AI Turns Glioblastoma Cells into Immune Cells for Cancer Treatment

In an innovative study, scientists have used artificial intelligence (AI) to reprogram glioblastoma cells, converting them into dendritic cells (DCs) that can identify and direct other immune cells to kill cancer cells. This groundbreaking research offers new hope for treating glioblastoma, the most common and deadliest brain cancer in adults.

The Challenge of Glioblastoma

Glioblastoma is notorious for its poor prognosis, with less than 10% of patients surviving five years post-diagnosis. The blood-brain barrier often shields these tumors from immune cell attacks, making immunotherapy less effective compared to other cancers.

Leveraging AI for Gene Identification

Supported partly by the National Institutes of Health and led by the Keck School of Medicine of USC, researchers employed AI to determine which genes control cell fate, such as transforming a glioblastoma cell into a DC. These reprogrammed cells within the tumor can effectively target and destroy cancer cells.

Promising Results in Mouse Models

In mouse models of glioblastoma, this approach improved survival rates by up to 75%, according to findings published in Cancer Immunology Research. "By turning the cancer's own cells against it, we are paving the way for more effective treatments," said Dr. David Tran, the study's lead author and associate professor of neurological surgery and neurology at the Keck School of Medicine.

Human Applications and Future Directions

Researchers also identified a set of genes capable of converting human glioblastoma cells into immune cells. The goal is to deliver this genetic material to patients via viral vectors. "AI is helping us answer critical questions and gives us a powerful way to manipulate a cell's fate," said Tran.

Addressing Specificity and Safety

A key concern is ensuring the specificity of cell conversion to avoid affecting healthy brain cells. The AI system developed by Tran's team can query tens of thousands of genes to identify those that specifically target glioblastoma cells, unlike previous empirical approaches.

Enhanced Immune Response

When combined with other immunotherapies, this AI-driven reprogramming significantly boosted the immune response and survival rates in mouse models. The new approach, paired with immune checkpoint therapy, increased survival chances by 75%, and when combined with a DC vaccine, it doubled the chance of survival.

Future Clinical Trials

The next steps involve fine-tuning the gene list, packaging the genetic material into a viral vector, and conducting safety and efficacy tests in animal models. If successful, clinical trials in patients could begin in several years. Tran and his team also hope to apply their AI model to reprogram other types of cancer cells.

Study Contributors and Support

In addition to Tran, the study's authors include Tianyi Liu, Son B. Le, Dongjiang Chen from the Keck School of Medicine of USC, and Dan Jin, Mathew Sebastian, Alberto Riva, and Ruixuan Liu from the University of Florida College of Medicine. The research was supported by the National Cancer Institute and the Bankhead Coley Research Program of the Florida Department of Health.