Titles and Affiliations
Chief, Computational Oncology Service
Department of Epidemiology and Biostatistics
Research area
Developing AI models to improve prediction of breast cancer recurrence risk and treatment response.
Impact
Artificial intelligence models trained on biological and medical data hold immense potential for assisting physicians in understanding and predicting patient outcomes. However, to ensure safety and utility, these AI models must be trained and deployed with the utmost care. Dr. Shah and his team have recently discovered that AI models trained on multimodal data –encompassing the tissue, cellular, and molecular levels—outperform models trained on a single data type. Dr. Shah aims to develop AI models using such multimodal data to predict the risk of recurrence after treatment in patients newly diagnosed with breast cancer and to assess the drivers of treatment resistance in metastatic breast cancer.
Progress Thus Far
Over the past year, Dr. Shah and his team made progress in using AI to improve how breast cancer is diagnosed and treated. They developed an AI model called Orpheus that can predict a patient’s risk of cancer recurrence by analyzing more than 4,000 tumor images from early breast cancer patients. Orpheus accurately identified high-risk patients across multiple hospitals and even performed better than a leading commercial test for identifying patients unlikely to develop metastatic disease. Unlike many AI tools, Orpheus is interpretable, meaning that clinicians can identify specific biological features, such as immune cell activity and tumor growth rates, that are driving the predictions. In parallel, the team created new computational tools to improve treatment for patients with HER2-positive metastatic breast cancer. One model uses tumor data to predict which tumors will respond best to a class of drugs called antibody-drug conjugates (ADCs), especially those with low HER2 levels who might otherwise not receive this therapy. They also studied why some tumors stop responding to ADCs by combining advanced imaging and gene expression data, uncovering patterns that reveal how cancers adapt to resist therapy.
What’s next
Dr. Shah and his team will continue to enhance the accuracy of their AI model’s risk recurrence predictions in early breast cancer. The team will integrate features from tumor sections that are stained to highlight both tissue architecture and specific molecular features. To leverage the complementary information, they will separately extract features and then merge them to train the AI model. They expect this “fusion” approach to improve accuracy in stratifying early breast cancer patients by recurrence risk, thereby enabling more precise identification of high-risk patients for tailored treatment or monitoring.
The team will continue to refine AI models for metastatic breast cancer to identify which patients are most likely to benefit from anti-HER2 ADCs. These models will analyze detailed images of tumor tissue taken before and after treatment and link them to patient outcomes, while also incorporating genetic data to make predictions more accurate. The team will also continue to explore why some tumors stop responding to ADCs. By studying tumor samples from patients before and after treatment, the team will track how cancer cells change at the molecular level as they become resistant. Finally, they will test and confirm these findings in larger patient groups.
Biography
Sohrab Shah, PhD is the Chief of Computational Oncology in the Department of Epidemiology and Biostatistics and Director of The Halvorsen Center for Computational Oncology at Memorial Sloan Kettering Cancer Center. He holds the Nicholls-Biondi Endowed Chair in Computational Oncology at MSK and is a Susan G. Komen Scholar. Dr. Shah has a joint appointment as a professor in the Department of Physiology, Biophysics, and Systems Biology at Weill Cornell Medical College. Dr. Shah holds a PhD in computer science and started his independent research laboratory in 2009. He oversees the research activity of the Computational Oncology program, which includes eight tenure-track principal investigators, and three laboratory-track members dedicated to advancing computational and data science research for cancer biology and clinical programs.
Dr. Shah’s research laboratory focuses on understanding the principles and processes of cancer evolution. To this end, his laboratory develops and applies computational approaches encompassing advanced machine learning, artificial intelligence, and Bayesian statistical methods combined with single-cell measurement technology. This work has led to advances in understanding drug resistance and tumor evolution in breast cancer and promising new directions to study how breast cancers evolve during and after therapy. Dr. Shah has recently initiated a new program of research in multimodal data integration to capitalize on the institution’s vast clinical and diagnostic data resources, including genomics, radiologic imaging, digital histology, treatment response, and high-resolution single-cell genomics. Dr. Shah is a principal investigator in several national and international collaborative research programs, including Breakthrough Cancer, the NIH-funded Center of Excellence in Genomic Sciences program called the Center for Integrated Cellular Analysis, and the CRUK Grand Challenge program entitled IMAXT: Imaging and Molecular Annotation of Xenografts and Tumors. His BCRF project will focus on artificial intelligence approaches to study the spatial biology of drug resistance in breast cancer.
