2424 - Machine Transfer vs. Replanning for Helical Pancreas Radiotherapy: Efficiency, Delivery Fidelity and Dosimetric Impact

Purpose/Objective(s): Treatment machine downtime during radiotherapy can occur and, depending on severity, may require rapid re-planning on another machine to avoid treatment delays. On helical delivery systems, the "machine transfer" process preserves the original optimization while translating machine-specific parameters, potentially avoiding replanning. However, it remains unclear whether machine transfer sufficiently preserves plan quality and delivery efficiency compared with replanning. This study evaluated whether machine transfer is adequate and when replanning may be necessary.

Materials/Methods: Ten pancreatic cancer patients treated using helical radiotherapy were retrospectively analyzed. For each patient, the original clinical plan was (1) machine transferred and (2) replanned after changing the machine model, using identical optimization settings with additional iteration rounds if any desired criterium was not met; target coverage was rescaled if required. Log files were analyzed to extract processing times for both workflows. Net beam-on time and dosimetric metrics were evaluated for original, transferred, and replanned plans. Deviations were calculated relative to the original plan. Comparisons between transferred and replanned plans were performed using paired Wilcoxon signed-rank test. Results are reported as median and interquartile range (IQR, Tukey method).

Results: Machine transfer significantly reduced processing time compared with replanning (median [IQR]: 62.5 s [57–78.5] vs 118 s [95–170]), corresponding to a median reduction of 46.7% (p = 0.002). Beam-on time deviations relative to the original plan were larger for machine transfer than replanning (median/IQR: 4.05/0.60 s vs 1.05/1.20 s, p < 0.001), all differences were clinically negligible and within a few seconds. Replanning achieved smaller median differences of beam-on time, whereas machine transfer demonstrated greater consistency. For target coverage, no significant difference in PTV V(5040 cGy) was observed between the two workflows (-0.31±1.37 vs 0.02±0.04%, p = 1.00), however, transferred plans may sometimes lead to slightly lower PTV coverage than desired. Machine transfer better preserved high-dose regions (PTV V(5300 cGy)), while replanning introduced greater variability. For OAR metrics, machine transfer demonstrated smaller mean deviations and reduced variability for both D0.03 cc (4±10 vs 28±61 cGy) and Vx (0.01±0.35 vs 0.28±1.30%) metrics compared with replanning.

Conclusion: Machine transfer provides a time-efficient and dosimetrically robust alternative to replanning for pancreas helical treatments, preserving target coverage and OAR dose with minimal clinical impact on delivery parameters. Replanning may be reserved for cases requiring further optimization when target coverage is compromised.