CETSA® and Tricky Targets: Improving GO/NO-GO Decisions When the Biology Won’t Cooperate
A program proposes to advance chemistry against a target where classical assays are slow to establish or weakly predictive—often a multipass membrane protein, a complex-dependent regulator, or a non-enzymatic scaffold. A cellular phenotype exists, but the mechanism remains uncertain. The governance question is simple: are we reducing uncertainty, or accumulating it?
Where uncertainty typically enters
- Assay–biology mismatch: Recombinant constructs may not reflect endogenous complexes, PTMs, localization, or state.
- Cellular exposure uncertainty: In vitro potency does not ensure access to the relevant compartment at realistic concentrations.
- Mechanism ambiguity in phenotypic hits: Phenotype without engagement evidence can drive optimization down an incorrect path.
- Modality complexity (e.g., degraders): Downstream effects can occur without the intended engagement chain being true.
What data changes the quality of decisions
Decisions improve when evidence narrows the gap between “activity” and “mechanism.” CETSA was developed to measure target engagement in cells and tissues by tracking ligand-associated changes in protein thermal stability. In governance terms, CETSA is most useful when it clarifies whether endogenous engagement is observed under relevant conditions and whether apparent activity is more consistent with direct engagement or downstream biology.
How earlier clarity prevents late-stage waste
For tricky targets, teams can spend months optimizing potency and DMPK around activity signals that later prove off-target, non-engaging, or context-limited. Earlier engagement evidence can tighten decision gates:
- Gate 1: Cellular engagement at relevant concentrations
If engagement is not observed within a plausible exposure window, pivot earlier (chemotype, permeability strategy, or target hypothesis) before resources compound. - Gate 2: Engagement in a context that resembles intended biology
Model selection is often an untracked risk. Engagement evidence in endogenous, disease-relevant contexts reduces reliance on engineered systems as primary proof. - Gate 3: Mechanism discrimination (direct vs downstream)
Structured comparisons (including intact vs lysate and, when appropriate, proteome-scale profiling) reduce the risk of advancing series driven mainly by pathway responses rather than target engagement. - Gate 4: Modality-specific validation (degraders)
For degraders, governance benefits from a disciplined evidence chain: engagement of protein of interest and/or E3 in cells, interpreted alongside downstream degradation in the same context.
Why this aligns with governance and accountability
Tricky targets are where narratives can outpace evidence. CETSA-type engagement data supports governance by:
- Making assumptions explicit (engagement evidence is distinct from functional activity evidence).
- Reducing time spent in ambiguous middle stages where spend rises faster than confidence.
- Providing a shared decision language across chemistry, biology, DMPK, and translation: “Do we engage the intended protein in the intended context?”
The goal is decision-quality confidence early enough to prevent avoidable late-stage waste.
Contact
If you are currently managing programs where evidence of activity is advancing faster than confidence in mechanism (especially for complex or difficult targets) we would be pleased to share practical examples of how teams structure decision-ready target engagement evidence early enough to reduce downstream uncertainty.
Pelago Bioscience has a representative in Japan, Stefan Sandstrom, and he is available to support your internal review conversations and help identify the most appropriate next questions to test, without disruption to existing workflows.
Contact: stefan.sandstrom@pelagobio.com