We used AI for action figures. They used it on cancer.
Merck and Moderna say their personalised cancer therapy passed a Phase 3 trial. A machine learning model picks up to 34 targets from a patient's own tumour, and those go into mRNA...
Read the original caption
Merck and Moderna say their personalised cancer therapy passed a Phase 3 trial. A machine learning model picks up to 34 targets from a patient's own tumour, and those go into mRNA so the immune system knows what to hunt. The companies released no efficacy numbers; those come later at a medical meeting. Not medical advice - the therapy is investigational.
The real news
The story, straight
Merck and Moderna said their personalized cancer therapy met the Phase 3 trial endpoints. A machine-learning model picks up to 34 targets from a patient's own tumour, and those targets go into an mRNA vaccine so the immune system knows what to hunt.
The efficacy numbers will come later at a medical meeting; the therapy is investigational.
The absurd
Why this is ridiculous
The vaccine is now a 34-person search list, and the same immune system that was taught one target is being taught thirty-four. The body's 'who's who' got a personal trainer.
The breakdown
How it breaks down
The change is not the drug, it is the list. Instead of one target for everyone, the model picks 34 targets from one patient's own tumour.
The machine's job is no longer to treat; its job is to remember what to look for.
The upside
Why it could matter
The upside is that personalized oncology moved from a conference abstract to a trial readout. If the efficacy numbers hold, the lesson is that the same platform can carry a disease-specific list without being rebuilt for every disease.
Sources
Merck and Moderna: Phase 3 INTerpath-001 met RFS and DMFS endpoints


