Caris Life Sciences' AI-Based Tissue of Origin Predictions Improve Diagnostic Accuracy for Cancers of Unknown Primary and Misclassified Tumors

Caris Life Sciences has published a study on its AI-based tissue of origin predictions, Caris GPSai, which uses deep learning to significantly improve diagnostic accuracy for cancers of unknown primary and misclassified tumors. The tool demonstrated 95% accuracy in identifying tumor tissue of origin and successfully reported a tissue of origin in 84% of CUP and 96.3% of non-CUP cases. In clinical use, GPSai changed the diagnosis in 704 patients, impacting treatment eligibility in 86.1% of these cases.

Caris Life Sciences has published a groundbreaking study on its AI-based tissue of origin prediction tool, Caris GPSai, in the AACR's Cancer Research Communications Journal. The study demonstrates that GPSai, which utilizes deep learning, significantly improves diagnostic accuracy for cancers of unknown primary (CUP) and misclassified tumors.

Caris GPSai was trained on over 200,000 cases and achieved 95.0% accuracy in identifying tumor tissue of origin in non-CUP cases. During validation across 97,820 cases, it successfully reported a tissue of origin in 84.0% of CUP and 96.3% of non-CUP cases. In clinical use over eight months, GPSai changed the diagnosis in 704 patients, impacting treatment eligibility in 86.1% of these cases based on Level 1 clinical evidence. Notably, 53.6% of surveyed physicians modified treatment plans based on these findings.

The enhanced Caris GPSai is a part of Caris' comprehensive molecular profiling platform, leveraging deep learning and whole exome and whole transcriptome sequencing (WES/WTS). This approach supports more informed treatment decisions and potentially improves patient outcomes by accurately identifying primary tumor sites.

Caris Life Sciences, a leading precision medicine pioneer, is actively developing innovative solutions to transform healthcare. The company's platform combines comprehensive molecular profiling with advanced AI and machine learning algorithms to analyze and unravel the molecular complexity of diseases.

The publication can be viewed in its entirety on the Caris Life Sciences website.

References:
[1] https://www.prnewswire.com/news-releases/caris-life-sciences-publishes-study-showing-its-multi-layer-ai-based-tissue-of-origin-predictions-are-best-in-class-and-identify-when-patients-have-been-misdiagnosed-302521431.html
[2] https://www.stocktitan.net/news/CAI/caris-life-sciences-publishes-study-showing-its-multi-layer-ai-based-ofzxwjioqhtk.html