PET Tracer Discriminates Alzheimer Disease From Other Neurodegenerative Disorders

September 21, 2018
Matt Hoffman
Matt Hoffman

Matt Hoffman, Senior Editor for NeurologyLive, has covered medical news for MJH Life Sciences, NeurologyLive’s parent company, since 2017, and previously wrote for its sister publication, HCPLive. Follow him on Twitter @byMattHoffman or email him at

The accuracy and potential utility of the [18F]flortaucipir tracer in patient care will require further research in clinically more representative populations.

Rik Ossenkopple, PhD

Amid patients with an established diagnosis from a memory disorder clinic, the use of positron emission tomography (PET) tracer [18F]flortaucipir provided a way to discriminate Alzheimer disease from other neurodegenerative conditions.

New study data has shown an estimated sensitivity of 89.9% and specificity of 90.6% for Alzheimer compared to other neurodegenerative diseases with [18F]floracuipir PET, while also outperforming the established volumetric magnetic resonance imaging (MRI) measures.

“[18F]flortaucipir is a relatively novel PET tracer that can be used to detect tau pathology in the living human brain," lead author Rik Ossenkopple, PhD, told NeurologyLive. "Previous studies have shown a robust signal in patients with Alzheimer’s disease, but in patients with other types of dementia, the signal was more variable. We aimed to assess the ability of [18F]flortaucipir PET to distinguish Alzheimer’s disease from other neurodegenerative diseases in more than 700 study participants."

The trial included a total of 719 patients from 3 centers in South Korea, Sweden, and the United States, of which 160 were cognitively normal controls, 126 had mild cognitive impairment, 179 had Alzheimer dementia, and 254 had various non-Alzheimer neurodegenerative disorders. Among those patient groups, 26.3%, 65.9%, 100%, and 23.8%, respectively, were amyloid-ß (Aß) positive.

The trial assessed the accuracy of the [18F]flortaucipir PET with standardized uptake value ratio (SUVR) in 5 predefined regions of interest: the entorhinal cortex, the inferior temporal cortex, the temporal meta-region of interest (consisting of the bilateral entorhinal, amygdala, fusiform, and the inferior and middle temporal cortices), the temporoparietal cortex, and Braak stages V and VI.

The PET tracers’ uptake in the medial-basal cortex and the lateral temporal cortex revealed those measurements of 89.9% sensitivity (95% CI, 84.6 to 93.9) and 90.6% specificity (95% CI, 86.3 to 93.9) with the threshold based on the controls (SUVR, 1.34). Using the Youden-Index derived cutoff (SUVR, 1.27) revealed a 96.8% sensitivity (95% CI, 92.0 to 99.1) and 87.9% specificity (95% CI, 81.9 to 92.4).

The area under the curve (AUCs) for all 5 of the [18F]flortaucipir regions of interest were higher (AUC range, 0.92 to 0.95) compared with the 3 volumetric MRI measures (AUC range, 0.63 to 0.75; all regions, P <.001). The diagnostic performance of the 5 [18F]flortaucipir regions of interest was lower in those with MCI due to AD (AUC range, 0.75-0.84).

The temporal meta-region of interest yielded a 90.3% accuracy (95% CI, 87.1 to 92.9), 89.9% sensitivity (95% CI, 84.6 to 93.9), and 90.6% specificity (95% CI, 86.3 to 93.9) using the threshold derived in controls (SUVR, 1.34).

Furthermore, when using the Youden Index-derived cutoff in the South Korea cohort (SUVR, 1.27) applied to the Sweden and US cohorts, accuracy was 91.7% (95% CI, 87.9 to 94.6), sensitivity was 96.8% (95% CI, 92.0 to 99.1), and specificity was 87.9% (95% CI, 81.9 to 92.4). Using the Youden Index-derived cutoff in the Sweden cohort (SUVR, 1.27) applied to the South Korea and US cohorts, accuracy was 87.7% (95% CI, 83.5 to 91.1), sensitivity was 92.1% (95% CI, 86.0 to 96.2), and specificity was 84.5% (95% CI, 86.0 to 96.2).

The voxel-wise analyses indicated that temporoparietal regions, including inferior and middle temporal cortices, medial temporal lobe structures, and posterior cingulate yielded the highest AUC (range, 0.85 to 0.95) for separating Alzheimer dementia from non-Alzheimer neurodegenerative conditions.

"We think that Tau PET imaging will improve the diagnostic workup at many memory clinics in the near future and might replace some less accurate assessments that are routinely used in current clinical practice," Ossenkopple said. "Tau PET tracers are currently solely used in investigational settings. The present study in conjunction with post-mortem studies verifying in vivo tau PET signal may facilitate the approval of tau PET tracers for clinical purposes in the next few years."

Along with Ossenkopple, Oskar Hansson, MD, PhD, and colleagues noted that while MRI, amyloid ß PET, and cerebrospinal fluid (CSF) are frequently used in addition to clinical examinations in those with cognitive impairment, [18F]flortaucipir PET’s utility has not been defined as a diagnostic biomarker.

“Because Aß is assumed to become abnormal approximately 15 to 30 years prior to dementia onset, many patients with a non-AD neurodegenerative disorder, especially at older age, will be in an Aß latency period, wherein Aß is detected by PET or CSF but is not the primary etiology underlying the clinical symptoms,” they wrote. “Due to the limited specificity, Aß biomarkers are therefore often used to rule out rather than rule in a diagnosis of [Alzheimer].”

Because of the tight temporal association of Alzheimer-like tau pathology with the clinical manifestation of the disease itself, Ossenkopple and colleagues noted that [18F]flortaucipir may be more discriminatively accurate in older populations in comparison with Aß biomarkers. This, they noted, is supported by the study findings, which showed an even more pronounced difference in specificity for [18F]flortaucipir compared to Aß status in those older than 69 years, in both non-Alzheimer neurodegenerative disorders (range, 17.8% to 23.3% increase; P <.001) and controls (range, 29.5% to 31.6% increase; P <.001). “Accordingly, an intended clinical use of [18F]flortaucipir PET might be to improve the diagnostic workup as an add-on test to Aß biomarkers in patients with early-onset dementia and possibly as a triage or even replacement test in patients with late-onset dementia in whom incidental Aß pathology is common,” they wrote.

Ossenhopple said that it was surprising that the sensitivity of [18F]flortaucipir PET dropped substantially at the prodromal stage of Alzheimer. "This suggests that, at least according to our threshold approaches and regions-of-interest, [18F]flortaucipir PET is most useful for the differential diagnosis of patients with dementia, rather than early disease detection," he said.


Ossenkopple R, Rabinovici GD, Smith R, et al. Discriminative accuracy of [18F]flortaucipir positron emission tomography for Alzheimer disease vs other neurodegenerative disorders. JAMA. 2018;320(11):1151-1162.


: 10.1001/jama.2018.12917.