Triggered plan adaptation using multi-image optimization for improved robustness in head-and-neck cancer proton...

Area of Science:

• Medical Physics
• Radiation Oncology
• Image-Guided Therapy

Background:

• Anatomical variations in head-and-neck (HNC) proton therapy can compromise target coverage and organ-at-risk (OAR) sparing.
• Daily anatomical changes necessitate adaptive strategies to maintain treatment efficacy and safety.

Purpose of the Study:

• To introduce and evaluate a triggered robust adaptation strategy using multi-image optimization for HNC proton therapy.
• To enhance treatment plan robustness by adapting to observed dose deviations in targets and OARs.

Main Methods:

• Retrospective analysis of five oropharyngeal cancer patients treated with proton therapy.
• Generation of synthetic CT scans from daily cone-beam CTs for dose recalculation.
• Comparison of four strategies: no adaptation (NA), triggered adaptation (TA), triggered robust adaptation with single (TRA-S), and multiple (TRA-M) triggered fraction images.

Main Results:

• Triggered robust adaptation (TRA-S and TRA-M) improved target coverage (median D98% up to +0.76 Gy) and reduced threshold violations compared to NA and TA.
• Unfavorable dose deviations were reduced, with OAR doses comparable to NA.
• TA showed similar OAR doses but did not consistently improve target coverage.
• Integral dose increased with TRA strategies, but differences between TRA-S and TRA-M were minimal.

Conclusions:

• Triggered robust adaptation effectively balances target coverage and OAR sparing in HNC proton therapy.
• This strategy requires fewer adaptations than TA and avoids reliance on predicted images or additional CT scans.
• It presents a viable approach for implementing anatomical robust optimization in clinical practice.