S2587
Radiobiology – Normal tissue radiobiology
ESTRO 2026
USA. 5 Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, USA Purpose/Objective: Late radiation-induced bladder injury, of which hematuria is a defining symptom, is a potential complication of prostate cancer radiotherapy that can negatively affect survivors' quality of life. Genome- wide association studies (GWAS) have previously explored the link between the risk of radiation- induced hematuria and single nucleotide polymorphisms (SNPs). In the current study, we aimed to validate SNPs identified in GWAS from the Radiogenomics Consortium (RGC). Material/Methods: Two GWAS datasets, originating from the prospective REQUITE and URWCI studies (N=2,034 men who received radiotherapy for prostate cancer), were merged after imputation using the TOPMed server. Based on the summary statistics from the RGC GWAS of hematuria (N=3,988 men who received radiotherapy for prostate cancer) [1-2], SNPs with p<0.001 were initially selected. These SNPs underwent clumping to identify independent signals within the regions of linkage disequilibrium. The resulting lead SNPs were then extracted from the merged REQUITE and URWCI genotype data. Statistically significant SNPs associated with hematuria risk (those showing p<0.05 in the merged REQUITE+URWCI cohort) were chosen for further predictive modeling and pathway analysis. The closest genes for these SNPs were identified using the topr tool. Finally, biological pathway analysis was performed on the annotated genes using the MetaCore database. Results: The combined cohort for analysis included 1,680 and 249 evaluable cases from the REQUITE and URWCI studies, respectively, encompassing 155 total gross hematuria events (patient-reported grade ≥ 2). After filtering out SNPs with a minor allele frequency < 5%, the two studies yielded 5,771,581 overlapping SNPs. Based on the RGC summary statistics, 5,619 SNPs were selected using a threshold of p < 0.001. Subsequent clumping reduced the candidate set to 992 lead SNPs, which were then extracted from the merged REQUITE and URWCI genotype data. A correlation test between these 992 SNPs and hematuria risk identified 43 significant SNPs, which mapped to 42 unique genes. Regularized cox regression analysis performed with 5-fold cross- validation yielded a predictive accuracy (c-index) of 0.72. Pathway analysis highlighted two key associated process networks (Figure): blood vessel morphogenesis (false discovery rate [FDR] = 8.5E-4) and neurophysiological process (FDR = 3.3E-3). Furthermore, gene ontology analysis indicated a strong association between hematuria risk and
tests. Results:
Under high-dose irradiation, we observed an early inflammatory response in dWAT preceding epidermal change. The cascade appeared to propagate outward from dWAT to dermis and epidermis, culminating in overt radiation dermatitis. Macrophage infiltration and TLR4 signaling were implicated in this early phase.
Conclusion: Compartment-resolved analyses show that adipocyte lipid-droplet loss and p-perilipin activation in dWAT precede and predict clinical radiation dermatitis. The ensuing inflammation propagates from dWAT to dermis/epidermis, with macrophage engagement and TLR4 signalling implicated. These findings reposition RD as an adipose-initiated injury and provide a quantitative rationale for mechanism-based prevention targeting adipocyte metabolism and innate sensing; validation across dose/fractionation schemas and human skin samples is warranted. Keywords: radiation dermatitis, dWAT, lipid metabolisim Digital Poster 4887 Genetically Informed Biological Pathway Analysis of Late Hematuria Risk Following Prostate Radiotherapy Jung Hun Oh 1 , Christopher J Talbot 2 , Petra Seibold 3 , Paul Auer 4 , William Hall 5 , Joseph O Deasy 1 , Sarah Kerns 5 1 Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, USA. 2 Department of Genetics and Genome Biology, University of Leicester, Leicester, United Kingdom. 3 Division of Cancer Epidemiology, German Cancer Research Center, Heidelberg, Germany. 4 Department of Biostatistics, Medical College of Wisconsin, Milwaukee,
Made with FlippingBook - Share PDF online