Advances in Predicting Immunotherapy Response for Glioblastoma Using Neural Networks, Machine Learning and Genetic/Omics Data: A Comparative Study

Abstract

Modern-day Glioblastoma (GBM) remains an aggressive brain tumor and has fairly few therapeutic options available to them. Immunotherapy has shown less promise; however, it is still difficult to identify the patients who would respond to this treatment in a clinical setting within the scope of GBM. This paper reviews recent advances in using neural networks, especially artificial & deep neural networks, and other machine learning models, to predict immunotherapy outcomes using genetic and/or omics data. Therefore, we compare the findings from several recent key articles, highlighting their applied predictive models’ strengths, limitations, and potential clinical applications. Indeed, future approaches should focus on developing more transparent/interpretable models and collecting representative data for better clinical applications. Finally, we open perspectives in using such prediction techniques in our ongoing interesting clinical research study while considering the interpretability of predictions essential to ensure trust in the model’s decisions.