In Vivo Dosimetric Verification of Volumetric Modulated Arc Therapy for Prostate Cancer using a Rectal Balloon

Abstract

The use of volumetric modulated arc therapy (VMAT) for prostate cancer is increasing rapidly. As part of patient-specific quality assurance, VMAT plans are verified in phantoms under the assumption that the surrogate geometry appropriately represents the scatter characteristics of the patient. However, for emerging, highly complex treatment delivery techniques, it is desirable to verify treatment delivery in the patient geometry. In this study, we examined an approach for verifying the dose delivered to the high dose volume of prostate patients during VMAT delivery using dosimeters affixed to a rectal balloon. Patient selection was limited to subjects with clinical disease profiles T1-3N0M0. Patients received irradiation to the prostate volume with or without concurrent nodal irradiation. All patients received prostate immobilization during simulation and treatment delivery using the Radiadyne water-filled rectal balloon system. VMAT plans consisting of 1 or 2 350-degree arcs were constructed using the Philips Pinnacle treatment planning system. VMAT plans were delivered using an Elekta Infinity accelerator. Prior to patient positioning, an LiF thermoluminescent dosimeter (TLD) was affixed to the exterior of the rectal balloon. Localization markers were placed 1-cm above and below the TLD along the long axis of the balloon. The balloon was encased in an ultrasound probe cover and inserted into the rectum such that the TLD was oriented along the anterior rectal wall. A daily cone beam CT (CBCT) was acquired and registered to the planning CT. Following treatment, absorbed dose measured by the TLD was compared to the planned dose at the TLD position, which was determined by correlating the position of the localization markers in the CBCT with the planning CT. VMAT plans provided better than 95% coverage of the planning target coverage for all patients with equal or better sparing of the rectum and bladder compared with a 7-field fixed-beam IMRT plan. The average delivery time for the VMAT plans was 1.8 ± 0.4 minutes. The average measured TLD dose was 191 ± 19 cGy, compared to an average planned dose at the TLD position of 193 ± 15 cGy. The average percent difference between the measured and planned doses was -1.0 ± 5% and did not exceed 7%. Repeat TLD measurements showed intra-patient variations of less than 5%. A subsequent sensitivity analysis revealed that dosimetric results were not sensitive to uncertainties in TLD localization. Clinical delivery of VMAT for prostate cancer has been verified in vivo using TLDs placed along the anterior rectal wall via immobilization balloon and localized on CBCT. The measurement procedure was valuable clinical tool appropriate for verification of a complex radiation therapy technology.