Cerebrospinal fluid sampling has been recommended as a biomarker for in vivo tau burden, but there have been conflicting results regarding the correlation of phosphorylated tau protein with neocortical neurofibrillary pathology in AD.1–3 Also, cerebrospinal fluid tau does not predict the spatial tau distribution in the brain. The recent development of 18F-labeled PET tracers that target tau in vivo has enabled imaging of the spatiotemporal distribution of tau,4–6 an important biomarker of Alzheimer disease (AD). Research Framework of the National Institute on Aging and Alzheimer's Association has defined AD on the basis of its underlying pathologic processes, which can be documented on postmortem examination or by in vivo biomarkers.7 Biomarkers for AD have been grouped into those indicative of β-amyloid deposition (A), pathologic tau deposition (T), and neurodegeneration (N). Based on the AT(N) classification, cognitive impairment in an individual can be classified into the AD continuum, enabling a more accurate understanding of the temporal course of the disease process. This moves the definition of AD toward a biological construct as of the clinical definition wherein AD was a syndromic construct.7 Further, this would improve our understanding of interventional agents targeting specific pathological pathways in vivo.
Hypothetical models for AD pathophysiology have suggested a sequential association of events in which amyloid precedes tau pathology, which in turn would precede neuronal dysfunction, as identified on 18F-FDG that would further affect downstream cognitive function.8
We undertook this study to investigate the relation between in vivo regional tau retention (using 18F-AD-ML104) with cerebral glucose metabolism (using 18F-FDG) and cognitive impairment.
PATIENTS AND METHODS
This was a prospective study undertaken between February 2016 and November 2017 in which patients referred from the cognitive clinic of Geriatric Medicine were included. Ours is an apex tertiary care hospital. Ethical clearance was obtained from Institute Ethics Committee (ECPG/175/27.01.2016, RT-35/24.02.2016). Written informed consent was obtained from each patient or the attendant after explaining the details of the tests he or she would be undergoing.
A total of 70 subjects (Table 1) were included in the study and divided into 3 groups based on their clinical and neuropsychological evaluation. Cognitive performance was assessed using Mini Mental State Examination (MMSE); those patients with an MMSE score less than 18 were classified as advanced AD, whereas those with an MMSE score between 18 and 24 were classified as early AD. The third group included healthy elderly individuals with no cognitive impairment. The diagnosis of AD was made using appropriate criteria9 by the geriatrician.
Healthy controls were included if they did not have any history of major depressive episodes and/or antidepressant treatment, subjective memory complaints, or cognitive impairment, as assessed on neuropsychological evaluation.
18F-FDG PET/CT Protocol
Each patient was injected 5 to 6 mCi (185–222 MBq) of 18F-FDG intravenously in euglycemic state. Images were acquired on a dedicated PET/CT scanner (Biograph mCT; Siemens, Erlangen, Germany) beginning 60 minutes after 18F-FDG injection. Patient was positioned supine with head in a headrest. An initial scout was followed by noncontrast head CT (110 mA, 120 kVp), following which a 3D emission PET was obtained, single-bed position for 15 minutes. PET images were reconstructed using ultra-HD PET (True X + TOF), 5 iterations, and 21 subsets.
18F-AD-ML104 PET/CT Protocol
The cassette for synthesis of 18F-AD-ML104 (18F-THK 5351) was obtained from Neptis-Ora, and labeling was carried out on site. PET/CT images were acquired with the patient in supine position beginning 60 minutes after IV injection 5 to 6 mCi (185–222 MBq) of 18F-AD-ML104 (Neptis-Ora). An initial scout was followed by noncontrast head CT (110 mA, 120 kVp), following which a 3D emission PET was obtained, single-bed position for 15 minutes. PET images were reconstructed using ultra-HD PET, 5 iterations, and 21 subsets.
Visual interpretation of each of tau and FDG scans was carried out independently by an experienced nuclear medicine physician as both scans were done at least a week apart. They were blinded to the structural imaging and clinical findings. The regional pattern of hypometabolism was evaluated on the 18F-FDG, and the patterns of tau retention were reported on the 18F-AD-ML104.
