Lethal Dose Determination for New Drugs: LD50 Testing Methods, Standards, and Applications

Lethal dose (LD50) determination is a fundamental component of preclinical safety evaluation for new drug candidates, providing critical data on acute toxicity profiles before human exposure. This article presents a comprehensive technical overview of lethal dose testing for new drugs, including sample types, testing items, internationally recognized methods such as OECD Test Guidelines 423 and 425, statistical approaches, and regulatory applications in pharmaceutical development.

Technical Overview

Lethal dose determination, commonly expressed as the median lethal dose (LD50), is the estimated dose of a substance that causes death in 50% of test animals under defined experimental conditions. For new drug candidates, this assessment forms part of the acute toxicity evaluation conducted during preclinical development. The resulting data support first-in-human dose selection, safety margin calculations, hazard classification, and the design of subsequent repeat-dose general toxicity studies.

Modern lethal dose testing has evolved significantly from the classical multi-dose LD50 protocol described in the original OECD Test Guideline 401, which was formally deleted in 2002. Current internationally harmonized approaches, including the Acute Toxic Class Method (OECD TG 423) and the Up-and-Down Procedure (OECD TG 425), are designed to reduce animal use while still providing sufficient information for regulatory classification and risk assessment, in line with the 3Rs principle of replacement, reduction, and refinement.

Under ICH Guideline M3(R2) on Nonclinical Safety Studies for the Conduct of Human Clinical Trials, formal single-dose lethality studies are no longer routinely required before first-in-human trials in most jurisdictions. However, acute toxicity information is still expected, typically obtained through escalating-dose studies, dose-range-finding studies, or dedicated acute toxicity protocols conducted in compliance with Good Laboratory Practice (GLP) principles at accredited third-party laboratories.

Testing Samples

New Chemical Entities (NCEs)

Synthetic small-molecule drug candidates, including active pharmaceutical ingredients (APIs) with defined structures and purities, represent the most common sample type. Batch characterization, including purity, impurity profile, and physicochemical properties, is typically documented prior to dosing.

Natural Product and Botanical Extracts

Plant-derived extracts, fractionated natural products, and traditional medicine preparations may undergo lethal dose testing when the active constituents are incompletely characterized and the safety profile cannot be inferred from existing data.

Biologics and Biotherapeutics

Recombinant proteins, monoclonal antibodies, peptides, and nucleic acid-based therapeutics may require species-specific acute toxicity evaluation, particularly when pharmacological activity is highly target-dependent and cross-reactive species are limited.

Formulated Drug Products

Final dosage forms, including tablets, capsules, solutions, suspensions, and injectable formulations, may be evaluated to assess the influence of excipients and formulation composition on the acute toxicity profile.

Impurities, Metabolites, and Degradation Products

Genotoxic or high-potency impurities, major human metabolites not present in animal plasma during standard studies, and forced-degradation products may warrant standalone acute toxicity characterization.

Testing Items

Lethal dose determination studies typically encompass the following core evaluation endpoints:

• LD50 (median lethal dose) with 95% confidence interval, expressed in mg/kg body weight
• Approximate lethal dose (ALD) or minimum lethal dose (MLD)
• Maximum tolerated dose (MTD) or maximum non-lethal dose
• Acute toxic symptom profile: onset time, type, severity, duration, and reversibility of clinical signs
• Mortality incidence and time-to-death recording
• Body weight monitoring at baseline and at regular intervals throughout the observation period
• Food and water consumption assessment
• Gross necropsy findings for all animals, including scheduled survivors and decedents
• Optional histopathological examination of target organs
• No observed adverse effect level (NOAEL) estimation where achievable

The standard observation period is 14 days following administration, with clinical observations performed at minimum once daily and more frequently during the first several hours after dosing, when acute effects are most likely to manifest.

Testing Methods

Acute Toxic Class Method (OECD TG 423)

The Acute Toxic Class method uses a stepwise dosing procedure with fixed dose levels (5, 50, 300, 2000, and exceptionally 5000 mg/kg) administered to groups of three animals, typically female rodents. The test substance is administered at a single step; based on the mortality outcome, the next step proceeds at a higher, lower, or the same dose level. This method does not calculate a precise LD50 value but assigns the substance to a Globally Harmonized System (GHS) acute toxicity category, substantially reducing animal use compared with the classical method.

