ESTRO 2025 - Abstract Book

S4025

Radiobiology - Tumour radiobiology

ESTRO 2025

Purpose/Objective: The standard treatment for locally advanced rectal cancer (LARC) is neoadjuvant chemoradiotherapy (nCRT) followed by total mesorectal excision. The percentage of patients with a good treatment outcome varies, and there is an unmet clinical need to profile patients to optimize treatment options and if possible extend the period between treatment and surgery (watch-and-wait). This study aimed to use publicly available transcriptomics datasets produced by NGS technology to conduct a comprehensive data analysis of phenotypic profiles of LARC, with the purpose of identifying predictive molecular features in this clinical context. Material/Methods: The investigated meta-cohort consisted of datasets GSE190826, GSE233517, GSE80606, GSE209746, and data from Toomey et al 1 . All included patients had pathologically confirmed locally advanced rectal cancer at stage II (T3/4N0M0) and stage III (T1-4 N+ M0). Based on their treatment outcomes, patients were categorized as responders (R; TRG 1 and TRG2, Mandard scale) and nonresponders (NR; TRG3-5, Mandard scale). The meta-cohort included 156 patients (92 R and 64 NR). Each dataset was initially analyzed independently, and the meta-cohort was created by merging datasets based on genes present in all studies. Differentially expressed genes (DEGs) that overlapped between individual studies and the meta-cohort were characterized as genes with predictive potential. Results: Of the seven genes that showed predictive potential for good and poor responses in individual datasets, three genes ( IGHG4 , HLA-DPB2 , TRIM54 ) confirmed their significance in the meta-cohort as well. TRIM54 (Tripartite Motif Containing 54), a ferroptosis activator, showed overexpression in the R group ( p = 1.98x10 ⁻ ⁵), while IGHG4 (Immunoglobulin Heavy Constant Gamma 4, secreted by B cells; p = 2.51x10 ⁻ ⁴) and the pseudogene HLA-DPB2 (Major Histocompatibility Complex, Class II, DP Beta 2; p = 3.03x10 ⁻ ³) were associated with poor treatment outcomes. Among all genes confirmed as predictors of R/NR, TRIM54 and IGHG4 were not previously linked to chemoradiosensitivity in colorectal cancer. Additionally, two lncRNAs, LOC105377847 ( p < 0.05) and LINC01610 ( p < 0.05), showed predictive potential for chemoradiosensitivity in all studies where their expression was measured; however, their predictive potential was not confirmed in the meta-cohort because one of the included studies did not measure their expression. Conclusion: Understanding the molecular response to treatment requires profiling the phenotypic features of the tumor and its microenvironment. Transcriptomic profiling is facilitated by meta-transcriptomic analysis; however, to build a predictive model based on multi-omics approaches, further research should incorporate proteomic and genomic profiling, integrating these data with other clinical and radiological features. References: This study was funded by the Horizon Europe STEPUPIORS Project (HORIZON-WIDERA-2021-ACCESS-03, European Commission, Agreement No. 101079217) , Ministry of Science, Technological Development and Innovation of the Republic of Serbia (Agreement No. 451-03-66/2024-03/200043) and COST action TRANSLACORE Translational control in Cancer European Network (European Cooperation in Science and Technology, Action nºCA21154) 1. Toomey S et al. Genomic and Transcriptomic Characterisation of Response to Neoadjuvant Chemoradiotherapy in Locally Advanced Rectal Cancer. Cancers (Basel). 2020 Jul 6;12(7):1808. doi: 10.3390/cancers12071808. PMID: 32640573; PMCID: PMC7408989. Keywords: LARC, nCRT, transcriptomics