Technical and Dosimetric Considerations in IMRT Treatment Planning for Large Target Volumes

Abstract

The maximum width of an IMRT treatment field is smaller than the conventional maximum collimator opening because of design limitations inherent in most multileaf collimators (MLCs). To increase the effective field width, IMRT fluences can be split and delivered with multiple carriage positions. Not all treatment planning systems and MLCs, however, support this technique, and even if they do, the maximum field width in multiple carriage position delivery is still significantly less than the maximum collimator opening. For target volumes with dimensions exceeding the field size limit for multiple carriage position delivery, such as liver tumours or other malignancies in the abdominal cavity, IMRT treatment can be accomplished with multiple isocenters or with an extended treatment distance. To study some dosimetric statistics of large field IMRT planning, an elliptical volume was chosen as a target within a cubic phantom centered at a depth of 7.5 cm. Multiple 3-field plans (one AP and 2 oblique beams with 160° between them to avoid parallel opposed geometry) with constraints designed to give 100% dose to the elliptical target were developed. Plans were designed with a single anterior field with dual carriage positions, or with the anterior field split into two fields with separate isocenters 8 cm apart with the beams being forcibly matched at the isocenter or with 1, 2, 3 and 4 cm overlap. The oblique beams were planned with a single carriage position in all cases. All beams had a nominal energy of 6 MV. In the dual isocenter plans, jaws were manually fixed and dose constraints remained unaltered. Dosimetric statistics were studied for plans developed for treatment delivery using both dynamic leaf motion (sliding window) and multiple static segments (step-and-shoot) with the number of segments varying from 5 to 30. All plans were analyzed based on the dose homogeneity in the isocenter plane, 2 cm anterior and 2 cm posterior to it, along with their corresponding dose-volume histograms (DVHs). As expected, all the dual isocenter plans had slight underdosage anterior to the match point and slight overdosage posterior to it, while the dual carriage plan had a nice blending of the dose distribution without the accompanying hot or cold spots. Based on the dose statistics, it was noted that the dual isocenter plans can be clinically acceptable if they have at least a 3 cm overlap. In case of step-and-shoot IMRT, the number of segments used in a dual carriage plan was found to also play an important role in the overall plan dosimetric indices.