Abstract
Objective
18F-flortaucipir (Tauvid™) is an 18F-labeled diagnostic positron emission tomography (PET) radiopharmaceutical, which was developed to detect tau pathology in the brain and to estimate the density and/or distribution of aggregated tau neurofibrillary tangles for patients being evaluated for Alzheimer’s disease (AD). The objective of this article is to report findings from two studies. The primary objective of the first (dose and biodistribution) study was to investigate the consistency of 18F-flortaucipir PET biodistribution between Japanese and non-Japanese individuals. In the second (reader) study, the comparability of visual reads performed by Japanese and US expert readers who followed the same training program was evaluated.
Methods
This article reports findings from two separate studies and datasets. In the dose and biodistribution study, whole-body distribution of 18F-flortaucipir was assessed (N = 9). Standardized uptake value ratios, normalized to the entire cerebellum, were calculated, and whole-body effective dose was calculated using a 73.7 kg bodyweight model. In the reader study (Study A27; I7E-AV-A27), five Japanese physicians visually interpreted 60 PET scans randomly selected from a previous US reader study (Study FR01; NCT03901092). Scan positivity was determined based on increased activity of 18F-flortaucipir in posterolateral temporal, occipital, or parietal/precuneus regions with or without frontal activity. Each scan was visually interpreted by all Japanese readers, who determined whether scans were consistent with an AD or non-AD pattern. Concordance between Japanese and US readers who underwent the same training on the interpretation of 18F-flortaucipir PET images was then evaluated (reader study primary endpoint).
Results
The biodistribution of 18F-flortaucipir was consistent between Japanese and non-Japanese individuals. When interpreting 18F-flortaucipir PET images to distinguish between AD versus non-AD patterns, Japanese and US readers who underwent the same training showed 100% agreement on the majority reads, with a Cohen’s Kappa value of 1.
Conclusions
18F-flortaucipir whole-body biodistribution for Japanese individuals was consistent with that of non-Japanese individuals. After completing the same training, Japanese readers showed high concordance with US readers on the interpretation of 18F-flortaucipir PET images to distinguish between AD and non-AD patterns. Data and scans were from previous studies registered with ClinicalTrials.gov (registration numbers: NCT04474405; NCT03901092).
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1007/s12149-026-02239-4.
Keywords: Alzheimer’s disease, Neurodegenerative disease, Positron emission tomography, Tauopathies, 18F-flortaucipir
Introduction
Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by a progressive decline in cognitive function and ability to perform activities of daily living, possibly leading to death due to complications [1]. Pathologic features of AD include amyloid beta plaques and neurofibrillary tangles (NFTs), which are composed of abnormal accumulations of hyperphosphorylated tau protein, and widespread downstream neurodegeneration [2].
Amyloid positron emission tomography (PET) imaging is a well-established diagnostic tool for the detection of cerebral amyloid pathology in patients with suspected AD [3]. Of note, amyloid accumulation is thought to potentially start well before onset of clinical symptoms, potentially reaching a plateau in the symptomatic phase [4]. In contrast, tau PET imaging has been shown to be more tightly associated with AD symptomatic onset [5], with potential value as a disease staging tool in the symptomatic phase [3].
18F-flortaucipir (Tauvid™) is an 18F-labeled diagnostic PET radiopharmaceutical developed to detect tau pathology in the brain, particularly in patients being evaluated for AD [6, 7]. After intravenous administration, 18F-flortaucipir crosses the blood–brain barrier and selectively binds to aggregated tau proteins, especially the paired helical filaments that form the NFTs, which are characteristic of AD [8]. Once bound to tau aggregates, the radioactive signal from 18F-fluorine can be detected using a PET scanner, allowing visualization and estimation of tau density and distribution in the brain.
18F-flortaucipir is the first and only PET tracer approved to detect tau pathology associated with AD in adults with cognitive impairment in various geographies, including the United States [9], European Union [10], and United Kingdom [11]. In Japan, 18F-flortaucipir has been approved to assist with administration of donanemab, an amyloid-targeting monoclonal antibody [12, 13]. Given 18F-flortaucipir uptake indicates tau deposition, which is closely related to impaired cognition and neuronal injury, it has the potential to predict the progression of AD stages and address the unmet medical need in AD care by contributing to clinically meaningful findings [5].
Two pivotal trials evaluated the performance of 18F-flortaucipir in estimating the density and distribution of aggregated tau NFTs [14]. Study 1 (A16; NCT02516046; Supplementary Fig. 1) assessed if visual reads of 18F-flortaucipir scans corresponded with B3-level (Braak stage V–VI [15]) tau pathology at autopsy. The study included individuals diagnosed with a terminal illness and who were aged > 50 years and had a projected life expectancy of < 6 months. All participants underwent 18F-flortaucipir PET imaging, and scans were independently interpreted by five nuclear medicine physicians or radiologists. The 18F-flortaucipir PET majority-read interpretations were compared against autopsy results, finding a 92.3% sensitivity and 80.0% specificity in predicting B3 tau pathology [14]. Furthermore, 18F-flortaucipir PET scans predicted high AD neuropathology changes (corresponding to advanced amyloid and tau pathology [15]) with 94.7% sensitivity and 80.8% specificity [14]. In Study 2 (Study FR01; NCT03901092; Supplementary Fig. 1), inter-reader agreement as reported using Fleiss' Kappa value was found to be > 0.8 with a new set of readers, evaluating a total of 241 scans, including an autopsy subgroup (N = 82) and a cognitively impaired subgroup from an observational study (N = 159; Study A05C, NCT02016560 [16]) [14].
