ESTRO 2024 - Abstract Book

S5288

Radiobiology - Normal tissue radiobiology

ESTRO 2024

Diagnosis of Radiation Exposure: A Metabolic Marker Approach Using Blood Serum Analysis

Reinhardt Krcek 1 , Christos T Nakas 2,3 , Patcharamon Seubnooch 2 , Katrin Freiburghaus 2 , Daniel M Aebersold 1 , Kristina Lössl 1 , Mojgan Masoodi 2 , Daniel H Schanne 1 1 Inselspital, Bern University Hospital and University of Bern, Department of Radiation Oncology, Bern, Switzerland. 2 Inselspital, Bern University Hospital and University of Bern, Institute of Clinical Chemistry, Bern, Switzerland. 3 University of Thessaly, Department of Agriculture, Nea Ionia, Greece

Purpose/Objective:

With rising concerns over nuclear incidents, fast and accurate diagnostic tools to determine radiation exposure in affected patients are needed. Existing diagnostic techniques either compromise on accuracy or are time- and resource-intensive. To address this critical gap, we analyzed blood serum samples from radiotherapy patients with high resolution mass spectrometry to identify metabolic markers of radiation dose exposure. The eventual goal is the development of a fast, cheap and accurate test for the diagnostic purposes and patient triage in case of a mass exposure event.

Material/Methods:

We prospectively enrolled 20 female breast cancer patients after breast-conserving tumor resection but no (neo- ) adjuvant chemotherapy. All participants received adjuvant radiotherapy (RT) to the affected breast, administered in 30 daily fractions of 2 Gray. Blood samples were drawn at six timepoints before, during and after RT. The study was funded by the Swiss Federal Nuclear Safety Inspectorate. Blood samples underwent extraction for both hydrophilic metabolites and lipids. We analyzed the samples using a targeted list of previously published metabolites with an untargeted approach. Internal standard mixtures were added during the extraction for normalization of the data. The calibration curves were performed for quantitative assessment of the detected metabolites where possible. Metabolite separation and detection was performed by ultra-high performance liquid chromatography coupled to high-resolution mass spectrometry (UHPLC-HRMS) using Vanquish UHPLC-Orbitrap Exploris 240 and Q-Exactive Plus (Thermo Fisher Scientific). In parallel, high throughput shotgun lipidomics was performed using high resolution Orbitrap Exploris 240 mass spectrometer coupled with direct infusion, a TriVersa NanoMate ion source.

The metabolic analysis was conducted using LipidXplorer software version 1.2.8.1 for shotgun lipidomics, and TraceFinder software version 5.1 for other metabolites.

Data were analyzed by contrasting metabolites before radiotherapy with those from week 5 and 6 (combined) of RT. Univariate analysis was performed with a mixed effects ANOVA model, treating individual subjects as random effects and radiation exposure as the fixed factor. Partial least squares-discriminant analysis (PLS-DA) was used in the multivariate analysis stage. Statistical analyses were performed using R (v 4.1.2).

Results:

Univariate analysis revealed five metabolites exhibiting statistically significant downregulation after RT exposure: hypoxanthine, 3-hydroxyisobutyric acid, L-lactic acid, pyruvic acid, and xanthine (all p<0.05).

Additionally, the PLS-DA led to a set of metabolites discriminating between irradiated and non-irradiated subjects. This set was composed of purine metabolites (xanthine, hypoxanthine, 7-methylguanine/1-methylguanine), glucose derivatives (pyruvic Acid, L-lactic acid), amino acid derivates (L-glutamic acid, L-tryptophan, L-