Up-and-Down Procedure (OECD TG 425)

The Up-and-Down Procedure (UDP) doses animals sequentially, one at a time, with each subsequent dose adjusted upward or downward based on the survival outcome of the preceding animal. The default dose sequence follows a half-log progression (1.75, 5.5, 17.5, 55, 175, 550, 2000 mg/kg). Testing is stopped when predetermined stopping criteria are met, and the LD50 with confidence interval is estimated using maximum likelihood statistical methods. The UDP may also be performed as a limit test at 2000 or 5000 mg/kg.

Classical LD50 Determination

Although no longer the preferred regulatory method, the conventional multi-group protocol may still be requested in specific cases. Animals are distributed across four or more dose levels, and the LD50 is calculated statistically using validated methods such as the Bliss method (probit maximum likelihood), the Spearman-Kärber method, or the Horn method, each providing point estimates with confidence intervals.

Limit Test

When a test article is expected to be of low toxicity, a limit test at 2000 mg/kg (or 5000 mg/kg under defined circumstances) is conducted. If no treatment-related mortality occurs at the limit dose, classification testing beyond that level is generally unnecessary.

Route of Administration and Species Selection

The intended clinical route of administration is normally used, with oral gavage, intravenous, subcutaneous, intraperitoneal, and inhalation routes being the most frequently applied. Rodents, particularly mice and rats, are the primary species; a second mammalian species, typically a non-rodent, may be included where scientifically justified, for example for biologics with species-specific pharmacology. Both sexes are evaluated unless a scientifically valid justification for single-sex testing is provided.

Applications

Lethal dose data generated through compliant acute toxicity studies serve multiple purposes across the drug development lifecycle:

• Supporting Investigational New Drug (IND) applications and clinical trial authorizations by establishing the acute toxicity baseline for first-in-human risk assessment
• Defining starting doses and dose escalation limits for Phase I clinical trials, including the NOAEL-based and minimum anticipated biological effect level (MABEL) approaches
• Calculating safety margins between animal lethal or severely toxic doses and projected human exposures
• Guiding dose level selection for 28-day and 90-day repeat-dose general toxicity studies
• GHS-based hazard classification and labeling of pharmaceutical substances, intermediates, and research compounds
• Occupational exposure assessment and containment strategy development for high-potency active pharmaceutical ingredients (HPAPIs)
• Safety characterization supporting drug master files and regulatory submissions in multiple jurisdictions

FAQ

What is the difference between LD50 and MTD?

The LD50 is the dose that produces mortality in 50% of treated animals, whereas the maximum tolerated dose (MTD) is the highest dose that does not produce overt toxicity, mortality, or clinically significant body weight loss. MTD determination is often more informative for drug development because it defines a non-lethal upper boundary for dose setting.

Is a formal LD50 study still required for new drug registration?

Under ICH M3(R2), precise LD50 values are generally no longer required before first-in-human studies. Regulators instead expect acute toxicity information sufficient to support clinical safety, which may be obtained through single-dose or escalating-dose studies integrated into the nonclinical program.

Which animal species are used in lethal dose testing?

Mice and rats are the standard rodent species. For biopharmaceuticals or where rodent data are not pharmacologically relevant, a non-rodent species such as rabbits, dogs, minipigs, or non-human primates may be justified following ethical and scientific review.

How long does an acute toxicity study take?

Following dosing, the standard observation period is 14 days, which allows assessment of delayed toxicity and recovery. Including acclimatization, formulation work, dosing, observation, necropsy, and reporting, a complete study typically requires four to eight weeks.

What quality standards apply to these studies?

Studies intended for regulatory submission are conducted in accordance with Good Laboratory Practice (GLP) principles, aligned with OECD GLP standards and applicable national regulations, and follow the experimental designs specified in OECD Test Guidelines and ICH safety guidance documents. Accredited third-party laboratories routinely perform these studies under such frameworks.

How are the results reported?

A final report typically includes the LD50 estimate or GHS classification category, mortality data, clinical observations, body weight trajectories, necropsy findings, dose-response characterization, and protocol deviations, presented in a format suitable for inclusion in regulatory submissions.