In this article, we report two studies designed to: (1) show the consistency of 18F-flortaucipir PET whole-body biodistribution between Japanese and non-Japanese individuals (dose and biodistribution study); and (2) assess the comparability of 18F-flortaucipir PET visual reads performed by Japanese expert readers or US expert readers who followed the same training program (reader study).
Materials and methods
Study design
Dose and biodistribution study
To compare the uptake and biodistribution of 18F-flortaucipir in Japanese and non-Japanese individuals, data from a phase I open-label exploratory study (NCT04474405) were used. Whole-body PET scans were conducted in healthy volunteers only (Supplementary Methods).
Reader study
This reader study (Study A27) was designed to compare reader agreement of the interpretation of 18F-flortaucipir PET images between majority visual read results by US readers and those by Japanese readers. Five qualified Japanese physicians were trained in person on the 18F-flortaucipir PET scan read methodology; the training consisted of teaching the readers the steps of interpretation, followed by a practice session using a set of demonstration and practice cases and then independently reading 60 scans. Of note, the training program was identical to the training program that had been previously administered to the US readers. The reader study was conducted on a single day (September 2, 2023); database lock was September 21, 2023.
Ethics
The studies were conducted in accordance with the principles of the Declaration of Helsinki, the International Council for Harmonisation (ICH) Good Clinical Practice (GCP) guidelines, the Code of Federal Regulations, and applicable laws and regulations. Written informed consent was obtained from all healthy volunteers and patients, and from caregivers/guardians if applicable. Participants consented to the use of their data (including scans) for future research purposes.
Participants
Dose and biodistribution study
No new participants were enrolled or imaged for the purposes of this study (Supplementary Methods).
Reader study
Five qualified Japanese PET nuclear medicine physicians who were certified by the Japanese Society of Nuclear Medicine and who had experience interpreting brain PET scans but had not received prior training on the 18F-flortaucipir read methodology were included in the study.
No new participants were enrolled or imaged for the purposes of this reader study (Study A27). 18F-flortaucipir PET scans were collected from a previous reader study (Study FR01, A05C cases; NCT03901092; NCT02016560; Supplementary Fig. 1) [14]. Participants from Study FR01 were male or female, aged ≥ 50 years, with mild cognitive impairment (MCI) or dementia with a suspected neurodegenerative cause, and a Mini Mental State Examination score of 20–27 inclusive.
Image acquisition
Dose and biodistribution study
Images from the phase I open-label exploratory study (NCT04474405) were used. Participants in this study received a single intravenous bolus administration of ~ 370 MBq (10 mCi) of 18F-flortaucipir immediately prior to the whole-body scans (Supplementary Methods).
Reader study
Scans (N = 60) were randomly selected from a previous reader study (Study FR01) [14] using stratified random sampling based on the majority visual read result (non-AD vs. AD). The scans were selected from the A05C cases of Study FR01 because they were comparable to the intended target population in the clinical setting (i.e., participants with MCI and clinically diagnosed AD being evaluated for cognitive decline). Methodology for PET imaging acquisition has been previously published [14]. Briefly, participants underwent PET imaging (4 × 5-minute acquisition frames) beginning ~ 80 min after administration of 370 MBq (± 10%) of intravenous 18F-flortaucipir. Frames were motion-corrected and summed from 80 to 100 min after injection.
Image analysis and assessment
Dose and biodistribution study
18F-flortaucipir whole-body effective dose in Japanese and non-Japanese individuals was determined. Radiation dose estimates measured in mSv/MBq for the whole body were obtained from Organ Level Internal Dose Assessment/Exponential Modeling (OLINDA/EXM) radiation dosimetry code. Results were calculated using a 73.7 kg bodyweight model.
Reader study
All readers were trained in person on the steps of image interpretation using a set of demonstration and practice cases prior to starting their independent reads. After measuring the cerebellar reference region, selecting an appropriate color scale, and scaling the image, regions of the neocortex were visually evaluated for increased tracer uptake.
A scan was deemed positive if there was increased neocortical tracer activity in either/both hemisphere(s) of the posterolateral temporal (PLT), occipital, or parietal/precuneus regions, with or without frontal activity [9]. A scan with a positive AD pattern showed disease progression if increased tracer activity was seen in the parietal (including precuneus) region(s) or frontal region with PLT or occipital regions of either hemisphere.
Each reader made a qualitative assessment of 60 randomly selected 18F-flortaucipir PET images, presented in random sequence, to determine AD pattern (τAD+ or τAD++) or non-AD pattern (τAD-) (Table 1). τAD- was defined as no increased neocortical activity, or increased neocortical activity isolated to the mesial temporal, anterolateral temporal, and/or frontal regions. τAD+ was defined by increased accumulation of ¹⁸F-flortaucipir in the PLT or occipital region(s) in either hemisphere. τAD++ was defined by either of the following: increased accumulation of ¹⁸F-flortaucipir in the parietal lobe, including the precuneus region(s), in either hemisphere, whether occurring in isolation or in combination with increased accumulation in the PLT or occipital lobe, or increased accumulation in the frontal lobe accompanied by increased accumulation in the PLT, parietal lobe, or occipital lobe. The reader independently reviewed a PET or PET/computed tomography brain image for each participant and then determined whether the scan was τAD-, τAD+, or τAD++. Readers were blinded to all participant demographic and clinical data and any previous PET scan interpretation (i.e., site investigator, US reader). Readers were also blinded to each other’s assessments.
Table 1.
