A marked dependence of the comparative effectiveness of neutrons on tumour line, and its implications for clinical trials.

The responses of five transplantable rat tumour lines to neutron and X irradiation have been compared by a method in which radiation-induced delay in tumour growth is used as a measure of effect. All the tumours were sarcomas and were irradiated at the same size, after growth in the same site, and under standard conditions. This group of similar tumours exhibited a large range in values of RBE in a dose range within which fractions of hypoxic cells did not detectably influence the result of X irradiation. Of the five tumour lines used, there were two pairs the members of which had a common origin and were histologically similar; the greatest differences in RBE values were between members of the pairs. These results suggest that the therapeutic use of high LET radiation cannot be expected uniformly to achieve local control better than conventional treatment of tumours at a given site, or even of a given histological type. Research is needed into methods that will have predictive value for the relative success of neutron therapy.