OncoHost to present proteomics-based biomarker model for clinical benefit to immunotherapy in NSCLC patients at SITC 2022

Original source here.

Unique model combining machine learning and proteomics predicts clinical benefit at twelve months based on one pre-treatment blood test.

OncoHost has developed a novel predictive model for clinical benefit of immune checkpoint inhibitors (ICI) in non-small cell lung cancer (NSCLC) patients.

The model is based on proteomic profiling of pre-treatment blood samples, combined with machine learning analysis that accurately predicts clinical benefit probability at twelve months, outperforming the predictive capabilities of the existing PD-L1 biomarker. The study was conducted using OncoHost's PROphet platform, and will be presented as a poster at the Society for Immunotherapy of Cancer's (SITC) 37th Annual Meeting.

The predictive model was developed on a cohort of 339 ICI-treated NSCLC patients taking part in OncoHost's ongoing, multicenter PROPHETIC clinical trial. Plasma samples and clinical data were collected, and deep proteomic profiling was conducted using SomaLogic's SomaScan platform. The model performance was assessed in a blinded validation.

The poster will be presented in the poster hall at SITC on November 10 and 11, 2022, between 9:00 and 21:00 EST and will be available in the Journal for Immunotherapy of Cancer (JITC) supplement.