Position Emission Tomography/Single‐Photon Emission Tomography Neuroimaging for Detection of Premotor Parkinson's Disease

Summary

Premotor Parkinson's disease (PD) refers to a prodromal stage of Parkinson's disease (PD) during which nonmotor clinical features may be present. Currently, it is difficult to make an early diagnosis for premotor PD. Molecular imaging with position emission tomography (PET) or single‐photon emission tomography (SPECT) offers a wide variety of tools for overcoming this difficulty. Indeed, molecular imaging techniques may play a crucial role in diagnosing, monitoring and evaluating the individuals with the risk for PD. For example, dopaminergic dysfunctions can be identified by detecting the expression of vesicular monoamine transporter (VMAT2) and aromatic amino acid decarboxylase (AADC) to evaluate the conditions of dopaminergic terminals functions in high‐risk individuals of PD. This detection provides a sensitive and specific measurement of nonmotor symptoms (NMS) such as olfactory dysfunction, sleep disorders, and psychiatric symptoms in the high‐risk patients, especially at the premotor phase. Molecular imaging technique is capable of detecting the dysfunction of serotonergic, noradrenergic, and cholinergic systems that are typically associated with premotor manifestations. This review discusses the importance of SPECT/PET applications in the detection of premotor markers preceding motor abnormalities with highlighting their great potential for early and accurate diagnosis of premotor symptoms of PD and its scientific significance.

Introduction

Parkinson's disease (PD) is a movement disorder manifested as bradykinesia, rigidity, resting tremor, and postural equilibrium dysfunction. In recent years, as a much‐heightened awareness of its complexity, PD has been recognized as a multisystem disorder with motor and nonmotor features. The widespread neurochemical and neuroanatomical changes are involved in not only the nigrostriatal dopaminergic system, but also serotonergic and noradrenergic systems in brainstem areas, and cholinergic system in frontal and brainstem regions 1. As such, recognition of these nonmotor symptoms, especially in the context of the prodromal and early stage of PD, has led to a rethinking of what is appropriate definition of PD; how we can accurately identify populations of people at stake either for development of the disease or its later complications through biomarkers; and whether we are able to find new targets for the diagnosis, prevention, and treatment of PD 1. However, it is difficult to make an early diagnosis for premotor PD in the clinical settings. Recent progresses in molecular imaging with position emission tomography (PET) and single‐photon emission tomography (SPECT) have offered a wide variety of tools for overcoming this difficulty. This review is specifically focused on this topic.

Classification of Premotor PD

Braak et al. 2 indicated that damage to the substantia nigra pars compacta was the primary cause of PD. Lesions initially occur in the dorsal motor nucleus, anterior nucleus, and bulbus olfactorius of the medulla oblongata, leading to an abnormal performance in olfaction (Pathologic phase I) 2, 3. Thereafter, the entire medulla oblongata, back cover of the pons, caudate nucleus, raphe nuclei, giant nucleus, basal forebrain nuclei, and intermediate cortex are gradually affected, resulting in sleep disorders and emotional changes (Pathologic phase II). The substantia nigra pars compacta becomes affected, giving rise to abnormal temperature regulation, cognitive disorders, and depression (Pathologic phase III). In pathologic phase IV, four major symptoms appear, including tremor, rigidity, bradykinesia, and impaired balance. Finally, the neocortex is affected, leading to motor fluctuations, fatigue, visual hallucinations, dementia, and other mental symptoms. This new staging theory provides a theoretical basis to explain the pathophysiology of the nonmotor symptoms 4.

Premotor PD can also be divided into several stages, based on the presence of risk, physiological, and/or clinical markers. As proposed by the UK Brain Bank 5, 6, working retrospectively from recognizable PD, premotor PD can be divided into the following stages: (1) the pre‐diagnostic phase, (2) the premotor phase, (3) the preclinical phase, and (4) the prephysiological phase. The term “Parkinson's Disease at Risk Syndrome” (PARS) 7, 8 describes a hierarchical classification pyramid for patients who do not yet have clinical PD. The details of each stage are as follows:

1. Patients in the prediagnostic phase have parkinsonian signs, such as classic clinical features and PD symptoms although these signs do not meet the diagnostic criteria for PD. With the help of dopamine transporter (DAT) imaging, patients with less obvious parkinsonian signs but with biomarker evidence of dopamine deficiency may come to be diagnosed with PD, and indeed be reclassified as having early and mild symptoms of the disease.

2. At the stage of premotor phase, patients precede the classical motor features of PD, including a decreased sense of smell, various gastrointestinal dysfunction, depression, and other systemic features 9. Neuroprotective treatments could potentially have a beneficial effect 10 on them, and their dopaminergic neurons are relatively spared in the brain 9. Before the diagnosis of PD, these features are present in most PD patients with variable degrees, although not universal in all patients, increased fatigue or anxiety has also been linked to PD.

