Positron emission tomography (PET) has revolutionized the diagnosis, staging, and monitoring of neurodegenerative diseases by enabling direct visualization of pathological proteins, metabolic changes, and neuroinflammatory processes in the living brain. Unlike structural imaging modalities such as computed tomography (CT) and magnetic resonance imaging (MRI), which primarily detect downstream consequences of neurodegeneration including brain atrophy and white matter changes, PET imaging provides molecular-level insights into the underlying disease mechanisms that drive clinical progression. This capability makes PET an indispensable tool for early diagnosis, differential diagnosis, disease staging, clinical trial enrichment, and therapeutic response monitoring in Alzheimer's disease (AD), Parkinson's disease (PD), and related neurodegenerative disorders[1].
The development of targeted radiotracers over the past two decades has transformed PET from a primarily research-focused modality into a clinically validated diagnostic tool. Radiotracers targeting amyloid-beta (Aβ) plaques, tau neurofibrillary tangles, dopaminergic terminals, translocator protein (TSPO) for neuroinflammation, and more recently, synaptic vesicular protein 2A (SV2A) for synaptic density, have each contributed unique insights into disease biology. The integration of PET with other biomarkers within frameworks such as the AT(N) system has further standardized biomarker-driven approaches to neurodegenerative disease classification and clinical care[2].
Positron emission tomography operates by detecting pairs of gamma rays emitted when a positron-emitting radionuclide undergoes annihilation with an electron. The most commonly used radionuclides in neurodegeneration imaging include carbon-11 (¹¹C, half-life 20.4 minutes), fluorine-18 (¹⁸F, half-life 109.8 minutes), and fluorine-18 labeled compounds for amyloid and tau imaging. The choice of radionuclide influences both the practical logistics of imaging (¹¹C requires on-site cyclotron facilities) and the kinetic properties of the resulting tracer[3].
PET imaging in neurodegeneration serves multiple clinical and research purposes that extend beyond traditional diagnostic applications. In clinical practice, amyloid PET has become a standard tool for evaluating patients with cognitive impairment when the etiology remains uncertain after comprehensive neurological assessment. The appropriate use criteria established by the Amyloid Imaging Task Force recommend amyloid PET in patients with subjective cognitive decline or mild cognitive impairment (MCI) when the cause remains ambiguous and AD is a plausible differential diagnosis[4]. Tau PET provides additional value for disease staging, as tau pathology follows a predictable spreading pattern that correlates with clinical severity, enabling more precise prognostication than amyloid PET alone[5].
In the research domain, PET biomarkers have become essential endpoints in clinical trials targeting pathological proteins. Amyloid PET enables enrollment of patients with confirmed amyloid pathology, enrichment of trial populations, and quantification of amyloid reduction in response to anti-amyloid therapies. The advent of anti-amyloid monoclonal antibodies such as lecanemab, donanemab, and Aduhelm (aducanumab) has elevated amyloid PET from a diagnostic tool to a companion diagnostic for treatment selection and monitoring[6]. Tau PET similarly supports trials of anti-tau therapeutics by enabling target engagement assessment and disease modification measurement[7].
The development of in vivo amyloid imaging represents one of the most significant advances in neurodegenerative disease research. In 2004, Klunk and colleagues at the University of Pittsburgh reported the first successful PET imaging of amyloid-beta plaques in living human brains using ¹¹C-Pittsburgh Compound B (PiB)[8]. PiB is a derivative of the amyloid-staining dye thioflavin T, modified to optimize brain penetration and specific binding to fibrillar Aβ plaques. The initial study demonstrated markedly elevated cortical retention of PiB in patients with Alzheimer's disease compared to cognitively normal controls, with retention patterns that corresponded to known distributions of amyloid pathology at autopsy[8:1].
The success of PiB-PET opened a new era in AD research by enabling visualization of amyloid pathology in vivo, years to decades before clinical symptom onset. Subsequent studies established that PiB positivity precedes clinical symptoms by approximately 15-20 years in autosomal dominant AD and by 5-10 years in sporadic AD, supporting the amyloid cascade hypothesis and informing models of disease progression[9]. However, the 20-minute half-life of carbon-11 limited PiB-PET to research centers with on-site cyclotron facilities, restricting widespread clinical adoption.
