ESTRO 2025 - Abstract Book

S3967

Radiobiology - Tumour radiobiology

ESTRO 2025

Consortium (DKTK), German Cancer Research Center, Heidelberg, Germany. 6 NCT/UCC), German Cancer Research Center (DKFZ), Faculty of Medicine and University Hospital Carl Gustav Carus, TUD, Helmholtz-Zentrum Dresden Rossendorf (HZDR), Dresden, Germany. 7 Department of Radiation Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA Purpose/Objective: In translational cancer research, selecting relevant preclinical endpoints is of high importance. For radiotherapeutic approaches with curative intent, tumor control is the gold standard, but large animal numbers are needed. Those numbers increase markedly, if more than one tumor model is used or if multiple drug candidates needs to be tested, i.e. if personalised approaches are evaluated. To test individualized treatment regimes, the mixing of several tumor models into a heterogeneous cohort was proposed, reflecting heterogeneous patient cohorts with tumors from several subtypes. An approach to estimate the required animal number was presented previously [1]. In this study, we show the first preclinical trial using such a strategy. We evaluated whether targeted therapies combined with fractionated radiotherapy improve local control rates over standard radiochemotherapy in multiple lung cancer models, while considerable reducing the number of animals required. Material/Methods: Inhibitors from the NCI Cancer Therapy Evaluation Program (CTEP) [2] were previously tested on a panel of genomically diverse non-small-cell lung cancer (NSCLC) cell lines using a spheroid-based high-throughput assay [3, 4]. A heterogeneous xenograft cohort was formed with n=394 NMRI (nu/nu) transplanted with three models: NCI-H1703, NCI-H441 and NCI-H23. Mice were randomized to receive the targeted therapies defined by the screen (Arm 1) or Carboplatin/Paclitaxel (Arm 2) each combined with clinically fractionated irradiation (30 fractions over 6 weeks) in 9 dose groups. 96 tumors were also imaged using MRI before treatment start. Tumor volumes were monitored for 180 days to assess tumor control rates (n=11 animals were excluded). Tumor control probability curves were fitted to a Poisson dose-response model, and the Dose Modifying Factor (DMF) at tumor control dose 50% (TCD50) between arms was calculated. Results: TCD50 of the targeted therapy arm (34.8 Gy) was significantly lower in the experimental arm (“personalised treatments”) compared to standard radiochemotherapy (45.2 Gy), with DMF=1.305 (p=0.022). Within the cohort, several sub-analyses are possible, although with less power. For example, targeted therapy significantly reduced TCD50 (63.9 Gy) compared to radiochemotherapy (96.9 Gy) with a DMF=1.52 (p<0.001) in the most radioresistant model (NCI-H1703). Radiomics analyses using the MRI data is ongoing. Conclusion: Targeted therapies showed better tumor control compared to conventional radiochemotherapy in our heterogeneous cohort consisting of xenografts originating from three different NSCLC models. This suggests a potential clinical benefit of individualized treatment based on in vitro screening results. Importantly, our experimental design allowed for a significant reduction in animal use without compromising statistical power. Funding: NIH-U01-CA220714 References: [1] Cecior, W. et al ., Sample-size calculation for preclinical dose-response experiments using heterogeneous tumour models. Radiotherapy and Oncology 158 (2021) 262–267. [2] https://ctep.cancer.gov/industryCollaborations2/agreements_agents_table.htm [3] H. Willers, X. Pan, I. Chamseddine, C. Grassberger and C.H. Benes. Landscape of the Radiosensitizing Properties of Targeted Agents from the NCI CTEP Portfolio across Genomically Diverse 3D Tumor Models. Int J Radiat Oncol Biol Phys 117 2S (2023) Oral Scientific Sessions S101. [4] H. Willers, et al. Screening and Validation of Molecular Targeted Radiosensitizers, Int J Radiat Oncol Biol Phys 111 (2021) e63–e74. Keywords: targeted therapy, preclincial trial, 3R