PURPOSE: To model the influence of hypoxic radioprotection in fractionated treatments over a range of fraction sizes. To determine whether there is a "therapeutic window" of dose per fraction where hypoxic radioresistance could be reduced, and if so, where it occurs in different cell lines.
MATERIALS AND METHODS: A mathematical model has been used to simulate the response of cells to low doses of radiation, in the region of clinical interest. We have used the inducible repair variant of the linear quadratic (LQ) equation, with a hypersensitive region (alphaS) at low doses that gradually transforms to the accepted "resistance" in the shoulder region (alphaR). It contains two new parameters, the ratio alphaS/alphaR, and D(C). We have accepted that the "induction dose" D(C) is modified by anoxia to the same extent as the other parameters. We have initially modeled using theoretical parameters and then checked the conclusions with 14 sets of published experimental data for cell lines investigated for inducible repair.
RESULTS: We have computed the clinical hypoxic protection (OER') as a function of dose per fraction in simulations of clinical fractionated schedules. We have identified a therapeutic window in terms of dose per fraction at about 0.5 Gy, where the OER' is minimized, regardless of the precise cell survival curve parameters. The minimum OER' varies from one cell line to another, falling to about 1.0 if alphaS/alphaR = 6-10 and even far below 1.0 if alphaS/alphaR > or = 20.
DISCUSSION: Hyperfractionation using 0.5 Gy fractions may therefore be more effective than oxygen mimetic chemical sensitizers, since it could even make some tumor cells more sensitive than oxic normal tissues. The tumor lines that benefit most from this type of sensitization are those with the highest intrinsic oxic radioresistance, i.e. those with high SF2 values.
1998. Vol. 42, no 4, 705-709 p.