The development of fluorine-18-labeled amyloid tracers with 110-minute half-lives enabled broader clinical implementation and led to FDA approval of three amyloid PET tracers:
Florbetapir (Amyvid): Approved by the FDA in 2012, florbetapir (also known as AV-45 or Amyvid) was the first ¹⁸F-labeled amyloid PET tracer to receive regulatory approval. The pivotal phase 3 clinical trial demonstrated high sensitivity (92%) and specificity (85%) for detecting histopathologically confirmed amyloid plaques in patients with AD compared to cognitively normal controls[10]. Florbetapir binds with high affinity to fibrillar Aβ plaques, and its cortical-to-cerebellar ratio correlates with amyloid burden measured at autopsy.
Florbetaben (Neuraceq): Approved in 2014, florbetaben similarly targets fibrillar Aβ plaques with high diagnostic accuracy. Phase 3 studies showed sensitivity of 86% and specificity of 88% for distinguishing AD patients from controls[11]. Florbetaben has demonstrated utility in both clinical settings for differential diagnosis and research settings for amyloid burden quantification.
Flutemetamol (Vizamyl): Also approved in 2013, flutemetamol is the ¹⁸F-labeled analogue of PiB with improved signal-to-noise properties. The phase 3 trial demonstrated sensitivity of 87% and specificity of 88% for detecting amyloid pathology[12]. Notably, flutemetamol has shown efficacy in detecting earlier stages of amyloid deposition, including in cognitively normal individuals with preclinical AD.
All three FDA-approved amyloid PET tracers provide binary visual reads (positive or negative) for clinical interpretation, while quantitative analysis enables continuous measurement of amyloid burden. The Centiloid scale, developed by Klunk et al. in 2015, standardizes quantification across tracers and analysis methods by anchoring values to a 0-100 scale where 0 represents typical amyloid values in young controls and 100 represents typical AD values[13].
The Centiloid scale was developed to address the need for standardized, cross-tracer comparability in amyloid PET quantification. Prior to its development, different tracers and analysis methods produced non-comparable SUVr (standardized uptake value ratio) values, limiting meta-analyses and direct comparison of results across studies. The Centiloid approach uses a linear transformation to convert tracer-specific SUVr values to a unified scale where:
Subsequent validation studies established that the Centiloid threshold of 30 Centiloids provides optimal discrimination between amyloid-positive and amyloid-negative individuals, while values below 20 Centiloids reliably indicate amyloid negativity[14]. This standardization has enabled large-scale multicentre studies, clinical trial meta-analyses, and longitudinal tracking of amyloid dynamics across different tracers and cohorts.
The clinical utility of amyloid PET was definitively established by the Imaging Dementia—Evidence for Amyloid Scanning (IDEAS) study, a large-scale, prospective investigation that enrolled over 16,000 Medicare beneficiaries with MCI or dementia of uncertain etiology[15]. The study demonstrated that amyloid PET changed clinical management in approximately 60% of patients, including alterations in diagnosis, treatment planning, and counseling. Importantly, the study showed that amyloid PET reduced the time to diagnosis and increased diagnostic confidence among treating physicians.
The New IDEAS study, launched in 2020, extends this investigation to more diverse populations, including African American and Hispanic participants who were underrepresented in the original IDEAS cohort[16]. Given the known disparities in AD diagnosis and the importance of biomarker confirmation across populations, New IDEAS aims to establish the clinical validity of amyloid PET in underrepresented groups and ensure equitable access to amyloid-targeted diagnostics.
Tau PET imaging represents the next frontier in neurodegenerative disease biomarkers, providing in vivo visualization of tau neurofibrillary tangle pathology that closely correlates with clinical symptoms and disease progression. Unlike amyloid PET, which detects a necessary but not sufficient condition for AD diagnosis, tau PET provides a direct window into the pathogenic protein that most closely tracks clinical decline.
The first successful tau PET tracer, ¹⁸F-Flortaucipir (also known as AV-1451 or T807), was developed by Avid Radiopharmaceuticals and received FDA approval in 2020 as a diagnostic tool for AD[17]. Flortaucipir demonstrates high binding affinity for paired helical filament (PHF) tau in AD, with retention patterns that closely follow the Braak staging scheme for tau neuropathology. Studies demonstrate strong correlations between Flortaucipir retention and clinical severity, CSF tau levels, cognitive performance, and post-mortem tangle density[18].
However, first-generation tau tracers have notable limitations. Off-target binding in the choroid plexus, basal ganglia, and meninges can produce spurious signals that complicate interpretation, particularly in the medial temporal lobe. Additionally, Flortaucipir shows limited binding to 3-repeat tau (3R) isoforms that predominate in primary tauopathies such as progressive supranuclear palsy (PSP) and corticobasal syndrome (CBS), limiting utility in non-AD tauopathies[19].
