In Vivo 45-min Dynamic Scan. Performance of 18F-DPA-714, 18F-VUIIS1009A and 18F-VUIIS1009B was evaluated during the 45-min dynamic PET scan using a focal cerebral ischemic rat model, in which TSPO expression was up-regulated with activated microglia, whereas the blood-brain barrier (BBB) was intact or not seriously disrupted[31]. As expected, the ipsilateral side (ischemia side) showed a higher uptake of the radiotracer than the contralateral side of the brain in the PET imaging for all of the three radiotracers, as shown in Fig. 2a, 2b for 18F-DPA-714, Fig. 2c, 2d for 18F-VUIIS1009A and Fig. 2e, 2f for 18F-VUIIS1009B. A fast uptake of the radiotracers was demonstrated according to the analysis of time-activity curves (TACs) on the ipsilateral side (Fig. 2g for 18F-DPA-714, Fig. 2h for 18F-VUIIS1009A and Fig. 2i for 18F-VUIIS1009B). Moreover, consistent levels of radiotracer uptake at the ipsilateral side were also achieved 2 min after the radiotracer injection as indicated by TACs (Fig. 2g for 18F-DPA-714, Fig. 2h for 18F-VUIIS1009A and Fig. 2i for 18F-VUIIS1009B). Compared to 18F-DPA-714 (%ID/cc = 0.73 at 45 min), 18F-VUIIS1009B (%ID/cc = 0.70 at 45 min) has a comparable %ID/cc in the neuroinflammation region, while 18F-VUIIS1009A (%ID/cc = 0.50 at 45 min) has a significantly lower uptake in the same region as shown in Fig. 2c, 2d and 2 h.
The contralateral side had a comparatively low uptake, and a consistent %ID/cc level was achieved at 10 min after the radiotracer injection for all three radiotracers (Fig. 2g, Fig. 2h and Fig. 2i). When compared to 18F-VUIIS1009A, 18F-VUIIS1009B and 18F-DPA-714 demonstrate a higher contralateral healthy brain uptake as indicated in the 45 min uptake (0.20 vs 0.15, ID%/cc at 45 min). For 18F-DPA-714, the %ID/cc ratio at 45 min between the ipsilateral side and contralateral side in this study was 3.50 as shown in the TACs (0.73 vs. 0.20%ID/cc at 60 min; Fig. 2g), 3.33 for 18F-VUIIS1009A (0.50 vs. 0.15%ID/cc at 45 min) and 3.50 for 18F-VUIIS1009B (0.70 vs. 0.2%ID/cc at 45 min), indicating a high and comparable signal-to-noise ratio of all the three TSPO radiotracers in this study.
In Vivo Displacement Study. The in vivo displacement study was also performed to evaluate TSPO binding specificity for both 18F-VUIIS1009A and 18F-VUIIS1009B in the same rats. 19F-VUIIS1009A and 19F-VUIIS1009B were used for the displacement assay and were injected at 20 min in the course of a 45-min dynamic scan. Based on the imaging results, both TSPO compounds could displace the radioactivity uptake on the ipsilateral side of the brain. As shown in Fig. 3a(18F-VUIIS1009A) and 3c(18F-VUIIS1009B), the uptake on the ipsilateral side was significantly higher than the uptake on the contralateral side for both radiotracers before the injection of their cold analogs. However, after the displacement, radiotracer uptake of the ipsilateral side dropped dramatically (Fig. 3b and 3d), which accounts to more than 66% decrease of the tracer uptake as shown in the TACs for both radiotracers (as shown in Fig. 3e and 3f). After displaced with their cold analogs, the uptake level of ipsilateral side was almost comparable to the contralateral side, both demonstrated by the imaging profiles and TACs in Fig. 3. All these indicate the significantly high binding specificity for both radiotracers in neuroinflammation region.
