Purpose: To evaluate plan quality longitudinally over the course of pancreatic MRI-guided adaptive stereotactic body radiation therapy (SBRT).
Methods: A script was developed to retrospectively extract SBRT plan quality metrics from dose volume histogram files generated during MRI-guided adaptive radiotherapy treatments on a ViewRay MRIdian Linac. Plan quality metrics, including conformity index (CI), ratio of the 50% isodose to the 100% prescription isodose (R50), and ratio of the 36% isodose to the 100% isodose (R36), were computed. At our institution, a planning target volume optimization (PTVopt) structure is generated for each adaptive case by subtracting away an organ-at-risk+5mm planning risk volume from the planning target volume. The CI was computed relative to the PTVopt. These plan quality metrics were computed for the initial plan prepared prior to treatment, as well as for each adapted plan. A paired t-test was used to compare initial plan quality to all subsequent adapted fraction metrics. A Cronbach’s alpha was calculated for each of the 3 dosimetric indices to determine consistency within adaptive fractions across all patients.
Results: Plan quality metrics were evaluated retrospectively for 42 pancreas patients. Initial results indicate that plan quality was stable throughout the course of radiotherapy. The average CI across all patients ranged from 1.03 to 1.05 throughout the course of treatment. The difference between the initial and all subsequent adapted CI values were not statistically significant. All dosimetric indices had Chronbach’a alpha ≥ 0.96, indicating high overall consistency across all adapted fractions. However, some differences between initial and adapted R50 and R36 values were statistically significant.
Conclusion: Initial results indicate that plan conformity was consistent throughout adaptive MRI-guided radiotherapy. However, lower isodose falloff varied more throughout the course of therapy. Work is underway to investigate other plan quality metrics and organ-at-risk sparing.