S2626
Radiobiology - Translational radiobiology
ESTRO 2026
1 UCL Cancer Institute, University College London, London, United Kingdom. 2 Department of Oncology, University College London Hospitals NHS Foundation Trust, London, United Kingdom. 3 Department of Genetics, Evolution and Environment, UCL Genetics Institute, London, United Kingdom. 4 Department of Medical Physics and Biomedical Engineering, University College London, London, United Kingdom Purpose/Objective: We investigated a proof-of-concept model on how radiotherapy or anticancer therapy combinations drive cell - type - dependent fate decisions within the colorectal tumour microenvironment and whether a patient - specific ex vivo platform provides actionable readouts to guide curative combination selection. Material/Methods: Fresh human colorectal liver metastases (CLRM) and adjacent normal liver were precision - cut into organotypic slices and maintained up to seven days. For Radiation, 8 Gy dose was selected based on radiobiological response assessment, while 1mM dose was chosen for ATR inhibitor Celarasertib. Slices were mounted in human organ size 3D - printed phantoms for CT - based Monte Carlo planning standardised proton (PBT) or photon delivery (Xray). Spatial transcriptomics guided by single - cell RNA - seq profiled baseline and post - irradiation time points, while γ H2AX foci mapped DNA damage (Figure 1). Figure 1: Workflow setup for human precision cut organotypic tumour slices derived from colorectal liver metastasis to investigate chemo/radiotherapy response. Results: γ H2AX mapping confirmed depth - and time- dependent dose delivery and DNA damage. it showed a more persistent DNA damage with protons versus photons at 24 hours in tumour (25 foci/cell at 30 min to 30 foci/cell at 24 hrs in CLRM PBT vs 20 foci/cell at 30 min to 15 foci/cell at 24 hrs in Xray). This difference was not noted in normal liver. Spatial transcriptomics identified common and rarer copy - number alterations and allowed stratification of tumour cells into cycling, stem - like, hypoxic and ECM - active states. Immune exhaustion and suppression pathways were upregulated in myeloid cells 24 hr after radiotherapy (both p<0.001), while dominant death pathways were suppressed or not activated in these cells, indicating immune shutdown without significantly affecting cell viability. Predominant cell - type - specific cell-fate programmes also emerged: myeloid cells underwent lysosome - dependent death; tumour cells activated
apoptosis plus lysosomal death pathways, stromal compartments senescence, whereas hepatocytes and cholangiocytes showed no statistically significant activation of death pathways at 24 hrs. Conclusion: This patient - specific organotypic slice workflow provides a reproducible high - dimensional molecular readout of the colorectal tumour microenvironment response to treatment. By defining primary cell - fate programmes in individual cell type, the model indicates selection of immunotherapy timing and cell - fate - directed combinations based on tumour microenvironment composition and beam choice. The modular workflow is parallelisable, deployable pre - or on - treatment in real-time, and adaptable to other tumours and agents, offering a bridge to prospective trials and adaptive decision making. Keywords: Colorectal cancer, PBT, precision-cut liver slices Integrated focused CRISPR screens reveal new modulators of radiotherapy response across diverse glioma subtypes Nareg Degirmenci 1 , Altar Ozbiyik 1 , Canan Bayraktar- Odabas 1 , Ipek Kok 1 , Ebru Yilmaz 1 , U ğ ur Selek 2 , Tugba Bagci-Onder 3 1 School of Medicine, Koç University, Istanbul, Turkey. 2 Radiation Oncology, Koç University School of Medicine, Istanbul, Turkey. 3 Medical Biology, Koç University School of Medicine, Istanbul, Turkey Digital Poster 5064 Purpose/Objective: Radiotherapy (RT) remains a core component of glioma treatment; however, intrinsic and adaptive radioresistance significantly limits clinical benefit. Therapy resistance arises through multiple mechanisms, including epigenetic rewiring, activation of DNA repair pathways, and cell-state adaptations. To uncover actionable regulators of RT response, we apply an unbiased functional genomics strategy using our focused CRISPR/Cas9 screening platforms that interrogate distinct biological pathways. This study aims to systematically identify novel genes that modulate RT response across diverse glioma subtypes, including both IDH–wild-type and IDH-mutant tumors, using clinically relevant primary cell models. Material/Methods: We generated focused sgRNA libraries optimized for parallel loss-of-function screening. The Epigenetic Knockout Library (EPIKOL) and the DNA Damage Response Knockout Library (DDRKOL) each target ~800 genes with 10 sgRNAs per gene. We also employed an available druggable genome sgRNA library. CRISPR screens were performed under
Made with FlippingBook - Share PDF online