Analysis of Radiation Therapy Quality Assurance in NRG Oncology RTOG 0848

Abstract

<h3>Purpose/Objective(s)</h3> NRG/RTOG 0848 is a two-step randomized trial to evaluate the benefit of the addition of concurrent fluoropyrimidine and radiotherapy (RT) following adjuvant gemcitabine-based chemotherapy (2^nd step) for patients with resected pancreatic head adenocarcinoma. Real-time quality assurance (QA) was performed for each patient who underwent RT. QA findings were reviewed, frequencies & reasons for resubmissions are reported. <h3>Materials/Methods</h3> Patients were treated using either a 3D-conformal RT (3DCRT) or intensity-modulated RT (IMRT) technique. In addition to a web-based contouring atlas, the protocol outlined step-by-step instructions for generating the clinical treatment volume (CTV) through the creation of specific regions of interest. The planning target volume (PTV) was established by expanding the CTV by 0.5 cm in all directions. Treatment planning data, including pre-operative imaging, surgical and pathology reports, simulation images with isodose lines, structure set, and dose volume histograms were submitted for review. One of two radiation oncology study chairs independently reviewed each plan. Plans with unacceptable deviations were returned for revision and then resubmitted for review until approved. Treatment started after final approval of the RT plan. <h3>Results</h3> Of the 354 patients who underwent second randomization, 180 were randomized to the RT arm. Of these, 160 received RT and were included in the QA analysis. Most patients (86%) were treated using IMRT. Resubmissions were more common for patients planned with 3D-CRT (43%) than with IMRT (31%). In total, at least one resubmission of the treatment plan was required for 33% of patients (see Table 1). Among patients requiring resubmission, the majority only needed to be resubmitted once (87%). The most common reasons for resubmission were unacceptable deviations with respect to the pre-operative gross target volume (GTV, 61%) and the pancreatojejunostomy (48%), leading to unacceptable deviations of the contours of the CTV and PTV (70% for each). Most patients requiring a resubmission had 5-9 unacceptable deviations (46%). Unacceptable deviations were similar for the few treatment plans that needed to be resubmitted more than once. <h3>Conclusion</h3> One third of patients required resubmission to meet protocol compliance criteria, demonstrating the continued need for expending resources on pretreatment QA for adjuvant pancreas cancer trials. Rigorous QA is of critical importance for clinical trials involving RT to ensure that the true impact of RT can be assessed. Moreover, RT QA serves as an educational process by training radiation oncologists on best practices.