Melanoma is one of the most immunogenic cancers, yet despite remarkable advances in immune checkpoint blockade (ICB) and targeted therapies, a substantial proportion of patients develop primary or acquired resistance, and durable responses remain elusive for many. Among the most promising strategies to overcome resistance is the development of personalized cancer vaccines targeting tumor neoantigens arising from somatic mutations. However, the mutational landscape is highly heterogeneous across patients, and high tumor mutational burden does not consistently translate into immunotherapy benefit, suggesting that mutational neoantigens alone are insufficient to explain immune recognition or escape.
Growing evidence indicates that the most immunologically relevant tumor-associated antigens (TAAs) may in fact originate from non-mutated protein sequences, including cryptic peptides derived from the so-called dark proteome: non-canonical open reading frames, transposable elements, introns, and other genomic regions previously considered translationally silent (1). These cryptic antigens may represent a largely untapped reservoir of shared, targetable epitopes with the potential to complement or even outperform mutation-derived neoantigens as vaccine targets.