FDA Preclinical Testing for Large-Molecule Drugs at IND Stage

Preclinical testing FDA expects for large molecule drugs at the IND stage

Scope. This overview covers the nonclinical (preclinical) program a sponsor is expected to have in place before an original Investigational New Drug (IND) application for a large molecule, meaning a biotechnology-derived pharmaceutical such as a monoclonal antibody, therapeutic protein, peptide, or fusion protein. It reflects the framework FDA applies through ICH S6(R1), the ICH safety and multidisciplinary guidelines, FDA's own guidance, and the IND content regulation at 21 CFR 312.23(a)(8). It does not cover CMC/quality, and it does not cover gene or cell therapies, which carry additional CBER-specific expectations.

The governing principle: a case-by-case, science-driven program

Unlike small molecules, biologics are not run through a fixed testing battery. ICH S6(R1) and FDA expect a program tailored to the specific product, its mechanism of action, the target biology, and the intended patient population and duration of use ¹. The standard tiered approaches used for chemical drugs are explicitly stated to be inappropriate for biotechnology-derived pharmaceuticals ¹. M3(R2) governs the timing of nonclinical studies relative to clinical development, but it defers the detail of what studies are appropriate for biologics to ICH S6; the small-molecule conventions in M3(R2), such as the 50-fold exposure margin, apply to small molecules only ².

The practical consequence is that two well-characterized elements have to be established before the toxicology program is even meaningful: the product's primary pharmacology and a pharmacologically relevant animal species.

Primary pharmacology and proof of concept

FDA expects the sponsor to characterize the drug's pharmacologic effects and mechanism of action in animals as part of the IND ⁷. For a biologic this means demonstrating the intended biological activity in a biologically relevant system: target binding, receptor or ligand occupancy and kinetics, and functional activity, ideally in species-specific cell systems and in in vivo pharmacology or disease models ¹. This work is not a formality; it defines which species are pharmacologically responsive and therefore usable for toxicology, and it feeds the pharmacologically active dose (PAD) used later in starting-dose selection.

Species selection: the pivotal decision

Species selection is the single most consequential nonclinical decision for a large molecule, because a toxicology study only has value in a species where the product is pharmacologically active. ICH S6(R1) lays out a stepwise approach ¹:

Start with sequence homology of the target across species.
Use in vitro assays to compare cross-species target binding affinity and receptor/ligand occupancy and kinetics.
Demonstrate functional activity, preferably in species-specific cell systems and/or in vivo.
Tissue cross-reactivity (TCR) in animal tissues is of limited value for species selection and is used only in specific cases when the other approaches cannot identify a relevant species ¹.

If two pharmacologically relevant species exist (one rodent, one non-rodent), both should be used for short-term toxicology up to one month; if findings are similar or the mechanism is understood, longer-term studies in a single species are usually sufficient, and the rodent is generally preferred absent a rationale for the non-rodent. Two non-rodent species are not appropriate. If only one relevant species exists, one species can suffice with justification. When no relevant species exists because the product does not engage the orthologous target in any animal, homologous molecules (a surrogate active in the animal) or transgenic models expressing the human target may be considered ¹.

Pharmacokinetics, toxicokinetics, and ADME

Before clinical studies, there should be information on absorption, disposition, and clearance in relevant animal models sufficient to predict exposure-based margins of safety ¹. ICH S6(R1) calls for targeted single- and multiple-dose PK/TK studies and tissue distribution studies in relevant species, using test material representative of the toxicology and clinical material and a route relevant to the planned clinical use, with systemic exposure monitored during toxicity studies whenever feasible ¹. Routine mass-balance ADME studies are not useful for these products and are not expected ¹.

