Introduction/Background Using clinically relevant imaging modalities in relevant animal models is crucial for strengthening the translational value of preclinical discoveries in endometrial cancer (EC). Imaging by ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG-PET/CT) is commonly used in diagnostic work-up in EC. ¹⁸F-FDG PET/CT in orthotopic mouse models of EC have been shown to be feasible, but standardized guidelines for image acquisition and interpretation is missing. Utilizing a large imaging database of orthotopic EC models, we aimed to characterize primary tumour ¹⁸F-FDG PET parameters and assess treatment response in a subset of mice.

Methodology The database consists of 91 ¹⁸F-FDG-PET-CT scans in 66 mice orthotopically implanted with patient-derived xenografts (n=30) or organoid-based patient-derived xenografts (n=36). A subset of mice was used for evaluation of treatment response (n=25). The mice were fasted for 12–16 hours prior to imaging, intravenously injected with ¹⁸F^-FDG and scanned for one hour. The following tumour parameters were extracted; max, mean and peak standardized uptake value (SUV_max/SUV_mean/SUV_peak), metabolic tumour volume, total lesion glycolysis, the 10 hottest voxels and metabolic rate of FDG. Interreader reliability between two readers were evaluated using intraclass correlation coefficient (ICC) test (n=25).

Results We utilized a 50% of SUV_max -segmentation threshold for tumour delineation, which correlated well with anatomical tumour volume extracted from MRI for a subset of mice (r²=0.74, n=25). There was a significant difference between treatment and control groups for the parameters SUV_max(p=0.020), SUV_peak (p=0.038) and the 10 hottest voxels (p=0.034) and the agreement between the readers were good (ICC; 0.89–0.97).

Conclusion ¹⁸F-FDG PET/CT in EC mouse models is feasible and multiple metabolic tumour features can be extracted. Using a clinically relevant imaging modality strengthens the potential for preclinical to clinical translation and reproducibility. Our work provides a basis for future studies on orthotopic mouse models of EC.