The 5% Problem: How the Biomedical Community Responded to the Animal-to-Human Translation Crisis, and the Case for...

A widely cited 2024 analysis reported that only 5% of therapeutic interventions tested in animals obtain regulatory approval for human use, reviving the debate about the predictive value of animal models. This review asks how the scientific community has interpreted that finding and what collective position is emerging.

The literature citing the original study was assembled from public citation indexes and classified by argumentative stance into five categories: defences of animal experimentation, critiques, methodological and disease-specific analyses, the author's own position, and work developing non-animal methods; no new animal data were generated.

Across the corpus, attempts to strengthen the validity of animal studies tended to reduce rather than reinforce their apparent translational signal; the low translation rate recurred across unrelated disease domains; and even defenders of animal models increasingly restricted their claims to specific contexts while conceding poor translation. In parallel, non-animal methods-including organoids, organ-on-chip systems, and computational models-were repeatedly described as equalling or surpassing animal models in several fields.

The scientific, ethical, economic and regulatory cases now converge, supporting a deliberately planned transition toward human-relevant methods rather than its deferral.