J. Gordon Chan

Amyloid Imaging with 18 F-Florbetaben in Alzheimer Disease and Other Dementias

Amyloid imaging with 18F-labeled radiotracers will allow widespread use, facilitating research, diagnosis, and therapeutic development for Alzheimer disease. The purpose of the study program was to compare cortical amyloid deposition using 18F-florbetaben and PET in controls and subjects with mild cognitive impairment (MCI), frontotemporal lobar degeneration (FTLD), dementia with Lewy bodies (DLB), vascular dementia (VaD), Parkinson disease (PD), and Alzheimer disease (AD). Methods: One hundred nine subjects in 3 clinical studies at Austin Health were reviewed: 32 controls, 20 subjects with MCI, and 30 patients with AD, 11 with FTLD, 7 with DLB, 5 with PD, and 4 with VaD underwent PET after intravenous injection of 300 MBq of 18F-florbetaben. Standardized uptake value ratios (SUVR) using the cerebellar cortex as a reference region were calculated between 90 and 110 min after injection. Results: When compared with the other groups, AD patients demonstrated significantly higher SUVRs (P < 0.0001) in neocortical areas. Most AD patients (96%) and 60% of MCI subjects showed diffuse cortical 18F-florbetaben retention. In contrast, only 9% of FTLD, 25% of VaD, 29% of DLB, and no PD patients and 16% of controls showed cortical binding. Although there was a correlation between Mini Mental State Examination and β-amyloid burden in the MCI group, no correlation was observed in controls, FTLD or AD. Conclusion: 18F-florbetaben had high sensitivity for AD, clearly distinguished patients with FTLD from AD, and provided results comparable to those reported with 11C-Pittsburgh Compound B in a variety of neurodegenerative diseases.

Head-to-Head Comparison of 11C-PiB and 18F-AZD4694 (NAV4694) for β-Amyloid Imaging in Aging and Dementia

11C-Pittsburgh compound-B (11C-PiB) is the benchmark radiotracer for imaging of β-amyloid (Aβ) plaque in Alzheimer disease (AD). 18F-labeled Aβ tracers subsequently developed for clinical use show higher nonspecific white matter binding and, in some cases, lower cortical binding in AD that could lead to less accurate interpretation of scans. We compared the cortical and white matter binding of a new 18F-labeled Aβ tracer, 18F-AZD4694 (recently renamed NAV4694), with 11C-PiB in the same subjects. Methods: Forty-five participants underwent PET imaging with 11C-PiB and 18F-AZD4694 (25 healthy elderly controls [HCs], 10 subjects with mild cognitive impairment, 7 subjects with probable AD, and 3 subjects with probable frontotemporal dementia). Images were coregistered so that region-of-interest placement was identical on both scans, and standardized uptake value ratios (SUVRs) using the cerebellar cortex as a reference region were calculated between 40 and 70 min after injection for both tracers. Results: 18F-AZD4694 showed reversible binding kinetics similar to 11C-PiB, reaching an apparent steady state at 50 min after injection. Both radiotracers showed a similar dynamic range of neocortical SUVR (1.1–3.3 and 1.0–3.2 SUVR for 11C-PiB and 18F-AZD4694, respectively) and identical low nonspecific white matter binding, with frontal cortex–to–white matter ratios of 0.7 ± 0.2 and 1.3 ± 0.2 for both radiotracers in HCs and AD subjects, respectively. There was an excellent linear correlation between 11C-PiB and 18F-AZD4694 neocortical SUVR (slope of 0.95, r = 0.99, P < 0.0001). Conclusion: 18F-AZD4694 displays imaging characteristics nearly identical to those of 11C-PiB. The low white matter and high cortical binding in AD indicate that this tracer is well suited to both clinical and research use.

Radiation Dosimetry of β-Amyloid Tracers 11C-PiB and 18F-BAY94-9172

β-Amyloid (Aβ) imaging has great potential to aid in the diagnosis of Alzheimer disease and the development of therapeutics. The radiation dosimetry of Aβ radioligands may influence their application; therefore, we calculated and compared the effective doses (EDs) of 11C-PiB and a new 18F-labeled ligand, 18F-BAY94-9172. Methods: Attenuation-corrected whole-body scans were performed at 0, 15, 30, 45, and 60 min after injection of 350 ± 28 MBq (mean ± SD) of 11C-PiB in 6 subjects and at 0, 20, 60, 120, and 180 min after injection of 319 ± 27 MBq of 18F-BAY94-9172 in 3 subjects. Coregistered CT was used to define volumes of interest (VOIs) on the PET images. The source organs were the brain, lungs, liver, kidneys, spleen, and vertebrae. The VOIs for the contents of the gallbladder, urinary bladder, lower large intestine, upper large intestine, and small intestine were also defined. Total activity in each organ at each time point was calculated by use of reference organ volumes. The resultant time–activity curves were fitted with constrained exponential fits, and cumulated activities were determined. A dynamic bladder voiding model was used. The OLINDA/EXM program was used to calculate the whole-body EDs from the acquired data. Results: For 11C-PiB, the highest absorbed doses were in the gallbladder wall (44.80 ± 29.30 μGy/MBq), urinary bladder wall (26.30 ± 8.50 μGy/MBq), liver (19.88 ± 3.58 μGy/MBq), and kidneys (12.92 ± 3.37 μGy/MBq). The ED was 5.29 ± 0.66 μSv/MBq. For 18F-BAY94-9172, the highest doses were also in the gallbladder wall (132.40 ± 43.40 μGy/MBq), urinary bladder wall (24.77 ± 7.36 μGy/MBq), and liver (39.07 ± 8.31 μGy/MBq). The ED was 14.67 ± 1.39 μSv/MBq. Conclusion: The estimated organ doses for 11C-PiB were comparable to those reported in earlier research. With the doses used in published studies (300–700 MBq), the EDs would range from 1.6 to 3.7 mSv. The ED of 18F-BAY94-9172 was 30% lower than that of 18F-FDG and, at the published dose of 300 MBq, would yield an ED of 4.4 mSv. The dosimetry of both Aβ radioligands is suitable for clinical and research applications.