3. Preclinical PD refers to physiological changes which can be detected by biomarker techniques when clinical features are insufficient. Neuroimaging changes, as detected by DAT, SPECT and [18F]‐fluorodopa PET, can be used to identify preclinical PD 11.

4. Prephysiological patients are those who have no evidence of PD, but possess a genetic mutation that can confer the individual to a high risk of PD in the future 12, 13, 14. The term “prephysiological” is intended to convey the idea that physiological probes, such as dopamine imaging, cannot detect signs of disease either clinically or physiologically.

PD is a progressive neurodegenerative disease and involves different systems in the brain. The recognition of nonmotor symptoms is crucial to ensure timely treatment 15. Several nonmotor symptoms are noted to precede the cardinal motor features of PD and thus have been termed “premotor”. These premotor symptoms may reflect the earliest pathological changes in the central nervous system that are associated with PD progression. The presence of some premotor features including sleep dysfunction, impaired olfaction, constipation, and mood changes strongly indicates the highly potential possibility developing into PD 7, 8, 10. Furthermore, other nonmotor symptoms such as cognitive impairment, autonomic dysfunction, and insomnia may occur in the early de novo patients with PD 1, 16, 17, 18, 19. Some of these nonmotor features such as depression and insomnia also reflect disturbances in the nondopaminergic systems. For example, characteristic of the cognitive impairment in early Parkinson's disease invokes dopaminergic frontal‐striatal and nondopaminergic posterior cortically based regions 1, while autonomic and sensory symptoms invokes involvement of the peripheral and central nervous system consistent with the proposed progressive neurodegenerative changes by Braak et al. 20. However, it is difficult for neurologists employing common imaging techniques to identify patients with PD when diagnosis of PD is gaining in attention. In recent years, rapid advances in imaging tests, such as PET/SPECT, have brought about further progress in the recognition and diagnosis of PD. With the help of radioactive tracers, molecular imaging techniques can detect changes in metabolism, circulation, neurotransmitter receptors, and transporters in the brain.

Detection of Neurotransmitter Deficiencies and Other Changes

In the past, the idea of detecting PD for purely diagnostic purposes met with limited enthusiasm by most scholars due to a lack of therapies with documented neuroprotective capacity. This has been changed in the wake of newer treatments becoming available.

There are many different approaches to detect PD with some of them assessing changes in the profile of metabolic activity or cerebral blood flow. Other approaches detect neurotransmitter deficiencies and nondopaminergic pathways, such as the bioactivity of striatal aromatic amino acid decarboxylase (AADC with ¹⁸F‐dopa PET), the density of vesicular monoamine (VMAT2 binding with DTBZ) or plasmalemmal dopamine transporters (DAT labeled with a variety of [¹¹C] [¹⁸F] ligands), the bioactivity of serotonin transporter ligands (with [¹¹C]‐DASB) or both dopamine and noradrenaline membrane transporters (with ¹¹C‐RT132).

18F‐FDG Imaging

¹⁸F‐fluorodeoxyglucose (¹⁸F‐FDG) can be used as a glucose analog to reveal the glucose metabolism processes of cells. A change in metabolic activity may indicate early lesions because glucose metabolism is altered with the development of PD 21. A lower level of ¹⁸F‐FDG was observed in the corpus striatum and leaf striatum of patients with PD compared to healthy subjects, especially among akinetic‐rigid subtype of patients with PD when comparing to tremor‐dominant subtype of patients with PD 22, 23. Meanwhile, the affected regions of the PD brain exhibited a reduced amount of ¹⁸F‐FDG as compared with that of the normal regions. However, some researchers have opposite views because they observed an increase, instead of decrease, in the metabolism, in the striatum of PD 24, 25.

Nevertheless, some studies have revealed that change in glucose metabolic rate is closely linked to the progress in the pathological stage, suggesting a potential to identify the pathological stage of patients with PD through ¹⁸F‐FDG‐PET 26, 27, 28. For example, ¹⁸F‐FDG‐PET studies examining cognitive dysfunction have shown significant metabolic reduction in the frontal temporoparietal association cortex in PD patients with mild cognitive impairment, as compared to that of PD patients without cognitive impairment 29, 30. Another study indicated that in cognitive subtypes of patients with PD, hypometabolism in the parieto‐temporo‐occipital cortex was found in PD patients with impaired nonamnestic single‐domain cognitive function, as compared to PD patients with intact cognition. Furthermore, widespread hypometabolism in the frontal, cingulate, and parieto‐temporo‐occipital cortices was observed in PD subjects with multidomain cognitive dysfunction 31. Huang et al. indicated that cognitive dysfunction correlated with increased posterior cingulate metabolism and decreased metabolism in the temporoparietal lobe. Moreover, depressive symptoms were correlated with increased metabolism in the amygdala; anxiety scores were associated with hypometabolism in the caudate, and apathy scores were connected with increased metabolism in the anterior cingulate and orbitofrontal lobe and decreased metabolism in the temporoparietal association cortex 25. Their findings suggest that cognitive and emotional dysfunction of patients with PD is associated with distinct patterns of cerebral metabolic changes.

