Non-motor abnormalities, including those related to autonomic failure, may precede motor signs and symptoms of Parkinson’s disease (PD) by years (Goldstein et al., 2002; Koller, 1992). Among the non-motor manifestations of PD, dementia, loss of sense of smell (anosmia), REM sleep behavior disorder, and orthostatic hypotension (OH) have all been associated with cardiac sympathetic noradrenergic denervation (Goldstein et al., 2010; Kim et al., 2017; Miyamoto et al., 2006). Cardiac sympathetic denervation in PD as well as in other forms of synucleinopathies (i.e., dementia with Lewy bodies and pure autonomic failure) is explained by neurodegeneration of catecholaminergic post-ganglionic neurons resulting from deficient vesicular storage of catecholamines and accumulation of toxic byproducts, known as the “catecholaldehyde hypothesis” (Goldstein et al., 2011, 2014).

Cardiac sympathetic denervation has been histologically confirmed by decreased tyrosine hydroxylase immunoreactivity in epicardial nerves in patients with PD and Lewy body disease (Orimo, 2002; Orimo et al., 2005). The staging concept introduced by Braak et al. (2004) suggests that degeneration of the autonomic nervous system in PD precedes dopaminergic cell degeneration at the level of the substantia nigra, and thus cardiac sympathetic denervation may be a prodromal biomarker of PD (Goldstein et al., 2007; Oka et al., 2006, 2011; Orimo et al., 2005). Others have shown that cardiac sympathetic denervation in PD occurs independently of the movement disorder and progresses over time. Thus it may precede the motor symptoms by years or appear as a later finding (Goldstein et al., 2011; Orimo et al., 2008). Cardiac sympathetic denervation in PD can be visualized by means of sympathetic neuroimaging with investigational positron emission tomography (PET) agents, such as 6-[18F] fluorodopamine (Goldstein and Holmes, 1997) and [11C]-hydroxyephedrine (Raffel et al., 2006), as well as with the routinely clinically used single-photon emission computed tomography (SPECT) agent [123I]-metaiodobenzylguanidine ([123I]-MIBG) (Iwasa et al., 1998; Sudmeyer et al., 2011; Uyama et al., 2017; Yoshii et al., 2017).

Most patients with multiple system atrophy (MSA) have normal findings on neuroimaging studies of cardiac sympathetic innervation due to a preganglionic lesion rather than postganglionic involvement (Braune, 2001; Druschky et al., 2000; Orimo et al., 2001). As such, cardiac sympathetic neuroimaging can help differentiate between PD and MSA. This is especially useful in the case of PD with OH, which almost invariably demonstrates sympathetic cardiac denervation (Goldstein et al., 2008, 2015). It has also been shown that cardiac sympathetic denervation may predict the development of PD in individuals with multiple risk factors for PD years before motor symptoms have developed (Goldstein et al., 2018). Other applications of imaging cardiac sympathetic denervation include assisting in differentiation between dementia with Lewy bodies from Alzheimer’s disease in patients without Parkinsonism, imaging diabetic autonomic neuropathy, as well as evaluating the risk of sudden cardiac death in patients with heart failure (Jacobson et al., 2010; Manabe et al., 2017).

L-3,4-dihydroxy-6-[18F] fluoro-phenylalanine ([18F] FDOPA) is a positron emitter and a true analog of L-dihydroxyphenylalanine (L-dopa), the immediate precursor of dopamine and norepinephrine (Nanni et al., 2007). L-dopa is carried into the brain by the large neutral amino acid transport system, converted into dopamine by aromatic L-amino acid decarboxylase, and then stored in intraneuronal vesicles, from which it is released when the nerve cell fires (Garnett et al., 1983; Lovenberg et al., 1962).

[18F] FDOPA PET is an established imaging modality for the assessment of striatal dopaminergic integrity in neurodegenerative disorders, such as PD. The present study aimed to assess whether [18F] FDOPA can be used to evaluate cardiac sympathetic denervation for differentiating between patients with and those without PD by quantifying cardiac [18F] FDOPA uptake.

Seventy-six patients were included in this study. Fifty-two patients were diagnosed with PD, and 24 subjects did not meet the UK Brain Bank criteria for PD (Hughes et al., 1992). Patients’ characteristics are summarized in Table 1. The PD patients were younger than the non-PD patients (P = 0.014), with median ages of 64 and 69 years, respectively. The other evaluated characteristics were not significantly different between the two groups of patients. [18F] FDOPA PET scans of the brain identified characteristic reduced presynaptic dopaminergic activity in all PD patients as opposed to preserved dopaminergic activity in the non-PD group. The mean [18F] FDOPA LV-SUV${}_{mean}$ was lower for the PD patients compared to the non-PD patients (1.08 $\pm$ 0.21 vs. 1.24 $\pm$ 0.32, P = 0.015). In the multivariate regression, only IHD and LV-SUV${}_{mean}$ remained in the last step of the regression, and a lower LV-SUV${}_{mean}$ was independently associated with PD (OR${}_{adj}$ = 0.033, 95% CI 0.002-0.484, $P$ = 0.013). Overall, the LV-SUV${}_{mean}$ was able to discriminate between PD and non-PD patients (AUC = 0.641, P = 0.049).

Examples of the brain and cardiac [18F] FDOPA PET scans of patients diagnosed with and without PD are depicted in Fig. 2.