Determination of AD pattern
| AD pattern (τAD) |
Not consistent with AD pattern τAD- |
|
|---|---|---|
| τAD+ | τAD++ | |
|
In either hemi-sphere, increased accumulation of 18F-flortaucipir in the PLT or occipital region(s) Tau positivity with moderate AD tau pattern |
In either hemisphere, increased accumulation of 18F-flortaucipir in the parietal lobe, including the precuneus region(s) OR Increased accumulation in the frontal lobe in addition to the PLT, the parietal lobe including the precuneus, or the occipital lobe Tau positivity with advanced AD tau pattern |
No increased neocortical activity, or increased neocortical activity isolated to the mesial temporal, anterolateral temporal, and/or frontal regions Tau negativity with negative AD pattern |
AD Alzheimer’s disease, PLT posterolateral temporal, τAD- neocortical uptake not consistent with AD pattern, τAD+ moderate AD 18F-flortaucipir pattern with increased neocortical activity in the PLT or occipital region(s), τAD++ advanced AD 18F-flortaucipir pattern with increased neocortical activity in the parietal/precuneus region(s), or increased activity in the frontal region(s) accompanied by increases in the PLT, parietal, or occipital region(s)
Objectives and endpoints
Dose and biodistribution study
The objectives were to evaluate the uptake and retention of 18F-flortaucipir in the brain, to obtain radiation exposure dose, and to assess whole-body distribution of 18F-flortaucipir in Japanese participants compared with non-Japanese participants.
Reader study
The primary objective was to evaluate the concordance between Japanese readers and US readers on interpretation of 18F-flortaucipir PET images obtained from Study FR01 to distinguish between AD pattern (τAD+/τAD++) and non-AD pattern (τAD-). Primary endpoints were majority visual read results classified as τAD- versus τAD+/τAD++ from Japanese readers and US readers.
The secondary objective was to evaluate the concordance between Japanese readers and US readers on interpretation of the 18F-flortaucipir PET images obtained in Study FR01 to distinguish between τAD++ and non-τAD++ (τAD-/τAD+). Secondary endpoints were majority visual read results classified as τAD++ versus non-τAD++ from Japanese readers and US readers.
Exploratory objectives were to evaluate the inter-reader reliability of the 18F-flortaucipir scan interpretation across the five Japanese readers for the AD pattern classifications of τAD+/τAD++ versus τAD- and τAD++ versus non-τAD++, and to evaluate the agreement between the majority-read result obtained in Study FR01 and the 18F-flortaucipir scan interpretation for each Japanese reader for τAD+/τAD++ versus τAD- and τAD++ versus non-τAD++. Exploratory endpoints were visual read results from Japanese readers classified as τAD+/τAD++ versus τAD- and τAD++ versus non-τAD++, and majority visual read results from US readers classified as τAD+/τAD++ versus τAD- and τAD++ versus non-τAD++.
Statistical analysis
Dose and biodistribution study
Descriptive statistics were applied to summarize the image analysis results.
Reader study
Sixty cases were randomly selected using stratified random sampling based on the majority visual read result from Study FR01 (~ 60% tAD+/tAD++ and ~ 40% tAD-). According to the majority rating from five Study FR01 readers, ~ 60% of the 159 Study FR01 (A05C) cases were classified as AD pattern (τAD+/τAD++). Assuming a 60% underlying AD pattern (τAD+/τAD++) rate and a Kappa value expected at 0.9 (almost perfect agreement), a sample of 60 cases would provide > 90% power to detect a difference from Kappa value 0.4 (lower bound of moderate agreement) under a two-sided 5% type I error rate.
The analysis dataset included the 18F-flortaucipir visual read results from the five Japanese readers in Study A27 and the corresponding 18F-flortaucipir visual read results from the five US readers in Study FR01 for all 60 scans. For analysis, the enrolled set included all 60 cases and the corresponding US readers’ visual read results selected for this reader study. The full analysis set (FAS) included all cases with a valid 18F-flortaucipir visual read result from Japanese readers. Data are summarized using descriptive statistics. Simple Cohen’s Kappa statistics, along with two-sided 95% confidence intervals (CIs), were provided to evaluate the consistency between the majority-read results from five Japanese readers and five US readers. The Kappa statistic can be interpreted such that 0 indicates no agreement, 0–0.20 indicates slight agreement, 0.21–0.40 indicates fair agreement, 0.41–0.60 indicates moderate agreement, 0.61–0.80 indicates substantial agreement, and 0.81–1.00 indicates almost perfect agreement [17]. Inter-reader reliability of the 18F-flortaucipir scan interpretation (τAD+/τAD++ vs. τAD-) across the five Japanese readers was assessed using Fleiss’ Kappa statistics. The overall percent agreement, Fleiss’ Kappa, and a 95% CI around the Kappa value are provided. Post hoc analyses were performed using the enrolled set to analyze concordance between Japanese readers and US readers on interpretation of 18F-flortaucipir images. All analyses were performed using SAS 9.2 or higher (SAS Institute Inc., Cary, NC, USA).
Results
Dose and biodistribution study
Mean (standard deviation) (mSv/MBq) effective dose equivalent and effective dose, estimated by the 73.7 kg bodyweight model, was 3.07E-02 (1.63E-03) and 2.36E-02 (1.17E-03) in the Japanese population, and 3.21E-02 (2.52E-03) and 2.35E-02 (1.88E-03) in the non-Japanese population (Table 2). The biodistribution of 18F-flortaucipir was consistent in the Japanese and non-Japanese populations. 18F-flortaucipir was distributed throughout the body, with < 10% of the injected 18F radioactivity remaining in the blood at 5 min and < 5% at 10 min post administration. During imaging, ~ 28–34% of circulating 18F was parent compound; the rest was metabolites. Clearance occurred primarily by hepatobiliary and renal excretion, with rapid bile excretion observed. An example of whole-body 18F-flortaucipir PET images of Japanese and non-Japanese male individuals when infused with 370 MBq of 18F-flortaucipir is shown in Fig. 1.
