On the dose-volume constraints based on radiobiological considerations

Purpose: To apply radiobiological principles in the estimation of dose‐volume constraints used in treatment planning. To evaluate NTCP distributions affiliated with certain multiple dose‐volume constraints. To estimate the probability PmD‐V that a dose‐volume histogram resulting in a given NTCP level also satisfies certain multiple dose‐volume constraints. Method and Materials: The reverse NTCP mapping method1 is used here to obtain physical dose‐volume constraints based on radiobiological indices. A procedure for random integral DVH sampling from the space of monotonously decreasing functions is developed. DVHs are randomly simulated and the ones producing an NTCP ε {5 ± 0.5%} are selected. An average DVH is produced from the selected DVHs. We propose that any point from the averaged DVH may serve as a physical dose‐volume constraint. A Monte‐Carlo method is used to estimate the probability PmD‐V for a number of these constraints. Results: Dose volume constraints for 16 organs selected based on the availability of parameter estimates for the Lyman and the Critical Volume NTCP models2,3 are obtained. The Emami4 constraints lay on the “upper boundary” of the DVH sub‐space defined by the condition NTCP=5±0.5%. The calculated probabilities PmD‐V are very low, indicating that the physical optimization uses a much smaller subspace of the possible solutions than the biological or the physico‐biological optimization. Conclusion: New dose‐volume constraints based on radiobiological considerations are proposed. DVHs passing through a combination of constraints1 are outside the range of the DVHs producing NTCP=5±0.5%. The physical RT optimization is more restricted in its choice of solutions than the biological one. © 2005, American Association of Physicists in Medicine. All rights reserved.