More than 90 percent of drug candidates that succeed in preclinical testing ultimately fail in human trials, a staggering rate of attrition that experts say points to a fundamental flaw in how medicines are developed. After nearly 80 years of relying on animal studies as the gold standard, a growing movement within the biopharmaceutical industry is pushing for a human-centric approach using new tools that better mirror our biology.
These tools, known as new approach methodologies or NAMs, include human induced pluripotent stem cell-derived cells, organoids, organ-on-chip systems, and computational models. Unlike traditional animal models, they are designed to predict drug responses using actual human genetics and physiology. The goal is not just to improve success rates, but also to replace, refine, or reduce animal testing, which is costly, ethically contentious, and often poorly predictive, especially in areas like central nervous system disorders and bone marrow toxicity.
Speaking at a recent industry forum, experts stressed that the shift must begin with human evidence. “If we want to understand human diseases and how drugs work, then we need to focus on the use of human cells, human tissue, and human data,” said Ross Dobie, founder of the Centre for Human Specific Research. He noted that researchers have become adept at artificially creating diseases in animals and curing them, yet those results “rarely translate” to patients.
However, introducing human cells is only part of the solution. Many lab workflows still depend on poorly characterized animal-derived materials like fetal bovine serum and Matrigel, which vary from batch to batch and can undermine experimental reproducibility. Dr. Eric Hill of Loughborough University urges researchers to consider the hidden costs of sticking with these materials. “Think about how many experiments you had to repeat because your data didn’t align with itself,” he said, adding that synthetic, animal-free consumables offer greater control and consistency.
Building Confidence Through Validation Frameworks
Adoption of NAMs, however, hinges on more than better technology. “We don’t need another 100 new NAMs; we need confidence,” said Dr. Pelin Candarlioglu Deacon, founder of 3D and 3Rs. “Industry doesn’t buy technology. They buy anything that will save them from uncertainty in a decision.” She argues that researchers must define what “better” means for each specific use, whether that is predictability, human relevance, speed, or cost, and then validate models accordingly. A liver model for metabolism, for example, cannot be judged by the same criteria as a blood-brain barrier model.
While universal standards for each model are impossible, experts say the process for evaluating them can be standardized. International initiatives are already developing qualification frameworks for microphysiological systems and organ-on-chip technologies. Professor Julie Frearson of IQVIA points to the cosmetics and chemical safety industries, which have successfully used NAMs for years in animal-free toxicity testing. “That should give us confidence that these in vitro models are capable,” she said. “The way they’re validated and used by regulators can accelerate the path for biopharma as well.”
The transition away from animal models will not happen overnight. Instead, human-specific systems will increasingly join a wider weight of evidence alongside computational tools and clinical data. What matters now, experts agree, is building the frameworks that give regulators and drug developers the certainty they need to trust these models with real decisions. The groundwork is underway, and the direction is clear: the next era of drug discovery will be built on human biology.