Table 2.
Estimated corrected exposure dose (mSv/MBq) using the 73.7 kg bodyweight model (Study A01/A15)
| Non-Japanese N = 6 |
Japanese N = 3 |
|||
|---|---|---|---|---|
| Mean | SD | Mean | SD | |
| Adrenal glands | 1.41E-02 | 9.03E-04 | 1.44E-02 | 6.93E-04 |
| Brain | 8.23E-03 | 1.24E-03 | 8.87E-03 | 1.24E-03 |
| Breast | 7.10E-03 | 3.13E-04 | 7.21E-03 | 2.80E-04 |
| Gallbladder wall | 4.12E-02 | 2.15E-02 | 3.16E-02 | 1.19E-02 |
| Lower large intestine wall | 3.56E-02 | 4.84E-03 | 3.32E-02 | 3.61E-04 |
| Small intestine | 8.68E-02 | 1.43E-02 | 7.99E-02 | 1.04E-03 |
| Stomach lining | 1.27E-02 | 3.20E-04 | 1.25E-02 | 3.06E-04 |
| Upper large intestine wall | 9.81E-02 | 1.61E-02 | 9.02E-02 | 1.13E-03 |
| Heart wall | 2.82E-02 | 1.72E-03 | 3.27E-02 | 4.69E-03 |
| Kidney | 4.18E-02 | 1.22E-02 | 3.61E-02 | 5.67E-03 |
| Liver | 5.46E-02 | 8.04E-03 | 6.25E-02 | 5.76E-03 |
| Lung | 3.40E-02 | 6.04E-03 | 3.35E-02 | 6.37E-03 |
| Muscle | 9.00E-03 | 3.41E-04 | 8.97E-03 | 1.87E-04 |
| Ovary | 2.10E-02 | 1.99E-03 | 2.00E-02 | 3.21E-04 |
| Pancreas | 1.43E-02 | 6.37E-04 | 1.45E-02 | 6.66E-04 |
| Red marrow | 1.01E-02 | 2.69E-04 | 1.00E-02 | 2.00E-04 |
| Bone formation cells | 1.15E-02 | 7.23E-04 | 1.15E-02 | 1.53E-04 |
| Skin | 5.97E-03 | 3.63E-04 | 5.98E-03 | 7.64E-05 |
| Spleen | 1.03E-02 | 5.68E-04 | 1.01E-02 | 3.00E-04 |
| Testis | 6.93E-03 | 5.38E-04 | 6.97E-03 | 1.98E-04 |
| Thymus | 8.52E-03 | 4.07E-04 | 8.69E-03 | 3.76E-04 |
| Thyroid gland | 6.71E-03 | 5.47E-04 | 6.72E-03 | 1.58E-04 |
| Urinary bladder wall | 3.57E-02 | 9.94E-03 | 4.15E-02 | 1.37E-02 |
| Womb | 1.84E-02 | 1.44E-03 | 1.79E-02 | 8.00E-04 |
| Systemic | 1.19E-02 | 2.07E-04 | 1.19E-02 | 3.61E-04 |
| Effective dose equivalent | 3.21E-02 | 2.52E-03 | 3.07E-02 | 1.63E-03 |
| Effective dose | 2.35E-02 | 1.88E-03 | 2.36E-02 | 1.17E-03 |
mSv/MBq millisieverts per megabecquerel, SD standard deviation
Fig. 1.
Comparison of Japanese and non-Japanese exposure dose (mean value) estimated by a 73.7 kg bodyweight model (Study A01/A15). Maximum intensity projection images showing representative radioactivity biodistributions after injection of ~10 mCi (370 MBq) of 18F-flortaucipir to (a) a 58-year-old non-Japanese male and (b) a 56-year-old Japanese male
Reader study
Demographic and baseline clinical characteristics
The enrolled set consisted of 60 cases, and the full analysis set consisted of 56 cases; four cases were uninterpretable by Japanese readers due to noise, low count density, or attenuation correction error and were therefore excluded from the FAS. Baseline demographics for the enrolled set are presented in Table 3. Mean age was 71.6 years, 98.3% of cases were of White race, 43.3% were female, 75.0% had ≥ 13 years of education, and 55.0% had a clinical diagnosis of MCI.
Table 3.
