Dosimetric investigation of multi-parametric 4D-MRI for radiotherapy in liver cancer

Sha Li, Xianggao Zhu, Haonan Xiao, Weiwei Liu, Yibao Zhang, Jing Cai, Tian Li, Yanye LuRadiation 2025 · DOI

Abstract

3D image-guided target delineation may be more likely to underestimate target volume and compromise dose coverage, suggesting that using multi-parametric 4D-MRI can provide more precise target contours and enhance target dose coverage.

Introduction

Radiotherapy for liver cancer faces a significant challenge: respiratory motion. The liver's proximity to the diaphragm causes it to shift and deform during the breathing cycle. Current standard practices often rely on 3D-MRI "snapshots" or 4D-CT. However, 4D-CT suffers from poor soft-tissue contrast, while gated 3D-MRI only captures the tumor at the extremes of inhalation and exhalation. This study investigates whether a full multi-parametric 4D-MRI approach, capturing the tumor's entire motion path, improves treatment planning accuracy.

Methods

The study involved 20 patients (17 male, 3 female; mean age 60.4) with primary liver cancer or metastases. The researchers developed a dual-supervised deformation estimation model (DDEM) to generate high-quality 4D-MRI from low-quality scans by leveraging high-resolution 3D-MRI "priors."

Two target volumes were delineated for each patient: Target_3D (based on standard gated 3D-MRI) and Target_4D (based on the multi-parametric 4D-MRI). The researchers then performed a cross-evaluation, applying 3D-based plans to 4D targets to assess potential dose coverage gaps.

Dosimetric evaluation criteria including target dose (DVH, CI, HI) and OAR dose metrics.

Results and Discussion

Volume Underestimation

The study revealed that 3D-MRI systematically underestimates tumor volume. Target_4D volumes were, on average, 32.47% larger than Target_3D volumes. This discrepancy was most pronounced in the left hepatic lobe—which is more susceptible to motion—where volume differences reached 181.54 cm³ compared to much smaller differences in the right lobe.

Dosimetric Impact

The "snapshot" approach of 3D planning led to significant coverage failures. When 3D-based plans were evaluated against the 4D motion model (Plan_3D’), the intended dose covered only 77.89% of the target volume. In contrast, plans designed using the 4D-MRI model (Plan_4D’) achieved 94.17% coverage.

Safety of Organs at Risk (OAR)

Crucially, the larger target volumes required by the 4D approach did not lead to increased radiation exposure for healthy tissues. There were no statistically significant differences in the doses delivered to OARs, such as the spinal cord, stomach, or kidneys, compared to 3D planning. This suggests 4D-MRI allows for better coverage without necessarily increasing side-effect risks.

Conclusion and Future Directions

Multi-parametric 4D-MRI offers a superior method for accounting for respiratory motion in liver radiotherapy, preventing significant target underdosing. While limited by a small sample size and the labor-intensive nature of manual 4D contouring, future integration of auto-segmentation tools could make this high-precision workflow a clinical standard. Further research is also needed to compare this approach directly with the current clinical standard, 4D-CT.

Summary updated

Abstract

3D image-guided target delineation may be more likely to underestimate target volume and compromise dose coverage, suggesting that using multi-parametric 4D-MRI can provide more precise target contours and enhance target dose coverage.
BibTeX
@inproceedings{li2025dosimetric,
  author = {Sha Li and Xianggao Zhu and Haonan Xiao and Weiwei Liu and Yibao Zhang and Jing Cai and Tian Li and Yanye Lu},
  title = {{Dosimetric investigation of multi-parametric 4D-MRI for radiotherapy in liver cancer}},
  booktitle = {Radiation},
  year = {2025},
  doi = {10.1186/s13014-025-02600-3},
}
Figure