Stanford Researchers Develop Tool to Predict Gene Activity in Certain Cancers
Researchers at Stanford Medicine have developed an "artificial intelligence-powered computational program that can predict the activity of thousands of genes within tumor cells based only on standard microscopy images of the biopsy.”
The research team was able to show that the tool “could use routinely collected biopsy images to predict genetic variations in breast cancers and to predict patient outcomes,” which could speed up clinical decision-making.
Which genes a tumor is using to grow and spread is an important data point when deciding which cancer treatment a patient should receive, but historically, “accessing this information often requires costly and time-consuming genomic sequencing.”
Researchers went to work integrating 7,584 cancer biopsies from 16 types of cancer into an AI model. For some cancer types, “the AI-predicted gene activity had a more than 80% correlation with the real gene activity data. In general, the more samples of any given cancer type that were included in the initial data, the better the model performed on that cancer type.”
The tool, nicknamed SEQUOIA, was able to provide the same type of genomic risk score as an FDA-approved tool used to “provide patients with a score of the risk their [breast] cancer” has to recur. As of now, the tool still needs to be “tested in clinical trials and…approved by the FDA” before it can be used in guiding treatment decisions.
About the Author
Matt MacKenzieMatt MacKenzie
Associate Editor
Matt MacKenzie is Senior Staff Writer and Associate Editor of Healthcare Purchasing News (HPN), where he leads daily news coverage and contributes in-depth features on healthcare supply chain, sterile processing, infection prevention and clinical operations. Through industry reporting and expert interviews, he examines the trends, technologies and operational challenges shaping healthcare purchasing and patient care.
Related
Dreamstime Xxl 153188199
