The potential and limitations of the inverse radiotherapy technique.

The objective of the work presented in this paper is to explore the scope of the applicability of the inverse radiotherapy technique for designing optimized intensity distributions to achieve a desired dose distribution. A specified desired uniform dose to the target volume is inverted, subject to constraints on the surrounding normal tissue dose, to produce optimum intensity distributions in a set of beams arranged around the target volume. We employed the inverse technique and software developed by Bortfeld and evaluated results both qualitatively and quantitatively using dose distribution displays, dose-volume histograms and biological indices including tumor control probability and normal tissue complication probabilities. So far we have applied this methodology to prostate and lung treatment plans. For prostate the inverse technique produces satisfactory approximations of the desired dose distributions. However, for lung its performance is considerably inferior. Our investigations point to a number of factors for this difference, the primary ones being differences in the tolerance doses of neighboring normal tissues, magnitudes of volume effect, tissue architectures, and the achievability of the specified desired dose distributions. We conclude that, for certain clinical situations, it is not sufficient to specify the objectives of optimization purely in terms of the desired pattern of the dose. The objectives must also include dose-volume effects and biological indices. Furthermore, the mathematics of optimization must be able to incorporate these factors into the process. We find that the inverse technique is not suitable for situations where dose-volume considerations and biological indices are important and that other methods of optimization of intensity distributions should be explored.