Stereotactic Comparison Study of 18F-Alfatide and 18F-FDG PET Imaging in an LLC Tumor-Bearing C57BL/6 Mouse Model

This study aimed to stereotactically compare the PET imaging performance of 18F-Alfatide (18F-ALF-NOTA-PRGD2, denoted as 18F-Alfatide) and 18F-fluorodeoxyglucose (FDG) and immunohistochemistry (IHC) staining in Lewis lung carcinoma (LLC) tumor-bearing C57BL/6 mouse model. 18F-FDG standard uptake values (SUVs) were higher than 18F-Alfatide SUVs in tumors, most of the normal tissues and organs except for the bladder. Tumor-to-brain, tumor-to-lung, and tumor-to-heart ratios of 18F-Alfatide PET were significantly higher than those of 18F-FDG PET (P < 0.001). The spatial heterogeneity of the tumors was detected, and the tracer accumulation enhanced from the outer layer to the inner layer consistently using the two tracers. The parameters of the tumors were significantly correlated with each other between 18F-FDG SUV and GLUT-1 (R = 0.895, P < 0.001), 18F-Alfatide SUV and αvβ3 (R = 0.595, P = 0.019), 18F-FDG SUV and 18F-Alfatide SUV (R = 0.917, P < 0.001), and GLUT-1 and αvβ3 (R = 0.637, P = 0.011). Therefore, 18F-Alfatide PET may be an effective tracer for tumor detection, spatial heterogeneity imaging and an alternative supplement to 18F-FDG PET, particularly for patients with enhanced characteristics in the brain, chest tumors or diabetes, meriting further study.

In this study, we performed micro-PET imaging using both 18 F-alfatide and 18 F-FDG in an LLC tumor-bearing mouse model. We aimed to stereotactically compare the PET imaging performance and standardized uptake value of 18 F-Alfatide and 18 F-FDG, and also used immunohistochemistry (IHC) staining in a Lewis lung carcinoma (LLC) tumor-bearing C57BL/6 mice model.

Results
Biodistribution Data by PET Imaging. A total of 15 C57BL/6 mice were transplanted with LLC in the right thigh successfully, and then they underwent PET imaging. During imaging, no significant adverse events were observed. The radiotracer biodistribution was measured in major organs at 1 hour after injection of 18 F-Alfatide and 18 F-FDG (Table 1). The highest accumulation activity was found in the kidneys and bladder with both radiotracers, demonstrating renal clearance. The blood also showed moderate uptake with the two radiotracers PET imaging. High radiotracer accumulation was found in the brain, heart, lung, and muscle in 18 F-FDG PET scans, whereas those of 18 F-Alfatide PET were minimal.
The SUV comparison results by layer are summarized in Table 3   Immunohistochemical Validations and their Correlation with PET Imaging. PET imaging was validated by IHC examination. The staining of both GLUT-1 and α vβ 3 was mainly cytoplasmic (Fig. 3). The expression levels were 9.11 ± 1.08 for GLUT-1 and 3.86 ± 1.10 for α vβ 3. Figure 4 shows a positive correlation between SUV FDG and the GLUT-1 expression level (R = 0.895, P < 0.001), and between SUV RGD and the α vβ 3 expression level in tumors (R = 0.595, P = 0.019).
In contrast, Fig. 4C shows a very strong positive correlation between SUV FDG and SUV RGD (R = 0.917, P < 0.001). A strong positive correlation was also found between the GLUT-1 expression level and α vβ 3 expression level (R = 0.637, P = 0.011) (Fig. 4D). These correlation studies showed that the tumor SUV changes of 18 F-FDG and 18 F-Alfatide are consistent with each other, as well as with GLUT-1 and α vβ 3 expression in rough calculation.

