Study to Develop a Kinetic Model for FDG and Me4FDG in Kidneys of Type 2 Diabetic Patients

Overview

In this study, using 18F-FDG and Gd-DTPA PET/MRI, we are aiming to perform a dynamic PET/MRI imaging using 18F-FDG and Me4FDG for a group of type 2 diabetic patients scheduled for Glifozine therapy due to the bad metabolic control to assess changes in renal function before and 1 to 2 weeks after initiating therapy with Gliflozine. Furthermore we aim to study the temporal behavior of 18F-FDG and Me4FDG activity in certain kidney regions of the diabetic participants to estimate basic kidney parameters using time activity curve. Further, we intend to find a kinetic model that describes the behavior of glucose in each part of the kidney can be acquired mathematically and to find out whether conclusions about the glucose reabsorption capability of the kidney in diabetes can be achieved in general. In addition, we aim to simultaneously determine renal lesions as well as obstructions with the fused, high definition, and three dimensional images of the kidney and estimate kidney function parameters from the dynamic Gd-DTPA MRI scan and compare them to the kidney function determined with the kinetic model.

Independent of insulin, inhibition of sodium-glucose transporter 2 (SGLT2) in the proximal tubular cells prevents glucose reabsorption and promotes glucose excretion by causing glycosuria. Therefore, SGLT2 inhibitors known as Gliflozine are recently authorized for the treatment of type 2 diabetes mellitus (DM), whether as monotherapy or in combination with other anti-diabetic medications including insulin. To our knowledge, there is still no established kinetic model that describes precisely the reabsorption mechanism of glucose in the proximal tubules and that shows the impact of Gliflozine on renal function in diabetic patients. Therefore, the temporal behavior of glucose in the various kidney regions needs to be studied. Currently, the most promising tool is a combined positron emission tomography and magnetic resonance imaging (PET/MRI) using radioactive glucose analog 2-deoxy-2-(18F)fluoro-D-glucose (FDG). The MRI scan in combination with the kidney-specific gadolinium based contrast agent diethylenetriaminepentacetate (Gd-DTPA) images the organ with high resolution allowing an estimation of kidney parameters; the PET scan on the other hand shows the dynamic behavior of the glucose analog FDG. However, the reabsorption process of FDG in the kidney is controversial. For this reason, a dynamic PET/MRI image using 18F-FDG will be performed for diabetic patients, need the Gliflozine therapy, directly before and 1 to 2 weeks after therapy initiation. We aim, generally, to study the chronological behavior 18F-FDG activity in kidneys and to show, particularly, the reabsorption process under the influence of SGLT2 inhibition. Furthermore, conclusions about the glucose reabsorption capability of the kidney in these patients might be achieved and a kinetic model that describes the exact behavior of glucose in each part of the kidney can be mathematically acquired. We aim, additionally, to compare the PET/MRI images with the levels of diabetic metabolic control before and 1 to 2 weeks after initiating therapy with Gliflozine to show whether conclusions about therapy response can be drawn with FDG among the participants which arises from a broader peak in case of medication. A possible explanation is that the Patlak slope is among other influenced by the re-absorption process: if re-absorption is lowered due to the medication, the peak gets broader leading to a lower Patlak slope. We therefore conclude that reabsorption might be studied with FDG. Furthermore, a first draft of a kinetic model could be developed with the collected data According to this model, the re-absorption is mainly covered by the rate-constant k3. However, there was no correlation found between k3 and the above mentioned Patlak slope which most likely is a measure for re-absorption. Furthermore, the fit algorithm not always leads to meaningful results. This leads to the assumption that the model is not yet finished and further input data are needed. The model allows calculating the TACs of the visible renal sub-regions (Cortex, Medulla and Pelvis). In comparison with the measured TACs, the calculated ones deliver similar shapes. Although it seems from these results promising to study diabetes type II with a routine tracer like FDG, the kinetic model is hampered by the low affinity of FDG for the SGL transporter, which makes a quantification difficult. Therefore, an alternative radiopharmaceutical shall be applied which is similar to FDG but mainly re-abosrbed via SGLT: alpha-Methyl-4-\[18F\]FDG (Me4FDG).