical context.

Ventricular Dysfunction and Heart Failure

A decline in left ventricular function can precede symptomatic heart failure. Conversely, dyspnoea or oedema may result from non-cardiac causes.

Assessment may incorporate ejection fraction, strain imaging, natriuretic peptides, symptoms, physical findings and treatment response. A numerical ejection-fraction change should not be interpreted without baseline variability and clinical context.

Myocarditis and Inflammatory Cardiac Injury

Immune-mediated or inflammatory myocarditis may present with chest pain, arrhythmia, troponin elevation, ventricular dysfunction or conduction abnormalities. Diagnosis may require integration of imaging, biomarkers, clinical presentation and occasionally tissue assessment.

Haemodynamic and Vascular Effects

Medicines can cause hypertension, hypotension, tachycardia, bradycardia, vasospasm, ischaemia or thrombotic/vascular effects. These phenotypes may be acute or cumulative and may indirectly increase cardiovascular risk without causing direct myocardial injury.

Mechanism Matters

Cardiac toxicity may be electrophysiologic, direct myocardial, immune mediated, vascular, haemodynamic or secondary to another organ toxicity. Renal dysfunction, hypoxia, anaemia, electrolyte disturbance and drug interactions can all modify susceptibility.

The safety assessment should therefore avoid grouping all cardiovascular events under one causal mechanism.

Individual-Case Assessment

A cardiovascular case should be reconstructed chronologically. Establish baseline disease and risk factors, treatment timing, symptoms, ECG/rhythm findings, biomarkers, imaging, haemodynamics, concomitant medicines, electrolyte abnormalities and recovery.

ECG and Rhythm Data

An ECG abnormality should be interpreted according to its phenotype. QT prolongation, QRS widening, atrioventricular block, atrial fibrillation and ventricular tachycardia have different clinical implications and mechanisms.

Automated ECG interpretations can support screening but clinically important findings may require expert review, repeat tracings or rhythm monitoring.

Cardiac Biomarkers

Troponin indicates myocardial injury but not its cause. Natriuretic peptides reflect myocardial wall stress and may support heart-failure assessment, but are influenced by age, renal function, atrial arrhythmias and other conditions.

Biomarkers should therefore be integrated with clinical findings rather than used as stand-alone causal tests.

Imaging

Echocardiography, cardiac MRI or other imaging may clarify ventricular function, wall motion, inflammation or structural disease. The appropriate modality depends on the suspected phenotype. Imaging findings can establish injury more convincingly than symptoms alone but still require causal interpretation.

Clinical-Development Assessment

Population-level review should examine:

• treatment-versus-control event rates;
• baseline cardiovascular risk;
• dose and exposure relationships;
• ECG and heart-rate distributions;
• blood-pressure effects;
• biomarker trajectories;
• ventricular-function changes;
• serious cardiovascular outcomes; and
• timing, reversibility and recurrence.

The analysis should be driven by the suspected mechanism. QT studies are appropriate for repolarisation risk; they are not substitutes for assessing myocardial dysfunction or vascular toxicity.

Post-Marketing Pharmacovigilance

Follow-up should target the information needed to define the cardiovascular syndrome: objective diagnosis, timing, ECGs, biomarkers, imaging, cardiovascular history, concomitant drugs, interacting factors, electrolytes, dechallenge and outcome.

Signal assessment should group clinically coherent phenotypes. Myocarditis, heart failure, bradyarrhythmia and myocardial infarction may all be cardiac events, but pooling them without mechanistic justification can obscure the signal.

Product-Level Interpretation

The product-level conclusion should explain whether evidence supports:

• an electrophysiologic liability;
• myocardial injury or inflammatory toxicity;
• ventricular dysfunction or heart failure;
• haemodynamic or vascular effects;
• an interaction-mediated risk;
• an unresolved signal; or
• a stronger alternative explanation.

The degree of certainty should match the evidence. A biomarker shift can justify further investigation without automatically establishing a clinically important adverse reaction.

Practical Medical Review Sequence

1. Define the cardiovascular phenotype precisely.
2. Establish baseline cardiovascular disease and risk factors.
3. Reconstruct timing relative to treatment and dose.
4. Review ECG/rhythm, biomarkers, imaging and haemodynamics according to the phenotype.
5. Assess electrolytes, organ dysfunction, interactions and competing causes.
6. Evaluate dechallenge/recovery and any naturally occurring rechallenge.
7. Integrate comparator, class and mechanistic evidence.
8. State the product-level conclusion and uncertainty clearly.

Potential Failure Modes

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

Governance and Inspection Considerations

For an important cardiac signal, assessors may ask how the organisation defined the phenotype, obtained objective cardiac data, evaluated competing causes and interactions, grouped cases and reconciled conclusions across clinical development, signal management, product information and risk management. There is no universal requirement for a single cardiovascular committee or fixed testing panel for all products.

Practical Review Checklist

1. Is the cardiac phenotype clearly defined?
2. Are baseline cardiovascular risks considered?
3. Are objective ECG, biomarker or imaging data interpreted correctly?
4. Are QT-specific issues separated from broader cardiotoxicity?
5. Are electrolyte, renal and drug-interaction effects considered?
6. Are alternative causes addressed?
7. Are cases grouped by coherent clinical mechanism?
8. Does the final conclusion calibrate causality and uncertainty appropriately?

Relationship With Other QPPV.com Articles

For the specialised ventricular-repolarisation framework see [[drug-induced-qt-prolongation-cardiac-safety]]. This article also complements [[causality-assessment-in-pharmacovigilance]] and [[signal-assessment]].

Key Takeaways

Cardiotoxicity is broader than QT prolongation and includes electrophysiologic, myocardial, inflammatory, haemodynamic and vascular effects.

ECGs, troponin, natriuretic peptides and imaging are evidence sources, not independent causality tests.

The strongest assessment defines the phenotype first, evaluates mechanism and competing causes, and then integrates case-level findings into a product-level conclusion.

References

1. International Council for Harmonisation. ICH S7A: Safety Pharmacology Studies for Human Pharmaceuticals.
2. International Council for Harmonisation. ICH E14: Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for Non-Antiarrhythmic Drugs.
3. International Council for Harmonisation. ICH S7B: Nonclinical Evaluation of the Potential for Delayed Ventricular Repolarization by Human Pharmaceuticals.
4. International Council for Harmonisation. ICH E14/S7B Questions and Answers. Step 4, 21 February 2022.
5. European Medicines Agency. GVP Module IX — Signal management (Rev. 1), with current EMA implementation material.

Regulatory Note

ICH E14 and S7B address QT/QTc and proarrhythmic potential and should not be presented as universal standards for all forms of cardiotoxicity. Broader cardiovascular monitoring should be proportionate to the product's pharmacology, nonclinical findings, clinical population and emerging safety evidence.