Richard Popple PhD, Presenter: Nothing to Disclose

John Britt Fiveash MD, Abstract Co-Author: Nothing to Disclose

To study the dosimetric effect of MLC leaf transmission and leaf gap on dynamic multileaf collimation (dMLC) IMRT treatment plans.

A head and neck re-irradiation case was selected from our clinical database. Such cases challenge the limits of IMRT delivery systems because of the stringent dose limits on the previously irradiated spinal cord. The planning goals for this case were to deliver 54 Gy to the PTV while minimizing the dose to the spinal cord. We developed IMRT treatment plans for odd numbers of 6 MV photon beams, ranging in number from 5 to 27, equally spaced in the transverse plan. Plans were developed using the Eclipse/Helios treatment planning system (Varian Medical Systems, Milpitas, CA) for dMLC delivery using the 120-leaf Millenium MLC (Varian Medical Systems, Milpitas, CA). The cost function parameters were identical to those used to develop the clinical plan. Eclipse/Helios generates optimal fluence maps without consideration of the delivery hardware and thus the beamlet fluence values can be in the range from zero to arbitrarily large. The Millenium MLC has an average leaf transmission of 2% for 6 MV and has rounded leaf ends, which result in an effective 1.7 mm increase in the gap between opposing leaf pairs. For dMLC delivery, the Eclipse/Helios leaf sequencing algorithm generates the leaf sequence that minimizes the difference between the optimal fluence and the actual fluence, which accounts for leaf transmission and the effective leaf gap. The actual fluence is used for dose calculation. For each plan, we calculated MLC leaf sequences using four sets of MLC parameters: an ideal MLC (no leaf transmission or leaf gap), and MLCs with 0%, 0.5%, and 2% leaf transmission and an effective leaf gap of 1.7 mm. The 0.5% leaf transmission corresponds to approximately 2 cm additional leaf thickness. The dose was calculated and the plans normalized such that the PTV volume encompassed by the prescription dose was the same as the clinical plan. The maximum dose received by 0.5 cc of spinal cord, the PTV mean dose, maximum dose, and standard deviation in dose were recorded for each plan and MLC parameter set

For the ideal MLC, the spinal cord dose (0.5 cc) was 7.2 Gy and was approximately constant as a function of number of fields. For the 0%, 0.5%, and 2% leaf transmission cases, the dose for the 5 field plans was 8.7, 11, and 16 Gy and increased approximately linearly at 0.1, 0.16, and 0.3 Gy per additional field, respectively. The PTV dose values were not correlated to the MLC parameters. The mean dose was approximately constant, decreasing from 54.5 Gy for 5 fields to 54.3 Gy for 27 fields. The standard deviation decreased from 8.5% for 5 fields to 5% for 27 fields.

The MLC leaf transmission and effective leaf gap have a significant effect on critical structure sparing. Although increasing the number of fields modestly improved the target dose homogeneity, an increase in field number also increased the critical structure dose.

Popple, R, Fiveash, J, Dosimetric Effect of Multileaf Collimator Leaf Transmission and Leaf Gap on Dynamic Multi-leaf Collimation IMRT Treatment Plans.  Radiological Society of North America 2006 Scientific Assembly and Annual Meeting, November 26 - December 1, 2006 ,Chicago IL. http://archive.rsna.org/2006/8001958.html