A team of researchers at Dana-Farber has developed an AI-based assessment tool that could help predict the response of clear cell renal cell carcinoma (ccRCC), a form of kidney cancer, to immunotherapy. Using image processing and deep learning, the tool evaluates two-dimensional pictures of tumor samples on pathology slides and identifies previously overlooked features, including tumor microheterogeneity, that are associated with immunotherapy response.
The results of the study, published in Cell Reports Medicine, suggest that pathology slides contain important biological information about ccRCC and potentially other types of tumors. This information could be valuable in understanding the biology of cancer and guiding cancer care.
Renal cell carcinoma is one of the most common cancers worldwide, with the clear cell subtype accounting for a majority of metastatic cases. While some ccRCC tumors respond to immune checkpoint inhibitors (ICIs), there are currently no measures to predict their response to immunotherapy.
The researchers initially trained their AI model to assess a tumor’s nuclear grade, which indicates how abnormal the tumor cells are compared to normal cells. The model was successful in assessing nuclear grade and identifying differences in grade across tumor samples. This inspired the team to expand the model to assess tumor microheterogeneity and immune properties, such as immune infiltration, on pathology slides.
The AI-based tool was then used to analyze pathology slides from patients involved in a randomized clinical trial. The analysis revealed that features such as tumor microheterogeneity and immune infiltration were associated with improved overall survival in patients treated with ICIs.
While the tool is not yet ready for clinical use, it presents a scalable approach to extracting valuable information from pathology slides. Further testing is underway in an ongoing clinical trial involving combination immunotherapy as a first-line treatment for patients with ccRCC.
This study highlights the potential of AI in revolutionizing cancer care by providing valuable insights into tumor biology and guiding treatment decisions.
– Clear cell renal cell carcinoma (ccRCC): A form of kidney cancer characterized by clear or pale cells.
– Immunotherapy: Treatment that uses the body’s immune system to fight diseases.
– Pathology slides: Thin sections of tissues on which samples are mounted for microscopic examination.
– Tumor microheterogeneity: The variation in nuclear grade within a tumor sample.
– Immune infiltration: The extent to which immune cells have penetrated a tumor.
– Cell Reports Medicine – “Artificial Intelligence Identifies New Features of Clear Cell Renal Cell Carcinoma Relevant for Immunotherapy”
– Authors: Jackson Nyman, Eliezer Van Allen, Sabina Signoretti, Toni Choueiri