Imaging the pancreas: from ex vivo to non-invasive technology
While many recently published reviews have covered non-invasive nuclear imaging techniques, the aim of this review is to focus on current developments in optical imaging technologies for investigating the pancreas. Several of these modalities are being developed into non-invasive, real-time monitoring routines for pancreatic diseases. However, they also provide pre-clinical ex vivo and/or intravital tools for three-dimensional quantitative assessments of cellular and molecular events, with levels of specificity and resolution difficult to achieve with other currently available modalities.
KeywordsBeta cell mass Diabetes Imaging Intravital Invasive Non-invasive Optical Tomography
magnetic resonance imaging
optical coherence tomography
optical projection tomography
positron emission tomography
single photon emission computed tomography
For pre-clinical and clinical assessments alike, technological limitations have hampered many areas of diabetes research. The scattered organisation of the islets of Langerhans within the much larger exocrine parenchyma puts high demands on the tools used, and even in small animals such as rodents, attempts to calculate the entire pancreatic beta cell content have been described as a true stereological challenge . Hence, it is generally accepted that technologies contributing to better quantitative and spatial analyses of the pancreas and its disease processes would greatly improve our ability to study important aspects of both type 1 and type 2 diabetes. During recent years, significant efforts have been made to overcome the technological hurdles associated with pancreas imaging. A major aspiration has been to develop a strategy for the non-invasive monitoring of dynamic processes in the pancreas, including changes in beta cell mass and the inflammatory process during the development of type 1 and type 2 diabetes. Such a technology would have tremendous clinical value, spanning from early prediction/diagnosis of type 1 diabetes to monitoring the response to potential beta cell restoration schemes.
In this review we will discuss recent advances in imaging technology with direct relevance for diabetes research. Since several excellent reviews have been published that focus on recent progress in the development of non-invasive pancreas imaging based on nuclear imaging techniques, such as magnetic resonance imaging (MRI) and positron emission tomography (PET) (see, for example [2, 3]), the focus here will be on optical techniques. These include ex vivo modalities that are currently applied primarily to pre-clinical diabetes research but have the potential to expand our scientific understanding of the disease, and may prove clinically relevant over the longer term. Generally speaking, across the range of techniques described here, the spatial resolution of the approach is inversely correlated with imaging depth. At one end of the spectrum, non-optical whole animal imaging displays the lowest resolution; at the other end, genuine microscopy (confocal) has the highest imaging resolution but the lowest depth penetration.
Towards non-invasive monitoring of the pancreas and its diseases
Optical projection tomography
Optical coherence tomography
While imaging modalities such as MRI and PET presently represent the methodologies of choice for non-invasive imaging of disease processes for the mouse and/or human pancreas, a technology such as OPT would appear to be more suited for monitoring 3D spatial relationships and quantitative processes across the whole organ, as required for the above type of applications. At present, high-resolution modalities such as confocal microscopy are required as a complement to, for example, OPT, to provide the means for cellular and subcellular dissection of these processes.
Other combinations of imaging modalities will provide the necessary tools for the further advancements of clinically applicable non-invasive techniques. Thus, ex vivo or intravital based technology platforms allowing whole organ 3D spatial analysis with high resolution and with specificity for molecular targets appear to be essential to drive this development. Further improvements in all of these techniques will be warranted and deserve continued efforts.
The authors are grateful to J. Sharpe (Centre for Genomic Regulation, Barcelona, Spain), A. Grapin-Botton (Polytechnic School of Lausanne, Switzerland), H. Edlund (Umeå Centre for Molecular Medicine, Umeå University, Umeå, Sweden) for helpful comments on the manuscript. At the latter institution, T. Alanentalo is acknowledged for help with figures and K. Loffler for editorial comments.
Duality of interest
The authors declare that there is no duality of interest associated with this manuscript.