Cedars-Sinai AI Tool Cuts Cost and Time for Cancer Gene Profiling
Researchers at Cedars-Sinai have developed an AI tool that predicts gene expression patterns in tumors from digital images of biopsy slides, potentially making personalized cancer treatment faster and cheaper for more patients.
The tool, called Path2Space, analyzes microscope images to estimate which genes are active at different locations within a tumor. Traditional spatial gene expression profiling takes weeks and costs thousands of dollars. Path2Space produces results in minutes at a fraction of the cost.
The team trained Path2Space using breast cancer patient data where both biopsy slides and spatial sequencing results were available. They then tested the system on three additional patient datasets to verify accuracy.
"For each sample, we predicted the spatial expression of almost 5,000 genes, and the predictions matched the measured expression well across all three patient groups," said Eldad Shulman, a research fellow at the National Cancer Institute who co-authored the study.
Larger Studies Now Possible
The high cost of traditional profiling has limited research to small patient groups. Before Path2Space, the largest spatial tumor dataset available contained about 30 patients. The new tool allows researchers to analyze slides from thousands of patients.
This scale matters because tumors vary internally. Gene expression differs across regions within the same tumor, and identifying which patterns predict treatment response requires analyzing many cases.
Path2Space also flags spatial biomarkers-patterns of gene activity that could guide treatment decisions or identify patients at higher risk of poor outcomes. Researchers found specific spatial patterns that predict how patients respond to therapy.
Next Steps: Clinical Trials and Other Cancers
The team is finalizing a study applying Path2Space to head and neck cancer. The tool can be adapted to other cancer types once trained on relevant patient data.
Eytan Ruppin, deputy director of the Translational Research Institute at Cedars-Sinai and senior author, said the next priority is clinical trials. "It represents an exciting development in a growing field and has to be tested carefully. But we are hopeful that it could make an impactful contribution to science and to patient care."
Currently, Path2Space analyzes groups of 10 to 20 cells together. The goal is eventually assessing individual cells for greater precision.
The research appears in Cell. For professionals working in cancer research or translational medicine, this represents a practical shift toward larger, faster studies. Learn more about AI for Healthcare and AI for Science & Research.
Your membership also unlocks:
