A new study published in the journal Nature Medicine has found that artificial intelligence (AI) can help doctors make better decisions for patients with glioblastoma, a deadly type of brain cancer.
What is glioblastoma?
Glioblastoma is the most common and aggressive form of brain cancer, affecting about 13,000 people in the US each year. It is also one of the most difficult to treat, as it often grows rapidly and invades surrounding brain tissue. The median survival time for patients with glioblastoma is only about 15 months, and less than 10% of patients survive longer than five years.
How can AI help?
The study, led by researchers from the University of California San Francisco (UCSF) and IBM Research, used AI to analyze genomic data from more than 1,000 glioblastoma patients. The AI was able to identify four distinct molecular subtypes of the cancer, each with different characteristics and outcomes.
The researchers then used the AI to predict the best treatment options for each subtype, based on the available clinical trials and drugs. They found that some subtypes responded better to certain therapies than others, and that some combinations of drugs were more effective than single agents.
For example, one subtype of glioblastoma was more sensitive to drugs that target a protein called EGFR, which is often overexpressed in this cancer. Another subtype was more responsive to drugs that inhibit a pathway called PI3K/AKT/mTOR, which is involved in cell growth and survival.
The AI also suggested some novel treatments that have not been tested in glioblastoma before, such as drugs that target DNA repair or immune system modulation.
What are the implications?
The study shows that AI can provide valuable insights into the molecular diversity and complexity of glioblastoma, and help doctors tailor personalized treatments for each patient. This could potentially improve the outcomes and quality of life for patients with this devastating disease.
The researchers hope that their AI model can be applied to other types of cancer as well, and that it can be integrated into clinical practice in the future. They also plan to validate their findings in prospective clinical trials and real-world data.
The study is an example of how AI can augment human intelligence and expertise, and how collaboration between academia and industry can advance medical research and innovation.
留言