SPECT Rivals PET in Head-to-Head Test of Parkinson’s Brain Scans

For millions of people worldwide living with tremor, stiffness, and slowness of movement, the question that matters most is deceptively simple: do they have a neurodegenerative parkinsonian syndrome such as Parkinson’s disease, or something else entirely? A new study published in the European Journal of Nuclear Medicine and Molecular Imaging offers one of the most direct answers yet, pitting three advanced brain imaging techniques against one another in the same group of patients whose diagnoses remained clinically uncertain. The results carry real weight for clinical practice, because Parkinson’s disease is the second most common neurodegenerative disorder and the neurological disease with the fastest growing prevalence and disability burden worldwide, yet clinical diagnosis alone remains notoriously error-prone even in experienced hands.

The research, led by Gerda Thomsen and Kristoffer Brendstrup-Brix of the Neurobiology Research Unit at Copenhagen University Hospital Rigshospitalet, together with colleagues including senior author Gitte M. Knudsen, enrolled 35 patients who had been referred for diagnostic dopamine transporter imaging between October 2021 and January 2024. Fifteen healthy controls were recruited for comparison. Every patient underwent two nuclear medicine scans and a magnetic resonance imaging session: single-photon emission computed tomography with the radiotracer [123I]FP-CIT, the long-standing gold standard known commercially as DaTscan, performed on a modern three-headed SPECT-CT camera equipped with high-resolution multi-pinhole collimators; positron emission tomography with the newer radioligand [18F]FE-PE2I; and a dedicated midbrain MRI protocol combining neuromelanin-sensitive gradient echo magnetization-transfer imaging with iron-sensitive STAGE sequences on a 3T Siemens Prisma scanner.

The biological logic behind all three approaches rests on the same target: the nigrostriatal pathway. Parkinson’s disease is characterized by widespread accumulation of intracellular alpha-synuclein aggregates and the progressive death of dopaminergic neurons in the substantia nigra, which denervates the caudate nucleus and putamen, the striatal structures that regulate movement. Dopamine transporter imaging measures the presynaptic dopaminergic terminals still surviving in the striatum, and reduced binding signals neurodegenerative parkinsonism. Neuromelanin MRI exploits the fact that the pigment neuromelanin, an iron chelator that accumulates in nigral dopaminergic neurons over a lifetime, brightens the substantia nigra on specially weighted sequences; as neurons die, that signal fades. Iron-sensitive techniques, including quantitative susceptibility mapping, attempt to capture the corresponding rise in paramagnetic iron deposition.

The head-to-head results were striking. Eighteen of the 35 patients were ultimately diagnosed with neurodegenerative parkinsonism after a median follow-up of 2.6 years, including 13 with Parkinson’s disease, three with Parkinson’s disease dementia or Lewy body dementia, and one each with progressive supranuclear palsy and a non-specific neurodegenerative basal ganglia disease. Dopamine transporter availability measured with SPECT and PET showed complete concordance across cases, with quantitative measures correlating strongly, Pearson coefficients ranging from 0.7 to 0.9 and reaching an almost perfect 0.9 for the ratio of putamen to caudate nucleus binding. Receiver operating characteristic analysis confirmed equally excellent diagnostic performance for both modalities, with areas under the curve between 0.84 and 0.92 for SPECT and 0.80 to 0.93 for PET across most regional measures.

Why does this equivalence matter? Conventional wisdom has held that PET, with its superior spatial resolution and shorter waiting times between tracer injection and scanning, would outperform SPECT for dopamine transporter imaging. But the Copenhagen team used the AnyScan Trio SC, a three-headed brain SPECT-CT camera whose multi-pinhole collimators deliver roughly 2.4 times the sensitivity of conventional systems at the edges of the field of view, with reconstructed voxel sizes of 1.8 cubic millimeters. The study suggests that this hardware leap has closed the gap, meaning that centers without on-site cyclotrons or radiochemistry facilities can achieve state-of-the-art diagnostic accuracy with SPECT, which is generally less costly and more widely accessible. PET retains an advantage where tracer logistics allow, and some prior literature hints at higher specificity, but the choice between modalities may now legitimately rest on local availability and cost.

