Standardisation of preclinical PET/CT protocols across multiple research centres

Abstract

Preclinical Positron Emission Tomography/Computed Tomography (PET/CT) is a well-established non-invasive imaging tool for studying disease development/progression, the development of novel radiotracers and pharmaceuticals for clinical applications. Over the last five years more than 8,200 preclinical studies using PET/CT were conducted. Despite this pivotal role, standardisation of preclinical PET/CT protocols, including CT absorbed dose guidelines, is essentially non-existent. Therefore, the first and second aims of this project were: (1) to quantitatively assess the variability of current preclinical PET and CT acquisition and reconstruction protocols in routine use across multiple centres and scanners; and (2) to propose optimized standardised acquisition and reconstruction PET/CT protocols for routine scanning procedures across all sites in a preclinical PET/CT laboratory. By assessing quantitative accuracy (known versus measured) and precision (reduced variability) of currently used routine protocols between five different sites/scanners (Bruker Albira, Mediso nanoPET/CT, Sedecal Super Argus, Siemens Inveon and Trifoil LabPET/CT), standard protocols were determined. Thereby, irrespective of scanner characteristics the least biased empirical quantitative and qualitative protocol results defined the standard protocol. In essence, neutralizing the manufacturers' difference, replacing scanner variability for scanner similarity to establish global standard protocols. The analysis of sites’ routine protocol results revealed significant quantitative differences between all five sites/scanners. Whereas the standard protocols put forth improved accuracy and precision across all sites. Additionally, the large disparity and measured amounts of CT absorbed ionising radiation amongst sites brought to light the lack of preclinical radiation guidelines and dose regulations. Unregulated CT radiation dose is of great concern. CT ionising radiation is known to have biological adverse effects. Thus, overexposure of radiation will potentially cause unnecessary animal suffering and confound research outcomes. Overall, the proposed standard CT protocol reduced radiation doses. The implementation of preclinical PET/CT standardised protocols, developed and tested in this project, will provide more robust, reliable and reproducible translational data sets for clinical applications. In accordance with the National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs) objectives, the refinement of PET/CT protocols and reduction of CT absorbed dose impacts animal welfare positively and potentially reduces the number of animals used. Reducing study variability in site and across sites through standardisation of protocols improves statistically significant results using less animals. For study specific imaging parameters in preclinical PET/CT rodents are commonly used to design the protocol. The third aim of this project strives to develop a tissue equivalent material (TEM) anthropomorphic rodent phantom for the replacement of animals when designing and optimizing varying in vivo rodent imaging protocols. Using a TEM phantom reduces potential biological experimental variability caused by the animals and increases reproducibility of findings. To address this aim, twenty-four commercially available 3D printing materials were X-rayed for the evaluation Hounsfield units (HU). A comparison of calculated 3D material attenuation coefficients and accepted tissue attenuation coefficient was also done. CT images were acquired using four CT protocols and the developed standard CT protocol. Based on measured material HUs compared to accepted tissue/organ HU values, four materials were chosen for testing and further evaluation in a 3D printed phantom prototype was undertaken. In order to obtain the anatomical features of the rodent a CT acquisition of a scheduled 1 rodent was acquired. The CT images were used for the 3D printing design. A 3D printed (TEM) anthropomorphic rodent phantoms was printed and tested. Measured HU analysis of the phantom TEM materials shows promise as a replacement strategy. This imaging protocol optimisation approach is also in line with the NC3Rs objective of replacing and/or avoiding the use animals.