ESTRO 2025 - Abstract Book

S2005

Clinical - Urology

ESTRO 2025

surface maps of rectal dose. Patients were categorised based on the presence of alternate allele(s) for candidate SNPs identified in the literature (rs1801516, rs17055178, rs7120482, rs17630638). For each SNP, patients were divided into two groups: those with an alternate allele and those without. Propensity score matching based on age, rectal volume, tumour volume and hormone treatment was used to identify control patients. Voxel-based Cox Proportional Hazards Models (CPHM) of dose surface maps was conducted for each group, with statistical significance determined through permutation testing. The effect of other covariates was tested using a multivariable CPHM, and internal validation was performed through bootstrapping. Results: Voxel-wise CPHM revealed an increased risk of acute and late bowel urgency in patients carrying alternate alleles for all examined SNPs (see Figure 1). Acute bowel control risk was higher in patients with alternate alleles rs17630638 and rs7120482. For these two SNPs, alternate alleles were also associated with a higher risk of late rectal bleeding, while a decreased risk was observed for acute rectal bleeding. Significant risk regions were consistently located in the lower posterior rectum. No significant associations were found for other endpoints.

Conclusion: This study validated GWAS findings for known SNPs using a novel dose surface mapping approach. Results show the benefit of correctly accounting for spatial dose distributions. The approach of this study could pave the way for combining traditional genomic studies with voxel-wise analysis, potentially informing tailored radiotherapy strategies in the future.

Keywords: radiogenomics, voxel wise analysis