Enzyme Engineering and Its Applications in Cancer Therapies: A Review of Machine Learning Approaches

Cancer remains one of the most challenging diseases to diagnose and treat, requiring the development of innovative therapeutic and diagnostic strategies. Cancer progression includes enzyme dysregulation, which orchestrates key biological processes ranging from metabolic reprogramming, epigenetic modifications to drug metabolism and immune evasion. Enzymes drive and modulate their roles in the tumor microenvironment, influencing how cancer cells adapt to stress, resist treatments, and evade immune surveillance. This review paper examines recent advances in enzyme engineering and its potential in cancer treatment. Computational tools, including artificial intelligence, have contributed significantly to enzyme optimization, enabling improvements in catalytic efficiency, stability, and specificity through structural modeling, functional annotation and generative design. Enzyme engineering can be optimized for targeted therapies, minimizing off-target effects while maximizing therapeutic potential. Challenges such as enzyme stability, delivery mechanisms, and immunogenicity persist, but recent advances offer promising solutions. Therefore, we also review the integration of computational approaches and experimental advances in enzyme engineering, thereby offering insights into future directions for optimizing enzyme engineering strategies in cancer treatment and diagnosis. In essence, this review provides an up-to-date synthesis linking the fundamental biological understanding of enzymes in cancer with the rapidly evolving field of enzyme engineering and the powerful capabilities offered by computational tools, highlighting both promising advances and remaining hurdles such as benchmarking and interpretability of machine learning offered solutions.