Published on November 17, 2023, 6:02 am

Artificial Intelligence And Epigenetic Factors: Predicting Patient Outcomes Across Different Types Of Cancer

Investigators from the UCLA Health Jonsson Comprehensive Cancer Center have developed an artificial intelligence (AI) model that can predict patient outcomes across different types of cancer. The AI model is based on tumor-associated epigenetic factors, which are factors that influence how genes are turned on or off. By examining the gene expression patterns of these epigenetic factors, researchers were able to categorize tumors into distinct groups and predict patient outcomes more accurately than traditional measures such as cancer grade and stage.

The researchers studied multiple cancer types and found that clusters based on epigenetic patterns were associated with significant differences in patient outcomes for 10 of the cancers evaluated. These differences were observed in measures such as progression-free survival, disease-specific survival, and overall survival. The clusters with poor outcomes tended to be associated with higher cancer stage, larger tumor size, or more severe spread indicators.

To train an AI model to predict patient outcomes, the researchers used epigenetic factor gene expression levels. The model was specifically designed for five cancer types that had significant differences in survival measurements. The AI model successfully divided patients into two groups: one group with a higher chance of better outcomes and another group with a higher chance of poorer outcomes.

The findings of this study have implications for developing targeted anticancer therapies aimed at regulating epigenetic factors. Epifactors, such as histone acetyltransferases and SWI/SNF chromatin remodelers, could serve as potential targets for drug development. Histone acetyltransferases were found to be enriched among prognostic genes and associated with improved patient outcome. In addition, the SWI/SNF family of chromatin remodelers was also enriched among the prognostic genes across different cancer types.

The authors noted that their research provides a roadmap for creating similar AI models using publicly available lists of prognostic epigenetic factors. These models could help identify influential factors in different types of cancer and potentially lead to the discovery of specific targets for cancer treatment.

Understanding the role of epigenetic factors in cancer progression and patient outcomes is crucial as it can contribute to the development of personalized treatments. By analyzing epigenetic patterns and using AI models, researchers can gain insights into why some patients respond differently to treatments and have varying outcomes. This knowledge can pave the way for more effective and targeted cancer therapies in the future.

Overall, this study highlights the importance of considering epigenetics in cancer research and treatment. The use of AI models based on tumor-associated epigenetic factors has the potential to revolutionize how we predict patient outcomes and develop tailored therapies for different types of cancer.

Share.

Comments are closed.