Patients themselves offered a surprising verdict on the scanning experience. Twenty-seven participants completed satisfaction questionnaires covering waiting time from injection to scan, tracer injection, comfort during the procedure, and overall experience, each rated on a five-point Likert scale. SPECT requires a three-hour wait after injection and a 30-minute acquisition, whereas PET scanning begins just 30 minutes after a 200 MBq [18F]FE-PE2I injection and lasts 10 minutes. Yet no statistically significant differences emerged for waiting time, injection, or the scanning procedure itself, and overall satisfaction was actually higher for SPECT, with a moderate effect size of Cohen’s d of 0.59, likely reflecting the extra clinical attention patients received during the add-on examination performed on the same day as their MRI.

The MRI results told a more sobering story. Substantia nigra neuromelanin content, quantified by summing z-scores within semi-automatically delineated hyperintense regions and normalized to reference areas in the cerebral crus, was significantly reduced in patients with neurodegenerative parkinsonism compared with individuals who had no neurological disease, with a large effect size of Cohen’s d of -1.03. A trend toward lower neuromelanin with longer symptom duration hinted at progressive nigral loss. But at the level of the individual patient, neuromelanin MRI achieved only an intermediate area under the curve of 0.7, well short of the nuclear medicine measures. Midbrain iron content fared worse still: it did not differ between any diagnostic groups and performed essentially at chance, with an area under the curve of 0.55, offering no diagnostic information in this early-stage cohort.

The authors interpret these MRI findings through the lens of disease timing. Their patients had a median symptom duration of just 29 months at the time of imaging, and accumulating evidence indicates that striatal dopaminergic dysfunction precedes measurable nigral neuromelanin loss and iron accumulation in the spatiotemporal evolution of Parkinson’s disease. Meta-analyses of neuromelanin and iron MRI similarly report lower diagnostic accuracy in patients within the first five years of symptoms. The reproducibility of the MRI quantification itself was excellent, with repeated region-of-interest delineations correlating at 0.95 for neuromelanin and 0.97 for iron, suggesting the method is technically sound and that its modest performance reflects biology rather than measurement noise. Standardized acquisition protocols, automated segmentation, and refined quantification strategies, the authors argue, will be needed before nigral MRI can serve as a reliable standalone clinical biomarker.

The study has limitations typical of real-world imaging research. Final diagnoses rested on best clinical practice, with movement disorder specialists integrating examinations, dopamine transporter imaging, and structural scans over follow-up periods of 20 to 48 months, rather than on neurohistological confirmation, which remains the definitive gold standard for synucleinopathies. Consecutive, unselected recruitment produced a realistic but heterogeneous cohort, and the small number of patients with atypical parkinsonism limits specific conclusions for that subgroup. Nevertheless, the practical takeaway is clear and consequential: in early, clinically uncertain parkinsonism, dopamine transporter imaging with modern multi-pinhole SPECT and with PET are equivalent, radiation-free MRI biomarkers of neuromelanin and iron are not yet ready to replace them, and hospitals can confidently choose the nuclear medicine pathway that best fits their infrastructure without compromising diagnostic accuracy for their patients.

Subject of Research: Comparative diagnostic performance of dopamine transporter SPECT, PET, and neuromelanin and iron-sensitive MRI in clinically uncertain parkinsonian syndromes.

Article Title: Nigrostriatal imaging in patients with clinically uncertain Parkinsonian syndromes: a head-to-head comparison between SPECT, PET, and MRI

Article References: Thomsen, G., Brendstrup-Brix, K., Svarer, C., Lindberg, U., Pinborg, L. H., Kettless, K., Law, I., Hasselbalch, S. G., Biering-Sørensen, B., & Knudsen, G. M. (2026). Nigrostriatal imaging in patients with clinically uncertain Parkinsonian syndromes: a head-to-head comparison between SPECT, PET, and MRI. European Journal of Nuclear Medicine and Molecular Imaging. https://doi.org/10.1007/s00259-026-08184-8

Image Credits: AI Generated

DOI: 10.1007/s00259-026-08184-8

Keywords: Parkinson's disease, dopamine transporter imaging, SPECT, PET, MRI, neuromelanin, iron deposition, substantia nigra, [123I]FP-CIT, [18F]FE-PE2I, parkinsonism diagnosis, multi-pinhole collimators