Second-generation tau tracers have been developed to address the limitations of Flortaucipir, with improved selectivity, reduced off-target binding, and broader coverage of tauopathies:
MK-6240: This ¹⁸F-labeled tracer demonstrates high specificity for 3R/4R tau in AD with minimal off-target binding. MK-6240 shows excellent correlation with cognitive measures and has emerged as a preferred tracer for anti-tau therapeutic trials[20].
PI-2620: Developed by Life Molecular Imaging, PI-2620 shows binding to both AD-type tau and non-AD tauopathies including PSP and CBS. This broader specificity makes PI-2620 valuable for differential diagnosis across tauopathies[21].
APN-1607 (Lu AF87908): This tracer demonstrates sensitivity to early tau deposition in the entorhinal cortex and has shown promise for detecting tau pathology in the preclinical stages of AD[22].
Tau PET has enabled biological staging of AD that complements clinical staging based on symptom duration. The tau PET pattern follows a predictable progression from the entorhinal cortex and hippocampus (Braak stage I-II) to inferior temporal and parietal regions (Braak stage III-IV), and finally to primary sensory and motor cortices (Braak stage V-VI)[23]. This topographic progression correlates with clinical progression from preclinical to mild cognitive impairment to dementia.
Longitudinal tau PET studies demonstrate that baseline tau PET retention predicts subsequent cognitive decline with greater accuracy than baseline amyloid PET, supporting the primacy of tau pathology in driving clinical symptoms[24]. Tau PET has thus become essential for patient stratification in clinical trials, enabling enrichment of populations most likely to show disease modification in response to anti-tau therapeutics.
Fluorodeoxyglucose (FDG) PET measures cerebral glucose metabolism as a proxy for neuronal and synaptic activity. In neurodegenerative diseases, characteristic patterns of hypometabolism precede structural atrophy and provide diagnostic information that complements amyloid and tau imaging. The FDG-PET pattern in AD shows posterior cingulate, precuneus, and temporoparietal hypometabolism, reflecting the characteristic distribution of AD pathology[25].
FDG-PET plays a particularly important role in differential diagnosis among neurodegenerative disorders. In frontotemporal dementia, FDG-PET shows frontal and anterior temporal hypometabolism that helps distinguish the behavioral variant FTD from AD. In Lewy body dementia, occipital hypometabolism provides a supportive diagnostic feature. In Parkinson's disease, FDG-PET can identify the PINK1/PARK2 phenotype and monitor disease progression[26].
Dopaminergic imaging using PET and single-photon emission computed tomography (SPECT) provides critical support for diagnosing Parkinsonian syndromes. ¹²³I-FP-CIT (DaTscan) SPECT, FDA-approved in 2011, images the dopamine transporter (DAT) and reliably distinguishes Parkinsonian syndromes from non-Parkinsonian tremor disorders such as essential tremor[27]. Reduced DaTscan binding in the striatum indicates presynaptic dopaminergic degeneration, supporting diagnosis of PD, PSP, or CBS.
¹⁸F-FDOPA PET provides complementary information by measuring dopa decarboxylase activity and dopamine turnover. FDOPA uptake correlates with disease severity and has been used to monitor disease progression and therapeutic response in PD[28].
Neuroinflammation, particularly microglial activation, is a hallmark of neurodegenerative diseases and represents both a therapeutic target and a biomarker opportunity. PET imaging of translocator protein (TSPO), which is upregulated in activated microglia, enables in vivo visualization of neuroinflammatory processes.
First-generation TSPO tracers such as ¹¹C-PK11195 demonstrated increased binding in AD, PD, and MS, confirming the presence of neuroinflammation in vivo. However, high non-specific binding and variable affinity across individuals (due to TSPO polymorphism) limited quantitative accuracy. Second-generation tracers including ¹⁸F-DPA-714, ¹⁸F-GE-180, and ¹¹C-ER176 have improved signal-to-noise properties and reduced affinity polymorphism sensitivity[29].
Emerging PET targets beyond TSPO include:
P2X7 Receptor: The P2X7 purinergic receptor on microglia represents a novel target for neuroinflammation imaging with potential for greater specificity than TSPO[30].
Monoamine Oxidase B (MAO-B): ¹¹C-L-deprenyl PET images MAO-B activity in astrocytes, providing complementary information about neuroinflammatory processes[31].
Imaging reactive astrocytes using PET represents an emerging frontier. The ¹¹C-BU99008 tracer binds to astrocytic MAO-B, which is upregulated in reactive astrocytes surrounding amyloid plaques in AD. This approach may provide unique insights into the neuroinflammatory component of AD pathogenesis[32].
