A new study published in the journal npj Breast Cancer demonstrates that an innovative artificial intelligence (AI) model offers more accurate predictions of recurrence risk in patients with early-stage breast cancer compared to the widely used 21-gene Recurrence Score. This research, conducted by a team from the ECOG-ACRIN Cancer Research Group (ECOG-ACRIN) and Caris Life Sciences®, specifically targets hormone-receptor-positive (HR+) and HER2-negative disease, the most common subtype of breast cancer, which accounts for about half of all breast cancer cases in the United States.
“Powered by artificial intelligence integrating clinical, molecular, and histopathology data, this new test provides more reliable prognostic information for breast cancer recurrence,” said lead author Joseph A. Sparano, MD , of the Icahn School of Medicine at Mount Sinai.
The new model, named IICM+, combines digitized pathology images of the tumor with features such as patient age, tumor size and grade, and molecular information from an expanded 42-gene panel. The team developed and independently validated IICM+ using tumor specimens and more than a decade of clinical outcomes from 4,429 participants in ECOG-ACRIN’s TAILORx breast cancer trial.
TAILORx established the value of the 21-gene Recurrence Score, also known as the Oncotype DX Breast Recurrence Score® test (Exact Sciences Corporation), for estimating the risk of recurrence and guiding chemotherapy use in patients with HR+, HER2-negative, lymph node-negative early breast cancer. However, the Recurrence Score is more effective at predicting recurrence within the first 5 years after diagnosis than for later recurrence. More than half of distant recurrences in this type of breast cancer occur more than 5 years after initial surgery, sometimes 10, 15, or even more than 20 years later.
Researchers developed the IICM+ model using data and tumor specimens from 2,808 participants in the TAILORx study. They then evaluated the completed model in an independent group of 1,621 TAILORx participants whose data had not been used to develop the model. Both groups had a median follow-up of over 11 years. Investigators evaluated overall distant recurrence (the return of cancer in other parts of the body), as well as early recurrence (within 5 years) and late recurrence (after 5 years).
Enhancing recurrence risk prediction
In the independent validation group, IICM+ significantly outperformed the 21-gene Recurrence Score in distinguishing between patients at different risks of recurrence, as measured by the C-index (higher scores indicate better performance):
Identifying differences in risk within Recurrence Score groups
In patients with a Recurrence Score of 0–25, which indicates a lower genomic risk, IICM+ identified a subset with a higher observed risk of distant recurrence. Conversely, in patients with a Recurrence Score of 26–100, indicating a higher genomic risk, IICM+ found a subset whose observed risk of recurrence was lower than what the Recurrence Score alone would have suggested.
“Using multi-modal data (clinical data, Next Generation Sequencing data, and data obtained from imaging), integrated through modern artificial intelligence techniques, allowed us to create a truly novel means of interrogating breast cancers for important prognostic information,” said George W. Sledge, Jr., MD , EVP and Chief Medical Officer of Caris.
Unlocking the long-term scientific value of TAILORx
The TAILORx biorepository contains tumor specimens, detailed information about each patient’s cancer, and meticulously collected treatment and long-term outcome data. As part of the public-private partnership with ECOG-ACRIN, Caris converted archived TAILORx pathology slides into digital images and then applied modern AI and deep learning methods to extract new information from specimens collected years ago.
“TAILORx continues to yield important new insights years after its original findings—a testament to the enduring value of Cooperative Group research,” said Peter J. O’Dwyer, MD , ECOG-ACRIN Group Co-Chair. “Applying today’s technologies to this exceptional research resource creates opportunities to further individualize breast cancer care, and with NCI’s concurrence and funding from Caris, this public-private partnership is a model for accelerating progress.”
The findings do not establish the use of IICM+ for selecting or changing treatment at this time. Further studies are needed—and are in development—to validate the model in other patient populations and determine whether its use can guide treatment decisions and improve patient outcomes.
The full study, “ An Artificial Intelligence (AI) model integrating multiscale foundation model histopathology representations with molecular and clinical features predicts early and late distant recurrence in TAILORx ,” is available in npj Breast Cancer (DOI: 10.1038/s41523-026-01022-y).
About TAILORx
The Trial Assigning Individualized Options for Treatment (Rx), called TAILORx, was one of the world’s most influential breast cancer clinical trials ever conducted. It involved 10,273 women with HR+, HER2-negative, lymph node-negative early-stage breast cancer and helped define which women could be treated effectively without chemotherapy.
TAILORx was designed and conducted by ECOG-ACRIN with funding from the National Cancer Institute, part of the National Institutes of Health, and participation from the Alliance for Clinical Trials in Oncology, the Canadian Cancer Trials Group, NRG Oncology, and the SWOG Cancer Research Network. Additional funding was provided by the Breast Cancer Research Foundation, Susan G. Komen, and the Breast Cancer Research Stamp (U.S. Postal Service).
About ECOG-ACRIN
The ECOG-ACRIN Cancer Research Group is one of the world’s leading scientific organizations, known for innovative studies involving adults with cancer or at risk of developing the disease. The group has an expansive membership that includes over 1,400 hospitals and cancer centers in the United States and around the worldwide, and nearly 21,000 researchers and advocates. Its research is funded primarily by the National Cancer Institute, one of the United States’ National Institutes of Health. To learn more, visit www.ecog-acrin.org and follow us on X , Facebook , LinkedIn , YouTube , Instagram , and BlueSky .
npj Breast Cancer
People
An Artificial Intelligence (AI) model integrating multiscale foundation model histopathology representations with molecular and clinical features predicts early and late distant recurrence in TAILORx
5-Aug-2026