Both scans (FDG and tau) of each patient were evaluated using regions of interest (ROI) drawn on the following regions: bilateral frontal, temporal, parietal, hippocampal, parahippocampal, anterior and posterior cingulate, precuneus, occipital (a total of 18 cortical ROI), and cerebellum. Ratios between the SUVmax of each of the regions to cerebellum (SUV ratios [SUVR]) were generated and used for intersubject comparison. We used SUV as it has been shown to correlate closely with binding potential estimates from kinetic studies.10
Group differences were assessed using χ2 test for discrete variables and Mann-Whitney U test for continuous variables. Linear regression models were performed to investigate association of tau retention and FDG metabolism with each other and with cognitive performance. Results were considered significant at P < 0.05.
Regional 18F-FDG PET Analysis Between Groups
Early AD Versus Normal Elderly
There were significant differences in SUVR (P < 0.0001) between early AD and normal elderly subjects among parietal, temporal, hippocampus, parahippocampus, anterior cingulate, posterior cingulate, and precuneal cortices, with the early AD group showing lower SUVRs.
AD Versus Normal Elderly
There were significant differences in SUVR (P < 0.0001) between AD and normal elderly subjects among parietal, temporal, hippocampus, parahippocampus, anterior cingulate, posterior cingulate, precuneus, and frontal regions, with the AD group showing lower SUVRs.
AD Versus Early AD
Significant differences in SUVR (P = 0.002) were noted among parietal, temporal, parahippocampus, precuneus, and frontal regions in AD versus early AD with lower SUVRs in AD.
Regional 18F-AD-ML104 PET Analysis Between Groups
Early AD Versus Normal Elderly
Early AD patients showed higher 18F-AD-ML104 binding, compared with normal elderly. The SUVRs were significantly higher in parietal lobe, temporal lobe, hippocampus, parahippocampus, anterior and posterior cingulate, and precuneus in early AD (P < 0.0001; Fig. 1).
AD Versus Normal Elderly
AD patients showed higher 18F-AD-ML104 tau SUVR in the frontal lobe, parietal lobe, temporal lobe, hippocampus, parahippocampus, anterior and posterior cingulate, and precuneus compared with normal elderly (P < 0.0001).
AD Versus Early AD
Significant differences in tau retention were seen in the parietal lobe (P < 0.0001), temporal lobe (P = 0.012), parahippocampus (P = 0.006), anterior (P = 0.018) and posterior cingulate (P = 0.002), and in the precuneus (P = 0.007) in AD versus early AD subjects with a higher binding in AD.
The maximum SUVR for tau retention was noted in the hippocampus in early AD and in the precuneus in the AD patients with maximum hypometabolism in these regions on corresponding FDG PET images (Fig. 2). The entire AD group was then divided into 2 groups based on age. The younger than 70 years age group showed higher tau binding, and more hypometabolism in the regions evaluated with a significant difference (P < 0.05) seen in the precuneus and posterior cingulate (individual region SUVR are in supplemental material upload).
There was a negative correlation between SUVRs on tau PET and 18F-FDG PET in AD patient with value of correlation coefficient close to −1 (r = −0.82 maximum). This negative correlation was seen in the frontal, parietal, temporal, hippocampus, parahippocampus, anterior and posterior cingulate cortices, and precuneus (Fig. 3).
No correlation was seen between the 2 tracers in normal elderly individuals with correlation coefficient almost equal to zero.
Correlation Analysis (Spearman ρ)
Correlation analysis (Spearman ρ) of hypometabolism on 18F-FDG PET using SUVR of the precuneus and cognition using MMSE scores revealed moderate positive correlation in AD patients (ρ = +0.42, P = 0.012).
Correlation analysis (Spearman ρ) of regional tau retention using SUVR of precuneus with cognition using MMSE scores revealed moderate negative correlation (ρ = −0.39, P = 0.022).
18F-AD-ML104 is the THK compound marketed by Ora, Neptis. As has been reported for 18F-THK 5351, there was good affinity of 18F-AD-ML104 for tau deposits in both early and advanced AD patients11,12 with a regional distribution that matched the pattern reported in AD patients.5,12
SUVRs for cortical tau binding in AD were higher than the early AD group. This enabled excellent discrimination between AD including early AD patients and healthy controls (significant P). The regions with high tau retention and hypometabolism corresponded to the regions involved in AD, namely, the parietal including precuneus and posterior cingulate, as well as temporal and frontal cortices. The highest binding of tau in the early AD group was in the hippocampus, and in AD, it was highest in the precuneus. This resembles the stereotypical pattern of the temporal spreading of tau pathology13 and implies that though the tau deposition starts in the mesial temporal regions with a temporal increase in the precuneus and posterior cingulate cortices, thus providing in vivo evidence of the utility of tau tracer for mapping the spatiotemporal pattern of tau deposition in AD. The precuneus as compared with other cerebral regions is a significant region that shows early and significant changes in glucose metabolism and tau radiotracer uptake in the early developmental period of AD as well as in advanced disease.
