FDA Signals a New Era for Drug Development — VeriSIM Life Is Already There
5.16.2025
The FDA’s recently released Roadmap to Reducing Animal Testing in Preclinical Safety Studies outlines a bold and long-awaited shift in regulatory science. With the agency now actively encouraging the use of New Approach Methodologies (NAMs) including artificial intelligence (AI)-based modeling, human organ-specific testing, and real-world human data — the path forward for safer, faster, and more ethical drug development is clearer than ever.
At VeriSIM Life (VSL), this is not new territory. It’s the very reason we exist.
Dr. Jo Varshney founded VeriSIM Life after witnessing firsthand the disconnect between preclinical models and real-world human outcomes. With a background in veterinary medicine, genomics, and translational science, she became deeply aware of how animal models too often failed to predict human drug responses—costing lives, time, and billions in R&D. Jo was driven by a simple but radical idea: what if we could simulate how drugs behave in the human body, before a single clinical trial?
That vision became VeriSIM Life.
From the beginning, VSL has pioneered a mission-driven approach rooted in the belief that advanced simulations can replace outdated, animal-centric paradigms that have limited applicability to humans. The FDA now explicitly affirms that AI-based computational modeling can reliably simulate how monoclonal antibodies (mAbs) distribute through the human body, predict side effects based on molecular features, and accelerate therapeutic delivery without compromising safety. That’s exactly what VSL already enables and more!
At VeriSIM Life, we leverage AI to simulate and predict monoclonal antibodies (mAbs), small molecules, and their conjugates (i.e., ADCs - see our publication Knowledge-enhanced AI to Supercharge ADC Development for Treatment of Cancer » ADC Review) distribution in human systems, not just animals. Our platform integrates structure, and amino acid sequences, and post-translational modifications to anticipate how biologics behave across human organs. From there, we map dose-exposure relationships and connect them with immunogenicity and toxicity thresholds, providing actionable insights that guide first-in-human dosing decisions. In addition, we integrate target-engagement and potency insights to guide efficacy considerations. Using our technology, we have advanced a client’s program in mAb development by providing them with over 82% accurate insights on mAb clearance in cynomolgus monkeys from its amino acid sequence. We have also developed accurate models for different immunogenic responses (T-cell activation - 95% accuracy, MHC binding - 82% accuracy, and others) and specific target binding to guide clients’ mAb safety and potency assessment programs - including for bispecific antibodies.
The result? A robust and clinically grounded Translational Index™ (TI) that balances dose-exposure, efficacy, and safety for each candidate. This TI isn’t just theoretical — we’ve validated it across over 60 diseases, with an 81% accuracy rate in predicting clinical success. And, we have gone beyond - we have integrated a generative chemistry engine with Translational Index™ to design novel compounds and biologics with improved likelihood of clinical success!
What makes our approach powerful is its generalizability. We’ve built AI-driven expert systems for drug distribution and safety that integrate disease biology, human physiology, and drug chemistry to drive our platform’s applicability to not only oncology but also extend to other disease areas including cardiometabolic, central nervous system (CNS), and other indications. Whether it's small molecules or antibody-based therapies, our platform has helped accelerate development timelines by several months, with four client programs now in clinical trials substantially ahead of schedule as a result of our work.
Moreover, our models provide critical granular insights into a mAb, biologic, or drug behavior in humans: we accurately predict ADME (Absorption, Distribution, Metabolism, Excretion) and immunogenicity risks with over 80% reliability, a critical threshold echoed in the FDA’s own NAM validation criteria.
The FDA’s statement that “we can get safer treatments to patients faster and more reliably, while also reducing R&D costs and drug prices” echoes the core belief that has guided VeriSIM Life since day one. This isn’t just regulatory progress—it’s long-overdue validation. The industry is finally recognizing what we've known all along: most animal models are no longer necessary—or sufficient—for predicting human outcomes.
As regulators, innovators, and drug developers align around this next-generation approach, VeriSIM Life isn’t adapting to a new standard—we built it.
AtlasGEN Novel Drug Designer efficiently identifies new molecules based on favorable chemical properties and simultaneously evaluates them for clinical safety and effectiveness based on biology using VeriSIM Life’s groundbreaking Translational Index™ technology.
Iterative ranking of target engaging hits by their Translational Index values reduces the time and cost associated with other computational hit discovery approaches, while compressing the hit-to-lead refinement and optimization process by pre-validating hits for translatability. This unique integration dramatically reduces the number of compounds to evaluate in subsequent experimental research.
The ability to generate target engaging compound hits depends largely on the size of the molecular search library. The AtlasGEN search space is supported by more than 1 trillion compounds. Generative AI and deep learning techniques deliver unprecedented discovery capability.
AtlasGEN accurately predicts protein-lygand binding affinity by combining geometric conformer analysis with machine learning based on experimental data 72% more efficiently than molecular docking-based approaches.
.png)
