A predictive model to guide management of the overlap region between target volume and organs at risk in prostate cancer volumetric modulated arc therapy

PURPOSE: The goal of this study is to determine whether the magnitude of overlap between planning target volume (PTV) and rectum (Rectum(overlap)) or PTV and bladder (Bladder(overlap)) in prostate cancer volumetric-modulated arc therapy (VMAT) is predictive of the dose-volume relationships achieved after optimization, and to identify predictive equations and cutoff values using these overlap volumes beyond which the Quantitative Analyses of Normal Tissue Effects in the Clinic (QUANTEC) dose-volume constraints are unlikely to be met. MATERIALS AND METHODS: Fifty-seven patients with prostate cancer underwent VMAT planning using identical optimization conditions and normalization. The PTV (for the 50.4 Gy primary plan and 30.6 Gy boost plan) included 5 to 10 mm margins around the prostate and seminal vesicles. Pearson correlations, linear regression analyses, and receiver operating characteristic (ROC) curves were used to correlate the percentage overlap with dose-volume parameters. RESULTS: The percentage Rectum(overlap) and Bladder(overlap) correlated with sparing of that organ but minimally impacted other dose-volume parameters, predicted the primary plan rectum V(45) and bladder V(50) with R(2) = 0.78 and R(2) = 0.83, respectively, and predicted the boost plan rectum V(30) and bladder V(30) with R(2) = 0.53 and R(2) = 0.81, respectively. The optimal cutoff value of boost Rectum(overlap) to predict rectum V(75) >15% was 3.5% (sensitivity 100%, specificity 94%, p < 0.01), and the optimal cutoff value of boost Bladder(overlap) to predict bladder V(80) >10% was 5.0% (sensitivity 83%, specificity 100%, p < 0.01). CONCLUSION: The degree of overlap between PTV and bladder or rectum can be used to accurately guide physicians on the use of interventions to limit the extent of the overlap region prior to optimization.