Photo Credit: iStock.com/Mark Kostich

The following is a summary of “Dosimetric investigation of multi-parametric 4D-MRI for radiotherapy in liver cancer,” published in the April 2025 issue of Radiation Oncology by Li et al. 

Accurate target delineation in liver cancer radiotherapy is often challenged by organ motion, which can lead to underdosing of the tumor and overdosing of adjacent healthy tissues. This study evaluated the clinical utility of multi-parametric four-dimensional magnetic resonance imaging (4D-MRI) in improving target delineation accuracy and dosimetric outcomes in liver cancer radiotherapy.

A cohort of 20 patients undergoing radiotherapy for liver cancer was prospectively analyzed. Each patient received a contrast-enhanced planning CT scan in free-breathing conditions, as well as contrast-enhanced T1-weighted (free-breathing) and T2-weighted (gated) 3D-MRI scans. Additionally, time-resolved imaging using the TWIST-VIBE sequence was employed to acquire low-quality 4D-MRI. A dual-supervised deformation estimation model was applied to the 4D-MRI dataset to construct a four-dimensional deformable vector field (4D-DVF), which was used to generate multi-parametric 4D-MRI by deforming prior images.

Target delineations were performed on planning CTs with the assistance of both 3D-MRI and the derived 4D-MRI, resulting in two target sets: Target_3D and Target_4D. Corresponding treatment plans (Plan_3D and Plan_4D) were generated. To evaluate dose variations stemming from different delineations without re-optimization, Plan_3D was applied to Target_4D (yielding Plan_3D’) and Plan_4D was applied to Target_3D (yielding Plan_4D’). The analysis compared target volumes, dose-volume histograms (DVHs), conformity index, homogeneity index, and organ-at-risk doses across all plans.

Volume discrepancies between Target_3D and Target_4D were most pronounced in the left hepatic lobe (mean difference: 181.54 cm³, SD: 68.50 cm³). CI and HI values were comparable between plans: 1.02 ± 0.04 and 0.108 ± 0.02 for Plan_3D, and 1.02 ± 0.01 and 0.107 ± 0.01 for Plan_4D, respectively. While OAR doses did not significantly differ across Plan_3D and Plan_3D’, or Plan_4D and Plan_4D’, significant discrepancies were found in target dose coverage between Plan_3D vs. Plan_3D’ (P = 1.47 × 10^-7) and Plan_4D vs. Plan_4D’ (P = 0.013). Notably, Plan_3D’ achieved full prescription dose coverage for an average of only 77.89% (SD: 10.13%) of Target_4D, whereas Plan_4D’ covered 94.17% (SD: 3.12%) of Target_3D.

These findings underscore the limitations of 3D-guided planning in capturing the full extent of liver tumor motion. Incorporating multi-parametric 4D-MRI significantly enhances the precision of target delineation and improves dose coverage, supporting its integration into liver radiotherapy workflows for better treatment accuracy.

Source: ro-journal.biomedcentral.com/articles/10.1186/s13014-025-02600-3