High-energy protons have physical properties that virtually always will res
ult in geometrically better dose distributions than can be achieved using p
hotons or electrons. The clinical gains in terms of the probability of high
er tumour control and/or the reduced probability of normal tissue complicat
ions are, however, not completely known. Comparative model dose planning st
udies using real patients offer the possibility of estimating the potential
gains using a new technique. Several recently completed model studies, inc
luding clinically relevant endpoints, indicate that protons map have advant
ages, even when compared with the conventional treatment that is likely to
be introduced at the most advanced hospitals world-wide within the next dec
ade. These advantages can be seen not only in well-demarcated targets close
to risk organs, but also when irradiating extended irregular tissue volume
s at risk of containing tumour cells.