PET/MR studies in metabolic and oncological diseases

Principal investigator: Håkan Ahlström

Project 1: Imiomics in whole-body PET-MRI studies of the obesity-diabetes-atherosclerosis disease process

The knowledge about the various genotypes and phenotypes of type 2 diabetes mellitus (T2DM), and the tissue-specific pertubations at different stages of the disease, are incomplete. More studies about the associations between white and brown adipose tissue (amount, localization and metabolic activity) and present and future atherosclerosis (location, amount and degree of inflammation), and brain infarcts as the consequence of vulnerable plaques, are needed. Imiomics is an automated whole-body image analysis concept, including an image registration method that deforms all image data to a common coordinate system, so that the signal in each voxel is comparable between individuals and within an individual over time. It will in the present application be applied for automatic statistical analyses of whole-body data in one  MRI and two PET/MR ongoing studies of T2DM and cardiovascular diseases. Whole-body PET-MRI scans will be performed for detailed analysis of the subjects’ body composition, lipid homeostasis (by MRI) and tissue-specific glucose metabolism/inflammation (by FDG-PET). The image data will be integrated with the non-imaging data, including clinical and molecular data from relevant tissues and blood, using Imiomics. By this approach new findings that are difficult, or impossible, to detect with other methods, and that are important for future prevention and therapy strategies of T2DM and cardiovascular diseases, are anticipated to be found.

Project 2: Whole-body PET/MRI for detection, staging, characterization, prognostic estimation, and early therapy evaluation of tumors

Uppsala University installed a whole-body PET/MR equipment, funded by VR, in October 2014. The whole body approach has the potential to give information about different cell populations, and their response to therapy, in spread cancer. Whole-body PET/MR generates a lot of information.

The overarching aim is to improve detection, staging, characterisation, prognostic estimation and therapy evaluation of tumors by using the fully integrated whole-body PET/MR equipment. To reach this aim the following projects can be defined:

A. Validation of different whole-body PET/ MR parameters and investigation of the additional clinical value of PET/MR in oncology using data from the PET/MR equipment, the U-CAN cohort and surgery as references.. Consecutive investigations with the PET/MR equipment in patients with different cancer diagnoes, i.e. gynaecological, colorectal and neuroendocrine tumours will be performed at diagnosis, during and after treatment according to the program of the U-CAN project.

B. Develop,and validate a new imaging concept, Imiomics, for holistic, objective and automated analysis of whole-body imaging and non-imaging biomarkers in oncology. The first step of this project is to create a reference person. The image data must then lie in the same space, i.e. the image data has to be registered/matched/fused. This fusing process can be applied to a large number of healthy subjects, to create a mean shape or a normal reference person. Each pixel in this reference person then contain the distribution of all "normal" imaging parameters measured in the entire cohort. The idea of the open ended and a priori driven computer aided image analysis is to calculate the difference using all image data available from one patient compared to this "normal reference person" to facilitate detection of tumors. For early therapy evaluation of tumors, a similar comparison is performed between image data at diagnosis and during treatment. The introduction of molecular imaging (i.e. MRI and especially PET/CT) has caused a shift of paradigm which has lead to the possibility for measuring the biochemical activity in vivo of tumors. The whole body approach gives information of different cell populations, and their response to therapy, in spread cancer. The integrated PET/MR equipment will be used in oncological projects within U-CAN, where PET and MRI used separately are established techniques, but the combination of modalities is anticipated to amplify the information. Validation of PET/MR methods and development of a new concept for computer aided diagnosis and therapy evaluation in oncology, using whole-body data from the PET/MR equipment, are key components for the project. PET/MR investigations will be performed within prospective protocols to guide sampling of tissues and to evaluate therapeutic strategies. We will also evaluate modern multimodality radiotherapy to explore combinations of radiation and targeted drugs in order to select the optimal therapy for individual patients and lesions within one patient. The information from the PET/MR investigations has also the potential to be of importance in supporting the national proton beam therapy facility, the Skandion Clinic, a collaboration between all counties of Sweden, which starts in Uppsala June 2015.The additional clinical value of PET/MR in oncology using data from the PET/MR equipment, the U-CAN cohort and surgery as references will be evaluated. 

Members of the group during 2016

Joel Kullberg, Ass Professor   
Jan Weis, Docent, Physicist 
Tomas Bjerner, Ass Professor, MD 
Charlotte Ebeling Barbier, MD 
Robin Strand, Ass Professor    
Filip Malmberg, Ass Professor   
Lina Carlbom, PhD student 
Per Hammar, PhD student   
Emil Johansson, PhD student 
Jonathan Andersson, PhD student 
Catrin von Below, PhD 
Simon Ekström, PhD student 
Elin Lundström, PhD student