The secret ingredient inside Moderna’s personalized cancer vaccine is an algorithm, STAT reported this week, a computational model that decides which mutations each patient’s vaccine should target.
STAT notes the melanoma shot, developed with Merck, would be the first approved medicine whose mRNA instructions a computer determined. The model reviews each patient’s tumor mutations, ranks the neoantigens most likely to provoke a strong immune response, and the mRNA is built to match.
Moderna’s program recently cleared its first phase 3 hurdle in melanoma, and the STAT report argues the real innovation is as much software as immunology. The design process compresses what was once a slow, manual selection exercise into a machine-readable pipeline that can be repeated for every patient.
The framing carries implications for drug development broadly: when the deciding factor in a medicine’s design is computational, regulators and patent offices need to think about AI in the review process. Moderna has long described its platform as algorithmic, but STAT puts the algorithm at the center of the vaccine’s story.
