BEUD/SECONDARY MALIGNANCY ANALYSIS: COMPARISON OF HT, MLC-BASED IMRT, AND CRT IN PROSTATE TREATMENT PLANNING

Abstract

Purpose: To show that radiobiological analysis offers a better clinical assessment of treatment outcome compared to physical dosimetric analysis and to apply this radiobiological analysis to characterize possible benefits of tumor tracking over conventional radiotherapy. Materials: 4DCT data from six (n=6) patients with tumor motion in the range of 2-10mmwere used. Ten image sets corresponding to various phases of the respiratory cycle were derived from each 4DCT image set. A single clinician contoured the target and organs at risk on all ten phase-dependent image sets. Two plans were developed per patient based on tracking and conventional radiotherapy. In the latter, larger margins were used to account for organ motion. First, a reference image set was chosen and a plan developed based on a composite target. The plan parameters were then exported to the remaining phases and 9 other plans were created. By applying a validated deformable image registration algorithm, a deformation field was derived between each phase and the reference image which was then used to deform the dose from each of the phases to the reference phase where a weighted sum of the dose was computed to constitute the 4D dose from which a 4D DVH was developed.Tracking was characterized by reduced margins as ten different plans were developed based on target location per image set. The 4D dose and DVH were derived from all plans in a similar manner.Radiobiological assessment was based on the linear quadratic-Poisson and relative seriality models. The radiobiological parameters for each tissue were the D50, which is the dose that cause response to 50% of the patients, γ, which is the steepness of the dose-response curve, and s, which is the relative seriality. These parameters are specific for each organ and injury type and have been calculated from clinical data. The radiobiological effectiveness of each plan was quantified by the complication free tumor control probability P+ = PB PI where PB and PI are the total target response (benefit) and normal tissue complication (injury) probabilities, respectively. Four different fractionation schemes were considered; 10Gy, 6Gy, 3Gy and 2Gy per fraction for 6, 10, 20 and 30 fractions, respectively. Results: The advantages of tracking over conventional delivery were best observed using radiobiological analysis (fig 1). Higher P+ values for tracking, which decreased towards hyperfractionated schedules, were observed across all patients. For patients with tumor motion extent greater than 5mm, the relative increase in P+ were 25%±10%; 15%±7%; 8%±4%; 6%±3%, respectively for the plans with 6, 10, 20 and 30 fractions. The corresponding values for two patients with tumor motion extent less than 5mm were 10%±5%; 7%±3%; 3%±1%; 2%±2%. Conclusions: Radiobiological analysis offers a more sensitive assessment of clinical outcome and show that tracking is advantageous over conventional method with significant improvements for hypofractionated schedules. 653 poster (Physics Track)