Researchers have harnessed artificial intelligence (AI) to reprogram glioblastoma cells into dendritic cells (DCs), which are crucial for detecting and directing the immune system to target cancer cells.
Glioblastoma, the most prevalent and lethal brain cancer in adults, has seen minimal advancements in treatment despite the progress of immunotherapy for other cancers. The challenge lies in the blood-brain barrier, which prevents immune cells from effectively targeting these tumors.
Supported by the National Institutes of Health and led by the Keck School of Medicine of USC, the study utilized AI to analyze genes that influence cell development. The team pinpointed genes capable of transforming glioblastoma cells into immune cells within the tumor, enabling them to attack other cancer cells.
In mouse models, this novel approach increased survival rates by up to 75%. The findings were published in Cancer Immunology Research, a journal of the American Association for Cancer Research.
“This innovative study uses AI to convert glioblastoma cells into immune-activating cells, representing a major leap in cancer immunotherapy,” said Dr. David Tran, MD, PhD, lead author and associate professor of neurological surgery and neurology at the Keck School of Medicine. Tran added, “By turning the cancer cells against themselves, we are advancing towards more effective treatments and offering renewed hope to patients facing aggressive cancers”.
Beyond animal models, the researchers applied AI to discover human genes capable of transforming human glioblastoma cells into immune cells. Future plans include delivering these genetic instructions to patients via viral vectors.
“Transforming a cell with 20,000 genes into something else is incredibly complex. Traditional methods would make this nearly impossible,” Tran noted. “AI provides a powerful tool to explore cell fate manipulation.”
DCs are essential for activating the immune response by presenting cancer antigens to other immune cells. While DCs have shown potential against glioblastoma, getting them through the blood-brain barrier has been a major hurdle. By reprogramming existing tumor cells, this method overcomes that barrier.
A key concern in cell reprogramming is ensuring specificity to avoid unintended effects, such as converting healthy brain cells into DCs. The AI system developed by Tran’s team analyzed thousands of genes and their interactions to pinpoint those that target glioblastoma cells specifically for reprogramming.
“The computational power of AI accelerates our discovery process,” Tran added. Combining this reprogramming technique with other immunotherapies significantly improved immune response and survival in mouse models. It increased survival rates by 75% when paired with immune checkpoint therapy and doubled the survival chances when combined with a traditional DC vaccine.
As reported by indiaeducationdiary.in, the next steps involve refining the identified human genes, packaging them into viral vectors, and conducting initial safety and efficacy tests in animal models. If successful, the team aims to seek approval for clinical trials in humans within a few years and hopes to apply this AI-driven method to other cancers as well.