Discussion
In this study, 18 F-Alfatide was shown to be a potentially effective tracer for the detection of brain, lung, and heart tumors with higher tumor-to-background ratios than 18 F-FDG in these organs. All xenografts were identified with high tumor-to-muscle ratios by both 18 F-Alfatide PET and 18 F-FDG PET imaging. 18 F-Alfatide PET and 18 F-FDG PET imaging were interrelated in tumor detection; SUV FDG , SUV RGD , GLUT-1, and α vβ 3 in tumors were correlated closely with each other, and the spatial heterogeneity of SUV FDG in different tumor layers was consistent with SUV RGD in rough calculation. 18 F-Alfatide, as a new RGD PET tracer, showed potential advantages for brain and chest tumors because of the high tumor-to-background ratio in vivo PET imaging. This finding was consistent with that in prior studies. In the application of lung cancer, Chen X et al. found that the primary lung boundary effects of RGD PET imaging is similar to FDG PET, and RGD PET provides better imaging for mediastinal lymph nodes and contralateral lung metastases 18 . Hiroshi Fushiki 19 et al. also reported that 18 F-FDG was specifically accumulated in tumors and the heart in the thoracic cavity, and there were some high background signals with 18 F-FDG PET in the chest, including the heart and skeletal muscle around the lung, indicating that 18 F-FDG PET showed difficulty in recognizing     the tumor. In our study, the same tumor-bearing mice imaged with 18 F-FDG not only showed a high uptake in chest tissues but also in the brain, thus making it difficult to delineate tumor metastases due to low tumor contrast; however, high tumor-to-background ratios exist in those tissues with 18 F-Alfatide PET images. Spatial heterogeneity in tumors was found in our study because the tracer accumulation was enhanced from the outer layer to the inner layer consistently with the two tracers. Tumor heterogeneity has been intensively investigated as a target to better serve tumor-individualized treatment. Different PET SUVs in different tumor areas represent different level of glycometabolism and angiogenesis and deserve different radiation dose for tumor control. This has been used in some clinical trial (For example, RTOG1106). Metz 20 et al. conducted a multi-image assessment of non-small cell lung cancer and other cancer patients, and simultaneously underwent RGD PET imaging for angiogenesis and FDG PET imaging for glucose metabolism. They found that the highest perfusion tumor sub-region was also the region with the highest α vβ 3 integrin expression (i.e., angiogenesis) and the highest glucose metabolism, and the hypoperfusion area was also the region with low α vβ 3 integrin expression and low glucose metabolism. Thus, this multi-mode imaging assessment for tumor heterogeneity is feasible, and this integrated multi-mode imaging can compensate for the lack of a single image. PET imaging would be a convenient method for noninvasive intratumor heterogeneity imaging and individualized radiotherapy. However, radiotracers' uptakes alone are not necessarily the only factor to consider in terms of tumor heterogeneity. Furthermore, clinical situations are far more complicated than the animal models. Whether spatial heterogeneity on PET imaging has correlation with tumor heterogeneity is still unclear and it deserves further study.
Hyperglycemia is associated with decreased FDG uptake by tumors as assessed by PET 21 . 18 F-Alfatide PET imaging is based on the expression level of α vβ 3, regardless of the glucose metabolism; thus, we believe that 18 F-Alfatide PET should have more advantages than 18 F-FDG PET in uncontrolled diabetic patients.
Many studies [22][23][24] have shown that 18 F-FDG uptake values have a strong relationship with the GLUT-1 IHC results. In patients with cervical 25 , ovarian 26 , and endometrial cancer 27 , it has been reported that a positive correlation exists between the SUV of the primary tumor and expression of GLUT-1. GLUT-1 expression, also related to tumor radioresistance at clinically relevant levels, has been reported in several studies 28 . Thus, FDG PET/CT is sensitive in detecting changes in tumor activity after treatment compared with conventional imaging methods 29 . In the current study, a strong positive correlation was found not only between FDG SUVs and GLUT-1 expression in tumors but also between 18 F-FDG SUV and 18 F-Alfatide SUV. The latter finding indicated that noninvasive 18 F-Alfatide PET imaging may play a similar role in evaluating tumor invasion, staging, or detecting changes in tumor activity after treatment using 18 F-FDG PET.
To further verify the conclusion as mentioned above, we explored the correlation between 18 F-Alfatide SUV and α vβ 3 IHC staining, and the correlation between IHC staining between α vβ 3 and GLUT-1. High 18 F-Alfatide uptake is due to high integrin expression in normal tissues and organs and consequently minimal nonspecific cardiac and lung activity accumulation with this radiotracer. In the current study, a significant positive correlation was found between 18 F-Alfatide SUV and α vβ 3 (R = 0.595, P = 0.019). Consequently, various radiolabeled RGD peptides have been developed for the noninvasive determination of α vβ 3 expression 30 . The most extensive use of RGD PET is the monitoring of tumor angiogenesis and anti-angiogenesis therapy 9,10 . In this study, a strong positive correlation was also found between GLUT-1 and α vβ 3 (R = 0.637, P = 0.011). Combined with the conclusions above, this study may further expand the application of 18   Conclusion 18 F-Alfatide PET may be an effective tracer for tumor detection, spatial heterogeneity imaging and an alternative supplement to 18 F-FDG PET, particularly for patients with enhanced characteristics with brain and chest tumors or diabetes, meriting further study. Although this preclinical study was performed successfully, the application value of 18 F-Alfatide PET and FDG PET needs further discussion.

