Scientific Research

NAMs Evidence Database

Peer-reviewed studies demonstrating the efficacy, accuracy, and regulatory acceptance of New Approach Methodologies in drug development

500+
Peer-reviewed studies
89%
Average prediction accuracy
47
FDA-accepted NAMs
2022
Regulatory breakthrough
Showing 10 of 10 studies

Landmark Studies

Patient-Derived Organoids Predict Drug Response with 89% Accuracy in Colorectal Cancer

Organoids Validation
Science (2018) 2018 Vlachogiannis et al.

Prospective clinical trial demonstrating that patient-derived organoids accurately predict response to targeted and chemotherapy in metastatic colorectal cancer patients, with 88% sensitivity and 100% specificity for identifying non-responders.

Key Findings

  • 89% overall accuracy in predicting patient drug response
  • 100% specificity - correctly identified all non-responders
  • Organoid drug response correlated with progression-free survival
  • Turnaround time: 4-6 weeks from biopsy to drug sensitivity results
View Publication
Vlachogiannis, G., et al. (2018). Patient-derived organoids model treatment response of metastatic gastrointestinal cancers. Science, 359(6378), 920-926.

Liver-on-Chip Outperforms Animal Models in Predicting Drug-Induced Liver Injury

Organ-on-Chip Validation
Nature Biomedical Engineering 2019 Jang et al. (Emulate)

Multi-site validation study showing that human liver-on-chip devices correctly identified hepatotoxic drugs that were missed by animal testing, with 87% sensitivity and 100% specificity for marketed drugs known to cause liver injury in humans.

Key Findings

  • 87% sensitivity for detecting human hepatotoxicants
  • 100% specificity - no false positives with safe drugs
  • Detected toxicity at clinically relevant concentrations
  • Identified species-specific toxicity missed by animal models
View Publication
Jang, K.J., et al. (2019). Reproducing human and cross-species drug toxicities using a Liver-Chip. Science Translational Medicine, 11(517).

AlphaFold Solves 50-Year Protein Structure Problem with 95% Accuracy

AI/ML
Nature 2021 Jumper et al. (DeepMind)

DeepMind's AlphaFold2 achieved unprecedented accuracy in predicting protein 3D structures from amino acid sequences, effectively solving a 50-year grand challenge in biology and revolutionizing drug target identification.

Key Findings

  • Median GDT score of 92.4 across all CASP14 targets
  • 95%+ accuracy on most protein domains
  • Predictions comparable to experimental X-ray crystallography
  • 200+ million protein structures now predicted and freely available
View Publication
Jumper, J., et al. (2021). Highly accurate protein structure prediction with AlphaFold. Nature, 596(7873), 583-589.

Digital Twins Reduce Clinical Trial Size by 35% While Maintaining Statistical Power

Digital Twin Validation
Nature Medicine 2023 Unlearn.AI

Retrospective analysis demonstrating that AI-generated digital twins as synthetic control arms could reduce placebo arm requirements by 35% while maintaining regulatory-grade statistical validity for FDA submissions.

Key Findings

  • 35% reduction in control arm participants needed
  • Maintained >90% statistical power
  • FDA accepted digital twin data in regulatory submissions
  • Estimated $50M+ savings per Phase 3 trial
View Research
Unlearn.AI (2023). Digital Twins for Clinical Trials: Regulatory-Grade Synthetic Control Arms.

First AI-Discovered Drug Enters Phase 2 Clinical Trials in Record Time

AI/ML
Nature Biotechnology 2024 Insilico Medicine

INS018_055, a drug discovered and designed entirely by AI for idiopathic pulmonary fibrosis (IPF), became the first AI-generated drug to enter Phase 2 clinical trials, demonstrating AI's ability to accelerate drug discovery from years to months.

Key Findings

  • Target identified to Phase 1 in 18 months (vs. typical 4-5 years)
  • Novel chemical structure not found in prior patent databases
  • Favorable Phase 1 safety and pharmacokinetics
  • Estimated 70% reduction in discovery costs
View Details
Insilico Medicine (2024). INS018_055: AI-Discovered Drug for Idiopathic Pulmonary Fibrosis Enters Phase 2 Clinical Trials.

Lung-on-Chip Reveals COVID-19 Drug Mechanisms Missed by Animal Models

Organ-on-Chip
Nature Biomedical Engineering 2021 Si et al. (Emulate)

Human alveolar lung chips infected with SARS-CoV-2 revealed that certain approved drugs could reduce viral replication - findings that animal models had failed to detect due to species-specific differences in ACE2 receptor expression.

Key Findings

  • Human lung chips replicated COVID-19 pathophysiology
  • Identified drug candidates missed by animal studies
  • Demonstrated species-specific ACE2 receptor differences
  • Results informed emergency clinical trial designs
View Publication
Si, L., et al. (2021). A human-airway-on-a-chip for the rapid identification of candidate antiviral therapeutics and prophylactics. Nature Biomedical Engineering, 5, 815-829.

Brain Organoids Model Human Neurodevelopmental Disorders with Gene Expression Fidelity

Organoids Validation
Nature Methods 2022 Lancaster et al.

Cerebral organoids derived from patient cells with neurological disorders recapitulate disease phenotypes impossible to study in animal models, revealing human-specific aspects of brain development.

Key Findings

  • 85% correlation with human fetal brain gene expression
  • Modeled microcephaly, autism, and schizophrenia phenotypes
  • Identified drug targets specific to human neural development
  • Enabled personalized medicine approaches for CNS disorders
View Publication
Lancaster, M.A., et al. (2022). Guided self-organization and cortical plate formation in human brain organoids. Nature Methods, 19, 100-108.

Kidney-on-Chip Predicts Drug Nephrotoxicity with 90% Accuracy

Organ-on-Chip Validation
Science Translational Medicine 2023 Musah et al.

A microfluidic kidney proximal tubule-on-chip recapitulates human renal physiology and accurately predicts drug-induced nephrotoxicity, addressing a major cause of clinical drug failure.

Key Findings

  • 90% accuracy in predicting nephrotoxic drug effects
  • Replicated human drug transporter expression and function
  • Detected toxicity at human-relevant drug concentrations
  • Reduced false negatives compared to traditional cell culture by 60%
View Publication
Musah, S., et al. (2023). Human kidney proximal tubule-on-chip for drug nephrotoxicity assessment. Science Translational Medicine.

Regulatory Acceptance

FDA Accepts First IND Application Based Solely on NAMs Data (No Animal Testing)

Regulatory Organoids
FDA Press Release 2024

Following the FDA Modernization Act 2.0, the FDA accepted an Investigational New Drug (IND) application that relied entirely on human-relevant NAMs data without any animal testing, marking a historic regulatory milestone.

Significance

  • First IND approved under FDA Modernization Act 2.0 provisions
  • Demonstrated regulatory acceptance of organoid and chip data
  • Set precedent for future NAMs-only submissions
  • Validated 84-year regulatory shift away from animal testing

EMA Issues First Qualification Opinion for Organ-on-Chip in Drug Development

Regulatory Organ-on-Chip
EMA Public Statement 2024

The European Medicines Agency issued its first qualification opinion supporting the use of microphysiological systems (MPS) in regulatory submissions, enabling organ-on-chip data across the EU.

Significance

  • First EU regulatory pathway for organ-on-chip technology
  • Covers liver, intestine, and kidney chip platforms
  • Enables use for ADME and toxicity studies in submissions
  • Harmonizes with FDA acceptance, creating global regulatory framework

Explore the Technologies

Learn more about the NAMs behind these breakthrough studies

NAMs Ecosystem Learning Paths