Longitudinal serum uric acid levels are not associated with dopamine transporter binding in progressive supranuclear palsy

Abstract

Progressive supranuclear palsy (PSP) causes rapid motor decline and severe dopaminergic dysfunction. While uric acid (UA) may act as a neuroprotective antioxidant in some neurodegenerative disorders (like Parkinson’s disease), its role in PSP remains unclear. This study evaluated the relationship between serum UA levels, measured cross-sectionally and longitudinally, and striatal dopamine transporter binding in PSP. A total of 33 PSP patients with repeated pre-[¹²³I]FP-CIT SPECT UA measurements, along with 30 healthy control individuals and 30 patients with Alzheimer’s disease (AD), were retrospectively analyzed. Group and sex effects were analyzed with t tests and ANOVA. Effects of mean UA and longitudinal UA trajectories on FP-CIT SPECT binding in PSP were modeled using linear mixed-effects models and regressed against binding in four regions (caudate and putamen), separated into more-affected and less-affected side for both sexes. A Bayesian two-stage measurement-error model provided sensitivity analysis. UA was significantly lower in PSP (4.98 mg/dl) and AD (4.69 mg/dl) compared to healthy controls (5.71 mg/dL; p = 0.001). Sex had a significant effect on UA (F(1, 89) = 9.38, p = 0.003, partial η² = 0.10), however, this effect was significant only in PSP (p< 0.001). Within PSP, UA–[¹²³I]FP-CIT SPECT correlations were weak and nonsignificant, and neither UA intercept nor slope predicted [¹²³I]FP-CIT SPECT binding (all p > 0.7). Bayesian estimates corroborated the absence of a credible relationship. In the present cohort, serum UA is reduced in PSP, primarily in females, but neither mean levels nor longitudinal changes are related to striatal [¹²³I]FP-CIT SPECT binding, suggesting no clear association with dopaminergic degeneration in PSP, without precluding a potential role of uric acid at other disease stages.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00702-026-03141-z.

Introduction

Progressive supranuclear palsy (PSP) is one of the most prevalent atypical Parkinsonian syndromes, characterized by early postural instability, supranuclear vertical gaze palsy, dementia and poor response to dopaminergic therapy (Litvan et al. 1997). Compared to idiopathic Parkinson’s disease (PD), PSP typically follows a more aggressive clinical course with rapid progression of motor and cognitive decline (McFarland 2016; Armstrong and McFarland 2019). These clinical differences are consistent with distinct underlying disease mechanisms, like different regional and cellular affection as well as aggregation and deposition of different proteins (Kovacs et al. 2020; Goedert et al. 2010).

Although the precise etiology of PSP is not understood, oxidative stress and neuroinflammation have been implicated as key contributors (Chirichigno et al. 2002). Uric acid (UA), the end product of purine metabolism, has attracted attention in this context due to its strong antioxidant properties, including scavenging of reactive nitrogen and oxygen species and stabilization of endogenous antioxidant systems (Squadrito et al. 2000). In PD, higher serum UA levels have been consistently associated with a reduced risk of disease onset and more favorable clinical outcomes, with several studies demonstrating correlations between serum UA levels and striatal dopamine transporter availability (Aerqin et al. 2022; Weisskopf et al. 2007; Zhang et al. 2024a; Jiang et al. 2024; de Lau et al. 2005; Chang et al. 2022; Moccia et al. 2015).

In contrast, the role of UA in PSP remains unclear. Some observational studies have suggested that higher serum UA levels may be inversely related to PSP risk (Oropesa-Ruiz et al. 2016; Schirinzi et al. 2017; Aerqin et al. 2022), whereas others have found no significant differences in serum and cerebrospinal fluid (CSF) UA levels between PSP patients and controls (Constantinescu et al. 2013; Sakuta et al. 2017; Brody et al. 2016). Nonetheless, a sex-specific reduction in serum UA levels has been observed in female PSP patients compared to male patients (Constantinescu et al. 2013). These discrepancies highlight the need for a more precise characterization of UA in PSP and its relationship with dopaminergic degeneration.

To date, no study has directly examined the association between serum UA levels and regional striatal dopamine transporter binding in PSP. In the present study, we examined this association by analyzing quantitative [¹²³I]FP-CIT SPECT imaging in a substantially sized PSP cohort.