Imaging and quantitation of the hypoxic cell fraction of viable tumor in an animal model of intracerebral high grade glioma using [18F]fluoromisonidazole (FMISO)

We have demonstrated that FMISO uptake is significantly higher in tumor tissue in the C6 intracerebral glioma rat model compared to normal brain, and that there is persisting hypoxia in gliomas independent of tumor size. FMISO uptake was observed homogeneously throughout viable glioma tissue in tumor sizes ranging from 2mm to almost 1cm. Quantitation of uptake of FMISO showed a tumor/brain ratio of 1.9 and a tumor/blood ratio of 2.6 at 2 hours post injection.

Implementation of the centiloid transformation for F-18 nav4694: Comparison with PiB

(ANTs; http://stnava.github.io/ANTs/) to structural MRI data from the ADNI project (http://adni.loni.usc.edu) in order to investigate EC thickness and atrophy in healthy older adults, individuals along the mild cognitive impairment (MCI) continuum, and AD patients. The technique uses diffeomorphic registration to calculate cortical thickness in 3D volumes and compares favorably with existing surface-based approaches (e.g. Freesurfer). We included only participants who had more than one timepoint and were followed for one year or longer. We assigned subjects to one of 4 groups (Control, MCI, Late MCI, and AD) based on their ADNI diagnosis status and examined baseline differences in EC thickness as well as differences in rate of atrophy, operationalized as annualized atrophy rate normalized to individual subjects’ baseline thickness. Results: First, we assessed EC thickness in subjects’ baseline scans and found no effect of ApoE4, however, all diagnostic groups differed significantly from each other in the following manner: Control > late MCI > MCI > AD, which was confirmed by posthoc tests. Next we assessed EC thickness annualized atrophy across all groups. Here, we found that the atrophy rates in healthy controls were significantly different from MCI groups (both MCI and late MCI), and MCI atrophy rates were significantly different from AD. Therewere a trend towards significance in the group difference betweenMCI and late MCI (p1⁄4.09). We also found a significant effect of ApoE4 status, where the presence of one or more ApoE4 alleles was associated with higher annualized decline across groups. Conclusions: The results suggest that differences in EC thickness are associated with AD pathology, both at baseline and during progression. They also suggest a role for ApoE4 status in neurodegenerative changes, across all diagnostic groups.

Head-to-head comparison of 11C-PiB and 18F-AZD4694 for beta-amyloid imaging in aging and dementia

The presence of diffuse plaques, vascular amyloid angiopathy and cerebellar amyloid did not affect correlation of scan results with neuritic plaque densities or the ability of scan readers to correctly predict the presence of moderate to frequent neuritic plaques. Conclusions: Both visual and quantitative binary interpretation of florbetapir PET scans demonstrated significant concordance with histopathological assessment of neuritic plaque density regardless of the presence of other potentially confounding amyloid Ab pathologies.

O1-04-06: Correlation of memory dysfunction and amyloid burden in mild cognitive impairment

and MCI patients needed to evaluate a putative disease-slowing treatment in a six-month clinical trial, using PET images from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (www.loni.ucla.edu ADNI). Methods: Baseline and approximately sixmonth follow-up ADNI PET images were compared in separate analyses of 16 AD patients, 27 MCI patients and 27 NC using SPM5. Using the atlas coordinate with the maximally significant CMRgl decline in the respective AD and MCI comparison, power analyses were performed to estimate the number of patients needed to evaluate a putative primary prevention therapy in a six-month multi-center clinical trial. Results: There was a trend for approximately six-month MMSE declines in the AD patients (P 0.06) and MCI patients (P 0.06) but not in the NC (P 0.86). The AD patients had six-month CMRgl declines in left temporal, parietal and precuneus regions (maximal reduction 3.0%), the MCI patients had six-month CMRgl declines in bilateral temporal, right parietal and right frontal regions (maximal reduction 2.0%), and the NC had had six-month declines in the left temporal cortex (P 0.001, uncorrected for multiple comparisons). Using the maximal, left temporal CMRgl declines in each patient group, we estimate the need for at least 224 AD patients or 642 MCI patients per treatment arm to detect a putative disease-slowing treatment efficacy to reduce sixmonth CMRgl declines by 25% with 80% power (one-tailed P 0.005, uncorrected for multiple comparisons) in a multi-center clinical trial. Conclusions: This study provides preliminary information about sixmonth CMRgl declines in AD and MCI patients and a preliminary estimate of the number of patients needed to detect CMRgl effects in six-month clinical trials of putative disease-slowing treatments.