However, this method lacks reliable specificity because the glucose metabolic rate may be decreased in other diseases, such as Alzheimer's disease (AD) and multiple system atrophy (MSA) 32, 33, 34, 35. Therefore, it is only when we are able to exclude the possibility of a patient having other diseases of the central nervous system, we can use this method to make an accurate diagnosis for patients with early symptoms of PD.

Cerebral Perfusion Imaging

¹⁵O‐H₂O, an ideal radio pharmaceutical mediator for measuring the blood flow in the brain because of its high absorption, is unaffected by metabolism and is normally used for very short duration due to its short radioactive half‐life of 2.05 min. By examination with 3D ¹⁵O‐H₂O‐PET, Boecker et al. showed that sensory‐evoked brain activation in patients with PD was reduced in cortical (parietal and frontal) and subcortical (basal ganglia) areas during the premotor phase of PD 36. Using PET with H₂ ¹⁵O, Feigin et al. assessed the effects of levodopa infusion on learning and memory in patients with PD during the premotor phase and found that Levodopa treatment appears to have subtle detrimental effects on nonmotor symptoms, such as cognitive function in nondemented patients with PD 37. However, PET with H₂ ¹⁵O was hampered, by the short half‐life of ¹⁵O‐labeled water, making studies quite difficult and sedation that in itself affects the cerebral blood flow 38.

Functional Imaging of Neurotransmitters

PET and SPECT have been extensively employed to investigate striatal dopamine terminal function in PD and atypical Parkinsonism 39. These approaches can measure (1) the bioactivity of striatal aromatic amino acid decarboxylase (AADC) with ¹⁸F‐dopa PET, (2) the density of presynaptic dopamine transporters (DAT) that are responsible for the high‐affinity transport of dopamine from the synaptic cleft with tropane‐based markers, and (3) the availability of vesicular monoamine transporter (VMAT2) with ¹¹C‐ or ¹⁸F‐dihydrotetrabenazine PET, reflecting the transport of monoamines from the cytoplasm to the vesicles 40 (Figure 1).

Figure 1.

Images of striatal beta‐CIT SPECT (DAT), FP‐CIT SPECT (DAT), ¹¹C‐DTBZ PET (VMAT2), and ¹⁸F‐dopa PET (DDC) uptake in healthy volunteers and early PD. Note that the four imaging modalities all show asymmetrically reduced posterior putamen dopaminergic function in PD 40.

PET/SPECT Imaging of Presynaptic Dopamine Terminal Function

¹⁸F‐dopa PET is a functional imaging method and is used to determine whether patients are suffering from PD. ¹⁸F‐dopa, like endogenous dopa, can be transformed to DA by AADC in the striatum. The amount of ¹⁸F‐dopa taken up by the basal ganglia may be used to reflect the bioactivity of AADC, which is closely linked to the severity of PD and reflects the degeneration of dopaminergic neurons in the substantia nigra. PET imaging has shown that the amount of ¹⁸F‐dopa taken up by the basal ganglia, in particular by the putamen, a part of the basal ganglia, is remarkably reduced in patients with PD, and the quantity of ¹⁸F‐dopa absorbed in affected regions of the brain was much lower than that in normal regions 41. The uptake of striatal ¹⁸F‐dopa was more rapidly and severely reduced in patients with PD compared to age‐matched controls 42, 43, 44, 45. Assessing dopaminergic presynaptic systems provides a method to detect preclinical dysfunction. Holthoff et al. evaluated 7 pairs of twins discordant for Parkinson's disease (PD), of whom the co‐twins showed no signs of motor impairment on neurological examination and revealed a significant correlation between scores obtained in Buschke's selective reminding test and striatal ¹⁸F‐dopa uptake, further substantiating the role of dopaminergic pathways in memory processing 11. Piccini et al. 46 observed putaminal ¹⁸F‐dopa uptake in relatives with familial PD and found that 25% of asymptomatic adult family members have an abnormal reduction in this process. Meanwhile, the ability to take up ¹⁸F‐dopa is lost by over 30% in the globus pallidus 47, 48. It has been reported that ¹⁸F‐dopa uptake, which may be responsible for both compensatory upregulation of AADC and uptake of ¹⁸F‐dopa into serotonergic and noradrenergic terminals, is increased in the anterior cingulate, dorsolateral prefrontal cortex, and globus pallidus interna of PD 49, 50, 51.

