As the originator of the project to develop a preventive breast-cancer vaccine, which is supported by the US National Breast Cancer Coalition's (NBCC) 2020 deadline, I question your dismissal of this new approach (Nature 491, 637; 2012).

You base your pessimism on the genomic complexity of tumours and the length of time that clinical trials would take to test such a vaccine. But the genomic complexity apparent in mature tumours is not relevant for developing a preventive vaccine: the immune system needs only to be preactivated with antigens presented by the nascent tumour. Unstable RNA processing in a tumour is likely to be a richer source of these antigens than genomic mutations.

It could indeed take a decade or more to validate the treatment using today's clinical standards. But trials could be made shorter and less expensive if, for example, there were definitive ways to detect very early tumours.

You say that “discovery does not answer to deadlines”, but accomplishment can. Given the alternatives, we should embrace this effort.