Researchers from The University of Texas MD Anderson Cancer Center demonstrated on June 30 that an artificial intelligence-based analysis of tumor biopsies can predict responses to immunotherapy in patients with rare cancers. The findings were published in the Journal for ImmunoTherapy of Cancer.
The study was led by Aung Naing, M.D., professor of Investigational Cancer Therapeutics. This work builds on previous research identifying features within the tumor microenvironment that may predict immunotherapy response, even among patients who do not have known markers typically associated with such responses.
Naing's earlier publication identified two main features indicating patient response to immunotherapy: the number of immune cells present within the tumor before treatment and changes in immune cell infiltration during treatment. Traditionally, manually counting these cells on pathology slides is labor-intensive and challenging to scale across large numbers of samples. However, AI tools can rapidly perform these analyses using standard pathology slides already collected as part of routine care. In this study, the AI-based approach quickly generated measurements and tracked changes over time across multiple biopsies from individual patients.
The results showed that both an increase in tumor immune infiltration and a decrease in tumor content independently predicted better outcomes for patients receiving immunotherapy. When combined, these signals provided a stronger indication—reflecting both an active immune response and a reduction in tumor burden. Patients displaying favorable signals had a 64% lower risk of disease progression or death, and lived nearly four times longer on average (median survival of 42 months versus 10 months) compared to those without these markers.
While researchers say these results are promising, they caution that further validation is needed before implementing this approach clinically. "While this AI-powered approach needs validation, this is an exciting step forward because it shows that meaningful insights can be extracted from routine pathology samples across a diverse group of rare cancers," Naing said.
