ESTRO 2021 Abstract Book

S1644

ESTRO 2021

objective was to provide a new stratification of the patients in relation 1. to the risk of developing radiation-induced side- effects, and 2. to the treatment response at 4 months. To that aim, we used functional multiplexed DNA repair assays on biochips that allow a comprehensive and precise investigation of the real capacities of the major DNA repair pathways. Materials and Methods 38 patients with H&N cancers were enrolled in 3 cancer centers. Three blood samples were collected before and after a few treatment sessions, and the Peripheral Blood Mononuclear Cells (PBMC) were isolated. Patients and treatment characteristics were recorded as well. A highly specific DNA repair profile was obtained from each PBMC extract using the DNA repair assays. They consist of specific DNA lesions immobilized on a biochip which are repaired by the patient’s enzymes that belong to different repair pathways (ExSy-SPOT, NEXT-SPOT). Repair of oxidative damage, alkylated bases, platinum adducts, and double strand breaks among others, was characterized. Results As first analysis, the DNA repair profile was assessed for each patient before the beginning of treatment. The biomarkers were associated with patients’ characteristics, tumor TNM staging, patient outcome and severe radiation-induced toxicities. Base and nucleotide excisions repair activities significantly increased with the lymph nodes staging of the tumor. An increased basal excision/resynthesis activity and platinum adducts repair was significantly associated with tumoral progression and death at 4 months after the treatment (p=0.0381 and p=0.0429, respectively). Patients without tumoral progression at 4 months showed an increased alkylated bases repair (p=0.0063). The risk of developing radiation-induced severe adverse events was associated with increased repair of 8-oxoguanine (p=0.0055), alkylated bases (p=0.0133), platinum adducts (p=0.0324), compared to patients who did not develop these events. Conclusion This comprehensive strategy based on the evaluation of DNA repair abilities enables to define the most important DNA repair based biomarkers associated with the tumor stage, treatment response and the risk of radiotoxicity. This profiling method could represent a powerful help for oncologists in the care of head and neck cancer patients, together with the biopsy DNA repair characterization. A larger cohort of patients is necessary to confirm these data. PO-1930 Plasmid DNA damages after FLASH vs conventional dose rate irradiations in various oxygen conditions N. Cherbuin 1 , J. Ollivier 2 , P. Jorge 1 , V. Grilj 1 , F. Chappuis 1 , L. Desorgher 1 , C. Bailat 1 , F. Bochud 1 , J.A. Jorge Pires 3 , M. Vozenin 2 1 CHUV, Institute of Radiation Physics, Lausanne, Switzerland; 2 CHUV, Laboratory of Radiation Oncology, Lausanne, Switzerland; 3 University of Applied Sciences and Arts Western Switzerland, HESAV, Lausanne, Switzerland Purpose or Objective Ultra high dose rate (UHDR) irradiation, several orders of magnitude higher than in conventional dose rate (CONV) radiotherapy, causes less damage to healthy tissue without impacting tumor control (Montay-Gruel et al, 2020). The physico-chemical and biological mechanisms underlying the FLASH effect are currently being investigated. In our work we thought to compare the effects of CONV and UHDR by quantifying DNA strand breaks (SB) using a plasmid (pBR322) under various oxygen concentrations. Materials and Methods pBR322 plasmid (4361 bp) was irradiated using a 6 MeV, FLASH-validated electrons beam (eRT6 Oriatron, PBM-Alcen) with increasing doses (1-100 Gy) and dose per pulse (0.01 Gy/s in CONV, 5.0*10 2 to 5.6*10 6 Gy/s in UHDR). pBR322 was irradiated dry or in water equilibrated at 21% (atmospheric level), 4% (physioxia) or 0.5% (severe hypoxia) using an hypoxia hood. The plasmid forms were quantified by agarose gel electrophoresis. Yields of radio-induced single (SSB) or double SB (DSB) were then computed using a mathematical model (McMahon & Currell, 2011). Results A very high dose was required (100 Gy) to produce measurable SB in dry pBR322 in absence of any free radicals. In this condition, UHDR and CONV irradiations produce similar yields of SB (Fig 1, 1st line). In aqueous solution and under atmospheric condition, 50% of relaxed form was produced at 2 Gy and 50% of linear form at 18 Gy (Fig 1, 2nd line). As expected, the plasmid was radio-protected in physioxia. Yet, the yield of SB was similar for UHDR and CONV (Fig 1, 3rd line). Interestingly, hypoxia revealed a difference between UHDR and CONV: while 5 Gy induced 50% of relaxed form regardless of dose rate, 50 Gy delivered in CONV induced more degradation of the plasmid and generation of smears. Furthermore, 50 Gy in UHDR produced less than 5% of linear form whereas it produces more than 20% of them in CONV (Fig 1, 4th line). Underlying SB yields are summarized in Table 1