AbstractTo support a request of very substantial resources to build and operate an experimental reactor such as ITER, it is necessary to show that such a device is well positioned on the route towards a reactor and not too far from the reactor in parameter space. For the reactor definition, we choose to start from ITER design and to identify, using the same design tools as in ITER, which enhancements in physics and technology, would be necessary and possible to reach the performance required in a ‘first of a kind’ reactor. The conclusion of our analysis is that the 30 years ITER programme if successful and complemented by a few specialized test beds (e.g. neutron source) constitute a solid and sufficient basis from which to move directly and safely to a ‘first of a kind’ reactor, avoiding a further experimental step.