In our view, it is unreasonable to condemn personalized medicine for oncology on the basis of the limited success of a few trials (see V. Prasad Nature 537, S63; 2016). We suspect that those failures were more likely to be caused by shortcomings in methodology.

With more than 40 precision-oncology drugs on the market, such therapies are helping tens of thousands of US patients by targeting specific molecular abnormalities. For example, mutations in the gene that encodes the epidermal growth-factor receptor are likely to occur in 10% of the 186,240 or so new cases of non-small-cell lung cancer predicted for 2016 (see go.nature.com/2fpxits). And the 8,220 people with chronic myeloid leukaemia (CML) predicted for this year will almost all carry the 'Philadelphia' chromosomal translocation (see go.nature.com/2fbbarj).

There are precision-oncology drugs to counteract both defects. Indeed, the life expectancy of people with CML is now starting to approach that of the general population (S. Saussele et al. Blood 126, 42–49; 2015).