ICH reached Step 4 on S5(R3), Detection of Reproductive and Developmental Toxicity for Human Pharmaceuticals, in February 2020, replacing the original 1993 S5 guideline. FDA adopted it as final guidance in May 2021. The revision did two things sponsors still under-scope today: it widened the stated scope to reach biopharmaceuticals and vaccines explicitly, and it pushed the field toward combination study designs that cover more than one reproductive stage per study. Neither change lowers the evidentiary bar. Both change how a nonclinical team should be building the package.

What actually changed in the 2020 revision

The parent guideline dates to 1993, with a male-fertility addendum added later. S5(R3) folded that addendum into a single, renamed guideline and did more than consolidate housekeeping. It elaborated how exposure margins should inform dose selection, added a dedicated risk-assessment section tying study findings back to a clinical risk conclusion, described qualification criteria for alternative, non-animal assays, and — the change with the widest practical reach — stated plainly that the guideline's scope now includes biopharmaceuticals and vaccines. A program built on the 1993-era assumption that reproductive toxicology guidance was a small-molecule concern is working from a superseded premise.

Feb 2020
ICH Step 4 adoption of S5(R3), replacing the 1993 S5 guideline.
May 2021
FDA's final guidance adoption, confirming current US applicability.
3
Standard study designs — FEED, EFD, and PPND — that combination designs can merge.

The combination designs the guideline actually endorses

The reproductive life cycle is conventionally split into three study types: Fertility and Early Embryonic Development (FEED), Embryo-Fetal Development (EFD), and Pre/Postnatal Development (PPND). Treating those as three mandatory, separate studies is the default assumption S5(R3) is written to correct. The guideline directs sponsors to minimize animal use where doing so does not weaken the overall assessment, and names specific combinations as the route there:

  • Combined FEED/EFD. A single study spanning fertility through the embryo-fetal period, typically paired with a separate PPND study to cover the full life cycle with one fewer standalone study than the traditional three-study battery.
  • Repeat-dose study combined with a fertility assessment. Males from an ongoing repeat-dose toxicity study are mated with separate, untreated, sexually mature females after a defined dosing period — useful specifically where long-term effects on male reproductive performance need evaluating, and a route that reduces the total animal count without adding a dedicated fertility study.
  • Single-sex designs where the science supports it. Where a mechanism of action or a repeat-dose finding predicts toxicity confined to one sex, the guideline supports testing that sex alone and mating against untreated animals of the other, rather than running a full two-sex design by convention.
  • The condition on all of it. Every combination is conditional on not weakening the risk assessment — the guideline frames animal reduction as a byproduct of good study design, not a goal that overrides the science.
Combining designs is not a shortcut through the data. It is the guideline's own answer to covering the same reproductive stages with fewer animals and the same evidentiary weight. Why combination is a design choice, not a waiver

Where S5(R3) ends and a different guideline takes over

S5(R3) is the default DART standard, not a universal one, and two boundaries on this site's own nonclinical coverage mark where it stops governing. ICH S9's anticancer waiver overrides the general reproductive toxicity requirement for advanced-stage oncology products against a cure-rate standard of its own — a program that qualifies for S9 does not default back to S5(R3) for its DART package. ICH S6(R1)'s species-selection logic governs the broader preclinical safety package for biotechnology-derived pharmaceuticals and sits beside S5(R3) rather than replacing it: a monoclonal antibody program can need an S6(R1)-scoped general toxicology package and a separate S5(R3)-scoped DART assessment, each answering a different question about the same molecule.

Scoping the DART package under S5(R3)
  1. Check for a modality override first. Confirm S9 or another modality-specific guideline doesn't already govern before defaulting to a full S5(R3) scope.
  2. Map stages before studies. Identify which reproductive stages the clinical population and route of administration actually require covered, rather than starting from a fixed three-study template.
  3. Evaluate combination designs on their merits. A combined FEED/EFD or a repeat-dose-plus-fertility design can meet the same evidentiary bar with fewer animals — document why a combination was or wasn't used.
  4. Carry the exposure-margin logic into the risk assessment. S5(R3) expects the dose-selection rationale to resurface in an integrated risk conclusion, not stay buried in the study report.

None of this is a reason to under-scope a program, including a vaccine development program now explicitly inside S5(R3)'s stated reach. It is a reason to design the package deliberately — naming the governing guideline, the stages that actually need coverage, and the combination designs that meet the bar — rather than inheriting a battery of studies sized for a different era of the guideline. Teams that treat nonclinical development as a fixed checklist tend to run more studies than the current guideline asks for, not fewer.

Frequently asked questions

Does ICH S5(R3) apply to biologics and vaccines?

Yes. The 2020 revision explicitly widened the guideline's scope beyond small molecules to reach biopharmaceuticals and vaccines, a change from the 1993 original. Whether a specific reproductive and developmental toxicity study is warranted for a given biologic still depends on the molecule's pharmacology and species cross-reactivity, but S5(R3) is now the governing framework for making that call rather than a small-molecule-only standard.

What is a combination study design under ICH S5(R3)?

A single study that covers more than one of the standard reproductive stages — most commonly a combined Fertility and Early Embryonic Development (FEED) plus Embryo-Fetal Development (EFD) study, or a repeat-dose toxicity study combined with a fertility assessment by mating treated males against separate, untreated females. The guideline endorses these combinations specifically to reduce animal use, provided the combined design does not weaken the overall risk assessment.

How is ICH S5(R3) different from ICH S9 and ICH S6(R1)?

S5(R3) is the reproductive and developmental toxicity standard that applies by default across modalities. ICH S9 overrides it for advanced-stage oncology products, where the general nonclinical reproductive package is typically waived against a cure-rate standard rather than run under S5(R3). ICH S6(R1) governs the broader preclinical safety package and species-selection logic for biotechnology-derived pharmaceuticals, and sits alongside S5(R3) rather than replacing it — a biologic can need both a S6(R1)-scoped general toxicology package and an S5(R3)-scoped DART assessment.

Sources & further reading

  1. FDA. S5(R3) Detection of Reproductive and Developmental Toxicity for Human Pharmaceuticals — guidance for industry (May 2021). fda.gov
  2. ICH. S5(R3) — Detection of Reproductive and Developmental Toxicity for Human Pharmaceuticals (Step 4, Feb 2020). ich.org

This article is provided for general informational purposes and reflects the regulatory landscape as of October 2026. It is not legal or regulatory advice. Confirm the current ICH S5(R3) guideline text and its regional adoption status with FDA, EMA, or qualified counsel before scoping a nonclinical program.