TAR DNA-binding protein 43 (TDP-43) pathology is a key feature of ALS, FTD, and limbic-predominant age-related TDP-43 encephalopathy (LATE). While no validated TDP-43 PET tracer exists currently, development efforts are underway to enable visualization of this important pathological protein in vivo[33].
The National Institute on Aging and Alzheimer's Association (NIA-AA) research framework employs a biomarker-based classification system that integrates PET imaging with other biomarkers. The AT(N) system classifies biomarkers according to what they measure: A for amyloid (Aβ), T for tau (tau), and (N) for neurodegeneration. FDG-PET and tau PET serve as (N) biomarkers reflecting downstream neuronal injury, while amyloid PET and tau PET provide evidence of upstream pathological processes[34].
This framework enables biologically defined AD diagnosis that does not require clinical symptoms, supporting research into preclinical disease and facilitating clinical trial enrichment. An individual can now be classified as having AD based on biomarker evidence even in the absence of cognitive impairment, opening opportunities for disease-modifying interventions before irreversible neuronal loss occurs.
The Amyloid Imaging Task Force established appropriate use criteria that guide clinical application of amyloid PET. Recommended indications include:
Amyloid PET is not recommended for patients with typical AD clinical presentation, as a positive result would not change management, or for asymptomatic individuals, as there is no disease-modifying therapy with demonstrated benefit in preclinical stages[35].
PET biomarkers have become essential tools in clinical trial design for neurodegenerative diseases. Amyloid PET enables confirmation of amyloid pathology for enrollment in anti-amyloid trials, reducing screen failure rates and ensuring target engagement. Tau PET supports patient stratification based on disease stage and provides a direct measure of target engagement for anti-tau therapeutics. FDG-PET serves as a secondary endpoint reflecting downstream neurodegenerative changes[36].
The use of PET biomarkers in trials has accelerated therapeutic development by enabling smaller, shorter trials with higher sensitivity to treatment effects. Regulatory agencies including the FDA and EMA now accept amyloid PET as a surrogate endpoint for accelerated approval in AD trials.
The field of PET imaging in neurodegeneration continues to evolve with several emerging opportunities:
Synaptic PET: Novel tracers targeting synaptic vesicle protein 2A (SV2A), such as ¹¹C-UCB-J, enable direct visualization of synaptic density, providing a more proximal measure of neuronal integrity than FDG-PET or structural MRI[37].
Alpha-Synuclein PET: Development of alpha-synuclein PET tracers would enable visualization of Lewy body pathology in PD and DLB, complementing dopaminergic imaging[38].
Multimodal Imaging: Integration of PET with MRI, especially arterial spin labeling (ASL) for cerebral blood flow and diffusion tensor imaging for white matter integrity, provides comprehensive characterization of neurodegenerative processes.
Artificial Intelligence: Machine learning approaches applied to PET data enable automated classification, prediction of progression, and identification of novel imaging phenotypes that may not be apparent through visual inspection alone[39].
Rowe CC, et al. Amyloid imaging: from PET tracer to biomarker. Lancet Neurol. 2024. ↩︎
Jack CR Jr, et al. NIA-AA Research Framework: Toward a biological definition of Alzheimer's disease. Alzheimers Dement. 2018. ↩︎
Pike KE, et al. Beta-amyloid imaging and memory in non-demented individuals: evidence for preclinical Alzheimer's disease. Brain. 2007. ↩︎
Johnson KA, et al. Appropriate use criteria for amyloid PET: a report of the Amyloid Imaging Task Force. Alzheimers Dement. 2013. ↩︎
Scholl M, et al. PET imaging of tau pathology in Alzheimer's disease and frontotemporal dementia. J Neurol Neurosurg Psychiatry. 2016. ↩︎
van Dyck CH, et al. Lecanemab in Early Alzheimer's Disease. N Engl J Med. 2023. ↩︎
