Quantitative assessment of the repeatability of [18F]F-DPA-714 PET imaging in the brain and heart in preclinical models

Introduction: [¹⁸F]F-DPA-714 has specific binding to the translocator protein (TSPO), an 18kDa protein that is primarily found on the mitochondrial membrane and modulates the transportation of cholesterol. TSPO upregulation within the brain is due to activated microglia, indicative of neuroinflammation. Additionally, TSPO modulates the Ca²⁺ transport within the heart, and alterations to calcium transport can cause severe cardiotoxicity. Further, TSPO expression within the tumor has also been found to be positively correlated to increased tumor malignancy. While TSPO PET began with neuroimaging, it has become more prevalent and applications are increasing and expanding to cardiac, oncology and more. Recent studies have assessed the efficacy of TSPO as a target to visualize changes within the brain, heart, and tumor through PET imaging. To understand the potential and limitations of [¹⁸F]F-DPA-714 more testing needs to be conducted to ensure the repeatability and accuracy of quantitative analysis of [¹⁸F]F-DPA-714-PET imaging while also gaining a better understanding of baseline and variability of uptake within these critical organs to guide future experiments.

Methods: Athymic nude mice (N=8) were used to quantify the repeatability of [¹⁸F]F-DPA-714. Two imaging session were conducted 2 weeks apart, during each imaging session mice were injected via tail vein with ~150 mCi of [¹⁸F]F-DPA-714 and imaged 50 minutes post injection for 20 minutes with a preclinical PET/CT. Standardized uptake value (SUV) was quantified and normalized to muscle SUVmean. Percent difference between imaging session for each SUV metric was calculated as change from the highest SUV reported for each mouse. Following the second imaging session, mice were euthanized and selected organs were collected for autoradiography and biodistribution. Organs were collected, weighed, and analyzed in a gamma counter. Following the biodistribution the brain was cut down the hemisphere with one half formalin flexed along with the tumor and heart, while the other half was used for autoradiography. The brains developed overnight for autoradiography and scanned (Amersham Typhoon Biomolecular Imager) and compared with imaging data. For statistical testing unpaired t-test was performed (GraphPad).

Results: Normalized SUV mean, max, and peak values for the brain over the test-retest imaging sessions were found to be 1.6 ±0.6, 4.7 ± 2.3, and 4.3 ± 2.0, respectively. In the heart, SUV mean, max, and peak averaged to 7.8 ± 4.2, 13.6 ± 7.4, and 13.3 ± 7.4 SUV, respectively. Changes in SUV mean, max, and peak within the brain and heart show no significant changes (p>0.05). SUV max provided the highest repeatability with 23.7% difference in the brain. Within the heart, SUV mean reported the highest repeatability with 24.9% difference. However, all SUV metrics for brain and heart ranged from 23.9%-26.1% difference between each imaging sessions. Autoradiography data indicates average uptake of 0.75 ± 0.1 nCi/g in the brain. Biodistribution results showed an average of 0.5 ± 0.3, 3.6 ± 2.6, 2.3 ± 3.7, and 0.55 ± 0.3 %ID/tissue in the brain, heart, tumor, and muscle, respectively. Bland-Altman plots indicate that all data points are within limits of agreement for each SUV metric for both the brain and heart.

Conclusions: Test-retest of [¹⁸F]F-DPA-714 reveals no significant differences in uptake within the brain and heart in normal subjects. Further, it can be concluded that [¹⁸F]F-DPA-714 can be used as a reliable tracer when assessing longitudinal changes > 35% within these critical organs. Voxel-wise comparisons will need to be completed in the future to understand variability in regional tissue changes. Moving forward, [¹⁸F]F-DPA-714 changes can be used to assess multiple organs and tissue simultaneously in the body during systemic therapies. This data has potential to guide understanding of the baseline expression of TSPO in these regions of interest and the repeatability of the quantitative metrics per region.