PII S0360-3016(99)00014-0 PHYSICS CONTRIBUTION PRACTICAL CONSIDERATIONS IN USING CALCULATED HEALTHY-TISSUE COMPLICATION PROBABILITIES FOR TREATMENT-PLAN OPTIMIZATION M. ZAIDER,PH.D.* AND H. I. AMOLS,PH.D.* *Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, NY Purpose: Healthy and neoplastic tissues are generally exposed nonuniformly to ionizing radiation. It is thus useful to develop algorithms that predict the probability of tumor control or normal tissue complication probability (NTCP) for any given spatial pattern of dose delivery. The questions addressed here concern: (a) the sensitivity of the NTCP predictions to the actual model used for extrapolation from uniform irradiation (where some clinical data exist) to nonuniform exposures, (b) its dependence on tissue type, and (c) consequences for treatment-plan optimization. Methods and Materials: Two (of several possible) NTCP formulations are used here: the Lyman model and a binomial equation. The effective volume-reduction scheme of Kutcher and Burman is used to obtain the NTCP for an arbitrary distribution of dose. NTCP was calculated for seven organs by postulating a dose distribution of maximum nonuniformity. Results: Both models fit available NTCP data well, but have very different extrapolations for exposures of small tissue volumes and very low values of NTCP (e.g., < 5%) where no data exist. Organs with pronounced volume effects (lung, kidneys) show substantial NTCP differences between the two models. Even in organs where the volume effect is small (e.g., spinal cord, brain), differences in NTCP due to the model selected may still have serious clinical consequences, as an actual example (for the spinal cord) indicates. Conclusions: NTCP calculations based on extrapolations to volume fractions and/or NTCP levels for which reliable data do not exist depend on the model used to fit the data and the degree of dose nonuniformity. If NTCP is to be used in treatment-plan optimization, the prudent approach is to design plans that reproduce the conditions under which available dose-volume data were taken (e.g., uniform dose distributions). © 1999 Elsevier Science Inc. NTCP, TCP, Treatment plan optimization INTRODUCTION Treatment-plan optimization for radiation therapy has here- tofore been driven largely by optimization of the dose distribution, with the goal being maximum separation be- tween the dose–volume histograms (DVH) for healthy and target tissues (as defined by the difference in area under the DVH for the two structures). Limitations of this technique include uncertainties on how to weigh the importance of different tissues and how to interpret two plans where the DVHs cross each other. An alternative approach to treat- ment-plan optimization is to include at least some informa- tion on normal tissue complication probabilities (NTCP) and tumor control probabilities (TCP), with the ultimate goal being to base all optimization decisions on complica- tion-free tumor control. The ability to achieve this in prac- tice—the subject of this paper—is currently limited by our incomplete knowledge of NTCP and TCP. NTCP, in par- ticular, can only be approximated for most clinical situa- tions. The only available comprehensive compilation of NTCP estimates (1) is actually a compendium of bona fide data supplemented for organs where little data exist by consensus estimation. It gives information only for uniform irradiations of 1/3, 2/3, or 3/3 or organ volume, and only for NTCP = 0.05 or 0.50. Thus, NTCP values for any other irradiation patterns, or for the more realistic case of non- uniform irradiation can only be estimated by interpolating or extrapolating data such as those published by Emami et al. (1) and also by defining a procedure for calculating an “effective” dose (see below). This, in turn, requires the selection of a mathematical model that fits the existing data, and allows one to predict NTCP for other irradiation con- ditions. Many functions, either phenomenological or semiempiri- cal, can be found to perform this function. Because the data of Emami et al. (1) have been presented without error bars, it is unlikely that a model can be selected based only on its ability to fit the data. In this study, we have selected two rather different models that fit these data equally well, and examine the extent to which the (basically arbitrary) selec- tion of the NTCP model affects predictions of NTCP in realistic treatment-planning situations. Put differently, we Reprint requests: Marco Zaider, Department of Medical Phys- ics, Memorial Sloan Kettering Cancer Center, New York, NY 10021 USA. Accepted for publication 30 December 1998. Int. J. Radiation Oncology Biol. Phys., Vol. 44, No. 2, pp. 439 – 447, 1999 Copyright © 1999 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/99/$–see front matter 439