Treatment selection in breast cancer is guided by molecular subtypes and clinical characteristics. However, current tools including genomic assays lack the accuracy required for optimal clinical decision-making. We developed a novel artificial intelligence (AI)-based approach that integrates digital pathology images with clinical data, providing a more robust and effective method for predicting the risk of cancer recurrence in breast cancer patients. Specifically, we utilized a vision transformer pan-cancer foundation model trained with self-supervised learning to extract features from digitized H&E-stained slides. These features were integrated with clinical data to form a multi-modal AI test predicting cancer recurrence and death. The test was developed and evaluated using data from a total of 8,161 female breast cancer patients across 15 cohorts originating from seven countries. Of these, 3,502 patients from five cohorts were used exclusively for evaluation, while the remaining patients were used for training. Our test accurately predicted our primary endpoint, disease-free interval, in the five evaluation cohorts (C-index: 0.71 [0.68-0.75], HR: 3.63 [3.02-4.37, p<0.001]). In a direct comparison (n=858), the AI test was more accurate than Oncotype DX, the standard-of-care 21-gene assay, achieving a C-index of 0.67 [0.61-0.74] versus 0.61 [0.49-0.73], respectively. Additionally, the AI test added independent prognostic information to Oncotype DX in a multivariate analysis (HR: 3.11 [1.91-5.09, p<0.001)]). The test demonstrated robust accuracy across major molecular breast cancer subtypes, including TNBC (C-index: 0.71 [0.62-0.81], HR: 3.81 [2.35-6.17, p=0.02]), where no diagnostic tools are currently recommended by clinical guidelines. These results suggest that our AI test improves upon the accuracy of existing prognostic tests, while being applicable to a wider range of patients.
View on arXiv@article{witowski2025_2410.21256,
title={ Multi-modal AI for comprehensive breast cancer prognostication },
author={ Jan Witowski and Ken G. Zeng and Joseph Cappadona and Jailan Elayoubi and Khalil Choucair and Elena Diana Chiru and Nancy Chan and Young-Joon Kang and Frederick Howard and Irina Ostrovnaya and Carlos Fernandez-Granda and Freya Schnabel and Zoe Steinsnyder and Ugur Ozerdem and Kangning Liu and Waleed Abdulsattar and Yu Zong and Lina Daoud and Rafic Beydoun and Anas Saad and Nitya Thakore and Mohammad Sadic and Frank Yeung and Elisa Liu and Theodore Hill and Benjamin Swett and Danielle Rigau and Andrew Clayburn and Valerie Speirs and Marcus Vetter and Lina Sojak and Simone Soysal and Daniel Baumhoer and Jia-Wern Pan and Haslina Makmur and Soo-Hwang Teo and Linda Ma Pak and Victor Angel and Dovile Zilenaite-Petrulaitiene and Arvydas Laurinavicius and Natalie Klar and Brian D. Piening and Carlo Bifulco and Sun-Young Jun and Jae Pak Yi and Su Hyun Lim and Adam Brufsky and Francisco J. Esteva and Lajos Pusztai and Yann LeCun and Krzysztof J. Geras },
journal={arXiv preprint arXiv:2410.21256},
year={ 2025 }
}