Demographics and baseline characteristics of cases selected for Study A27 (FAS)
| Majority interpretation results from US readers | |||
|---|---|---|---|
| τAD+/τAD++ (N = 35) |
τAD- (N = 25) |
Overall (N = 60) |
|
| Female sex, n (%) | 18 (51.4) | 8 (32.0) | 26 (43.3) |
| Mean age, years (SD) | 73.5 (7.4) | 68.9 (11.2) | 71.6 (9.3) |
| Race, n (%) | |||
| Black or African American | 0 | 1 (4.0) | 1 (1.7) |
| White | 35 (100) | 24 (96.0) | 59 (98.3) |
| Ethnicity, Hispanic/Latino, n (%) | 2 (5.7) | 1 (4.0) | 3 (5.0) |
| Education, ≥ 13 years, n (%) | 27 (77.1) | 18 (72.0) | 45 (75.0) |
| Clinical diagnosis group, n (%) | |||
| MCI | 15 (42.9) | 18 (72.0) | 33 (55.0) |
| AD | 20 (57.1) | 7 (28.0) | 27 (45.0) |
AD Alzheimer’s disease, FAS full analysis set, MCI mild cognitive impairment, PLT posterolateral temporal, SD standard deviation, τAD- neocortical uptake not consistent with AD pattern, τAD+ moderate AD 18F-flortaucipir pattern with increased neocortical activity in the PLT or occipital region(s), τAD++ advanced AD 18F-flortaucipir pattern with increased neocortical activity in the parietal/precuneus region(s), or increased activity in the frontal region(s) accompanied by increases in the PLT, parietal, or occipital region(s)
Primary endpoint
The majority reads from Japanese and US readers showed 100% agreement with a Cohen’s Kappa value of 1 when interpreting 18F-flortaucipir PET images to distinguish between AD pattern (τAD+/τAD++) versus non-AD pattern (τAD-) (Table 4). The majority imaging visual interpretation results of all Japanese and US readers for AD pattern (τAD+/τAD++) and non-AD pattern (τAD-) are shown in Table 4.
Table 4.
Concordance between Japanese readers and US readers on interpretation of 18F-flortaucipir images (τAD+/τAD++ vs. τAD-) (FAS)
| Majority read: Japanese readers | Majority read: US readers |
||
|---|---|---|---|
| τAD+/τAD++ | τAD- | Total | |
| τAD+/τAD++ | 35 | 0 | 35 |
| τAD- | 0 | 21 | 21 |
| Total | 35 | 21 | 56 |
| Overall agreement | |||
| Cohen’s Kappa (95% CI) | 1 (1.0–1.0) | ||
| p-valuea | < 0.001 | ||
| Overall percent agreement (95% CIb) | 100 (93.6–100.0) | ||
aThe p-value corresponds to the test with null hypothesis: Kappa = 0
bThe 95% CI is based on the Wilson score method
AD Alzheimer’s disease, CI confidence interval, FAS full analysis set, PLT posterolateral temporal, τAD- neocortical uptake not consistent with AD pattern, τAD+ moderate AD 18F-flortaucipir pattern with increased neocortical activity in the PLT or occipital region(s), τAD++ advanced AD 18F-flortaucipir pattern with increased neocortical activity in the parietal/precuneus region(s), or increased activity in the frontal region(s) accompanied by increases in the PLT, parietal, or occipital region(s)
Secondary endpoint
The majority reads from Japanese and US readers showed 100% agreement and a Cohen’s Kappa value of 1 when interpreting 18F-flortaucipir images to distinguish between τAD++ and non-τAD++ (τAD+/τAD-) (Table 5).
Table 5.
Concordance between Japanese readers and US readers on interpretation of 18F-flortaucipir images (τAD++ vs. non-τAD++) (FAS)
| Majority read: Japanese readers | Majority read: US readers |
||
|---|---|---|---|
| τAD++ | τAD+/τAD- | Total | |
| τAD++ | 31 | 0 | 31 |
| τAD+/τAD- | 0 | 25 | 25 |
| Total | 31 | 25 | 56 |
| Overall agreement | |||
| Cohen’s Kappa (95% CI) | 1 (1.0–1.0) | ||
| p-valuea | < 0.001 | ||
| Overall percent agreement (95% CIb) | 100 (93.6–100.0) | ||
aThe p-value corresponds to the test with null hypothesis: Kappa = 0
bThe 95% CI is based on the Wilson score method
AD Alzheimer’s disease, CI confidence interval, FAS full analysis set, PLT posterolateral temporal, τAD- neocortical uptake not consistent with AD pattern, τAD+ moderate AD 18F-flortaucipir pattern with increased neocortical activity in the PLT or occipital region(s), τAD++ advanced AD 18F-flortaucipir pattern with increased neocortical activity in the parietal/precuneus region(s), or increased activity in the frontal region(s) accompanied by increases in the PLT, parietal, or occipital region(s)
Exploratory endpoints
When distinguishing between AD pattern (τAD+/τAD++) versus non-AD pattern (τAD-) images, Japanese readers displayed an overall percent agreement of 95.7% and a Fleiss’ Kappa value of 0.91, indicating almost perfect consistency between readers and image interpretations (Table 6). Inter-reader reliability across the five Japanese readers is shown in Supplementary Table 1. When distinguishing between τAD++ and non-τAD++ (τAD-/τAD+) images, Japanese readers displayed an overall percent agreement of 95.4% and a Fleiss’ Kappa value of 0.91, indicating almost perfect consistency between readers and image interpretations (Table 6). Inter-reader reliability across the five Japanese readers is shown in Supplementary Table 2.
Table 6.
Inter-reader reliability across Japanese readers (FAS)
| Overall agreement between Japanese readers | |
|---|---|
| τAD+/τAD++ vs. τAD- | |
| Fleiss’ Kappa (95% CI) | 0.91 (0.83–0.99) |
| p-valuea | < 0.001 |
| Overall percent agreement (95% CI) | 95.7 (93.7–97.1) |
| τAD++ vs. non-τAD++ | |
| Fleiss’ Kappa (95% CI) | 0.91 (0.83–0.99) |
| p-valuea | < 0.001 |
| Overall agreement (95% CI) | 95.4 (93.3–96.8) |
aThe p-value corresponds to the test with null hypothesis: Kappa = 0
AD Alzheimer’s disease, CI confidence interval, FAS full analysis set, PLT posterolateral temporal, τAD- neocortical uptake not consistent with AD pattern, τAD+ moderate AD 18F-flortaucipir pattern with increased neocortical activity in the PLT or occipital region(s), τAD++ advanced AD 18F-flortaucipir pattern with increased neocortical activity in the parietal/precuneus region(s), or increased activity in the frontal region(s) accompanied by increases in the PLT, parietal, or occipital region(s)
Each Japanese reader showed high concordance with the majority-read results obtained from Study FR01 when interpreting 18F-flortaucipir images to distinguish between τAD+/τAD++ versus τAD- (Table 7) and τAD + + versus non-τAD++ (Table 8).
