Optimisation of Adaptive Radiotherapy for Cervical Cancer

Abstract

Cervical cancer (CC) presents unique radiotherapy challenges due to substantial inter￾and intra-fraction anatomical variations, including changes in tumour size and organ motion from bladder, rectum, and bowel filling. Adaptive radiotherapy (ART) aims to address these variations to improve target coverage while minimising dose to organs at risk (OARs). However, the optimal frequency, methodology and criteria for ART implementation remain unclear. This PhD thesis aims to identify the optimal ART strategy for CC using magnetic resonance (MR) guidance on the Elekta Unity MR-Linac, focusing on geometric, dosimetric, and an investigation into biological adaptation. The first study is a systematic review of the current evidence of ART in CC, which demonstrated substantial heterogeneity in imaging methods, adaptation strategies, and dosimetric outcomes. It highlighted the absence of standardised reporting and clinical consensus, reinforcing the need for evidence-based protocols to facilitate wider adoption. The second study quantified the dosimetric benefits of daily online Adapt-to-Shape (ATS) compared with non-adaptive workflows. The results demonstrate that daily online ART significantly improves dosimetric outcomes, with a measurable benefit from adapting in 82% to 100% of fractions at a 2% dose deviation and in 25% to 84% at 5% deviations, reflecting the variation observed across the five patients included in the study. The third study compared Library of Plans (LOP) with daily online ART (ATS), assessing plan selection accuracy, dosimetric performance, and clinical workload. While LOP offers a pragmatic approach, daily ART consistently achieved superior target coverage and OAR sparing, particularly in patients with high anatomical variability. The fourth study investigated the feasibility of incorporating diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) measurements into daily or weekly workflows to assess tumour biological response. ADC changes during treatment demonstrated feasibility and potential as non-invasive biomarkers for early treatment response, supporting future biologically adaptive radiotherapy strategies.