Neoantigen mRNA Cancer Vaccines: How Deep Learning Is Advancing Design and Breaking Barriers

mRNA-based cancer vaccines have emerged as a transformative approach to combating cancerous cells among all cancer immunotherapy options. mRNA vaccines that encode tumor-specific antigens (TSAs) or neoantigens are responsible for a robust response to cancer. Unlike DNA vaccines, neoantigen-based mRNA vaccines are the preferred choice due to ease in scalable production, flexible neoantigen design, improved stability, structural optimization for enhanced translation efficiency, and decreased chances of genomic integration. Encapsulation of nonreplicating mRNA into a lipid nanoparticle-based delivery system further augments mRNA uptake and promotes endosomal escape. Recent advancements in machine learning and deep learning models can assist in designing solutions to the recurring problems of neoantigen mRNA vaccines by addressing the challenges such as tumor heterogeneity, neoantigen variability, the discovery of novel biomarkers, antigen presentation, and modulation of tumor microenvironment (TME) through initial experimental validations. This chapter highlights the convergence of mRNA technology with AI-optimized neoantigen prediction using machine learning and deep learning tools for effective therapeutic options.