Fractionated radiation therapy after Strandqvist.

Models for predicting the total dose required to produce tolerable normal-tissue damage in radiation therapy are becoming less empirical, more realistic, and more specific for different tissue reactions. The progression is described from the 'cube root law', through Strandqvist's well known graph to NSD, TDF and CRE and more recently to biologically based time factors and linear-quadratic dose-response curves. New applications of the recent approach are reviewed together with their implications for non-standard fractionation in radiation therapy. It is concluded that accelerated fractionation is an important method to be investigated, as well as hyperfractionation; and that more data are required about the proliferation rates of clonogenic cells in human tumours.