Serial CTRS (CT Response Score) improves prediction of overall survival compared to RECIST 1.1 and Tumor Volume Change in advanced NSCLC

CHICAGO: Onc.AI, a digital health company developing advanced AI-driven clinical management solutions for oncology, today announced that findings from a recent collaboration with global biopharma company GSK using Onc.AI’s FDA breakthrough-designated Serial CTRS AI model will be presented at the 2025 American Association for Cancer Research (AACR) Annual Meeting. The study externally evaluated Serial CTRS in GSK’s GARNET Phase I clinical trial (NCT02715284) Cohort E, that enrolled patients with advanced non-small cell lung cancer (NSCLC) treated with dostarlimab, GSK’s anti-PD-1 checkpoint inhibitor.

Results, highlighted in a poster presentation during the Predictive Biomarkers 2 session (poster #21, April 27, 2025, 2-5pm), demonstrated that the Serial CTRS biomarker, leveraging routine CT imaging without manual annotations, improved prediction of overall survival (OS) compared to traditional surrogates including RECIST 1.1 response criteria and tumor volume. The analysis was conducted through an independent and blinded retrospective validation study carried out by GSK. In particular, Serial CTRS showed the ability to distinguish patients with intermediate versus high probabilities of 12-month OS (HR: 2.91, 95% CI: 1.16–7.31), outperforming RECIST 1.1 (HR: 1.34, 95% CI: 0.57–3.13) and tumor volume change assessments (HR: 1.00, 95% CI: 0.43–2.34). This enhanced predictive performance supports Onc.AI’s commitment to establishing Serial CTRS as a new standard for automated, AI-based imaging endpoints in the early assessment of treatment response, seamlessly integrating into standard imaging workflows across diverse therapeutic regimens.

Key findings include:

• Serial CTRS improved discrimination between intermediate and high probabilities of overall survival compared to RECIST 1.1 and tumor volume changes.
• Serial CTRS remained a significant predictor of overall survival after adjusting for known prognostic factors, such as age, baseline tumor volume, and PD-L1 Tumor Proportion Score (TPS).

“This important milestone for Serial CTRS builds on a recent breakthrough-designation from FDA,” said Akshay Nanduri, CEO of Onc.AI. “The success of this validation study and ongoing continued collaboration with GSK reflects the strength and robustness of model performance. This can only be achieved using advanced Deep Learning combined with methods for harmonizing diverse imaging data and the breadth of our training data. Onc.AI’s high-quality data on thousands of patients has been sourced from dozens of healthcare systems representing hundreds of clinics across the United States and other international cancer centers, ensuring unparalleled diversity in training, test and validation.”

“As the past head of multiple Phase I clinics at top cancer centers, I believe that Onc.AI’s innovation with Serial CTRS could transform the pharma clinical development process from Phase I to Phase III studies,” said George R. Simon, MD, FACP, FCCP, Vice President of Oncology at OhioHealth.

About Onc.AI

Onc.AI is a digital health company developing AI-driven oncology clinical management solutions using advanced Deep Learning applied to routine diagnostic images. The company’s platform is applied at the point of care by medical oncologists and is also leveraged by global pharmaceutical leaders to accelerate oncology drug development. Onc.AI is backed by premier institutional investors, including: Sandbox/Blue Venture Fund, Action Potential Venture Capital, MassMutual Alternative Investments, Accomplice, Digitalis, KdT, and Life Extension Ventures. Onc.AI is also supported by the National Cancer Institute SBIR program (1R44CA291456-01A1).

For more information, please visit: www.onc.ai.

Fonte: Business Wire