Iterative Optimization Method for IMRT Treatment Planning of Patients with Prosthetic Hardware
D Peskorski*, R Lalonde, D3 Radiation Planning, Pittsburgh, PASU-E-T-658 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: To present a new and improved method of IMRT treatment planning for patients with one or more prosthetic hip implants and compare the results to traditional treatment planning for these patients.
Methods:Four patients with unilateral or bilateral hip prostheses were planned for the 6MV beam of a Varian Truebeam linac using the Varian Eclipse planning system (AAA algorithm). Traditional treatment plans were generated with IMRT treatment fields that completely avoided entering through the prosthetics. A second set of plans was generated with 9 evenly spaced fields by iteratively editing the IMRT fluence for each field, so that in the final plan the prosthetics were completely blocked by the MLC during IMRT delivery. All plans were evaluated by normalizing the dose such that 95% of the target volume was treated to 100% of the dose.
Results:In the traditional plans, the severe restrictions on beam angle limited the conformality and uniformity of dose delivery to the target volumes, and increased the dose to the organs at risk (bladder, rectum, bowel) anterior and posterior to the target. In the 4 traditional plans, the conformality index (CI)(V100/VPTV) averaged 1.23, while in the iteratively optimized plans the CI average was 1.07. The target dose uniformity index (Dmax/Dprescribed) for the same 4 plans averaged 1.15 in the traditional plans, and 1.12 in the iteratively optimized plans. Average volumes of the organs at risk receiving the prescribed dose were reduced in the iteratively optimized plans by 11.3%, 18.2%, and 41.8% for the rectum, bladder and bowel, respectively.
Conclusion:This work presents a new and relatively simple method for IMRT planning of patients with prosthetics. With this method, substantial improvements can be made in target dose coverage and unformity and sparing of organs at risk.