Nonetheless, ¹⁸F‐dopa is extensively metabolized in peripheral tissues and its radio‐labeled metabolites can cross the blood–brain barrier. Therefore, corrections for metabolites crossing the blood–brain barrier are required 52. Instead, ¹¹C‐dopa with a half‐life decay of 20.3 min is not readily and rapidly metabolized outside the brain, making it possible to challenge different substances or perform multiple scans in the same individual 52. However, due to its low productivity, ¹¹C‐dopa is still far from clinical use 53. ¹⁸F‐dopa PET reflects both neuron loss and a failure of AADC upregulation; however, it does not strictly provide remaining terminal density 40. As a consequence, ¹⁸F‐dopa PET still contributes to improving diagnosis and facilitating timely therapeutic interventions.

PET/SPECT Imaging of Dopamine Transporters (DAT) on Presynaptic Membranes

DATs, located at the presynaptic membrane of dopaminergic neurons in the central nervous system, mainly function to remove DA from the synaptic cleft after release from the presynaptic neuron, thus terminating the signal transmission between nerve cells. DAT imaging shows the function of the presynaptic membrane in the nerve endings and the amount of neurons. Several lines of evidence indicate that DAT declines in patients with PD, and the decrease in DAT is an earlier, more sensitive, and more direct indicator than the change in DA receptors 54, 55. The integrity of the nigrostriatal pathway can also be detected by DAT imaging. However, DAT imaging cannot reflect the synthetic function of dopamine such as the bioactivity of AADC 56. DAT imaging may be a more sensitive and objective method used to examine the capability of dopaminergic neurons for patients. The rate of decline of DAT is faster in patients with PD, although DAT decreases each year. PET imaging using tropane‐based tracers such as ¹¹C‐CFT, ¹⁸F‐FP‐CIT, and ¹¹CRTI‐32 or with ¹¹C‐methylphenidate can be employed to assess the dopamine transporter function in PD. SPECT ligands include ¹²³I‐b‐CIT, ¹²³I‐FP‐CIT, ¹²³I‐IPT, ¹²³I‐altropane, and ⁹⁹ mTc‐TRODAT‐1 57.

SPECT neuroimaging showed that ⁹⁹mTc‐TRODAT‐1 selectively binds to the DAT in patients with PD and healthy subjects. Regions of interest (ROIs) in the caudate and putamen were calculated by the TRODAT‐1 distribution volume ratio, which is a reflection of DAT availability. Weintraub et al. examined the association between the nonmotor symptoms of PD (anxiety, depression, and fatigue) and DAT availability in two groups and then assessed the impact of the disease severity in the PD group and showed the correlation between striatal DAT imaging results and anxiety and depression symptoms in PD 58, 59. Several animal studies have also shown that the ability of the striatum to take up ¹²³I‐PE2I falls significantly in PD models that have not developed PD symptoms 60, 61. Some reports have shown that the levels of ¹²³I‐β‐CIT absorbed by the bilateral striatum in patients with PD were lower than those in the healthy subjects, and the difference in the absorbed amount could be observed in the bilateral striatum, indicating that the quantity in the putamen in the affected side is reduced remarkably compared to the untreated side 62. Moreover, the caudate nucleus is less affected as compared to the putamen. The capacity to take up ⁹⁹Tcm‐TRODAT in the front and back parts of the caudate nucleus and putamen is seriously impaired during the early stages of PD 63. Recently, a new computer‐assisted image analysis system adopted by Goebel et al. 64 has been employed to remarkably improves the diagnostic accuracy of PD through analysis of [¹²³I] β‐CIT SPECT images.

Hyposmia has been recognized as a common nonmotor manifestation of PD that usually precedes motor symptoms 65. Studies have demonstrated abnormalities of DAT binding in hypoxic 1st degree relatives of patients with PD, some of whom then go on to develop PD at follow‐up 66, 67. Ross et al. found that hyposmia was correlated with the occurrence of PD by detecting the olfactory ability of 2263 elderly people from 1991 to 1996 68. In a two‐year follow‐up study, Ponsen et al. screened 40 anosmatic cases from 361 PD patients' relatives without any symptoms and found that 10% of them developed PD, while in the control group of 38 osmotic relatives, none developed PD 67. Becker et al. 69 used cranial ultrasonic examination to observe 30 cases of patients with a diminished sense of olfaction or loss of olfaction without any obvious cause. Using single‐photon emission computerized tomography, they found that the echogenicity of the substantia nigra was increased in 11 cases, and dopamine transporters were decreased in 5 cases, with three cases developing into PD during the follow‐up observation. The above studies confirmed the Braak theory and verified that olfactory dysfunction was one of the symptoms of the early stage of PD. Some researchers have proposed that dopamine exists as a neurotransmitter in some regions of the brain, including the olfactory bulb and especially the caruncula mammillaris, and hyposmia may be the result of a lack of dopamine in these areas.

