Simultaneous tumor and surrogate motion tracking with dynamic MRI for radiation therapy planning

Area of Science:

• Medical Physics
• Radiotherapy Technology
• Oncology Imaging

Background:

• Respiration-induced tumor motion complicates precise radiotherapy for thoracic and abdominal cancers.
• Current surrogate-based methods for tracking tumor motion lack quantified correlation and accuracy.

Purpose of the Study:

• To develop and evaluate a method for simultaneously tracking lung tumors and external surrogates using dynamic MRI.
• To quantitatively assess the spatial correlation between tumor motion and external surrogates during radiotherapy.

Main Methods:

• Utilized dynamic MRI with MRI-compatible fiducials for real-time, non-ionizing radiation tracking.
• Reconstructed 4D-MRI from multi-slice cine MRIs for breathing phase templates.
• Employed 3D-2D template matching for 3D tumor position calculation and fiducial matching for surrogate trajectories.

Main Results:

• Demonstrated noticeable phase mismatch and significant cycle-to-cycle variation in 3D tumor trajectories.
• External surrogates were insufficiently sensitive to capture these tumor motion variations.
• Significant phase mismatch was observed between surrogate signals from different fiducial locations.

Conclusions:

• The proposed dynamic MRI method offers a more accurate assessment of tumor motion compared to traditional surrogates.
• Current external surrogates may not adequately represent actual tumor motion, impacting radiotherapy precision.
• Further refinement of surrogate strategies is necessary for improved lung cancer radiotherapy.