In non-small cell lung cancer (NSCLC), immunce checkpoint inhibitors (ICIs) that block PD-1 or PD-L1 checkpoint proteins obtain durable responses, but only 20% to 30% of patients respond to this treatment. Today, PD-L1 expression detected by IHC is a widely used, but imperfect, biomarker.
Owkin has developed AI-driven biomarker models to predict treatment response to ICIs and prognosis in NSCLC patients from clinical and histological data.
Our multimodal prognostic model significantly outperforms current FDA-approved biomarkers (PD-L1 expression) in predicting both progression-free (PFS) and overall survival (OS) with an increase of 15 points on the C-index.
Pharma companies can use this model to select high-value subgroups of patients that are most likely to respond to the ICI being tested, therefore improving the statistical power of the trial.