PubMed1996
Intensity modulation provides greatly increased control, leading to superior dose distributions with a potential for improved clinical outcome. It also allows us to compensate for deviations from the expected patterns in dose distributions caused by the lateral transport of radiation. This is important not only to produce more homogeneous dose distributions in the target volume but also, more importantly, to allow a reduction in the margins for the penumbra and a corresponding reduction in the volume of normal tissue irradiated. Potentially, this would permit escalation of doses to higher levels and further improve local control for the same or lower normal tissue complications. The intensity-modulated treatment design process, regardless of the specific method used, involves the tracing of rays from the source of radiation through the target volume. Intensities of rays are adjusted iteratively in an attempt to produce a desired homogeneous dose within the target volume, while at the same time striving to maintain normal tissue exposure within the limits of tolerance. If the lateral transport of scattered radiation is ignored, as is commonly done because of the complexities of incorporating it, the resulting dose distribution within the target volume may be considerably different from the anticipated pattern. This would be particularly true if there are high gradients in fluence patterns within the field to shield a normal anatomic structure. Similarly, the lateral transport of radiation would lead to a dose deficit just inside the boundary of the planning target volume (PTV) and a dose excess just outside it. The conventional remedy to make up for the loss of dose near the boundaries, thereby ensuring complete coverage of the target volume, would be to employ a margin for the "penumbra." We demonstrate that, with intensity modulation, we have an important new tool to improve target coverage, namely an appropriate increase in fluence just inside the boundary. Most suitably, a combination of increased fluence and a smaller than conventional margin should be employed. We have used an iterative scheme to compensate for lateral transport in the intensity modulation optimization process. In each iteration, the intensity distribution is first designed ignoring lateral transport. At the end of each iteration, the dose distribution is calculated using a pencil beam convolution method, thereby incorporating lateral transport and revealing the deviations from the anticipated dose distribution caused by lateral transport. In the next iteration, ray intensities are further adjusted to rectify the deviations. In general, only a few iterations are needed to adequately account for the lateral transport of radiation. We have applied this method to intensity-modulated prostate treatment plans and demonstrate that this methodology allows the use of smaller margins, improves target dose homogeneity, and provides greater protection for normal tissues. We examine the variation of the magnitude of the gain from one patient to another. The methodology described in this paper has been introduced into routine clinical use.