Artificial intelligence-, organoid-, and organ-on-chip-powered models to improve pre-clinical animal testing of vaccines and immunotherapeutics: potential, progress, and challenges

Résumé fourni par la source

Vaccines and immunotherapies against infectious diseases and cancers have been a great success of the medical sciences over the last century. Pre-clinical testing in animal models has played a crucial role in the development of vaccines and immunotherapies, informing subsequent clinical trials. The current practices in pre-clinical animal model research must be approved by committees with strict policies and assessments on animal experiments including the “three Rs”: (1) Replacement, which assesses the scientific justification and rationale for using a live animal in biomedical research; (2) Reduction, which determines whether the number of animals required in an experiment is adequate to achieve scientifically valid results while reducing costs; and (3) Refinement, which ascertains that any given animal procedure will cause no to minimal pain or distress. The recent initiatives by the United States NIH and FDA to reduce or phase out animal testing in biomedical research underscore a growing interest in artificial Intelligence (AI), deep learning (DL), organoid, and organ-on-chip-powered models to slash the time and cost of preclinical animal research. This review highlights the strengths, progress, and limitations of these alternative pre-clinical research approaches, with a focus on vaccine and immunotherapeutic development. While the implementation of AI- and DL-, organoid-, and organ-on-chip-powered models will certainly help accelerate pre-clinical discoveries, modeling the safety, immunogenicity, and protective efficacy of vaccines and immunotherapeutics as they occur in vivo is not yet comprehensive enough to fully replace or replicate the complexity of living systems, in both animals and humans. Thus, these models should be viewed as powerful complementary tools that combine hybrid human and artificial intelligence and must be validated through animal model testing. This review discusses the path forward and the scientific challenges that persist in investing in AI- and DL-human hybrid validation systems, regulatory reforms, and the development of interconnected platforms that bridge digital models with biological reality.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.