Table 7.
Concordance between Japanese readers and Study FR01 majority-read results (τAD+/τAD++ vs. τAD-) (FAS)
| Japan reader | Agreement between Japanese and majority read from Study FR01 | ||
|---|---|---|---|
| % Agreement with FR01 reads (95% CIa) |
Cohen’s Kappa (95% CI) |
p-valueb | |
| 1 | 96.4 (87.9–99.0) | 0.92 (0.82–1.0) | < 0.001 |
| 2 | 98.2 (90.6–99.7) | 0.96 (0.89–1.0) | < 0.001 |
| 3 | 94.6 (85.4–98.2) | 0.89 (0.77–1.0) | < 0.001 |
| 4 | 100.0 (93.6–100.0) | 1.0 (1.0–1.0) | < 0.001 |
| 5 | 98.2 (90.6–99.7) | 0.96 (0.89–1.0) | < 0.001 |
aThe 95% CI is based on the Wilson score method
bThe p-value corresponds to the test with null hypothesis: Kappa = 0
AD Alzheimer’s disease, CI confidence interval, FAS full analysis set, PLT posterolateral temporal, τAD- neocortical uptake not consistent with AD pattern, τAD+ moderate AD 18F-flortaucipir pattern with increased neocortical activity in the PLT or occipital region(s), τAD++ advanced AD 18F-flortaucipir pattern with increased neocortical activity in the parietal/precuneus region(s), or increased activity in the frontal region(s) accompanied by increases in the PLT, parietal, or occipital region(s)
Table 8.
Concordance between Japanese readers and Study FR01 majority-read results (τAD++ vs. non-τAD++) (FAS)
| Japan reader | Agreement between Japanese and majority read from Study FR01 | ||
|---|---|---|---|
| % Agreement with FR01 reads (95% CIa) |
Cohen’s Kappa (95% CI) |
p-valueb | |
| 1 | 94.6 (85.4–98.2) | 0.89 (0.77–1.0) | < 0.001 |
| 2 | 94.6 (85.4–98.2) | 0.89 (0.77–1.0) | < 0.001 |
| 3 | 98.2 (90.6–99.7) | 0.96 (0.89–1.0) | < 0.001 |
| 4 | 98.2 (90.6–99.7) | 0.96 (0.89–1.0) | < 0.001 |
| 5 | 100.0 (93.6–100.0) | 1.0 (1.0–1.0) | < 0.001 |
aThe 95% CI is based on the Wilson score method
bThe p-value corresponds to the test with null hypothesis: Kappa = 0
AD Alzheimer’s disease, CI confidence interval, FAS full analysis set, PLT posterolateral temporal, τAD++ advanced AD 18F-flortaucipir pattern with increased neocortical activity in the parietal/precuneus region(s), or increased activity in the frontal region(s) accompanied by increases in the PLT, parietal, or occipital region(s)
Post hoc analyses
Japanese and US readers showed 100% agreement and a Cohen’s Kappa of 1 when interpreting 18F-flortaucipir images to distinguish between AD pattern (τAD+/τAD++) and non-AD pattern (τAD-) (Supplementary Table 3 [enrolled set]), and between τAD++ and non-τAD++ (τAD-/τAD+) (Supplementary Table 4 [enrolled set])
Discussion
The studies reported in this article evaluated the radiation exposure dose of 18F-flortaucipir in Japanese participants and provide evidence that Japanese readers’ performance in interpreting 18F-flortaucipir PET images was comparable to that observed for US expert readers on the same dataset after completing the same training. In the dose and biodistribution study, imaging data obtained from Japanese individuals were similar to those of non-Japanese individuals pertaining to dose of radiation exposure and biodistribution. In the reader study, majority-read results from Japanese readers showed high concordance with US expert readers on the interpretation of 18F-flortaucipir PET images to distinguish between AD and non-AD patterns, and between τAD++ and non-τAD++ (τAD+/τAD-). Japanese readers showed high inter-reader reliability on interpretation of 18F-flortaucipir PET images, and each Japanese reader also showed high concordance with the majority-read results obtained from Study FR01. The results of the current study demonstrate that Japanese readers have high concordance with US readers on the interpretation of 18F-flortaucipir PET images when given the same training, supporting the safe use of 18F-flortaucipir in Japan.
Alzheimer’s disease is characterized by amyloid beta plaques and aggregated tau NFTs, which can appear years before symptoms. These pathologic markers, previously only detectable post mortem, can now be identified in vivo using biomarkers, thereby allowing for a definitive diagnosis [7]. 18F-flortaucipir has been extensively validated in various research and clinical settings, highlighting its diagnostic and prognostic value in typical and atypical AD, MCI, and pre-symptomatic stages, with limited evidence for other neurodegenerative disorders [7]. Despite varied interpretation methods, the literature supports 18F-flortaucipir’s high value in AD diagnosis and prognosis [7]. Recently, expert recommendations for the implementation of amyloid and tau PET in clinical practice for guiding the diagnosis and management of patients who have, or who are at risk for, cognitive decline have been reported [18], further supporting the clinical application of 18F-flortaucipir. Although 18F-flortaucipir has quantitatively been shown to be potentially sensitive enough to identify Braak stage IV in the inferior temporal region [19], the visual read method has been validated to detect advanced tau pathology (B3-level, Braak stage V or VI [15]).