Repeat-dose toxicology

The repeat-dose toxicity study is the core of the IND-enabling package. ICH S6(R1) expects ¹:

Species: relevant species, generally two when available, but one relevant species may suffice with justification.
Route and regimen: reflect the intended clinical use or exposure, with toxicokinetics included when feasible.
Dose selection: define a dose-response relationship including a NOAEL and, where possible, a toxic dose. Under the S6(R1) addendum, the high dose should be set on PK/PD principles, targeting either the maximum intended pharmacological effect or roughly a 10-fold exposure multiple over maximum anticipated clinical exposure, using the higher of the two unless a lower dose is justified.
Duration: driven by intended clinical exposure and indication. Typical durations are 1 to 3 months for most biologics; up to 2 weeks may suffice for short-term or acute life-threatening use; 6 months is generally appropriate for chronic indications, and the addendum confirms 6 months in rodents or non-rodents is sufficient for chronic-use products when the high dose is appropriately selected.
Recovery: a recovery period should generally be included to assess reversibility, worsening, or delayed effects; complete recovery is not essential, and the non-dosing period is for reversibility rather than to detect delayed toxicity.

Anti-drug antibodies should be measured in these studies to aid interpretation ¹.

Safety pharmacology

The ICH S7A core battery evaluates effects on vital functions across the cardiovascular, respiratory, and central nervous systems, and these effects should generally be investigated before first administration in humans, ordinarily under GLP ³. For biologics the approach is pragmatic: products that achieve highly specific receptor targeting can often have safety pharmacology endpoints folded into the toxicology and/or pharmacodynamic studies, reducing or eliminating stand-alone studies ³. A biologic that represents a novel therapeutic class or does not achieve highly specific targeting warrants a more extensive safety pharmacology evaluation ³. Consistent with this, ICH S6(R1) allows functional indices (cardiovascular, respiratory, renal, CNS) to be monitored within toxicity studies rather than as separate studies ¹.

Immunogenicity in nonclinical studies

Because most biologics are immunogenic in animals, antibody responses should be measured in repeat-dose toxicity studies, but the purpose is to interpret the study, not to predict human immunogenicity, which animal data do not forecast ¹. ICH S6(R1) recommends characterizing responses by titer, number of responders, and neutralizing versus non-neutralizing status, and correlating antibody appearance with changes in PK/PD, adverse effects, complement activation, new toxicities, or immune-complex pathology ¹. Antibody detection alone should not be the basis to stop or shorten a study unless the response neutralizes the pharmacologic or toxicologic effect in a large proportion of animals, and a recovery period is not required solely to assess immunogenicity ¹.

Immunotoxicity

Nonclinical immunotoxicity evaluation is weight-of-evidence and largely triggered rather than routine. FDA's current guidance expects a first-tier integration of general toxicity study findings with pharmacology, the intended population, structural similarity to immune-active compounds, disposition, and clinical findings, and, for immunomodulators, the target's role in immune function ⁸. If that assessment flags a potential immunotoxicity hazard, more specific immune-function testing may be warranted; the concepts of ICH S8 sit behind this framework ⁸⁵.

Studies that are generally not required for biologics

Several standard small-molecule studies are explicitly not expected for biotechnology-derived pharmaceuticals ¹:

Genotoxicity: the standard genotoxicity battery is not applicable and is not needed.
Carcinogenicity: standard rodent carcinogenicity bioassays are generally inappropriate; a product-specific, weight-of-evidence assessment may still be needed depending on the product, population, duration, and biological activity.
Tissue cross-reactivity for species selection: of limited value, though for monoclonal antibodies a TCR study across a panel of human tissues is used to characterize antigenic specificity and any unintended human-tissue reactivity.

Reproductive and developmental toxicity

Reproductive and developmental toxicity is handled case by case and is scientifically justified rather than run to a fixed schedule ¹. When a non-human primate is the only relevant species, endpoints are often combined in an enhanced pre- and postnatal development (ePPND) study, with dosing extended through gestation to parturition and offspring assessed after birth ¹. These studies generally support later clinical development rather than initial first-in-human entry, and their timing relative to enrollment of women of childbearing potential follows ICH M3(R2) ².

Local tolerance and use of the clinical route and formulation

Local tolerance should be evaluated, ideally with the intended commercial formulation, though a representative formulation can be acceptable when justified ¹. Separate local tolerance studies may be unnecessary if the endpoints can be captured within the single- or repeat-dose toxicity studies using the intended clinical route ¹.

