Models for estimating the risk of ulcers in the small intestine after localized single or fractionated irradiation

Abstract

Subacute and chronic ulcerations of the intestinal mucosa are important causes of serious complications following radiation therapy for abdominal or pelvic tumours. We describe dose-response models for estimating the risk of mucosal ulcers in the small intestine after uniform, localized single or fractionated (once-daily) X-ray exposure. The models were fitted to data for ulceration incidence, based on a 26 week post-irradiation follow-up of male Sprague-Dawley rats which received a wide range of single and fractionated once-daily 250 kV X-ray doses to a short loop (partial volume) of transposed, but functionally intact, small intestine. The models presented for single (Weibull (W)) and fractionated (modified Weibull (MW)) exposures of a partial volume of tissue allow estimation of the risk of radiation-induced injury. While the W model is not new, its adaptation to partial volume irradiation and the MW model are. Isoeffect relationships are presented for the uniform fractional dose Ds(i%) associated with an i% (e.g. 0%, 5%, 10%, 50%) risk of intestinal mucosal ulcers as a function of the number of once-daily dose fractions, where Ds(0%) represents the threshold fractional dose. Although the Ds(5%) and Ds(0%) estimates provided for intestinal mucosal ulcers are based on animal data, the ratio Ds(0%)/Ds(5%) and more generally ratios Ds(j%)/Ds(i%) (where i not equal to j), are presumed to apply to humans. The indicated ratios are predicted to be independent of the partial volume irradiated and the number of once-daily dose fractions, and may be independent of radiation quality. Isoeffect equations are also presented that apply to circumstances where different partial volumes within the same reference volume (i.e. the total volume of tissue considered) receive different doses, but the dose within a given partial volume is uniformly distributed. These isoeffect equations provide a means of converting non-uniform dose within a reference volume to uniform isoeffect dose to the total reference volume and may have applications outside the field of radiation therapy (e.g. evaluating effects of non-uniform exposure of the small intestine or skin by a hot particle).