A key strength of the dose and biodistribution study is the diverse set of images sourced from multiple studies, offering a realistic reflection of those encountered in daily clinical practice. The inclusion of Japanese participants in the dose and biodistribution study shows valid evidence of its safe use in the Japanese population. However, we acknowledge that the study included participants of a relatively young age range (> 50 years), and only three out of a total of nine participants were Japanese, which limits the extent to which the dose and biodistribution findings can be generalized to the Japanese population. Nevertheless, exposure dose results from Choi et al. [20], whose study comprised six healthy Korean participants, yielded similar findings. Taken together, the consistent results observed in both Japanese and Korean participants support the safe use of 18F-flortacuipir in the Japanese population.
In vitro, titrated flortaucipir has been reported to bind with low nanomolar affinity to monoamine oxidase (MAO)-A and MAO-B, which could contribute to off-target binding [9]. While in vitro studies have reported higher binding affinity of ¹⁸F-flortaucipir for MAO-A than MAO-B, the in vivo relevance of MAO-A–related binding remains limited. Enzyme inhibition studies have demonstrated that MAO-related binding is not expected to significantly affect cortical ¹⁸F-flortaucipir signals [21], and modeling work corroborates that MAO-related binding is unlikely to meaningfully impact typical clinical interpretation [9]. Importantly, a recent voxel-to-voxel histology-to-PET study [22] provides direct spatial evidence that, in AD, phosphorylated tau burden is the dominant contributor to the ¹⁸F-flortaucipir signal, with an ~ 1.5-fold greater regression coefficient than ferric iron, and with MAO-B showing an inverse association in cortical regions. The same study demonstrated that MAO-B signal is predominantly concentrated in white matter, thereby limiting its influence on neocortical ¹⁸F-flortaucipir interpretation. These findings collectively support the conclusion that, while off-target binding to iron and MAOs exists, it does not substantially compromise the ability of ¹⁸F-flortaucipir to reflect AD tau pathology in neocortical regions relevant to clinical visual read interpretation.
The near-perfect consistency of read results by Japanese readers underscores the technical robustness of 18F-flortaucipir for clinical use. Like in any reader study, however, this study was carried out in a controlled environment, which may not fully capture the complexity and pressures of actual clinical practice; nevertheless, similar high concordance for the 18F-flortaucipir visual read method was also reported in Europe [23]. Together, these results support the reliability of using 18F-flortaucipir as a PET tracer regardless of geography and associated clinical environment. As such, despite limitations, it is reasonable to infer that 18F-flortaucipir PET tracer visual read can be performed with precision by Japanese readers given the relevant training.
Conclusion
The biodistribution of 18F-flortaucipir throughout the body was equivalent in Japanese and non-Japanese populations. Japanese readers showed high concordance among Japanese readers, and with US readers, on the interpretation of 18F-flortaucipir PET images to distinguish between AD and non-AD patterns, thereby demonstrating that given the same training, Japanese physicians were able to read 18F-flortaucipir PET images with the same high precision as US physicians.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgments
The authors would like to thank all the readers who participated in the reader study.
Author Contribution
All authors participated in the interpretation of study results, and in the drafting, critical revision, and approval of the final version of the manuscript. AKA and ML were involved in the study design, and AM conducted the statistical analysis.
Funding
The studies reported in this article were funded by Eli Lilly and Company and Avid Radiopharmaceuticals, Inc (a wholly owned subsidiary of Eli Lilly and Company), manufacturer of 18F-flortaucipir. Eli Lilly and Company was involved in the study design, data collection, data analysis, and preparation of the manuscript. Medical writing assistance was provided by Prudence Stanford, PhD, CMPP, and Koa Webster, PhD, CMPP, of ProScribe KK, part of Envision Ignite, an Envision Medical Communications agency, a part of Envision Pharma Group, and was funded by Eli Lilly Japan K.K. ProScribe’s services complied with international guidelines for Good Publication Practice.
Data availability
Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.
Declarations
Conflict of interest
All authors are employees of Eli Lilly Japan K.K. or Eli Lilly and Company and are minor shareholders of Eli Lilly and Company.
