227 - Quality Assurance Analysis of Radiation Treatment Plans on EA2182: Lessons for Future Anal Cancer Clinical Trials
Presenter(s)
J. E. Meyer1, J. Fredette2, H. Chapman3, J. Santiago3, S. R. Amarnath4, A. B. Chakravarthy5, P. Das6, P. B. Romesser7, T. Fitzgerald8, and J. A. Dorth9; 1Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, PA, 2Department of Biostatistics and Bioinformatics, Fox Chase Cancer Center, Philadelphia, PA, 3Imaging and Radiation Oncology Core, Lincoln, RI, 4Department of Radiation Oncology, Cleveland Clinic Foundation, Cleveland, OH, 5Department of Radiation Oncology, Vanderbilt University Medical Center, Nashville, TN, 6Department of Gastrointestinal Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, 7Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, 8Image and Radiation Oncology Core-Rhode Island, Lincoln, RI, 9Department of Radiation Oncology, University Hospitals Seidman Cancer Center, Case Western Reserve University, Cleveland, OH
Purpose/Objective(s):
On RTOG 0529, 81% of submitted IMRT plans required replanning. Since this trial, IMRT for anal cancer (AC) has gone from novel to routine. We performed a retrospective review of prospectively evaluated treatment plans submitted on EA2182 (DECREASE), a randomized phase II clinical trial for early-stage AC patients testing deintensification (DI) vs standard of care (SOC). We hypothesized that the deviation rate would be inversely related to the number of patients enrolled by an institution (INST) or investigator (INV).Materials/Methods:
All radiation plans were required to be submitted for pretreatment review. A contouring atlas was included in the protocol. All plans were scored as major deviation (MAJ - replan required), minor deviation (MIN - contour errors that did not result in dosimetric failure) or plan approved (PA). Deviations were categorized by location (Primary (P), Presacrum, Mesorectum, Internal Iliac, External Iliac (EI), Obturator (OB), Inguinal, Bladder, Bowel, Vagina, Superior border, PTV margin, or GTV size error) and whether a plan’s errors were protocol-specific (PS), SOC or both. Variables analyzed predicting deviation included study arm – identical for contouring except for the lower superior border at the bottom of the sacro-iliac joints for the DI arm – and number of patients enrolled by INST or INV. Continuous variables were compared across treatment arms using Wilcoxon Rank Sum tests. Categorical variables were compared using Fisher’s Exact Tests. Logistic regression was used to examine the association between INV and INST experience and odds of needing a replan.Results:
Plan reviews were performed for 234 subjects enrolled at 75 INSTs and submitted by 126 INVs. 123 plans (53%) were PA while 30 plans (13%) were MIN and 76 (33%) of plans were MAJ, with 5 (2%) requiring 2 resubmissions. The most common sites of deviation were: P (56,24%), EI (45,19%), and OB (29,12%). Most errors in plans were SOC contouring (81,73%), while 9 (8%) were PS and 20 plans (18%) had errors in both categories. Fifteen plans (14%) had errors in determining the size of the primary tumor with 5 causing a change in stage and one patient becoming ineligible for the trial when corrected. Trial arm was not associated with MAJ (p=0.577) or MIN (p=0.539). Successive submitted cases by the same INV decreased the odds of requiring a replan by 18.6% (p=0.046). Increased institutional enrollments did not change the risk of requiring a replan (p=0.906).Conclusion:
This analysis found a large number of submitted plans contained contouring errors that required replanning. The most common sites of errors included the P, EI and OB regions. Experience of the INV improved contour accuracy. Although errors in contouring have improved since the landmark analysis of RTOG 0529, one third of plans on EA2182 had major deviations. As anal cancer represents a rare disease, more education may be helpful in improving the quality of care both on and off clinical trials.