Prof Fred Hirsch speaks to ecancer about the new data validating an artificial intelligence–based approach to PD-L1 scoring in non-small cell lung cancer using samples from the landmark Blueprint studies.

The AI model demonstrated performance comparable to expert pathologists, achieving high agreement across commonly used PD-L1 assays. Notably, agreement was particularly strong at clinically relevant high-expression thresholds, supporting its reliability in identifying patients most likely to benefit from immunotherapy.

These findings highlight the potential of artificial intelligence to enhance consistency, scalability, and efficiency in PD-L1 testing, offering a promising tool to support both clinical decision-making and research applications.