Megavoltage CT–Assisted Mitigation of Metallic Implant–Induced Uncertainties In Intensity-Modulated Proton Therapy for Head and Neck Cancer
Abstract
Purpose
Metallic implants within the clinical target volume (CTV) introduce substantial uncertainties in intensity-modulated proton therapy (IMPT) for head and neck cancer (HNC). This work introduces megavoltage CT (MVCT)-assisted uncertainty-mitigation framework that quantifies stopping power ratio (SPR)-driven dosimetric discrepancies between kVCT and MVCT and investigates a planning strategy that minimizes reliance on explicit implant segmentation and density overrides.
Methods
Eighteen HNC patients with dental and/or surgical metallic implants within the CTV were retrospectively analyzed. Standard and extended HU–SPR calibration curves for kVCT and MVCT were generated using tissue-equivalent substitutes, with and without titanium, following consensus guidelines. SPR discrepancies were quantified using mean absolute error (MAE) and root mean square error (RMSE). Clinically approved kVCT-based IMPT plans served as references and were recalculated and reoptimized on kVCT and MVCT with and without explicit implant segmentation using standard and extended HU–SPR calibration. Target coverage (V98%, V95%), high-dose volume (V105%), robustness, and organ-at-risk (OAR) doses were compared across eight planning strategies.
Results
Bone exhibited the largest SPR discrepancies, with higher errors for kVCT (MAE/RMSE: 3.47%/4.47%) than MVCT (0.41%/0.45%). Titanium substantially increased bone-related errors for kVCT (9.53%/11.64%) but had minimal impact on MVCT (0.95%/1.24%). Lung tissue showed higher errors with MVCT than kVCT, while soft and adipose tissues were minimally affected. MVCT-based dose recalculation reduced V98% and increased V105% relative to kVCT reference plans. Reoptimization restored V98% and V95% and reduced high-dose volumes across all strategies (Fig1). OAR doses remained stable, with median variations within ±1.5 GyRBE (Fig2).
Conclusion
MVCT significantly reduces implant-related SPR uncertainties, particularly in bone. Although direct MVCT dose recalculation may reduce target coverage, reoptimization restores coverage and improves dose homogeneity. MVCT-assisted optimization with extended HU–SPR calibration offers a robust and clinically practical IMPT planning strategy for patients with metallic implants.