Ethical approval and consent to participate
The studies were conducted in accordance with the principles of the Declaration of Helsinki, the International Council for Harmonisation (ICH) Good Clinical Practice (GCP) guidelines, the Code of Federal Regulations, and applicable laws and regulations. Written informed consent was obtained from all healthy volunteers and patients, and from caregivers/guardians if applicable. Participants consented to the use of their data (including scans) for future research purposes.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Zheng Q, Wang X. Alzheimer’s disease: insights into pathology, molecular mechanisms, and therapy. Protein Cell. 2025;16(2):83–120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Knopman DS, Amieva H, Petersen RC, Chételat G, Holtzman DM, Hyman BT, et al. Alzheimer disease. Nat Rev Dis Primers. 2021;7(1):33. [DOI] [PMC free article] [PubMed]
- 3.Jack CR Jr., Andrews JS, Beach TG, Buracchio T, Dunn B, Graf A, et al. Revised criteria for diagnosis and staging of Alzheimer’s disease: Alzheimer’s Association Workgroup. Alzheimers Dement. 2024;20(8):5143–69. [DOI] [PMC free article] [PubMed]
- 4.Johnson KA, Schultz A, Betensky RA, Becker JA, Sepulcre J, Rentz D, et al. Tau positron emission tomographic imaging in aging and early Alzheimer disease. Ann Neurol. 2016;79(1):110–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Villemagne VL, Doré V, Burnham SC, Masters CL, Rowe CC. Imaging tau and amyloid-β proteinopathies in Alzheimer disease and other conditions. Nat Rev Neurol. 2018;14(4):225–36. [DOI] [PubMed] [Google Scholar]
- 6.Jie CVML, Treyer V, Schibli R, Mu L. Tauvid™: the first FDA-approved PET tracer for imaging tau pathology in Alzheimer’s disease. Pharmaceuticals (Basel). 2021;14(2):110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Burnham SC, Iaccarino L, Pontecorvo MJ, Fleisher AS, Lu M, Collins EC, et al. A review of the flortaucipir literature for positron emission tomography imaging of tau neurofibrillary tangles. Brain Commun. 2023;6(1):fcad305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Chien DT, Szardenings AK, Bahri S, Walsh JC, Mu F, Xia C, et al. Early clinical PET imaging results with the novel PHF-tau radioligand [F18]-T808. J Alzheimers Dis. 2014;38(1):171–84. [DOI] [PubMed] [Google Scholar]
- 9.Eli Lilly and Company. TAUVID™ (flortaucipir F 18 injection), for intravenous use [highlights of prescribing information]. https://pi.lilly.com/us/tauvid-uspi.pdf. Accessed 12 Jun 2025.
- 10.European Medicines Agency. Tauvid flortaucipir (18F). Accessed 12 Jun 2025. https://www.ema.europa.eu/en/medicines/human/EPAR/tauvid
- 11.Eli Lilly and Company. Tauvid 1 900 MBq/mL solution for injection [summary of product characteristics]. https://mhraproducts4853.blob.core.windows.net/docs/40278c788f4441b96ce5b225fe2b4a3bbf79f125. Accessed 17 Dec 2025.
- 12.Sims JR, Zimmer JA, Evans CD, Lu M, Ardayfio P, Sparks J, et al. Donanemab in early symptomatic Alzheimer disease: the TRAILBLAZER-ALZ 2 randomized clinical trial. JAMA. 2023;330(6):512–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.PDRadiopharma Inc. TAUVID® Injection [package insert] [in Japanese]. https://www.pdradiopharma.com/wpeptip/wp-content/uploads/2025/01/TAU_PackageInserts.pdf. Accessed 10 Oct 2025.
- 14.Fleisher AS, Pontecorvo MJ, Devous MD Sr., Lu M, Arora AK, Truocchio SP, et al. Positron emission tomography imaging with [18F]flortaucipir and postmortem assessment of Alzheimer disease neuropathologic changes. JAMA Neurol. 2020;77(7):829–39. [DOI] [PMC free article] [PubMed]
- 15.Hyman BT, Phelps CH, Beach TG, Bigio EH, Cairns NJ, Carrillo MC, et al. National Institute on Aging-Alzheimer’s Association guidelines for the neuropathologic assessment of Alzheimer’s disease. Alzheimers Dement. 2012;8(1):1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lu M, Pontecorvo MJ, Devous MD, Sr., Arora AK, Galante N, McGeehan A, et al. Aggregated tau measured by visual interpretation of flortaucipir positron emission tomography and the associated risk of clinical progression of mild cognitive impairment and Alzheimer disease: results from 2 phase III clinical trials. JAMA Neurol. 2021;78(4):445–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159–74. [PubMed] [Google Scholar]
- 18.Rabinovici GD, Knopman DS, Arbizu J, Benzinger TLS, Donohoe KJ, Hansson O, et al. Updated appropriate use criteria for amyloid and tau PET: a report from the Alzheimer’s Association and Society for Nuclear Medicine and Molecular Imaging Workgroup. Alzheimers Dement. 2025;21(1):e14338. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kotari V, Southekal S, Navitsky M, Kennedy IA, Lu M, Morris A, et al. Early tau detection in flortaucipir images: validation in autopsy-confirmed data and implications for disease progression. Alzheimers Res Ther. 2023;15(1):41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Choi JY, Lyoo CH, Lee JH, Cho H, Kim KM, Kim JS, et al. Human radiation dosimetry of [(18)F]AV-1451(T807) to detect tau pathology. Mol Imaging Biol. 2016;18(4):479–82. [DOI] [PubMed] [Google Scholar]
- 21.Wright JP, Goodman JR, Lin Y-G, Lieberman BP, Clemens J, Gomez LF, et al. Monoamine oxidase binding not expected to significantly affect [18F]flortaucipir PET interpretation. Eur J Nucl Med Mol Imaging. 2022;49(11):3797–808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Chen Y, La Joie R, Pereira FL, Blazhenets G, Zhu L, Spina S, et al. Disentangling on and off-target binding in flortaucipir PET: a voxel-to-voxel P-tau, ferric iron, and MAO-B histology-to-flortaucipir PET comparison. Acta Neuropathol. 2026;151(1):21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Coomans EM, de Koning LA, Rikken RM, Verfaillie SCJ, Visser D, den Braber A, et al. Performance of a [18F]flortaucipir PET visual read method across the Alzheimer disease continuum and in dementia with Lewy bodies. Neurology. 2023;101(19):e1850–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.