Generative AI in neoantigen cancer vaccine design — a review of promise and limits

A review published in Biotechnology Advances (1 September 2026) sets out where generative models can assist in designing therapeutic neoantigen cancer vaccines: tumour antigen discovery, ranking candidates by predicted immunogenicity, and assembling multi-epitope constructs. The authors also state why clinical efficacy of these vaccines remains constrained — imperfect antigen prioritisation, incomplete modelling of immunogenicity, tumour heterogeneity and immune evasion. They stress that generative models should complement rather than replace established prediction tools and experimental validation, with the shortage of functionally validated immunogenicity datasets as the core obstacle. This is a review without primary clinical data; the authors themselves conclude that the clinical value of the approach remains to be established.