Elevating Drug Development with NAMs: Meeting the FDA’s Call for Change

Presented by: Alice Stanton, PhD, Harvard Medical School

Neurological conditions are the leading cause of illness, but over 92% of clinically tested CNS drug candidates fail to become treatments. Contributing to this high failure rate is a lack of understanding of human disease mechanisms, technologies to address them, and the restrictive blood-brain barrier (BBB), which most compounds fail to cross. New models are critically needed that more faithfully recapitulate human neurological disease. Recently, we have developed multicellular human mini-brain and Brain-on-Chip platforms, combining iPSC-derived cells of each of the six major brain cell types into an engineered 3D tissue utilizing a brain-mimicking biomaterial scaffold and novel microfluidic platform. Our mini-brain platform forms 3D immuno-glial-neurovascular units with enhanced cell- and tissue-scale phenotypes inclusive of myelinated neuronal networks, microglial immune cells, and BBB, which is perfusable in the chip version. This technology provides potential advantages at each step in preclinical drug development: a tool for discovery of biomarkers and targets, evaluation of therapeutic efficacy and delivery, and personalized drug screening.

Presented by: PJ Devine, PhD, Bristol Myers Squibb

Increasing use of New Approach Methodologies (NAMs) in the pharmaceutical industry has become a major topic of discussion, especially to reduce reliance on animal models and improve success in drug discovery and safety evaluations. The recent FDA roadmap to reducing animal testing in preclinical safety studies and the recent draft guidance from the FDA on streamlining nonclinical safety studies for monoclonal antibodies are cited as endorsements for pharmaceutical companies to rely more on alternative models. In spite of this common goal, animal studies will still be needed for regulatory decision-making for the foreseeable future.

The IQ MPS Affiliate, a collaboration of pharmaceutical and biotechnology companies, is dedicated to advancing complex in vitro models (CIVM) and microphysiological systems (MPS) for drug discovery and development. The IQ MPS influences the field through key activities, including scientific articles, regulatory engagement, pilot qualification projects, and strategic partnerships. To accelerate adoption of CIVM, development of animal cell-based MPS is advocated by the pharmaceutical industry. Animal cell-based CIVM enable cross-species comparisons, investigating potentially species-specific mechanisms, and help establish confidence in the predictive nature of CIVM, since more data are available from non-clinical animal studies than from clinical trials.

NAMs are already widely used for internal drug discovery decisions, with regulatory filings increasingly incorporating them. In both situations, utility depends on clearly defining the NAM’s context(s) of use and providing supporting data to establish its reliability. Through collaboration across the field, NAMs use will continue to evolve and further the reduction of animal use in drug discovery.

Presented by: Shiny Rajan, PhD, Javelin Biotech

Gene therapy has demonstrated remarkable therapeutic potential, yet clinical translation remains limited by the inadequacy of conventional preclinical models. Animal models exhibit species-specific toxicity and poor prediction of vector tropism, failing to reliably forecast human outcomes for either the delivery vector or therapeutic payload. This translational gap is particularly acute for hepatotoxicity, the foremost safety concern in gene therapy, as the liver serves as both the primary site of therapeutic accumulation and a key target tissue.

To address this, we developed a multicellular liver microphysiological system (MPS) with in vivo-like architecture capable of capturing 5 clinical liver failure mode - hepatotoxicity, immune-mediated toxicity, cholestasis, steatosis, and fibrosis, while assessing AAV transduction efficiency and ASO target engagement along with comprehensive safety profiling under physiologically relevant conditions. Extending Liver Tissue Chip (LTC) to a coupled liver-kidney platform (LTC+) under dynamic perfusion further enables multi-organ toxicity assessment, overcoming the inter-organ interaction gaps for predicting systemic distribution, clearance, and organ-specific adverse effects.

AAV serotypes (AAV5, AAV8, engineered capsid) tested on Javelin platform displayed distinct hepatocyte tropism and dose-dependent transient hepatocellular stress, with differential cytokine profiles reflecting serotype-specific immune interactions monitored via 10-biomarker panel. These findings correlated well with published clinical data. For ASOs, the LTC recapitulated receptor-mediated endocytosis over extended exposure periods, demonstrating improved ASO uptake and greater target protein suppression compared to spheroids. The LTC further detected mipomersen-induced DILI at clinically relevant doses and dosin regimen, including an accelerated inflammatory failure in a disease-relevant inflamed model, while spheroids required 100-fold higher concentrations to show any hepatotoxic signal. The LTC+ uniquely captured inotersen-driven kidney-dependent hepatotoxicity, a signal absent without the coupled kidney compartment.

As gene therapy advances toward increasingly complex modalities and broader indications, human-relevant preclinical tools become a necessity. Javelin platform provides physiologically relevant, long-term (>21 days) tools for capturing vector tropism, PK/PD target engagement, immune-mediated responses, and multi-organ toxicity. This system substantially outperforms current models in predicting clinical-relevant outcomes for gene therapy vectors and payloads, positioning them as essential de-risking platforms in gene therapy development pipelines.

Presented by: Karissa Cottier, PhD, BioIVT

Hepatic steatosis is influenced by gene‑dependent pathways that regulate lipid storage and droplet dynamics. Using the long‑term HEPATOPAC micropatterned hepatocyte co‑culture, siRNA was used to examine how knockdown of two steatosis‑associated genes, PNPLA3 and CIDEB, modulates lipid loading under free‑fatty‑acid and high‑sugar conditions. In this system, siRNA delivered at the time of hepatocyte seeding produced robust, sustained gene suppression, allowing target effects to be evaluated within a stable hepatocellular environment. Silencing PNPLA3, reduced lipid loading selectively in I148M donor hepatocytes, demonstrating genotype‑dependent modulation of steatotic responses. Knockdown of CIDEB, a regulator of lipid droplet fusion and growth, significantly decreased overall lipid content and shifted droplet morphology toward smaller populations, indicating reduced capacity for droplet expansion. These findings highlight the utility of siRNA in HEPATOPAC cultures for defining gene‑specific drivers of steatosis and for validating therapeutic targets influencing hepatocellular lipid burden.

Presented by: Bhagwat Prasad, PhD, Cincinnati Children’s Hospital Medical Center and Precision Quantomics, Inc.

As new approach methodologies (NAMs) increasingly replace traditional animal studies, the credibility of complex in vitro systems hinges on rigorous molecular qualification. Quantitative proteomics offers a powerful framework for transforming these models from “black boxes” into benchmarked, biologically validated platforms. By enabling absolute and relative measurement of drug‑metabolizing enzymes, transporters, nuclear receptors, and structural proteins, proteomics provides a direct assessment of the fidelity of microphysiological systems (MPS), including organoids, spheroids, co‑cultures, and organ‑on‑chip platforms, relative to human tissues. Such measurements identify pathway‑level gaps, capture cellular heterogeneity, and reveal fetal‑like or incomplete maturation signatures that limit translational relevance.

This talk will highlight how proteomics-derived scaling factors strengthen in vitro–to–in vivo extrapolation, improve PBPK and QSP/QST model parameterization, and support population-level and longitudinal assessment of model stability. Examples include evaluating DME and transporter abundance in liver and kidney MPS, detecting temporal drift during chronic cultures, and leveraging proteomics to differentiate physiologically relevant platforms from underperforming ones. By establishing quantitative, tissue‑anchored benchmarks, proteomics elevates NAM qualification to a data-driven discipline, accelerating regulatory acceptance, improving cross-platform reproducibility, and ensuring that complex in vitro models authentically reflect human biology.