1361

Development of contrast agents for simultaneous PET/MRI of murine tumor models

Samuel C. Gilmore¹, Abbie Shepard², Christine Howison³, Joshua M. Goldenberg², Julio Cárdenas-Rodríguez³, and Mark D. Pagel³

¹University of Arizona, Tucson, AZ, United States, ²Pharmaceutical Sciences, University of Arizona, Tucson, AZ, United States, ³Medical Imaging, University of Arizona, Tucson, AZ, United States

Synopsis

PET/MRI contrast agents represent a new field of molecular imaging that provide an outstanding opportunity to employ simultaneous PET/MRI instrumentation for quantitative imaging. Over 90 responsive MRI contrast agents have been reported that change their contrast based on a biomarker and agent concentration, limiting their utility in vivo because a change in signal could arise from the presence of the biomarker and/or a change in the concentration of the agent. We propose to use PET to quantify the agent concentration, and a comparison of PET and MRI contrast can quantitatively evaluate the biomarker in a concentration-independent manner. Herein we describe the synthesis and characterization of ¹⁸F-radiolabeled responsive MRI contrast agents that are designed to measure pH and redox state using simultaneous PET/MRI for small animal studies.

Purpose

We sought to develop PET/MRI contrast agents that can quantitatively evaluate biomarkers such as pH and redox state. These agents consist of an ¹⁸F radioligand that is rapidly conjugated to a T₁-relaxation or T₂-relaxation-based MRI contrast agent. The radiolabel can be monitored with PET to measure the concentration of the agent. The MRI contrast agent changes image contrast based on the agent’s concentration and dependence on the biomarker. The simultaneous acquisition of PET and MRI images can be used to quantitatively measure the biomarker more specifically.

Methods

We have developed a synthesis method that conjugates an aminooxy ligand to the a-amino group of a Gd-DOTA derivative. ¹⁸F-fluorodeoxyglucose (¹⁸FDG) can selectively couple to an aminooxy group to form a stable oxime (Figure 1).¹ We are using this method to couple ¹⁸FDG to Gd-DOTA-4AMP that can generate T₁-weighted contrast that is dependent on pH and the concentration of the agent.² The PET and MRI signals can be used to estimate the T₁ relaxivity, and the pH can be calculated using a calibration curve between relaxivity and pH. Importantly, a ¹⁹F version of the agent can be synthesized using the same approach. A mixture of ¹⁸F- and ¹⁹F-labeled agent can be injected that have identical pharmacokinetics, so that ¹⁸F can be detected at ~10 nM concentration and the MRI contrast agent can be detected at ~100 mM concentration.

We have developed similar synthesis schemes that incorporate an aminooxy ligand in SLENU (Figure 2).³ The same radiolabeling of this MRI contrast agent with ¹⁸FDG leads to T₁-weighted contrast that depends on redox status and agent concentration, and PET signal that is related to the agent concentration. A comparison of the simultaneous PET and MRI detection can lead to quantitative measurement of redox status. To simultaneously detect the PET signal and T₁-weighted MRI contrast from our PET/MRI contrast agents, we have obtained a NuPET^TM PET detector (Cubresa, Inc., Winnepeg MN) that inserts into a 7T magnet with a 20 cm bore of a Biospec MRI instrument (Bruker Biospin, Inc., Billerica, MA). To demonstrate simultaneous PET/MRI with contrast agents, we have simultaneously performed ¹⁸FDG PET imaging with Dynamic Contrast Enhanced (DCE) MRI, using the following protocols (Figures 3,4): PET Protocol, each mouse was injected with 1.50 mCi ¹⁸F-FDG 30 minutes before starting any imaging procedure; next, each mouse was anesthetized with 1.5% isoflurane and secured to a customized cradle manufactured by Cubresa, Inc. PET data was acquired in list mode continuously immediately after placing each mouse in the imaging scanner. A final single PET image was reconstructed using a back projection method and all of the acquired data. MRI Protocol: The following MRI scans were acquired for all mice: a) Anatomical localizer with an standard spin echo MRI protocol, b) Saturation recovery sequence used to estimate T₁ time, c) T₁-weighted MRI at a temporal resolution of 6.4 seconds. Next, Multihance was injected at 10 mM/Kg in 100 L of saline after ~60 seconds of initiation the DCE MRI scan and continued for 30 additional minutes. The MR images were analyzed with the Linear Reference Region Model.⁴

Results

We have synthesized DOTA with the aminooxy ligand. We are currently refining our synthesis if DOTA-4AmP and SLENU with an aminooxy ligand (Figures 1,2). We are also refining methods to chelate Gd(III) and optimize the yield of coupling ¹⁹F-FDG and ¹⁸F-FDG to aminooxy-DOTA, aminooxy-DOTA-4AmP, and aminooxy-SLENU. The spatial distributions of the ¹⁸FDG uptake and the relative permeability of the tumor and muscle (RKtrans ) from DCE MRI were overlaid on an anatomical MR image (Figures 3,4). The results showed good ¹⁸FDG uptake and strong vascular permeability in the tumor rim relative to the core. Overall, these results demonstrated that simultaneous PET/MRI is feasible with PET and MRI contrast agents, and is therefore feasible with coupled PET-MRI contrast agents.

Discussion

More generally, over 90 MRI contrast agents have been reported that are responsive to molecular biomarkers.^5,6 Our research shows that many of these contrast agents can be easily modified to become PET/MRI contrast agents. In particular, MRI contrast agents that can measure oxygen saturation and can detect enzyme activity can be converted into PET/MRI agents. T₂-relaxation MRI contrast agents can also be augmented to be PET/MRI contrast agents. A key development is the availability of simultaneous PET/MRI that can track these contrast agents, which provides an outstanding rationale for the continued development and dissemination of simultaneous PET/MRI instrumentation.