Material and methods

Subjects

Patients with PSP (n = 33) included in this study were selected retrospectively at the Movement Disorder Unit, Department of Neurology, Medical University of Vienna, Austria, between January 2015 and March 2025. All patients with PSP underwent ¹²³I-FP-CIT SPECT imaging for diagnostic purposes at the Department of Biomedical Imaging and Image-Guided Therapy. PSP diagnoses were established according to the latest criteria of the Movement Disorder Society and classified as probable by an experienced movement disorder specialist (Hoglinger et al. 2017). As non-Parkinsonian controls, patients with Alzheimer’s dementia (AD, n = 30) were included. AD diagnoses were established according to the revised criteria for the diagnosis and staging of Alzheimer’s disease proposed by the Alzheimer’s Association Workgroup (Jack et al. 2024). Healthy controls (HC, n = 30) were retrospectively identified from general outpatient clinics among individuals presenting with non-neurological conditions (e.g., routine check-ups or minor somatic complaints). All included participants were systematically reviewed for medical comorbidities and medications known to influence serum uric acid levels (e.g., renal dysfunction, thyroid disease, malignancy, alcoholism, gout, or treatment with urate-lowering medication), and none were identified in any group.

For all included PSP patients, the last documented motor status and clinical information prior to ¹²³I-FP-CIT SPECT imaging were reviewed to categorize disease severity. Patients were classified as having mild disease when they were ambulatory without major postural instability (including PSP with oculomotor variant, PSP with Parkinsonism or early/mild PSP with Richardson syndrome) and as having severe disease when marked postural instability, frequent falls, or the need for walking aids or wheelchair use was documented (PSP with corticobasal syndrome, late PSP with Richardson syndrome). All available serum UA levels prior to ¹²³I-FP-CIT SPECT imaging were collected retrospectively from our internal patient data system and the available archived hospital database. We accepted values up to 30 days prior to ¹²³I-FP-CIT as associated with the investigation. None of the included PSP patients were receiving medications known to significantly affect dopamine transporter binding (e.g., bupropion, psychostimulants, or selective serotonin/norepinephrine reuptake inhibitors) at the time of ¹²³I-FP-CIT SPECT imaging, as verified through review of the available medical records and patient-reported documentation prior to ¹²³I-FP-CIT SPECT (Chahid et al. 2023). The study was approved by the local ethics committee (EK Nr. 1265/2020).

¹²³I-FP-CIT SPECT image acquisition, processing and image analysis

Three to 4 h prior to the imaging start ~ 185 MBq ¹²³I-FP-CIT was injected intravenously (specific activity > 185 MBq/nmol; radiochemical purity > 99%; produced as DaTSCAN according to GMP criteria at GE Healthcare, Eindhoven, The Netherlands). For SPECT imaging, a three-head gamma camera (IRIX 465, Philips Medical Systems, MN Divion) equipped with a fan beam, low-energy, and high-resolution collimators was used. Acquisition parameters, including the step-and-shoot mode for over 30 min, were used. Over 360° 60 projection angles were taken using a 128 × 128 matrix. Reconstruction was performed using filtered projections using a Shepp and Logan filter. The triple-energy window method for scatter compensation and a uniform Chang attenuation correction were applied to compensate for photon attenuation using a uniform attenuation coefficient of 0.15/cm. The reconstructed I-FP-CIT images were subsequently reoriented and normalized automatically using BRASSTM (HERMES Medical Solutions, Stockholm, Sweden). Following automatic alignment, all scans were inspected visually and manually realigned to fit the predefined template where necessary. The template volumes of interest (VOIs) were predefined using BRASS software, Version 4 (Booij et al. 2001; Oh et al. 2012). For count concentration estimation in striatal (S) and background occipital (O) VOIs. Using these activity concentrations in VOIs, striatal uptake ratios defined as (S–O)/O of specific tracer binding were calculated. Similar procedures were used to split the striata in the caudates and putamina. We defined the clinically more severely affected side based on the measured data and clinical observations. As PSP typically affects both hemispheres in a largely symmetric manner, the striatal side with lower ¹²³I-FP-CIT SPECT binding was considered to be the more severely affected side.

Statistics

All analyses were conducted using R (version 4.3.2; R Foundation for Statistical Computing, Vienna, Austria) within the RStudio environment (version 2023.09; Posit Software, Boston, MA). Data distributions were assessed using histograms and density plots. Categorical variables were summarized as frequencies and percentages, and continuous variables were reported as means with 95% confidence intervals or medians with interquartile ranges (IQR). The assumptions of normality and homogeneity of variance were verified using the Shapiro–Wilk and Levene’s tests, respectively. Bonferroni correction was applied to adjust for multiple comparisons.