Timing relative to clinical phases

ICH M3(R2) sets when studies must be complete ². It defers the biologics-specific study set to ICH S6, but its timing logic still applies: for example, a male fertility study should be complete before large or long trials such as Phase III; embryo-fetal endpoints can sometimes be deferred for short early trials in women of childbearing potential when pregnancy risk is intensively controlled; juvenile animal studies, when needed for pediatric development, should be complete before those trials; and additional immunotoxicity studies, if indicated, should be complete before exposing a large Phase III population ².

Deriving the first-in-human starting dose

FDA expects the starting dose to be justified from the nonclinical data ⁶. The conventional NOAEL-based path is:

Identify the NOAEL from adequate toxicology studies.
Convert the NOAEL to a human equivalent dose (HED), usually by body-surface-area scaling.
Apply a safety factor, generally at least 10, to obtain the maximum recommended starting dose (MRSD = HED ÷ safety factor); the factor can be raised for safety concerns or study limitations, or lowered where class knowledge supports it ⁶.

For biologics FDA cautions that the starting dose is often based on a dose expected to produce no adverse biological effect, with attention to species specificity and receptor-avidity differences between animals and humans ⁶. Where toxicity is driven by exaggerated pharmacology, the pharmacologically active dose (PAD) may be more sensitive than the NOAEL and can lower the MRSD; this is the basis of the minimal anticipated biological effect level (MABEL) approach used for higher-risk biologics such as immune agonists ⁶.

Special case: anticancer biologics for advanced cancer (ICH S9)

For biologics intended for patients with advanced cancer, ICH S9 permits a reduced, risk-based package to support first-in-human and early trials ⁴. Stand-alone safety pharmacology studies are not called for unless a specific risk is identified; genotoxicity, embryo-fetal, fertility, pre/postnatal, and carcinogenicity studies are not considered essential for this setting; and studies to determine a NOAEL/NOEL are not essential ⁴. Species selection still follows ICH S6, and for immune-agonistic biopharmaceuticals a MABEL-based starting dose should be considered ⁴.

What FDA expects in the IND submission itself

Under 21 CFR 312.23(a)(8), the pharmacology/toxicology section of an original IND must contain ⁷:

A description of pharmacologic effects, mechanism of action, and ADME in animals (a summary report usually suffices; state what is unknown).
An integrated summary of the toxicologic effects in animals and in vitro, with the specific studies depending on the product and the clinical phase.
Full data tabulations for each toxicology study supporting the proposed trial, suitable for detailed review.
A GLP compliance statement for each study subject to GLP, or an explanation of any noncompliance.

The integrated toxicology summary should describe study designs and dates, present findings systematically across body systems, flag signals an informed expert would view as possible human risk (with reasons where a signal is judged not to indicate human risk), identify and give the qualifications of the person who concluded it is reasonably safe to begin, and state where studies were conducted and records are held. Final quality-assured study reports should be available within 120 days of starting the human study ⁷.

Bottom line for RA planning

For a typical monoclonal antibody or therapeutic protein entering the clinic outside oncology, the IND-enabling nonclinical core is: proof-of-concept pharmacology and mechanism characterization; a documented species-relevance assessment; PK/TK and tissue distribution in relevant species; GLP repeat-dose toxicology (commonly up to 1 to 3 months, in one or two relevant species, with recovery, TK, and ADA monitoring); safety pharmacology endpoints usually embedded in the toxicology studies; local tolerance; and a justified starting dose using NOAEL/HED/MRSD, shifting to a PAD- or MABEL-based approach for higher-risk targets ¹³⁶. The standard genotoxicity and carcinogenicity batteries are omitted, and reproductive toxicology is generally deferred and shaped by the intended population and M3(R2) timing ¹². Anticancer biologics for advanced disease follow the leaner ICH S9 package ⁴.

A useful follow-up to ask Rhizome directly: how a specific modality (for example a bispecific antibody, an antibody-drug conjugate, an Fc-engineered agonist, or a cytokine) changes species selection, MABEL justification, and the depth of the toxicology program.