Three-Step Image Guidance Strategy for Rectal Anterior Wall Margin Reduction in Hypofractionated Prostate Radiation Therapy.

Gao, Yan; Huang, Lei; Zhao, Bo; Ma, Lin; Li, Xuesong; Deng, Ximeng; Ma, Mingwei; Ren, Xueying et al. · Pract Radiat Oncol · 2026

other

Where this comes from

Abstract

This study evaluates a 3-step image guidance strategy (bone-prostate-rectal anterior wall alignment) to reduce posterior clinical target volume-planning target volume (CTV-PTV) margins in hypofractionated prostate radiation therapy while preserving target coverage and minimizing rectal toxicity. Twenty-three patients (575 daily cone beam computed tomography scans) underwent a 3-step alignment protocol: bony alignment, prostate soft-tissue matching, and rectal anterior wall fine-tuning. A dose calculation and hybrid deformable image registration-based workflow was created to accumulate the delivered dose. A 5-mm CTV-PTV margin (0 mm posteriorly) with 70 Gy/25 fractions and strict rectal constraints was used. The deformable image registration uncertainty has been thoroughly evaluated, covering geometric indicators, deformation vector field physical properties, spatial uncertainty, and dosimetric uncertainty indicators. Normal tissue complication probability for grade ≥2 gastrointestinal (GI) and genitourinary (GU) toxicity was calculated using the Lyman-Kutcher-Burman model. Statistical comparisons employed paired t-tests or Wilcoxon rank tests (P < .05). The dice similarity coefficient for CTV, rectum, and bladder exceeded 0.8, with mean distance to agreement under 3 mm. Jacobian determinant analysis showed biomechanical validity, with most voxels deforming within physiological ranges (92.23% for CTV, 89.63% for rectum, and 88.96% for bladder) and no nonphysical deformations. The distance discordance metric varied from 0.01 mm to 10.25 mm, and the δ index showed a 90.04% average passing rate within the PD80%. Compared to prostate-only alignment, rectal normal tissue complication probability was significantly reduced with the 3-step strategy (7.95% vs 13.61%, P < .01), with comparable CTV coverage (D95%:69.98 Gy delivered vs 70.98 Gy planned, P = .08). Deformable dose accumulation confirmed adequate dominant intraprostatic lesion (DIL) coverage (median D100 %: 71.51 Gy), except for DILs <0.7 mm from the rectal wall, or the rectal volume overlapping with DIL<sub>2 cm</sub> (DIL extends uniformly outward by 2 cm) >1.5 cm³ (requiring 2-3 mm margin expansion). A 3-step image guidance protocol enhanced rectal protection while maintaining target coverage compared to prostate-only alignment. Reducing the posterior PTV margin preserved CTV coverage, but DIL coverage varied by location. If the DIL is less than 0.7 mm from the rectum or overlaps more than 1.5 cm³ within 2 mm of the rectum, the reduced margin may not be suitable, potentially compromising DIL coverage, suggesting the need for personalized margins and adaptive radiation therapy to prevent underdosage.

Medical subject headings