Recent studies show that hyposmia may appear 2–7 years before PD is diagnosed 70. As a result, it may become one of the indicators of PD. The dopamine system in patients with olfactory abnormalities and a family history of PD has been explored in numerous studies. For example, in a two‐year follow‐up study of nonparkinsonian normosmic and hypoxic individuals with a positive family history, Ponsen et al. found that approximately 10% of the hypoxic individuals (who also had reduced DAT binding determined by SPECT) had a high likelihood of developing PD 67. However, with regard to dopamine system abnormalities, it is necessary to screen other risk factors in people with hyposmia.

Rapid Eye Movement sleep behavior disorder (RBD) has also been recognized as another risk factor of PD. In general, 15–27% of patients with PD exhibit RBD 71, which can occur 10 years earlier than the onset of motor symptoms 72, 73. Longitudinal analysis indicated that in more than 40% of patients with PD, RBD predicts the impending motor symptoms, and some subjects with RBD can even develop PD 73. In addition, RBD has also been recognized as a neurodegenerative disease and a risk factor of PD with dementia and cognitive dysfunctions. Iranzo et al. 74 carried out a prospective study based on 20 patients with IRBD (idiopathic RBD) undergoing serial DAT imaging with (¹²³)I‐FP‐CIT SPECT at baseline and again after 1.5 years and 3 years. Their findings suggest that ¹²³I‐FP‐CIT SPECT in IRBD identifies progressive nigrostriatal changes that also arise in PD patients with recent onset of Parkinsonism. DAT imaging abnormalities in the putamen of patients seemed to be more pronounced than in the caudate nucleus both at baseline and 3 years later, which suggests that the rate of reduction of tracer uptake in the putamen is a more sensitive measure of the progression of dopaminergic dysfunction than the rate in the caudate nucleus. As there is a lack of prospective studies on the natural course of sleep disorder 75, 76, 77, this study provides a reliable test to determine whether the process of aging alone could lead to a declined uptake of ¹²³I‐FP‐CIT in striatal and a prospective longitudinal design in a cohort of patients with RBD.

In addition, at least 50% of patients with PD are affected by depression that can affect motor symptoms 78 and even accelerate the progress of PD. A number of studies have shown that depressive symptoms appear earlier than motor symptoms 79, 80, 81. In approximately 20% of patients with PD, depressive disorders appear earlier than motor symptoms and can last up to 20 years before motor symptoms appear. Epidemiological surveys have shown that in the 3–6 years before PD diagnosis, the incidence of depression is relatively higher, and the risk of individuals with depression developing PD is 2.2–3.2 times that of nondepressed individuals 82. A transcranial Doppler study found that among depressed individuals without PD, the hyperechogenicity rate of the substantia nigra was three times than that of controls 83. Accordingly, depression can also be a risk factor of PD pathogenesis. Kitamura and Marsh 84, 85 demonstrated that the substantia nigra is affected in the later stages, and the raphe nuclei are affected in the early stages, which suggests that depression is a prodromal symptom of PD. Clinical studies demonstrate the efficacy of medications and psychotherapies for PD depression, underscoring the importance of timely detection and concerted management. PET studies have demonstrated that reuptake of the dopamine ligand [¹¹C]RTI‐32 is significantly reduced in the ventral striatum, thalamus, anterior cingulate cortex, and locus coeruleus of PD subjects with depression when compared with those without depression 86.

Vriend et al. 87 found that depressive symptoms were related to lower DAT binding in the right caudate nucleus, while motor symptoms were associated with decreased [¹¹C]RTI‐32 binding in the right putamen. This double dissociation was most pronounced in early‐stage patients with PD. These results suggest that depression of PD is related to dopamine loss in the caudate nucleus, which is associated with degeneration of dopaminergic projections from the ventral tegmental area, while motor symptoms are associated with low dopamine signaling to the putamen and loss of nigrostriatal projections. In addition, Rektorova et al. 88 demonstrated that imaging with ¹²³I‐FP‐CIT SPECT appears to be sensitive for detecting dopaminergic deficit associated with mild depressive symptoms and specific cognitive dysfunction in patients with PD, yet without a current depressive episode and/or dementia.

PET/SPECT Imaging of Type II Vesicular Monoamine Transporter (VMAT‐2)

VMAT‐2, the membrane protein of vesicles in the presynaptic terminal, transports monoamines, such as 5‐hydroxytryptamine (5‐HT) and dopamine, from the cytoplasm to the vesicle for storage to prevent these monoamines from decomposing. The binding of VMAT‐2 can be measured to determine the loss of monoamines in dopamine nerve terminals. PET, using DTBZ (dihydroxytetramethylene benzo quinolizine) as a tracer, is usually used in clinical studies to measure vesicles containing monoamines. The decreases in the amount of DTBZ absorbed are closely linked to the drop in the number of dopamine neurons. Some studies have reported that the amount of DTBZ binding to the caudate nucleus, putamen, and substantia nigra is decreased, especially in the back of the putamen 89, 90. However, this method lacks reliable specificity because the amount of DTBZ absorbed may be decreased in the caudate nucleus and putamen of patients with progressive supranuclear palsy (PSP) and MSA 91, 92. Therefore, this measurement may not be appropriate for differential diagnosis.