Cell Culture and Animal Tumor Model Preparation. Murine LLC cells, recently used in several
high-profile preclinical studies 31,32 , were purchased from the Type Culture Collection of the Chinese Academy of Sciences, Shanghai, China. LLC cells were grown in RPMI 1640 (Sigma Chemicals Aldrich, Milan, Italy), supplemented with 10% fetal bovine serum and 1% penicillin streptomycin antibiotic mixture (Life Technologies, Inc.-Invitrogen, Grand Island, NY) in a humidified incubator (Heraeus, Hanau, Germany) at 37 °C with 5% CO 2 atmosphere. The LLC tumor model was generated by subcutaneous injection of 2 × 10 6 cells into the right hind leg of C57BL/6 mice (Charles River Lab). Animals were housed in a limited-access animal facility. The animal room temperature and relative humidity were set at 22 ± 2 °C and 55 ± 10%, respectively. Artificial lighting provided a 24-h cycle of 12-h light/12-h darkness (7 a.m. to 7 p.m.).
All animal procedures were approved by the Shandong Cancer Hospital & Institute Ethical Committee Guide for the care and use of Laboratory Animals. The methods were carried out in accordance with the approved guidelines.
PET Imaging. The mice were subjected to PET studies when the tumor diameter reached approximately 1 cm.
The simple lyophilized kit for labeling the PRGD2 peptide was purchased from the Jiangsu Institute of Nuclear Medicine, and the synthesis process was carried out in accordance with previous studies 33 . The radiochemical purity of the 18 F-Alfatide exceeded 95%, and its specific radioactivity exceeded 37 GBq (1,000 mCi)/μ mol. PET data acquisition was performed using an Inveon microPET scanner (Siemens Medical Solutions USA, Inc). With the assistance of the Inveon system's positioning laser, each LLC tumor-bearing mouse was placed with its tumor located at the center of the field of view, where the highest imaging sensitivity can be achieved. 18 F-FDG and 18 F-Alfatide images were performed 1 hour after tail-vein injection under isoflurane anesthesia. Each mouse underwent 18 F-FDG (2.6-3.6 MBq) PET within 2 days of the 18 F-Alfatide (2.4-3.5 MBq) PET scan. Before 18 F-Alfatide PET scanning, no specific subject preparation was applied, and the mice did not need fasting. Before the 18 F-FDG PET examinations, each mouse had been fasted for at least 4 hours. During the acquisition period, a thermostat-controlled thermal heater maintained the body temperature of the mice. PET emission images were taken from the head to the tail. The images were reconstructed using a 2-dimensional ordered-subsets expectation maximization algorithm. The 10-min static PET scans were then acquired at 1 hour after injection.  tumor were first evaluated by visual analysis, and then a quantitative analysis was performed by determining the SUV. Every of the outlined slices for each tumor was divided into outer, middle, and inner layers based on the luminance signal in the coronal plane by the images (Fig. 5). Then ROI was positioned around the tumor area of interest slice by slice and obtained a set of data such as ROImax, mean. The SUVs were calculated according to the following formula: [measured activity concentration (Bq/mL) × body weight (g)]/injected activity (Bq).
Immunohistochemistry. The mice were sacrificed to harvest the whole xenografts within 1 day after their micro-PET scans and processed routinely for integrin α vβ 3 and GLUT-1 IHC. In all cases, each tumor sample was fixed in 10% formalin and embedded paraffin. Each tumor sample was sectioned sequentially and transversely with a macrotome (Microm HM 450; GMI, Ramsey, MN) into approximately 3-to 5-μ m-thick 5 slices at 0.6-mm intervals. One slice was stained with HE, and the others were used for IHC studies. The expression levels of α vβ 3 and GLUT-1 were detected by IHC (pv-6000 two-step) (Zhongshan Golden Bridge Biotechnology Corporation, Beijing, China). Integrin α vβ 3 (1:400, Sigma) and GLUT-1 (1:250, Abcam) were used as primary antibodies. A section of normal leg muscle was used as the positive control, and negative controls were obtained by omitting the primary antibody.
Integrin α vβ 3-positive cells were stained with brown-yellow granules or masses, specifically in the cytoplasm. GLUT-1-positive expression was observed mainly in the cytoplasm. Both the intensity and percentage of positive cells were measured. The staining intensity was determined using the following four classes: 0 = undetectable; 1 = faint buff; 2 = moderate buff; and 3 = high buff or sepia. Stained cell sections in each case were randomly selected, and five high-power fields were counted under the microscope up to 400. We counted 200 cells in each region for a total number of 1,000 cells, calculated the percentage of positive cells, and then calculated the score. Using the percentage of stained cells × staining intensity, the integrated scoring was assessed. Cell expression was stratified as follows: 0 (negative) for no immunoreactivity, 1 for less than 25% positive cells, 2 for less than 50% positive cells, 3 for less than 75% positive cells, and 4 for more than 75% positive cells. The product of the staining intensity and positive cell scores determined the final result for each section.
Statistical Analysis. All quantitative data are expressed as the mean ± standard deviation (SD). Differences between continuous variables and dichotomous variables were tested by one-way ANOVA. Student's paired t-test was used to detect differences between the two sample means. Multiple comparisons were compared using LSD. Linear regression analysis was used to evaluate the correlation studies. All statistical tests were carried out using SPSS, version 17.0. Statistical significance was assumed for P values less than 0.05. All P values were 2-tailed.