Artificial intelligence accurately predicts cancer patients’ odds of survival

Artificial intelligence accurately predicts cancer patients’ odds of survival

Angels – A new study shows that artificial intelligence can predict cancer patients’ chances of survival with a high degree of accuracy. Scientists have created an artificial intelligence model capable of predicting survival outcomes for patients with different types of cancer.

By analyzing gene expression patterns of epigenetic factors — which influence gene activation and deactivation — in tumors, researchers were able to classify tumors into distinct groups. These classifications have proven to be more effective in predicting patient outcomes than traditional methods.

A team from the UCLA Health Jonsson Comprehensive Cancer Center believes their work could pave the way for targeted therapies that regulate these epigenetic factors in cancer treatment.

“Traditionally, cancer has been viewed primarily as a result of genetic mutations within oncogenes or tumor suppressors,” says study co-senior author Hilary Koller, professor of molecular, cell and developmental biology, in a media release. “However, the advent of advanced next-generation sequencing technologies has made more people realize that the state of chromatin and the levels of epigenetic factors that maintain this state are important for cancer and cancer development.”

Scientists have also noted that some aspects of chromatin, such as histone protein modifications or DNA methylation, can influence cancer outcomes.

Professor Koller adds: “Understanding these differences between tumors can help us learn more about why some patients respond differently to treatments and why their outcomes vary.”

Previous research has shown that mutations in genes associated with epigenetic factors can increase cancer susceptibility, but the effect of their levels on cancer development is still poorly understood. Addressing this knowledge gap is essential, according to Kohler, to fully understand how epigenetics affects a patient’s odds of survival.

(Photo by Funlamai on Shutterstock)

The research team evaluated the expression patterns of 720 epigenetic factors, and classified tumors from 24 different cancer types into distinct groups. For 10 of these cancers, these groups were associated with significant differences in patient outcomes, including progression-free survival, disease-specific survival, and overall survival. The groups with lower results had indications of advanced stages of cancer, larger tumor sizes, or more extensive spread.

“We saw that the prognostic efficacy of an epigenetic factor depends on the tissue of origin of the cancer type,” says Dr. Mithun Mitra, co-lead author of the study. “We saw this association in the few childhood cancers we analyzed. This may be helpful in determining how important it is to target these factors therapeutically for cancer.”

The team then developed and validated an AI model using gene expression levels of epigenetic factors to predict patient survival, with a particular focus on the five cancer types with notable disparities in survival. The model successfully divided patients into two groups: those with a greater likelihood of positive outcomes and those with a greater risk of negative outcomes. The most important genes for the AI ​​model predictions showed significant overlap with the signature genes identified for the group.

The researchers suggest that although the AI ​​model was trained and tested on adult patient data from the TCGA cohort, further testing on independent data sets is necessary to determine its wider applicability. This approach could also be adapted to pediatric cancers to understand how decision-making factors differ from those in adult cancers.

“The roadmap demonstrates how to identify some of the factors influencing different types of cancer, and has exciting potential for predicting specific targets for cancer treatment,” concluded the study’s first author, Michael Cheng, a graduate student in the UCLA Interdepartmental Bioinformatics Program.

The results are published in the journal Communication biology.

Southwest News Service writer Stephen Beach contributed to this report.

YouTube video

You may also like...

Leave a Reply