Effect-independent measures of tissue responses to fractionated irradiation.

Tissue repair factors measure the sparing that can be achieved from dose fractionation in the absence of proliferation. Four repair factors are analysed in these terms: FR, Frec, the ratio of linear-quadratic survival model parameters beta/alpha, and the half-time T1/2 for intracellular repair processes. A desirable feature of any repair factor is that it be independent of the level of injury induced in the tissue (or its single-dose equivalent, D1), since the comparison of tissues on the basis of the factor would not be meaningful, if they were characterized by differing D1S. Theoretically, FR and Frec are increasing functions of D1, and thus depend on level of effect. This is confirmed by analysis of skin reactions after multifractionated radiation. By contrast, beta/alpha is effect-independent as a measure of repair capacity in skin, gut, and bone marrow, tissues for which it is reasonable to assume that survival of identifiable target cells is the primary determinant of the endpoint. For a functional endpoint not clearly connected with the depletion of a specific target-cell population (late fibrotic reactions in the kidney), there was an increase in beta/alpha with increased levels of injury, but this was statistically insignificant. Effect-independence is defined for T1/2 as independence from size of dose per fraction. T1/2 is independent of fraction size in skin, gut, and spinal cord, and is longer (1.5 hours) in the late-reacting tissues (lung and spinal cord) than in those that react acutely (T1/2 less than 1 hour), with skin as the exception (T1/2 approximately 1.3 hours). Therefore, early and late-responding normal tissues may be distinguished in terms of both repair capacity and repair kinetics: repair is slower in late-responding tissues, which are also more sensitive to changes in dose fractionation. If generally true, these results imply that the potential for a therapeutic gain from hyperfractionation to spare late effects differentially would be limited by the number of daily doses that could be given, allowing sufficient time for repair between fractions in late-reacting tissues. In practice, dose fractions should be separated by 6 hours or more to permit repair to approach completion in late-responding tissues. Similar considerations lead to the conclusion that late effects would be more sensitive to changes in dose rate in low dose-rate brachytherapy.