Join Danielle Thomas, PhD from Elsevier to see how a mechanistic AI framework built on LeapSpace and Elsevier’s agentic connectors connects ADC design choices, such as linker and payload, to organ-level toxicity, with every conclusion attributed to its source.
Safety-driven attrition remains a major barrier in oncology drug development, and ADCs add complexity. Toxicity depends on target biology, linker stability, payload mechanism, and physicochemical properties, and the evidence for each sits in a different chemistry, biology, toxicology, or clinical source.
Danielle Thomas shares an approach that retrieves curated evidence, evaluates its relevance and quality, and builds traceable hypotheses linking linker and payload design to tissue exposure and clinical adverse events, so teams can act earlier on safety, efficacy, and PK decisions.
What you’ll see:
- Why ADC toxicity is hard to anticipate when evidence is split across separate chemistry, biology, toxicology, and clinical silos
- How a mechanistic framework links linker and payload design to tissue exposure and organ-level toxicity
- A worked example: how linker instability and payload lipophilicity align with hepatic and hematologic adverse event patterns seen in similar ADCs
- How preclinical mechanistic signals map to clinical adverse event patterns
- How evidence-attributed outputs can guide design levers such as payload class, DAR, and linker stability
Who should join
Anyone in pharma R&D exploring AI for oncology and ADC discovery, including scientists, informatics and data science teams, and R&D decision-makers
Availability note: Elsevier connectors are available through enterprise licensing.