It has been shown that isolated RBD in elderly nonparkinsonian individuals was associated with reduced striatal uptake of DTBZ 93. In patients with early PD, REM sleep duration has a negative correlation with dorsal and lateral mesopontine fluoro‐deoxyglucose uptake 94. The increased activity in the locus coeruleus and dorsal raphe is abnormal because nondopaminergic structures are included in these regions. The author indicated that there is a compensatory upregulation of decarboxylase activity in these neurons, which might suggest that the activities in more caudal brainstem structures are impaired.

PET/SPECT Imaging of Postsynaptic Dopaminergic Function (DA Receptor)

The detection of dopaminergic D2 receptors is often used for a diagnosis judgment of PD because pathological damage of the D2 receptors is always observed in patients with PD 95, 96. Recently, SPECT using [¹¹C]raclopride as a specific D2 receptor marker has been employed to detect D2 receptors. The quantity of D2 receptors in terms of the number and function can be obtained through the ratio of the D2 receptors in the basal ganglia to those in the cerebellum and frontal and occipital brain regions.

Sun et al. 97 pointed out that D2 receptor may show a dynamic change with the PD progression. They also showed that in medial forebrain bundle lesioned rat model, D2 receptor binding following 6‐month lesion was significantly lower than that at 4‐week lesion, while D2 receptor binding in striatal model was decreased to the same extent at both 4 weeks and 6 months after lesion 98. Antonini et al. 99 observed a higher [¹¹C]raclopride uptake in the putamen in early and untreated patients with PD, which is called “upregulation effect of dopamine receptors”. The significant increase in [¹¹C]raclopride uptake in the putamen is reversed with the increase in PD severity, indicating the long‐term downregulation of striatal dopamine D2 receptor binding in PD 100. Sawle et al. 101 combined the ¹⁸F‐dopa and [¹¹C]raclopride studies and found that the putamen with the lowest ¹⁸F‐dopa uptake had the highest [¹¹C]raclopride binding in patients with PD, suggesting the upregulation of postsynaptic D2 receptors. This finding implies the decrease in dopamine in the nigralstrial pathway leads to hypersensitivity of the postsynaptic D2 receptors.

A limitation of the studies with D2 antagonist radiotracers such as [¹¹C]raclopride is the failure to provide specific information on D2 receptors configured in a state of high affinity for the agonists. Therefore, further study is needed for dopaminergic D2 receptors 102.

The cognitive impairment of PD has traditionally been linked to the presence of dementia in later stages of the disease 103. However, recent studies show that cognitive disorders can appear in early stages 16, 17, 104. Knowing the cognitive profile of PD can deepen our understanding of the clinical phenotype and render it easier to reach a timely diagnosis and favoring intervention on the symptoms from the initial stages. Studies of incident PD cohorts affirm that cognitive dysfunction is no longer solely a complication of advanced disease. In early PD, characteristics of the cognitive phenotype such as deficits in attention, executive function, verbal fluency, and visuospatial domains, are mild, which thereby invoke dopaminergic frontal‐striatal, as well as, nondopaminergic posterior cortically based regions 1. In studying a mouse model of cognitive performance and synaptic plasticity of PD, Bonito‐Olivia et al. showed that partial dopamine depletion leads to impairment of long‐term recognition memory accompanied by abnormal synaptic plasticity in the dentate gyrus. They also demonstrate that activation of dopamine D1 receptors corrects these deficits through a mechanism that requires intact extracellular signal‐regulated kinases signaling 105.

Additionally, different studies also report a decline in olfactory performance after a short period of sleep deprivation 106, 107. The mechanisms underlying the dysfunctions are poorly understood though the impairment of dopamine (DA) neurotransmission in the olfactory bulb and the nigrostriatal pathway may be relevant to the olfaction and REM sleep disturbances 106. Rodrigues et al. 106 demonstrated that modulation of the dopaminergic D2 receptors in the olfactory bulb could provide a more comprehensive understanding of the olfactory deficits in PD and REM sleep deprivation. The bulbar dopaminergic D2 receptor produced by periglomerular neurons plays an essential role in olfactory discrimination processes, as the Substantia Nigra Pars Compacta (SNpc), and the striatum. In recently diagnosed PD patients with mild symptoms, severity of olfactory deficits has been found to correlate with dopaminergic dysfunction as measured by the dopaminergic D2 receptors, despite a lack of correlation between either measurement and severity of motor dysfunction 108.

