Skip to content
Review

AI-integrated human-relevant alternatives to animal experimentation in drug development: Toward a predictive translational ecosystem, from in vitro and in silico models to clinical and real-world evidence

Sep 2026 · Alternatives to laboratory animals : ATLA · Vol 54, pp. 338 - 356 · 1 citation · 118 references
Medicine

TL;DR

An AI-centred, multi-tier decision-support framework, to enable a phased shift toward animal-reduced and, ultimately, animal-free drug development and to address gaps in the ‘AI-orchestrated Human-Relevant Translational Ecosystem’ (AI-HRTE).

Abstract

Animal experimentation has long been associated with biomedical research and drug development. However, ethical concerns, limited translatability to humans and high clinical attrition rates (> 90%) necessitate a transition toward human-relevant approaches. Although diverse in vitro, in silico and artificial intelligence (AI)-enabled methodologies have emerged, their integration into a unified, regulatory-acceptable framework remains limited. This review proposes an AI-centred, multi-tier decision-support framework, to enable a phased shift toward animal-reduced and, ultimately, animal-free drug development. Human-relevant platforms (e.g. organoids, microphysiological systems (MPS) and 3D bioprinting), are evaluated for their potential to reproduce selected aspects of human pharmacokinetics, toxicity and disease phenotypes. AI-driven strategies, including pharmacokinetic/pharmacodynamic modelling, quantitative systems pharmacology, digital twins and virtual clinical simulations, support the prediction of individualised responses with multi-omics, imaging and real-world data. Special emphasis is placed on neuropharmacology, where AI-integrated human neural models enhance mechanistic understanding and therapeutic optimisation. Human-based approaches, such as microdosing, non-invasive imaging and real-world evidence, provide validation pathways and strengthen translational reliability. Despite progress, challenges related to technical limitations, data integration, cost and regulatory acceptance persist. To address these gaps, we propose the ‘AI-orchestrated Human-Relevant Translational Ecosystem’ (AI-HRTE), a conceptual framework integrating experimental, computational and clinical data, to support iterative, AI-assisted decision-making. While this framework offers a structured approach to enhance predictive modelling and reduce animal use in line with the Three Rs principles (i.e. replacement, reduction and refinement), its effectiveness remains hypothetical and requires prospective validation.

View source

Similar papers

Review Sep 2026

In Silico Medicine: The Regulatory Gap in AI-Driven Drug Development

In silico medicine, which depends on computational modeling and simulation, is becoming more popular in drug discovery worldwide. Biomedical research is being rapidly transformed by Artificial Intelligence (AI), but there are risks associated with regulatory ambiguity. The integration of AI offers significant oppor...

S. Mutta, Naifa Faiz, T. N. et al. · 0 citations
Open access Oct 2026

Organs-on-Chips for Disease Modeling and Drug Development: Engineering, Validation and Medical Translation

Introduction: about nine in ten drug candidates that enter clinical development fail, partly because preclinical models, including animal models, predict human responses poorly. Organs-on-chips (OoCs), microfluidic devices lined with living human cells under fluid flow, have emerged as engineered alternatives that repr...

Maikro Osvaldo Chávez Moya, Leodanis Hernández Cabrera, Lauren Martínez Ramírez et al. · 0 citations
Review Oct 2026

Organoids in drug discovery and development: emerging technologies to transition beyond animal models

ABSTRACT Introduction High translational failure rates and ethical concerns surrounding animal models are driving the development of technologies to improve the drug development process. Organoids, three-dimensional self-organizing structures derived from stem cells, provide scalable, human-relevant platforms that reca...

Jolene Phelps, Amanda Orr, Stephanie M. Willerth · 0 citations
Review Open access Oct 2026

Cancer therapy-related cardiovascular toxicity testing: an overview of experimental preclinical in vitro models and how to best use them in combination with animal models, omics and AI-based in silico approaches

Cancer therapy-related cardiovascular toxicity (CTR-CVT) remains a major contributor to morbidity and mortality among cancer survivors. While advances in cancer therapy have significantly improved survival rates, many of these treatments come at the cost of unintended cardiovascular damage. This creates a complex and...

Anaëlle Fratard, Georgia Kalogianni, N. Kindt et al. · 0 citations
Open access 2026

The role of PK/PD modeling and in silico approaches in drug development: A pharmaceutical industry perspective

Pharmacokinetic/pharmacodynamic (PK/PD) modeling and in silico approaches have become central to modern drug development, providing a quantitative framework for evidence-based decisions across non-clinical, translational, and clinical stages. Model-informed drug development integrates approaches such as population PK/P...

B. Golubovic · 0 citations
Review Open access Oct 2026

Target‐Engagement Prediction From Molecular Structure, In Vitro and Animal Data: A Tutorial for Early Assessment of Clinical Drug Exposure and Therapeutic/Adverse Potential Using the ANDROMEDA and PLATO Software on the InSilicoTrials Platform

ABSTRACT The target‐engagement predictor, implemented within the InSilicoTrials cloud‐based, AI‐powered platform for model‐informed drug discovery and development, enables early exploration of time‐dependent pharmacological responses. Predictions can be generated using inputs ranging from molecular structure to in vitr...

Niccolò Totis, Fianne Sips, Simona Catozzi et al. · 1 citation

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.