Translation Gap

The translation gap (also translational gap) refers to the systematic failure of scientific findings obtained in laboratory and preclinical research settings to produce the anticipated effects when tested in human clinical trials. It is most acute in pharmaceutical drug discovery, where results in cell culture and animal models have historically predicted human clinical outcomes poorly — with the majority of drugs that enter Phase II trials failing on efficacy grounds despite performing well preclinically.

The gap is not a random failure rate. It is structured: certain disease areas (oncology, neurology, psychiatry) show far higher translational attrition than others (infectious disease, metabolic disease in specific indications). This structure reflects genuine differences in how well animal models recapitulate human disease biology. A mouse model of acute bacterial infection is a reasonable biological analog; a mouse model of Alzheimer's disease is a genetic artifact that replicates only a small fraction of the pathological complexity of the human condition.

The translation gap is ultimately an epistemological failure: it reflects the systematic generation of knowledge in a context (the laboratory) that is not representative of the context in which that knowledge is expected to apply (the human patient in a heterogeneous clinical population). Addressing it requires not just better experimental models but better philosophy of science about what constitutes valid evidence for intervention effects — a problem the reproducibility crisis in biomedical research has made newly urgent.