PET/SPECT Imaging and Nondopaminergic Pathways

The majority of the imaging investigations to date have concentrated on the dopamine system. However, a growing number of studies have shown that nondopaminergic neurotransmission is also impaired, giving rise to nonmotor manifestations including depression, dementia, sleep disturbance, fatigue, and autonomic dysfunction. Despite this investigation has less specificity than dopaminergic neurotransmission for differential diagnosis of PD, assessment of the function of serotonergic, cholinergic, and noradrenergic neurons caudal to the midbrain during the disease by SPECT and PET ligands is becoming increasingly important.

PET/SPECT Imaging of Brain Serotonergic Function

¹¹C‐DASB PET, a tracer of serotonergic terminal function, is used to evaluate brain serotonin transporter (SERT) availability, while early ¹²³I‐beta‐CIT reuptake in the brainstem is a reflection of raphe SERT binding 40.

Changes in serotonergic neurotransmission have been implicated in idiopathic depression. PET studies with the brain serotonin transporter ligand [¹¹C]‐DASB have demonstrated an 7% reduced amount of tracer reuptake in the dorsolateral prefrontal cortex, an area implicated in major depression 109. Additionally, because midbrain raphe 5‐HT1A receptors can regulate serotonin release, Doder et al. investigated if levels of depression and personality traits are correlated with 5‐HT1A receptor binding by ¹¹C‐WAY 100635 PET in patients with PD. PET examination was performed on 9 healthy volunteers, 12 patients with PD, and 12 equally severely disabled PD patients without dementia, either suffering from a major depressive episode or with a personal or family history of depression. They found that there was a 28% decrease in 5‐DEDGGHT1A binding in the median raphe of PD patients with depression or euthymic patients with PD, arguing against the loss of serotonergic function playing a crucial role in mood determination 110.

5‐HT1A receptors are also found in pyramidal neurons of limbic areas, and there is a significant reduction of 5‐HT1A binding in the left inferior temporal gyrus of patients with PD. Conversely, there is a relative decrease in the rate of 5‐HT1A binding in the anterior cingulate and superior/medial frontal gyrus of PD patients with depression. Individual levels of inferior frontal 5‐HT1A binding are closely correlated with depression severity in patients with PD 111, 112. These findings indicate that, in frontal and anterior cingulate areas, vulnerability to depression in PD is correlated with decreased postsynaptic 5‐HT1A receptor binding.

Recently, a severe reduction in striatal ¹⁸F‐dopa was found in patients with PD who presented with chronic fatigue syndrome 113. As ¹¹C‐WAY 100635 binds with similar affinity to serotonin 5‐HT1A receptors, and the magnitude of ¹¹C‐WAY100635 uptake reduction is not different between PD patients with and without depression, this reduction suggests that the loss of serotonergic neurons may not be related to the development of depression in patients with PD 114.

PET/SPECT Imaging of Brain Cholinergic and Noradrenergic Functions

¹¹C‐methyl‐4‐piperidinyl acetate (MP4A) and ¹¹C‐methyl‐4‐piperidinyl propionate (PMP) PET 40 have been commonly used to investigate the acetylcholine esterase (AChE) activity in PD patients with and without dementia. Bohnen et al. studied the AChE activities in PD patients with and without dementia and patients with Alzheimer's disease with ¹¹C‐PMP PET 115. They reported that PD patients with dementia (PDD) had a significant reduction in cortical AChE activity (Figure 2). Recently, more studies have investigated the mutual relationship between cholinergic and dopaminergic function and the occurrence of recurrent falls in patients with PD 116. In these studies, ¹¹C‐PMP PET was used to assess brain AChE activity in two groups of patients with PD based on Unified Parkinson's Disease Rating Scale (UPDRS) ratings 115, 116, 117. The results show that the thalamic AChE activity in PD patients with high UPDRS scores was significantly decreased compared to controls and PD patients with lower scores.

Figure 2.

¹⁸F‐dopa and ¹¹C‐NMPA PET images of dopamine storage and acetylcholine esterase activity in a normal control, PD, and PD dementia case. Note that the loss of cortical AchE activity in PD parallels the loss of striatal dopamine storage 40.

[¹¹C]RTI 32 is a marker used to evaluate the bioactivity of both dopamine and noradrenaline membrane transporters 118. Remy et al. used [¹¹C]RTI32 PET 86 to identify differences between depressed and nondepressed patients with PD. They studied 8 PD patients with a personal or family history of depression and 12 euthymic patients with PD were studied. [¹¹C]RTI 32 uptake in the dorsal striatum, which reflects nigrostriatal DAT binding, showed no difference in the depressed PD cohort compared to the nondepressed PD cases. However, [¹¹C]RTI 32 uptake was lower in the locus coeruleus and the thalamus, the noradrenergic areas, as well as in several other regions including the anterior cingulate cortex and the ventral striatum in the depressed cases 86. Another study revealed that [¹¹C]RTI 32 binding in most of these regions had a negative correlation with the severity of anxiety ratings in patients with PD, and [¹¹C]RTI 32 binding in the ventral striatum was inversely correlated with apathy. This finding suggests that both noradrenaline and limbic dopamine dysfunctions not only participate in the pathophysiological process of motor symptoms in patients with PD, but also play an important role in the pathogenesis of depression in patients with PD.

