ESTRO 2024 - Abstract Book

S5349

Radiobiology - Tumour biology

ESTRO 2024

selectivity of MEK inhibitors for tumor models with KRAS/LKB1 co-mutations (n=6), compared to mutant KRAS/wild type LKB1 (n=11; p=0.03) and all other (n=15; p=0.002) genotypes. In contrast, DNA damage response (DDR) inhibitors constituted the most active class of agents in these tumor models, with DNA-PKcs, ATM, ATR, and PARP being the top targets. We found that radiosensitization by PARP inhibitors correlated with the PARP trapping potential of these agents, with the broadest radiosensitization seen for talazoparib, which has the highest PARP trapping potential. When performing unsupervised hierarchical clustering of drug effects across all models, PARP trapping effects could be clearly separated from catalytic PARP inhibition. Further, gene expression analysis showed that cell lines sensitive to PARP trapping upregulated genes that are related to cellular proliferation pathways and DNA repair compared to cell lines that were PARP inhibitor resistant. Importantly, PARP-targeted radiosensitization could be achieved even for tumor genotypes that are associated with radioresistance in the clinic, such as mutant KRAS, LKB1, or KEAP1. In 21 HPV-negative HNSCC models, grown in 3D extracellular matrix, targeting DDR genes as well as targeting the PI3K-AKT pathway for radiosensitization showed promise; however, for the latter only in a smaller subset of models (n=7, 33%). PI3K and AKT inhibition showed similar effect patterns in parallel experiments. While PI3K-AKT pathway inhibition was radiosensitizing in a PIK3CA mutant but not a wild-type HNSCC xenograft model, a significant enrichment for PIK3CA mutations unexpectedly was not observed in the 3D models that were successfully radiosensitized. An early phase clinical trial of the AKT inhibitor ipatasertib with concurrent chemoradiation therapy in locally advanced HPV negative HNSCC has been initiated and will explore correlations of genomic tumor markers with outcome (NCT05172245).

Conclusion:

We present unique insight into the landscape of radiosensitizing drug effects across diverse genomic tumor backgrounds. The observed patterns strongly argue for the use of larger panels of clinically relevant tumor models rather than a few non-representative cell lines when examining potentially radiosensitizing targeted drugs. Our data support clinical translation of select agents, such as PARP and AKT inhibitors, and highlight the need for development of predictive genomic biomarkers. Supported in part by National Cancer Institute U01CA220714 (HW, CB, AH, MK).

Keywords: screening, radiosensitization, targeted drugs

871

Proffered Paper

ABCC10 blocks STING dependent DNA damage pathway as a novel target for radiotherapy

Zhengyang Zhang 1,2 , Hsiang-i Tsai 1,2 , Dongqing Wang 1,2 , Haitao Zhu 1,2

1 Jiangsu University, Institute of Medical Imaging and Artificial Intelligence, Zhenjiang, China. 2 The Affiliated Hospital of Jiangsu University, Department of Medical Imaging, Zhenjiang, China

Purpose/Objective:

The predominant cause of radiotherapy failure and unfavorable patient prognosis stems from the presence of radioresistant molecule. The metabolic reprogramming of tumor cells stands as a major hallmark of cancer cells.