# Time-saving potential of daily online adaptive proton therapy for head and neck cancers by reducing number of beams

## TL;DR

This study shows that daily adaptive proton therapy can reduce treatment time and improve outcomes for head and neck cancer patients.

## Contribution

The study demonstrates that daily adaptive proton therapy with fewer beams can maintain treatment quality while improving efficiency.

## Key findings

- DAPTRBC workflow reduced treatment time by 24% compared to conventional plans.
- DAPTSBC reduced normal tissue dose by 13% and xerostomia risk by 7%.
- Adaptive approaches improved target coverage compared to offline methods.

## Abstract

•Daily adaptive proton therapy may outperform offline triggered adaptation.•Reducing beam numbers in adaptive workflows can shorten treatment time.•Reducing margin alongside beam reduction and adaptation maintains treatment quality.

Daily adaptive proton therapy may outperform offline triggered adaptation.

Reducing beam numbers in adaptive workflows can shorten treatment time.

Reducing margin alongside beam reduction and adaptation maintains treatment quality.

Standard care for head and neck cancer (HNC) treatment with proton therapy typically involves a 4–6 field Intensity Modulated Proton Therapy plan to enhance robustness towards anatomical changes and patient misalignments. This study aimed to evaluate whether a more efficient plan with fewer beams, designed for faster delivery, can be combined with online daily adaptation (DAPT) to provide treatment of comparable quality, and improve treatment outcomes.

We retrospectively analyzed five HNC patients with available daily 3D imaging treated at our institution. To simulate DAPT, synthetic Computed Tomography (CT) images were generated by deforming planning CT to each daily Cone-Beam CT with targets and organs-at-risk (OARs) propagated to daily images. Three plans were created per-patient: OfflineSBC and DAPTSBC with standard, and DAPTRBC with reduced, beam configuration. DAPTSBC and DAPTRBC were reoptimized on daily synCTs, while OfflineSBC followed clinical workflow, with offline replanning as needed.

OfflineSBC showed >5% target underdosage in 15% of fractions, with both adaptive approaches significantly improving coverage. Although DAPTRBC outperformed OfflineSBC for target coverage, its advantage in OARs sparing was less definitive. DAPTSBC reduced pooled average normal tissue dose across patients and fractions by 13% and pooled average normal tissue complication probability for xerostomia by 7%. Delivery of DAPTRBC with fewer beams was 24% faster than plans with conventional arrangement.

Our delivery efficiency study shows that DAPT can allow fewer beams to achieve faster delivery, as shown in case of DAPTRBC workflow, and a reduction in the dose to normal tissue.