Preclinical imaging: improving translational power in oncology.
Imaging continues to provide crucial datasets to scientists in neurology, cardiology and metabolic disease, for example. However, in this article we focus on the impact preclinical imaging is having in oncology – strengthening the options for in vivo visualisation of cancer-related processes over time.
Importantly, technological developments both in instrumentation and in probe synthesis and labelling have resulted in imaging systems with increased potential for basic research, as well as for translational and clinical applications. In addition, more sophisticated models are now available to address cancer-related research and therapeutic questions.
Initially using clinical-scale instrumentation, imaging provided a non-invasive means of assaying biological structure and function in vivo. The development of dedicated small animal imaging systems followed and, more recently, techniques in molecular imaging have been established to allow imaging modalities to be combined into multi-modal methods. Among these, the combination of positron emission tomography (PET) and computed tomography (CT) is a successful imaging strategy and has become an important tool in clinical practice. Technological approaches that combine magnetic resonance imaging (MRI), optical modalities and PET have now been introduced. PET/MRI and the resulting combination of molecular, morphological and functional information will pave the way for a better understanding of physiological and disease mechanisms in the preclinical setting (Figure 1).
A convincing model
Small animal imaging provides quantitative, spatially and temporally-indexed information on normal and diseased tissues such as tumours. Importantly, because of the non-invasive nature of the technique, imaging allows longitudinal (serial) study of animal models of human cancer.
This allows monitoring of disease progression from inception to progression, as well as of treatment options over a period of time. Each animal acts as its own control, reducing biovariability. Not only does this minimise the number of experimental animals required, it also gives results in real time. Moreover, in contrast to cell or tissue culture-based experiments, studies in intact animals incorporate the interacting physiological factors present in a complex living system. Looking forward, as drug development continues on the path towards personalised medicines, such longitudinal studies will offer scientists extremely valuable insights.
Advances in PET Technology
However, PET imaging is not without limitations, even with these advanced systems. Many available systems lack good spatial resolution3. Small anatomical structures cannot be distinguished or accurately analysed with low resolution PET. Additionally, PET imaging lacks anatomical context.
Technological advances have emerged to tackle these limitations, bringing significant improvements to the resolution issues faced by researchers. Using instruments that utilise innovative crystal technology and high sensitivity detectors, researchers can now obtain complete Full Field of View Accuracy (FFVA), which offers precise, homogenous sub-millimetre volumetric PET resolution in all three axes across the whole field of view. The latest breakthrough in PET detector technology enables rings to be located in-line with either MRI or CT, a design which also supports simultaneous PET and MR imaging. Advanced depth-of-interaction (DOI) detection enables precise 3D localisation of events without the constraint of having to work with conventional crystals configured in discrete layers. The result is the generation of an area of optimum resolution up to 10 times larger than conventional systems, providing unprecedented clarity.
This new technology has been successfully implemented into a tri-modal PET, SPECT, CT imaging system*, and a bi-modal PET/MR system**. Moreover, there is now the ability to transport a sedated subject to another instrument for further functional and/or anatomical study, for example to an MRI system. The use of multimodal animal beds removes any requirement to disturb the animal during study. This accurate positioning facilitates the layering of the three images to provide better understanding of the molecular mechanism or interaction of interest, with a correct anatomical reference. A further layer of imaging can therefore be added to the molecular visualisations achieved, with intelligent software ensuring precise automatic co-registration of images.
There is growing interest in combining PET imaging with MRI in this manner, as MRI provides superior soft tissue contrast, one of the key challenges for PET imaging. MRI techniques have also advanced with developments including a new 3T cryogen-free magnet, and now incorporate functions such as diffusion weighted imaging (DWI) that can be included into preclinical studies. These technology combinations are being applied in a number of clinical areas including oncology, neurology, cardiology and metabolic disease. In addition to cross-platform methodologies, integrated PET/MR solutions** are employed to streamline workflows.
In practice – cross-platform MRI/PET or MRI/SPECT imaging and coregistration
To illustrate the power of multimodal imaging using cross-platforms, here we report results from a study that employed the Bruker Multimodal Animal Bed (MMAB)4. In this instance preclinical imaging was undertaken using PET or SPECT (Albira™, Bruker) together with MRI (ICON 1T MR, Bruker).
To achieve cross-platform imaging, researchers frequently employ makeshift animal transports. While this approach is generally useful, there are typically limitations with animal care, anaesthesia, stable positioning and image registration between scans. Nelson et al (2011) reported on an immobilisation bed for cross-platform (PET/CT) imaging in a tumour xenograft model. Interestingly, when the immobilisation bed was used, inter-user variability for SUV analysis fell from 9.4% to 0.7%. This illustrates the importance of stable animal positioning and registration.
To ensure the highest quality results, scientists conducting this study employed MMAB which is equipped with a snug immobilisation shell that maintains the specimen animal positioning. Fiducial markers were employed for simple image registration.
This cross-platform imaging method was first tested for 18FDG-PET/MR using a Foxn1nu mouse without tumour grafting or other experimental treatments. Cross-platform PET/MR images were neatly registered using the MMAB solution (Figure 2).
Next, the protocol was evaluated for cross-platform 99mTc-SPECT/MR imaging in a HCT 116- hNIS-NEO tumour model. This imaging protocol and registration method resulted in excellent SPECT and MR tumour signal/contrast registration (Figure 3). These results provide for excellent tumour margin contrast with MRI and reliable cross-platform PET or SPECT registration. This protocol should also allow for accurate production of volume-of-interest (VOIs) based on tumour margins identified in MR images and application to functional PET or SPECT images.
It is apparent that there are considerable benefits of preclinical imaging – and the new PET technology in particular – aiding researchers as they seek to translate their work from in vivo models into the clinical situation. The sub-millimetre spatial resolution now available is a key improvement in comparison with conventional imaging systems, which will allow researchers to produce much higher quality images for analysis. This is particularly valuable for scientists looking at small anatomical features.
In addition, the introduction of a multimodal animal platform allows for considerably better image co-registration when using cross-platform imaging protocols. This again, adds significantly to the quality of data now available.
There are further benefits in cross-platform imaging; it maximises access to equipment, and has the potential to minimise down-time for maintenance. But perhaps most importantly – given that all-modality imaging systems are not currently available – is how the developments described here, that allow full integration of all modality combinations, are advancing the state of the art and bringing more powerful data to the research community as they search for ways to combat cancer and improve patient outcomes.
Dr Todd Sasser is a Field Applications Scientist for Bruker Preclinical Imaging. He provides application support for in vivo imaging across a wide variety of disciplines from infection imaging, cancer biology and probe development. He currently focuses on application development for the Albira PET/SPECT/CT system. Dr Sasser studied at The University of Liverpool and The University of Hawaii and is currently a visiting scholar at The University of Notre Dame.
1 O’Farrall et al. Non-invasive molecular imaging for preclinical cancer therapeutic development. Br J Pharmacol 2013; 169(4): 719-735.
2 Massoud, TF, Gambhir, SS. Molecular imaging in living subjects: seeing fundamental biological processes in a new light. Gene Dev 2003; 17(5):545-80.
3Wehrl, H et al. Preclinical and Translational PET/MR Imaging. J Nucl Med 2014; 55:11S-18S.
4 Sasser, T et al. Cross- Platform MRI/PET or MRI/SPECT Imaging, and Co-Registration.