Conclusion

PD affects not only the nigrostriatal dopaminergic systems, but also many other systems in the brain. It is important for neurologists to recognize the NMS of PD. The correct identification of NMS in PD will allow an accurate diagnosis and ensure timely and efficient medical treatment. Indeed, timely recognition of NMS is essential for improving the quality of life in patients 119, 120, 121, although NMS are often neglected by medical practitioners. Dopaminergic dysfunctions can be identified in high‐risk individuals with a family history of PD using the PET and SPECT techniques to evaluate dopaminergic terminals. In addition, information regarding VMAT2 and DAT expression obtained through molecular imaging techniques is a simple and objective indicator for sensitive and specific evaluation of NMS, such as olfactory dysfunction, sleep disorders, and depression, in high‐risk patients (Table 1).

Table 1.

Major methods in the diagnosis of Premotor Stages in PD

Stages of Premotor PD	PET/SPECT Imaging Methods	Target of PET/SPECT Imaging Methods	Findings	References
1) The prediagnostic phase	18F‐FDG	Glucose metabolism	Exhibited a change amount of 18F‐FDG compared with the normal regions	122, 123
	15O‐H2O‐PET	Cerebral blood flow (rCBF)	rCBF in prediagnostic patients with PD were reduced in cortical (parietal and frontal) and subcortical (basal ganglia) areas	36
	DAT	Dopamine reuptake	Decline of DAT is faster in patients with mild parkinsonian signs	54, 55
2) The premotor phase
Hyposmia:	DAT	Dopamine reuptake	Reduced DAT binding in the olfactory bulb and especially the caruncula mammillaris	66, 67
	DA receptor	D2 dopamine receptor	Dopamine D2 receptors increase in the olfactory bulb in Prediagnostic PD	124
RBD	DAT	Dopamine reuptake	reduction of tracer Dopamine reuptake in the putamen	75, 76, 77
	DTBZ(VMAT‐2)	Vesicular monoamine transporter	REM sleep duration is inversely correlated with dorsal and lateral mesopontine FD uptake in patients with early PD	94
Depression	18F‐FDG	Glucose metabolism	An increased amount of 18F‐FDG in amygdala metabolism	25
	DAT	Dopamine reuptake	Reuptake of the dopamine ligand significantly reduced in the ventral striatum, thalamus, anterior cingulate cortex, and locus coeruleus	86
	5‐HT1A receptor	Brain serotonergic function	An insignificantly reduced amount of serotonin transporter tracer reuptake in the dorsolateral prefrontal cortex	125, 126, 127, 128
	11C‐RT132	AChE activities	An reduced amount of AChE activity in the locus coeruleus and the thalamus and also lower in the anterior cingulate cortex and the ventral striatum	86
Anxiety	18F‐FDG	Glucose metabolism	Hypometabolism of the caudate	25
	DAT	Dopamine reuptake	Dopamine ligand reduced in basal ganglia impairment, particularly involving the left hemisphere	59
Cognitive dysfunction	18F‐FDG	Glucose metabolism	Significant metabolic reductions in the frontal temporoparietal association cortex	29, 30
	DA receptor	Dopamine receptor	Dopamine D2 receptors reduce in the dentate gyrus	11
3) The preclinical phase	DAT	Dopamine reuptake	The relative reduction of dopamine ligand uptake in patients with early PD was greater in the putamen than in the caudate	129
	18F‐dopa	L‐Dopa metabolism	significant reduction of striatal 18F‐dopa uptake	11, 46
4) The prephysiological phase	No signs of disease can be detected by physiological probes

The function of nondopaminergic systems can also be examined by PET and SPECT in patients with PD. Dysfunctions of the serotonergic, noradrenergic, and cholinergic systems can be detected and are typically associated with nonmotor manifestations including hyposmia, RBD, depression, sleep disorders, and dementia. We acknowledge that the same symptoms such as depression may not be caused by the same mechanisms between premotor and motor disabled PD. Therefore, more studies on PET/SPECT are needed to explore the precise mechanisms in terms of the same symptoms such as depression occurring pre‐ and motor stages in PD. Above all, functional imaging techniques play a crucial role in monitoring and assessing the efficacy of neuroprotective and therapeutic agents.

Conflict of Interest

The authors declare no conflict of interest.