Lowe VJ, et al. An approach to validate FLT-PET as a biomarker in the Alzheimer's disease spectrum. Alzheimers Res Ther. 2022. ↩︎
Klunk WE, et al. Imaging brain amyloid in Alzheimer's disease with Pittsburgh Compound-B. Ann Neurol. 2004. ↩︎ ↩︎
Bateman RJ, et al. Clinical and biomarker changes in dominantly inherited Alzheimer's disease. N Engl J Med. 2012. ↩︎
Clark CM, et al. FLUTEORAPIR: PET imaging of beta-amyloid pathology in Alzheimer's disease and mild cognitive impairment. JAMA. 2011. ↩︎
Barthel H, et al. First-in-human PET study with 18F-AV-45: assessment of amyloid burden in healthy controls, MCI, and AD patients. J Nucl Med. 2011. ↩︎
Wolk DA, et al. Comparison of 18F-Flutemetamol and 11C-PiB for PET imaging of amyloid pathology. J Nucl Med. 2013. ↩︎
Klunk WE, et al. The Centiloid scale: standardized quantification of cerebral amyloid burden in imaging studies. Alzheimers Dement. 2015. ↩︎
Mintun MA, et al. 18F-AV-45 PET and cognitive decline in cognitively normal individuals and patients with MCI. Neurology. 2019. ↩︎
Rabinovici GD, et al. Impact of amyloid PET on clinical management: the IDEAS study. JAMA. 2019. ↩︎
Nation DA, et al. New IDEAS: imaging dementia-refinement of amyloid scanning in diverse populations. Alzheimers Dement. 2021. ↩︎
Xia J, et al. Structure-activity relationship and pharmacokinetic studies of PET probes for imaging tau pathology. J Med Chem. 2013. ↩︎
Maruyama M, et al. Imaging of tau pathology in a tauopathy mouse model and in Alzheimer patients compared to normal controls. Neuron. 2013. ↩︎
Bu G. Apolipoprotein E and its receptors in Alzheimer's disease: pathways, pathogenesis and therapy. Nat Rev Neurosci. 2009. ↩︎
Lohith TG, et al. Brain imaging of tau pathology in Alzheimer's disease: 18F-MK-6240 PET. J Nucl Med. 2018. ↩︎
Song M, et al. 18F-PI-2620: a novel tau PET tracer for Alzheimer's disease. J Nucl Med. 2018. ↩︎
Teng E, et al. Preliminary evaluation of APN-1607 PET in Alzheimer's disease and controls. J Prev Alzheimers Dis. 2020. ↩︎
Braak H, et al. Staging of Alzheimer disease-associated neurofibrillary pathology using paraffin sections and immunocytochemistry. Acta Neuropathol. 2006. ↩︎
Brier MR, et al. Tau and Aβ imaging, CSF measures, and cognition in Alzheimer's disease. Sci Transl Med. 2016. ↩︎
Minoshima S, et al. Metabolic reduction in the posterior cingulate cortex in very early Alzheimer's disease. Ann Neurol. 1997. ↩︎
Eckert T, et al. FDG-PET in the differential diagnosis of Parkinsonian disorders. Neuroimage. 2005. ↩︎
Marshall V, et al. Parkinsonism in patients with behavioral variant frontotemporal dementia: 123I-FP-CIT SPECT findings. Mov Disord. 2013. ↩︎
Whone AL, et al. Slower disease progression in Parkinson's disease with early versus delayed start of dopaminergic treatment: evidence from the 18F-FDOPA PET study. Lancet Neurol. 2003. ↩︎
Cagnin A, et al. In vivo measurement of activated microglia in dementia. Lancet. 2001. ↩︎
Junker A, et al. P2X7 receptor expression in microglia: a potential target for treatment in neurodegeneration. J Neural Transm (Vienna). 2019. ↩︎
Fowler JS, et al. Monoamine oxidase B (MAO-B) inhibitors: implications for brain-imaging in Parkinson's disease. Nat Rev Neurol. 2015. ↩︎
Carter SF, et al. Evidence for astrocytic dysfunction in early Alzheimer's disease: a PET study with 11C-BU99008. J Cereb Blood Flow Metab. 2019. ↩︎
Josephs KA, et al. Updated TDP-43 in Alzheimer's disease staging: a new classification system. Acta Neuropathol. 2020. ↩︎
Jack CR Jr, et al. A/T/N: an unbiased descriptive classification scheme for Alzheimer disease biomarkers. Neurology. 2016. ↩︎
Minoshima S, et al. Appropriate use criteria for amyloid PET: a systematic review. J Nucl Med. 2016. ↩︎
Sehlin D, et al. Antibody-based PET imaging of amyloid and tau pathology in Alzheimer's disease. Nat Rev Neurol. 2023. ↩︎
Finnema SJ, et al. Imaging synaptic density in the living human brain. Sci Transl Med. 2016. ↩︎
Foulds PG, et al. Spreading of alpha-synuclein pathology in the brain: implications for diagnosis and treatment. J Parkinsons Dis. 2019. ↩︎
Ding Y, et al. Deep learning-based classification of Alzheimer's disease with multimodal neuroimaging and clinical data. Neuroimage Clin. 2021. ↩︎