Robust Treatment Planning of Intensity Modulated Proton Therapy for Pelvic Lymph Node Irradiation in High-Risk Prostate Cancer Patients

Abstract

Using intensity modulated proton therapy (IMPT) for pelvic nodes irradiation in high risk prostate cancer patients can achieve better sparing of the normal tissues in the abdomen than photon plans. However, patient anatomic variation, such as increased bowel free air, can create large, sometimes unacceptable uncertainties. This work proposes a method of creating IMPT plans that are robust against dosimetric uncertainty due to anatomic variation. Data from 5 representative patients were used. The originally-treated IMPT plans were created in XX treatment planning system (TPS) without robustness optimization (RO). The plan robustness was evaluated by recalculating the dose with isocenter offset of 5mm combined with a range variation of ±3.5%. All the treated plans (without RO) had acceptable robustness per our institutional guidelines. During the treatment, 3-5 re-scan CT images were acquired for each patient and were registered with the planning CT image. Doses were recalculated on the re-scan CTs to evaluate the dosimetric variations due to patient anatomy change. In this work, treatment plans with RO were created in YY TPS. In addition to the native planning CT, two copies of the CT images were created. On the first copy, rectum, large bowel, and small bowel are assigned the density of air; and on the second copy, these structures are assigned the density of tissue. All 3 CT images are included in the robustness optimization process, which also includes a 5mm isotropic setup uncertainty and a ±3.5% range uncertainty. The RO plans are normalized to the same target coverage as the treated plans, and doses are also recalculated on the re-scan CTs for the RO plans. Table 1 lists the maximum doses (Dmax) of the treated plans (without RO) and the RO plans on the planning CT as well as the Dmax range of those respective plans when recalculated on the 3-5 re-scan CTs. Without RO, the Dmax can be as high as 145.2%, and the hot spot can be in or close to the critical structures, such as the rectum and bowel. With RO, the Dmax are no greater than 111.2% for all 5 patients on all re-scan CTs, and the 105% doses are usually limited to small volumes. Our method of including modified copies of the planning CT images in the RO to simulate the potential bowel filling variations greatly improved the plan robustness as evaluated by the doses recalculated on the re-scan CTs. The improved plan robustness eliminated the need of re-plan due to patient anatomy variation.Abstract 3778; Table 1Maximum doses for the treated plans (without RO) and RO plan on the original CT and recalculated on the re-scan CTsPt.Number ofDmax (%) ofTreated planDmax(%) ofRO planre-scan CTsOn Planning CTOn re-scan CTOn Planning CTOn re-scan CT 13104.9105.1 - 120.1105.7105.7 - 107.0 24107.3107.4 - 134.0108.9109.4 - 111.2 34107.3107.6 - 123.2106.5106.6 - 108.7 44105.9108.7 - 145.2107.3108.0 - 110.0 55105.3106.7 - 132.3106.0106.8 - 108.0 Open table in a new tab