Gastrointestinal Motility-Induced Interplay In Abdominal Proton Therapy: Patient-Specific Delivery Replay In Pancreas and Liver
Abstract
Purpose
To quantify the impact of gastrointestinal (GI) motility on pencil-beam scanning (PBS) proton therapy for abdominal cancers, and assess how fractionation and motion amplitude mitigate motility-induced interplay effects.
Methods
A physiology-informed simulator generated peristalsis, rhythmic segmentations, and high-amplitude propagated contractions along stomach/duodenum/small-bowel centerlines, producing motility-only, time-resolved anatomies. Ten clinically treated PBS plans (consecutive pancreas (n=5) and liver (n=5)) were replanned with delivery-accurate spot timing using an in-house 4D dose-accumulation framework. For every spot timestamp, we computed dose on the deformed frame of motility scenarios and accumulated to reference using invertible deformations. We studied two motion amplitudes (MAX, MODERATE) and compared their interplay effects of single-fraction versus full-course accumulation. Endpoints were dosimetric differences (Δ) relative to static, e.g., CTV D95% and OAR D2%. Dosimetric Δ and dose distribution (the width of CTV-D95% or OAR-D2% between two extreme scenarios) are reported as mean [min, max] in Gy.
Results
Single-fraction MAX-motion produced the largest interplay effect. For pancreas patients, Δ was −0.75 [−2.98, 0.05] Gy for CTV D95%, indicating target under-coverage. The large distribution widths for OAR D2% (3.73 [2.02, 7.99] Gy for duodenum, 4.35 [3.10, 6.33] Gy for stomach, and 4.40 [0.63, 13.81] Gy for small bowel) suggest OAR hotspot risks. In liver patients, this interplay effect was overall smaller: CTV/OAR distribution widths were typically within 1 Gy and contracted further with fractionation. Compared to single-fraction and MAX motions, full-course accumulation and MODERATE motion mitigated interplay uncertainties. Across the combined cohort, risk concentrated when the distal fall-off intersected motile bowel. When distal edges were remote, interplay caused modest deviations from static reference.
Conclusion
GI motility is a clinically relevant, geometry-dependent source of interplay uncertainty in abdominal PBS. We present a patient-specific method for replaying clinical PBS deliveries on motility-only time-resolved anatomies. Fractionation reduces, but does not eliminate these interplay effects.