The lowest SUVRs on FDG PET were observed in the hippocampus in early AD and precuneus in AD patients, which corresponded to the increased 18F-AD-ML104 tau radiotracer binding in these regions. The increased tau deposition showed negative correlation with glucose metabolism in all brain regions known to be affected in AD patients. These differences were also seen in early AD with significant reductions in brain glucose metabolism and corresponding maximum tau radiotracer uptake (18F-AD-ML104). The AD patients with age younger than 70 years showed higher tau retention (higher SUVR) and greater hypometabolism (lower SUVR) on FDG than those with age older than 70 years. A higher tau binding has been reported in younger AD patients previously14 along with higher retention in early-onset AD,6 which corresponds to a greater pathological burden in early-onset AD.15
The highest 18F-AD-ML104 retention in the cerebral cortices was noted in the AD patients with a lower MMSE compared with early AD, and this corresponded to the temporal distribution of hypometabolism on FDG PET. Tau retention as assessed by 18F-AD-ML104 in AD patients revealed a significant negative association with glucose metabolism and cognition. This is similar to the results of Aubert et al,12 which was the first study investigating the relationship between regional 18F-THK 5317 retention and cognition in vivo. Our findings support the hypothesis that regional tau deposition in AD is closely related to cognitive performance.16 Negative associations between tau retention and cognition were noted in the temporal (among the first to be affected by NFT), parietal (both AD groups), and frontal lobes (in advanced AD groups). Tau retention was noted in neocortical regions and was associated with cognitive impairment similar to that reported by Aubert et al,11 wherein this association was seen for all the cortical regions primarily affected by tau pathology in AD.17,18 Aubert et al12 reported the association between 18F-THK5317 retention, and the full-scale intelligence quotient score was seen in several ROIs, whereas the association with the MMSE score was only seen in the inferior temporal gyrus. Other tau tracers have also reported this association for the inferior temporal gyrus and parahippocampal gyrus.19 We however found the association between tau retention and MMSE for multiple regions, and therefore global cognitive scores with MMSE also reflect tau progression.
The off-target binding seen in control subjects in our group was also seen in the AD group, and this was previously reported for all tracers.5,20,21 A recent study examining the correlation of in vivo THK-5351 retention with histopathology in AD brain demonstrated that the PET signal reflected a combination of tau pathology and reactive astrocytes, and correlated with neocortical paired helical filament-tau and MAO-B levels. This tracer would thus reflect tau-associated neuroinflammatory changes in AD.22 An in vivo blocking study showed substantial reduction of THK retention after administration of MAO-B inhibitor selegiline.23 Striatal and midbrain 18F-AD-ML104 retention was observed in all clinical groups, including healthy controls, in accordance with postmortem data. These findings limit the specificity of the THK analogs for selective detection of tau pathology, which is expected to be overcome the second-generation tau tracers. Early reports on the second-generation tracers have in fact suggested reduction of off-target binding with high binding to cortical regions with neurofibrillary tangles in AD.24,25
The limitations of this study included the small number of subjects/patients in each of the subgroups. We relied on cross-sectional data to provide explanations for a longitudinal process. AD was diagnosed clinically as we were not able to perform amyloid PET, which is the other important biomarker for AD. Our results were not supported be histological correlation. Further, we did not include corrections for partial volume effect in PET data.
This study shows that 18F-AD-ML104, a tau-specific PET tracer, can image the expected extent and regional distribution of tau pathology in patients at different clinical stages of AD. In support of studies in literature, we found negative correlation between 18F-FDG uptake and 18F-AD-ML104 retention in AD patients. There was a negative correlation between tau retention and global cognition assessed by MMSE with a positive correlation between hypometabolism on FDG PET and MMSE scores. Thus tau PET can be a potentially useful imaging biomarker for determining risk of developing clinical AD, and its regional distribution can complement FDG PET for the same.
