Accelerating Human-Centric Translational Research using Computational Approaches
Advancing New Approach Methodologies for Efficiency and Accuracy in Drug Development
To date, over 90% of drug candidates still fail in clinical development—largely because traditional animal models fall short in predicting human outcomes. As new therapeutic modalities like antibodies, cell and gene therapies emerge, this translation gap is widening.
Human-relevant NAMs —including advanced in vitro systems, in silico mechanistic models, and AI-driven analytics—offer a more predictive, ethical, and efficient alternative. They not only reduce reliance on animal testing but also enable earlier, data-informed decisions that improve the success rate of clinical translation.
Backed by growing regulatory support, the time to integrate NAMs into drug development is now—transforming preclinical science into a more human-relevant and responsible science.
Physiologically-Based Pharmacokinetic (PBPK) Modeling
PBPK is a computational approach that predicts how a drug moves through the body—how it is absorbed, distributed, metabolized, and eliminated. Unlike traditional methods that rely heavily on animal studies, PBPK models can use lab-based (in vitro) and physicochemical property data to simulate drug behavior in humans before any in vivo testing. This makes PBPK a powerful New Approach Methodology (NAM) for reducing animal use in early drug development.
How PBPK Can Help Reduce Animal Studies?
PBPK models can be built “from the bottom up,” using only in vitro measurements and known biological information. As shown in case studies discussed Mehta et al., 2025, these bottom-up models have successfully predicted real in vivo drug levels for several small molecules and biologics. Also, as discussed, in a priori workflow, PBPK models can be built early using only in vitro and existing biological knowledge to predict drug behavior before any animal studies. By generating these predictions upfront, PBPK helps design only the minimal, most targeted in vivo studies—reducing the number of animals needed while improving study efficiency.
Mechanistic Insights Not Feasible From Animal Data Alone
PBPK can also predict drug levels in specific organs—such as the brain, tumors, or other hard-to-sample tissues. Studies have shown PBPK models accurately estimating brain and tumor exposure where direct measurement in animals would be invasive, limited, or impossible. This helps researchers understand drug behavior while avoiding difficult or high-burden animal procedures.
When Combined With Advanced In Vitro Systems
PBPK can become even more useful when paired with modern in vitro technologies such as organ-on-chip systems. These systems generate human-relevant data that PBPK can translate into whole-body predictions. For example, combining liver-on-chip data with PBPK has accurately predicted human drug clearance and variability across different individuals [Aravindakshan et al., 2025].
PBPK: A Robust Tool NAM for Preclinical Development
Together, these features make PBPK one of the most practical and ready-to-use tools in the NAM space. It supports smarter, smaller, and more ethical preclinical study designs while improving confidence in human-relevant predictions.
Learn more about PBPK models using the following resources (more to come):
A Priori NAM Workflow for Preclinical Studies
In “a priori NAM workflow”, we propose the use of non-animal methods at the very start of preclinical development to generate human-relevant insights before any in vivo studies. By predicting drug behavior upfront, this approach helps reduce the number and scope of animal experiments needed later.
Mechanistic modeling — PBPK and QSP — enables simulation of human pharmacokinetics, pharmacodynamics, efficacy, toxicity, drug-drug interactions, and anti-drug antibody responses, integrating physiological, biochemical, and molecular data to support species translation and optimize study design [Tieghi et al., 2025; Mehta et al., 2025]. Concretely, PBPK models simulate a drug’s absorption, distribution, metabolism, and excretion in virtual human bodies, accounting for species-specific biology, and have already been used to inform FDA drug labels. QSP goes further by capturing human-specific biological networks, allowing clinicians to test dose–response scenarios and combination therapies in “virtual patients”, an approach that is particularly powerful for endpoints where animal models are notoriously poor predictors, such as immunotoxicity, cytokine release, and complex systemic responses [Tieghi et al., 2025].
AI is now accelerating mechanistic approaches further; machine learning is already helping process complex datasets, inform parameter values, and reduce manual calibration steps, making QSP more scalable across programs [Androulakis et al., 2025].
Crucially, these tools are most powerful when deployed before animal studies are designed, not after. Mehta et al. propose an “a priori in silico” workflow where in vitro data and prior knowledge first feed PBPK/QSP models to generate in vivo predictions — which then guide the design of smaller, targeted animal studies rather than the reverse. This iterative loop, in vitro → in silico → optimized in vivo, is precisely the paradigm shift Tieghi et al. envision: computation and human-relevant biology driving the decision, with animal data as the exception rather than the default [Mehta et al., 2025; Tieghi et al., 2025].