#United States #Caris Life Sciences #AI Models #ECOG-ACRIN #Philadelphia and Irving

ECOG-ACRIN and Caris Life Sciences: Pioneering AI in Breast Cancer Prognosis

Today at the San Antonio Breast Cancer Symposium (SABCS), groundbreaking findings were unveiled from a collaboration between the ECOG-ACRIN Cancer Research Group and Caris Life Sciences. This project is set to revolutionize the way recurrence risks for early-stage breast cancer patients are assessed by utilizing advanced artificial intelligence (AI) technologies.

Enhancing Recurrence Risk Assessment

The collaboration aims to enhance the accuracy of predicting recurrence risks in early-stage breast cancer patients by harnessing the power of AI to integrate various data forms, including histopathologic imaging, clinical information, and molecular profiling. This holistic approach comes from Caris's extensive database derived from the TAILORx tissue biorepository, which is well-known for its rigorous annotation.

ECOG-ACRIN's deep expertise in clinical trials synergizes with Caris's innovative MI Cancer Seek® platform, showcasing the transformative potential of AI in oncological care. The aim is to facilitate more precise and personalized treatment choices based on comprehensive patient data.

A Unique Challenge

The challenge is significant, as approximately 310,720 new breast cancer cases are diagnosed annually in the U.S., with 60% categorized as early-stage, representing a diverse patient population where treatment decisions often rely on uncertain risk assessments.

Dr. Peter J. O'Dwyer, Co-Chair of ECOG-ACRIN, emphasized that this collaboration is a methodological interplay of datasets from the crucial TAILORx trial, showcasing a commitment to personalizing medicine and improving treatment outcomes through the power of AI.

Results of Multimodal AI Models

The innovative AI models developed showed significant enhancements in prognostic performance over existing methods, marking a step forward towards individualized treatment strategies. As Caris's Chief Medical Officer, Dr. George W. Sledge, Jr., puts it, the integration of these various types of data allows for a much richer understanding of recurrence risks in breast cancer, moving beyond simplistic diagnostic methods.

The models were critically evaluated during today’s sessions at SABCS. For instance, one study led by Dr. Joseph A. Sparano showcased a model that prospectively validated the integration of pictorial pathology, clinical factors, and an expansive molecular dataset from TAILORx's 4,462 tumor specimens. This study notably included a panel of 42 tumor genes related to breast cancer recurrence.

Creating Better Diagnostic Tools

The ultimate goal is to develop next-generation diagnostic tests particularly for women with HR-positive, HER2-negative, and node-negative breast cancer. The combined approaches present an opportunity to accurately evaluate recurrence risk, especially for late recurrences that might occur five or more years post-diagnosis. Dr. Sparano remarked that while the Oncotype DX test has proven useful, its limitations necessitate the advancement of more precise tools.

Another project discussed involved a multitask deep learning model aimed at determining distant recurrence risk among early-stage HR+ breast cancer patients. This model, originally developed and validated in the NSABP B-42 trial, demonstrates the efficacy of using AI to not only hone in on clinical risk factors but also offer insights for extended therapy beyond the conventional five years.

The Road Ahead

As these innovative approaches are validated, they signify a potential shift in standard practice in breast cancer prognostication. Both ECOG-ACRIN and Caris Life Sciences underscore a firm commitment to bridging the gap between advanced research and clinical application, leading towards improved outcomes for breast cancer patients.

For more information about ECOG-ACRIN and Caris Life Sciences, visit eco.acrin.org and carislifesciences.com respectively.