Fig. 2.

[18F] FDOPA PET scans of two patients who presented with tremor and referred for differentiation between essential tremor and PD. (A & C) Transaxial brain [18F] FDOPA PET. (B & D) Polar view of cardiac [18F] FDOPA PET generated by Carimas${}^{TM}$ software. (A & B) A 69-year-old patient with normal [18F] FDOPA uptake in the basal ganglia, compatible with essential tremor. The cardiac LV-SUV${}_{mean}$ was 1.49. (C & D) A 38-year-old patient with markedly decreased [18F] FDOPA uptake in the left striatum compatible with the dominant right-sided clinical manifestations and consequently diagnosed as PD. The cardiac LV-SUV${}_{mean}$ was 0.7.

This proof-of-concept study demonstrates the feasibility of utilizing quantitative [18F] FDOPA PET scans of the heart for discriminating between patients with PD from patients without PD. To date, the gamma-emitting tracer [123I]-MIBG is the only tracer in current clinical practice for noninvasive demonstration of cardiac denervation (Dhawan et al., 2002; Sudmeyer et al., 2011; Uyama et al., 2017; Yoshii et al., 2017). However, [123I]-MIBG is a false analog of norepinephrine that differs from true catecholamines. Therefore it is not as efficiently removed by sympathetic nerves and does not undergo intracellular metabolism by monoamine oxidase or by catechol-O-methyltransferase (Balogova et al., 2013; Neurauter et al., 2008). The main limitation of [123I]-MIBG is that it is not a positron emitter, but is instead imaged employing a gamma camera which has inherent limitations, such as reduced spatial resolution, increased radiation dose, and longer acquisition, compared with PET imaging (Goldstein, 2013).

6-[18F]-fluorodopamine, is an investigational positron emitter that has been effectively used for visualizing cardiac sympathetic denervation (Goldstein and Holmes, 1997). However, it is not widely available, and, unlike [18F] FDOPA, it does not pass the blood-brain-barrier. It, therefore, cannot be used for imaging striatal dopaminergic uptake. [18F] FDOPA, an analog of L-dopa, the immediate precursor of dopamine, is a PET tracer used in nuclear departments for the evaluation of striatal dopaminergic integrity as part of the diagnostic work-up for PD. The ability to use this tracer for both striatal and cardiac imaging is appealing since it enables a “one-stop-shop” imaging procedure for patients being evaluated for PD.

In the present trial, quantitative analysis of [18F] FDOPA uptake in the left ventricle myocardium had only a weak ability to differentiate between patients with PD and those without. One possible explanation is the heterogeneity of the comparison group. Another explanation for this could be the lack of clinical data available on symptoms related to autonomic failure in the study cohort. Previous studies have shown that about 30-40% of PD patients have OH (Merola et al., 2018; Velseboer et al., 2011) and that imaging cardiac sympathetic denervation is best for differentiating patients with PD and OH from patients with MSA (Goldstein et al., 2008). In contrast, cardiac sympathetic denervation in PD patients without OH is variable (Li et al., 2002). Thus, the current study cohort probably consisted of PD patients, both with and without OH, showing varying degrees of cardiac denervation.

Nevertheless, these results show promise that cardiac [18F] FDOPA uptake may be able to aid in the workup of patients with suspected PD, and allow differentiation between those with idiopathic PD and non-neurodegenerative parkinsonism. Another limitation is that a group of patients with an established clinical diagnosis of MSA was not available as part of the study population. One would expect such patients to generally have normal values of cardiac [18F] FDOPA uptake, and that those values would be higher than those measured in PD patients. Such an analysis should be pursued in future studies to verify whether [18F] FDOPA can genuinely serve as a clinically relevant biomarker for cardiac sympathetic denervation that could differentiate between PD and MSA patients.

In conclusion, this proof-of-concept study demonstrates the feasibility of differentiating between patients with PD and without PD by quantifying cardiac [18F] FDOPA uptake. This finding should be verified in more extensive prospective trials.

AUC: Area under the receiver operating characteristic curve; ET: Essential tremor; L-dopa: L-dihydroxyphenylalanine; LV-SUV${}_{mean}$ : Left ventricular mean standardized uptake value; MSA: Multiple system atrophy; OH: Orthostatic hypotension; PD: Parkinson’s disease; PET: Positron emission tomography; PET/CT: Positron emission tomography/computed-tomography; SPECT: Single photon emission computed tomography; [123I]-MIBG: [123I]-metaiodobenzylguanidine; [18F] FDOPA: L-3,4-dihydroxy-6-[18F] fluoro-phenylalanine.

JK contributed to the conception of the study, data collection, and analysis, study design, statistical analysis, writing and revising the manuscript; AL contributed to data collection and analysis and revising the manuscript; MK, TS, and NG contributed to the study design, data analysis, writing and revising the manuscript; NF contributed to data collection and analysis, study design, statistical analysis and revising the manuscript; HL and EES contributed to the conception of the study, data collection, study design, writing and revising the manuscript; All authors read and approved the final manuscript.

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

This study was approved by the Tel-Aviv Sourasky Medical Center IRB (approval number 0606-14-TLV). All subjects signed informed consent.

We thank our dedicated technicians for their contribution to this work.

The authors declare no conflict of interest.