1. Buerger K, Alafuzoff I, Ewers M, et al. No correlation between CSF tau protein phosphorylated at threonine 181 with neocortical neurofibrillary pathology in Alzheimer's disease. Brain
2. Buerger K, Ewers M, Pirttila T, et al. CSF phosphorylated tau protein correlates with neocortical neurofibrillary pathology in Alzheimer's disease. Brain
3. Engelborghs S, Sleegers K, Cras P, et al. No association of CSF biomarkers with APOEepsilon4, plaque and tangle burden in definite Alzheimer's disease. Brain
. 2007;130(Pt 9):2320–2326.
4. Lockhart SN, Baker SL, Okamura N, et al. Dynamic PET measures of Tau accumulation in cognitively normal older adults and Alzheimer's disease patients measured using [18
F] THK-5351. PLoS One
5. Chiotis K, Saint-Aubert L, Savitcheva I, et al. Imaging in-vivo tau pathology in Alzheimer's disease with THK5317 Pet in a multimodal paradigm. Eur J Nucl Med Mol Imaging
6. Cho H, Choi JY, Hwang MS, et al. In vivo cortical spreading pattern of tau and amyloid in the Alzheimer disease spectrum. Ann Neurol
7. Jack CR Jr, Bennett DA, Blennow K, et al. NIA-AA Research framework: towards a biological definition of Alzheimers disease. Alzheimers Dement
8. Jack CR Jr., Knopman DS, Jagust WJ, et al. Hypothetical model of dynamic biomarkers of the Alzheimer's pathological cascade. Lancet Neurol
9. McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement
10. Shcherbinin S, Schwarz AJ, Joshi A, et al. Kinetics of the tau PET
F-AV-1451 (T807) in subjects with normal cognitive function, mild cognitive impairment, and Alzheimer disease. J Nucl Med
11. Lemoine L, Saint-Aubert L, Marutle A, et al. Visualization of regional tau deposits using (3)H-THK5117 in Alzheimer brain tissue. Acta Neuropathol Commun
12. Saint-Aubert L, Lemoine L, Chiotis K, et al. Tau PET
imaging: present and future directions. Molecular Neurodegeneration
13. Braak H, Braak E. Neuropathological staging of Alzheimer-related changes. Acta Neuropathol
14. Pontecorvo MJ, Devous MD Sr, Navitsky M, et al. Relationships between flortaucipir PET tau binding and amyloid burden, clinical diagnosis, age and cognition. Brain
15. Marshall GA, Fairbanks LA, Tekin S, et al. Early-onset Alzheimer's disease is associated with greater pathologic burden. J Geriatr Psychiatry Neurol
16. Nelson PT, Alafuzoff I, Bigio EH, et al. Correlation of Alzheimers disease neuropathologic changes with cognitive status: a review of the literature. J Neuropathol Exp Neurol
17. Braak H, Alafuzoff I, Arzberger T, et al. Staging of Alzheimer disease-associated neurofibrillary pathology using paraffin sections and immunocytochemistry. Acta Neuropathol
18. Delacourte A, David JP, Sergeant N, et al. The biochemical pathway of neurofibrillary degeneration in aging and Alzheimer's disease. Neurology
19. Johnson KA, Schultz A, Betensky RA, et al. Tau positron emission tomographic imaging in aging and early Alzheimer disease. Ann Neurol
20. Maruyama M, Shimada H, Suhara T, et al. Imaging of tau pathology in a tauopathy mouse model and in Alzheimer patients compared to normal controls. Neuron
21. Ossenkoppele R, Schonhaut DR, Schöll M, et al. Tau PET
patterns mirror clinical and neuroanatomical variability in Alzheimer's disease. Brain
. 2016;139(Pt 5):1551–1567.
22. Harada R, Ishiki A, Kai H, et al. Correlations of 18
F-THK5351 PET with postmortem burden of tau and astrogliosis in Alzheimer disease. J Nucl Med
23. Ng KP, Pascoal TA, Mathotaarachchi S, et al. Monoamine oxidase B inhibitor, selegiline reduces F-18 THK 5351 uptake in human brain. Alzheimers Res Ther
24. Betthauser TJ, Cody KA, Zammit MD, et al. In vivo characterization and quantification of neurofibrillary tau PET
F-MK-6240 in humans from Alzheimer disease dementia to young controls. J Nucl Med
25. Lohith TG, Bennacef I, Vandenberghe R, et al. Brain imaging of Alzheimers dementia patients and elderly controls woth F-18 –MK-6240, a PET tracer targeting neurofibrillary tangles. J Nucl Med