Macro Parameters Determined Using Graphic Analysis
Although both 18F-VUIIS1009A and 18F-VUIIS1009B have significantly higher in vitro binding affinities when compared to 18F-DPA-714, they did not demonstrate a more promising imaging characterized by the semi-quantitative uptakes (like %ID/cc) as shown in Fig. 2. In order to further evaluate their in vivo performances, macro parameters like VT, DVR and BPND were determined using graphic analysis with AIFs or reference tissue TACs input. According to our previous studies, TSPO PET imaging can be analyzed using a two-tissue, four-parameter model as shown in Fig. 4a. Based on this model, graphic analysis using AIFs can also be employed to determine the VT for both neuroinflammatory region and the normal brain region. As shown in Table 1, VT determined for inflammation are higher than the corresponding values for normal brain for all the three radiotracers, indicating all these radiotracers have the tendency to be uptake by inflammation region instead of the healthy brain. Direct comparison of the inflammation VT from three radiotracers indicates 18F-VUIIS1009B has a higher VT than the other two radiotracers as shown in Table 1. Moreover, for the contralateral healthy brain, 18F-VUIIS1009B also demonstrates a slightly higher VT than the other two radiotracers. The ratios of the VT values (noted as DVR) between inflammation region and the contralateral healthy brain indicate a higher DVR for 18F-VUIIS1009B (8.53 ± 1.06) when compared with 18F-DPA-714 (6.00 ± 0.41) and 18F-VUIIS1009A (4.37 ± 0.82). Further study using contralateral healthy brain as a reference region was also performed to determine DVRs and BPND for all three radiotracers (as shown in Fig. 4 and Table 1). Logan plots using contralateral brain as a reference region was plotted with BPND and DVR determined (Fig. 4b, 4c and 4d), demonstrating a good fit and linearity for all three radiotracers with r > 0.96. As shown in Table 1, 18F-VUIIS1009B (7.55 ± 0.65) demonstrating a much higher DVR than 18F-DPA-714 (5.37 ± 0.36) and 18F-VUIIS1009A (3.91 ± 0.50). Similarly, as for the relationship between DVR and BPND (noted as DVR = BPND + 1), 18F-VUIIS1009B also demonstrates a much higher BPND value than other two radiotracers as shown in Table 1. In sum, the macro parameter analysis revealed that 18F-VUIIS1009B has a superior imaging potential than the other two radiotracers on VT, DVR and BPND.
Parametric Image Analysis
Macro parameter analysis using Logan plot revealed 18F-VUIIS1009B is superior to 18F-DPA-714 and 18F-VUIIS1009A as shown in dynamic PET imaging. In this study, we further compared the performance of the three radiotracers with parameter images (VT and BPND) generated (Fig. 5). In detail, voxel-wise VT images were generated via a Logan graphic method with AIFs input. Voxel-wise BPND images were generated using a Logan reference model with the contralateral healthy brain input as the reference region. As expected, compared to 18F-DPA-714, 18F-VUIIS1009B has similar biodistribution profile for neuroinflammation region in both VT and BPND parameter images (Fig. 5), but with higher VT, BPND in the neuroinflammatory region, demonstrating its promising characteristics for TSPO imaging. Interestingly, although 18F-VUIIS1009A and 18F-VUIIS1009B feature the comparably high in vitro binding affinity, 18F-VUIIS1009A did not show a similar imaging profile as 18F-VUIIS1009B in both VT and BPND parametric images (Fig. 5). Compared to the parametric image of 18F-VUIIS1009B, 18F-VUIIS1009A imaging analysis generated a noisier parametric image as shown in Fig. 3a and Fig. 3b.
Correlation between %ID/cc, BPND and VT at the Voxel Level
This study also determined and evaluated %ID/cc, BPND and VT at the voxel level, with the aim to better elucidate the relationship among these parameters derived from the dynamic PET images. As shown in Fig. 6, with %ID/cc, BPND and VT determined for the voxels in the neuroinflammatory region, we made different plots for the three probes and measured the correlation coefficient r and p value for each data set. As shown in Fig. 6a and 6c, a strong correlation was elucidated for 18F-DPA-714, 18F-VUIIS1009B among all three parameters (%ID/cc, BPND and VT). Compared to 18F-DPA-714, 18F-VUIIS1009B demonstrates a higher r value (0.99 vs. 0.78) and thus a more positive linear relationship between VT and BPND as shown in Fig. 6a and 6c. Furthermore, 18F-VUIIS1009B also demonstrates a stronger correlation between semi-quantitative parameter %ID/cc and BPND or VT, as shown by the plots in Fig. 6a and 6c, which is probably due to the lower non-specific binding profile for 18F-VUIIS1009B instead of 18F-DPA-714. While for 18F-VUIIS1009A, a weaker correlation is found between %ID/cc and the other two parameters (BPND and VT) as shown in Fig. 6b.
Correlation Analysis between BPND for 18F-VUIIS1009B and 18F-DPA-714
In this study, both 18F-VUIIS1009B and 18F-DPA-714 PET imaging were performed using the same rats with the aim to more accurately reflect their BPND and in vivo performance. In order to compare the performance of 18F-VUIIS1009B and 18F-DPA-714, we co-registered the PET images from the same rats and calculated BPND for 18F-VUIIS1009B and 18F-DPA-714 both at the regional level and voxel level with the same ROIs. As shown in Fig. 7a, 18F-VUIIS1009B demonstrates a strong positive correlation with 18F-DPA-714 (r = 0.88) at the voxel level in the same rat. The slope of the fitting curve is 1.35, which indicate the BPND for 18F-VUIIS1009B is more sensitive to TSPO expression when compared to 18F-DPA-714. Regional BPND was also obtained and analyzed for both 18F-VUIIS1009B and 18F-DPA-714 for the same rats (n = 6). As shown in Fig. 7b, 18F-VUIIS1009B also demonstrates a stronger correlation of 18F-DPA-714 in the regional BPND (r = 0.89) for a cohort of rats (n = 6). The slope for the fitting curve is 1.45, which also demonstrates the higher sensitivity of 18F-VUIIS1009B to profile TSPO expression when compared to 18F-DPA-714.