SAN CARLOS, Calif. & WATERLOO, Ontario--(BUSINESS WIRE)--Onc.AI, a digital health company developing AI-driven oncology clinical management solutions, announced that results from a study evaluating the use of serial imaging in advanced non-small cell lung cancer (NSCLC) will be presented at the upcoming Society for Immunotherapy of Cancer (SITC) annual meeting. The multi-institutional study demonstrated improved prediction of overall survival for patients receiving immunotherapy compared to traditional assessment tools.
Early identification of patients likely to benefit from immune checkpoint inhibitors is critical for optimizing cancer treatment. Traditional methods that rely on tumor size alone may not adequately capture early treatment effects. The study showed that using pre-treatment (baseline) and three-month follow-up CT scans to predict overall survival can serve as a novel way to accurately predict long-term outcomes after a few cycles of treatment.
- Serial CT Response Score achieved a C-index of 0.734, outperforming RECIST (0.631) and tumor volume change measurements (0.679) in predicting overall survival.
- Serial CT Response Score remained a significant predictor of overall survival after multivariate adjustment with other known predictors, including tumor volume change, PD-L1 TPS, age, sex, and line of therapy.
- For patients identified as having stable disease, the serial CT Response Score provided better prognostication over tumor volume change, with a 12-month OS ROC-AUC of 0.74 (0.65-0.82) compared to 0.62 (0.52-0.72).
“Our serial imaging response score is setting a new standard in early outcome assessment for NSCLC immunotherapy, allowing us to move beyond conventional metrics and deliver a more precise, predictive insight into patient survival,” added Petr Jordan, Chief Technical Officer of Onc.AI
About Onc.AI
Onc.AI is a privately held digital health company developing a precision oncology clinical management platform with the goal of transforming clinical decision-making. Onc.AI was founded based on a simple idea: radiomics-based machine learning models can radically improve the ability to predict patient response to PD-(L)1 immunotherapy. Onc.AI’s platform leverages deep learning to extract and quantify complex imaging features that are not visible to the naked eye, revealing critical insights into tumor biology. By utilizing these subtle radiographic markers, Onc.AI’s advanced biomarker provides oncologists with more precise information to help guide personalized treatment decisions. Onc.AI aims to develop a first-in-class radiomic solution to address the unpredictable response and non-response to PD-(L)1 ICI therapy to improve outcomes for patients and reduce the cost burden on healthcare systems and public/private payers.
