Development of a nasopharyngeal carcinoma (NPC) tumor model platform for quantitative evaluation of 18F-fluorodeoxyglucose (FDG) on a micro positron emission tomography-magnetic resonance imaging (PET/MRI) system

Abstract

Nasopharyngeal carcinoma (NPC) is one of the most prevalent malignant diseases in southern China. Despite most NPC is sensitive to radiation therapy, the recurrent rate is high. Therefore, investigations of novel therapy in NPC tumor models are still needed. To choose appropriate models for assessment of novel therapies, it is necessary to have a comprehensive understanding of the tumor growth characteristics and metabolism. MicroPET/MRI is a novel imaging modality that has been increasingly used in preclinical and translational studies due to the advantages of excellent soft tissue contrast and functional information. In this thesis, I aim to establish a NPC animal model PET/MR imaging platform and demonstrate the preclinical value of this platform in monitoring tumor growth and for treatment response assessment.

Evaluation of the robustness and reliability of this microPET/MR system was done by a test-retest study in healthy mice (n = 7). To investigate whether experimental parameters may affect [18F]FDG quantification, three experimental conditions were applied: Longitudinal study at 1-month and 5-month of age; warming of 32-35°C or 35-38°C; overnight fasting or non-fasting. Regions of interest (ROIs) in the limb muscle, liver, lung, myocardium, kidney, brown adipose tissue (BAT), cerebellum and Harderian gland were manually drawn on the images. The effect on uptake quantification measured by standardized uptake value (SUV) was evaluated. Liver was found to be the most suitable organ for normalisation for single timepoint test-retest studies and for longitudinal studies from 1-month to 5-month of age. I found overnight fasting is not necessary for mouse models. BAT and Harderian gland are the most susceptible to temperature, aging and fasting conditions, therefore conditions should be standardized for studies evaluating BAT and Harderian gland.

With a reliable PET/MR system and a comprehensive knowledge of [18F]FDG uptake in the mice, I evaluated five NPC mice models using the PET/MR system for consecutive 3-7 weeks. Two well-established NPC tumor models (C666-1, C17) and three novel models (Xeno76, Xeno23 and NPC43) were included (n = 5). SUVmean tumour-to-liver ratio (SUVRmean) and histological characteristics were analyzed and compared across the five NPC models. Cisplatin was administrated to one selected optimal tumor model, C17 to evaluate our imaging platform. I found variable tumor growth and metabolic patterns across different NPC tumor models. C17 has an optimal growth rate and higher tumor metabolic activity compared with C666-1. C666-1 has a fast growth rate but is low in SUVRmean at endpoint due to necrosis as confirmed by H&E. NPC43 and Xeno76 have relatively slow growth rates and are low in SUVRmean, due to severe necrosis. Xeno23 has the slowest growth rate and a relatively high SUVRmean. Cisplatin showed the expected therapeutic effect in the C17 model in a marked reduction of tumor size and metabolism.

The results demonstrated the ability of microPET/MR to produce reliable and robust results in healthy mice and multiple animal models. This study established a powerful investigation platform for these well-characterized NPC tumor models that will facilitate novel drug discovery and evaluation of NPC translational studies.