Currently, animal models are used along with a limited in vitro human models to predict the first-in-human safe dose and estimating the ideal clinical efficacious dose. Nevertheless, clinical attrition rates following IND nomination is very high (75% or higher depending on therapeutic area) with $2.5 billion cost for one drug to reach the market. Animal efficacy and toxicity models are poorly predictive of clinical efficacy and safety, therefore, requiring novel alternate models.
In the last two decades, there is extensive interest in building 3D human models those mimic human physiology and pharmacology, for effective clinical translation and for rational clinical trial design. Two major factors those impact drug disposition in humans are drug metabolism and drug transport. The drug metabolizing enzymes (DME) and drug transporters (DT) of humans are functionally different from animal equivalents and thus contribute to variation in drug pharmacokinetics (PK) from animals to humans and associated efficacy and safety predictions in humans. Additionally, human DME and DT have intra-ethnic and inter-ethnic polymorphic differences as well as can be modulated by environment and life style factors. In many cases the disease state, in particular, inflammatory mediators down-regulate the gene expressions of both DME and DT. Drug-drug interactions, where poly pharmacy of geriatric patients is associated with serious and fatal adverse events or modifying efficacy has been another serious post-marketing issue and major cause of withdrawal of drugs from market (or black-box warnings limiting their use). 3D human models can retain these properties of DME and DT, therefore, next generation models to predict human PK, efficacy and toxicity as well as for rational clinical trial design should be based on human models rather than animal models. The author will walk through the paradigm shifts in drug discovery and development, and need for human derived models to bridge the translational gaps.