A Test for Dose-Modifying Factors

Abstract

Methods have been developed for the analysis of low- and high-dose dose-response data to test for a dose-modifying factor and to estimate its value. Subsets of the response data (e.g., cell-survival data in the clinical dose range) may be analyzed separately. While dependent on an underlying dose-response model, the method gives similar results when different sets of cell-survival data are analyzed with two commonly used survival models. The use of the method is illustrated for the relative biological effectiveness (RBE) by analyzing survival data for the comparison of γ rays and neutrons and for the comparison of two neutron beams. The latter comparison leads to a constant RBE in the clinical dose range, whereas the constancy of the RBE is rejected when combined low- and high-dose data are analyzed.