2265 - Developing Standardized Online Adaptive Automated Templates for Ongoing Clinical Trials: Methodology from the Janus Rectal Cancer Trial

Purpose/Objective(s): Online adaptive radiation therapy (OART) enhances treatment precision by adapting to daily anatomical variations. However, the lack of standardized OART templates and guidelines has impeded its adoption in clinical trials. We hypothesize that a standardized OART planning template can be developed and integrated into an existing trial without compromising radiation compliance or treatment efficiency.

Materials/Methods: A multidisciplinary team, including our adaptive program’s radiation oncologists, physicists, and national quality assurance (QA) leadership, collaborated to develop a unique adaptive template for the JANUS Rectal Cancer Trial, a randomized phase II/III multi-institutional study on nonoperative management following total neoadjuvant therapy for locally advanced rectal cancer. Intensity modulated radiation was 4500 cGy in 25 fractions, plus a 900 cGy boost in 5 fractions. Three primary steps guided OART template design and integration. First, the adaptive workflow was defined by listing all procedural and technical requirements compliant with JANUS. Second, automated contours were designed per the trial’s radiation directives (e.g., CTV_54 = [GTV_54 + 1.5 cm margin] + mesorectum - sacrum - bladder - muscle). Lastly, constraints were prioritized in the automated template based on protocol-defined tumor and organs at risk (OAR) dose limits, ensuring consistent plan quality for adaptive fractions. Time from initial CBCT to beam-on was recorded and dosimetric comparisons between adaptive and non-adaptive plans over 30 fractions were performed using a two-sided t-test (a = 0.05).

Results: OART was successfully implemented for a JANUS trial patient, meeting all protocol requirements and receiving QA approval. The time from initial CBCT to beam-on ranged from 16 to 19 minutes. Using the adaptive plan, dosimetric comparisons revealed significantly improved target coverage for both PTV_4500 (p < 0.001) and PTV_5400 (p < 0.001), with the median V100% for PTV_5400 increasing from 89.3% (interquartile range: 86.4–91.7%) to 99% (98.7–99.1%). Bladder sparing was also enhanced (p < 0.001), as V3500 decreased from 40.3% (37.5–42.3%) to 29.7% (27.9–31.1%). Deformable-registration–based accumulated dose analysis was performed for validation and trial reporting. Target coverage improved, and OAR dose was lower in accumulated adaptive plans versus scheduled plans. The patient achieved a clinical complete response based on MRI, sigmoidoscopy and ctDNA assay with normal anorectal and urinary function.

Conclusion: Our findings demonstrate the feasibility and potential clinical benefit of incorporating OART using a standardized template in a complex radiation trial. This practical framework underscores the importance of structured guidelines for widespread OART adoption.