Differences in serum UA levels between males and females within each diagnostic group were tested using Welch's independent-samples t-test, one- or two-way analysis of variance (ANOVA) with group and sex as fixed factors, including the interaction term. For this analysis, individual mean values were calculated within subjects using the arithmetic means of all valid observations. Pairwise comparisons of the estimated marginal means were performed using the emmeans package (Tukey adjustment for multiple testing). To complement significance testing, effect sizes were calculated and reported as partial eta squared (η²), reflecting the proportion of variance attributable to each factor. Associations between mean UA levels and striatal dopamine transporter availability in the more-affected and less-affected caudate and putamen regions, as measured by ¹²³I-FP-CIT SPECT were examined using Pearson’s correlations, interpreted as weak (r < 0.3), moderate (0.3 ≤ r ≤ 0.5), or strong (r > 0.5).

We adopted a two-stage modeling framework to first estimate individual longitudinal UA trajectories (Stage 1) and then relate these subject-specific parameters to striatal dopamine transporter availability measured by ¹²³I-FP-CIT SPECT (Stage 2), thereby accounting for both within-subject variability and measurement errors in the predictors.

UA trajectory modelling (stage 1)

We first characterized each participant’s longitudinal UA profile by fitting a linear mixed-effects model. Longitudinal UA profiles were modelled using linear mixed-effects models with random intercepts and slopes (lme4). For each participant, best linear unbiased predictors of the UA intercept and slope, and the standard error of the slope, were extracted (arm package). Sex differences in slopes were tested within diagnostic groups, and an additional linear model including a sex × group interaction was fitted.

Endpoint regression (stage 2)

To relate UA trajectories to striatal dopamine transporter availability, ¹²³I-FP-CIT SPECT–derived SUVs in the four striatal regions (more affected/less affected caudate and putamen) were regressed on individual UA intercept and slope estimates. Models were weighted by the inverse variance of the UA slope, so that participants with more precisely estimated slopes contributed more strongly to the analysis.

To validate our simple two-stage- regression, we implemented a Bayesian two-stage measurement error model in brms as a sensitivity analysis. To this end, we implemented a Bayesian linear mixed-effects model using the brms package in R (Bürkner 2017). Each participant’s standardized UA slope (derived from the Stage 1 longitudinal model) was entered as a predictor of regional DAT values (caudate more affected, caudate less affected, putamen more affected, and putamen less affected). Sex and region were included as fixed effects, and subject-specific intercepts were modeled as random effects to account for repeated measures across regions. All predictors were standardized before the analysis. The model used weakly informative priors (normal[0, 0.5] for fixed effects, Student-t[3, 0, 2.5] for the intercept, and exponential [1] for variance parameters) and was estimated using four Markov chains (4,000 iterations per chain; 1,000 warm-up). Convergence was assessed using the potential scale reduction statistic (R̂ < 1.01) and effective sample size diagnostics. Posterior summaries are reported as medians with 95% credible intervals (CrI). Practical equivalence was evaluated using a region of practical equivalence (ROPE) of ± 0.1 SD and the probability of direction (pd) index, as implemented in the bayestestR package.

Results

After excluding PSP patients with insufficient UA values (at least three documented values prior to ¹²³I-FP-CIT SPECT), 33 patients diagnosed with progressive supranuclear palsy (PSP) were identified, with the mean disease duration at the time of DAT imaging of 39 months (SD 25). Further, 30 healthy controls and 30 patients with Alzheimer’s disease (AD) with at least 3 UA values (without ¹²³I-FP-CIT SPECT) were selected. The demographic characteristics of the cohort are shown in Table 1.

Table 1.

The demographic characteristics and mean ¹²³I-FP-CIT SPECT values per region

Group	Total (n)	Mean age (y)	Median age (y)	Caudate more affected	Caudate less affected	Putamen more affected	Putamen less affected	Male (%)	Female (%)
Control AD	30 30	69.1 70.9	70.7 72.6	– –	– –	– –	– –	18 (60.0%) 14 (46.7%)	12 (40.0%) 16 (53.3%)
PSP	33	67.0	66.6	1.34 ± 0.65	1.63 ± 0.71	0.93 ± 0.57	1.24 ± 0.69	18 (54.5%)	15 (45.5%)

Values are presented as mean ± standard deviation unless otherwise stated. More affected and less affected sides denote the striatal hemispheres with lower and higher dopamine transporter binding, respectively. Striatal dopamine transporter values were not available in healthy controls and are therefore not shown (–). Abbreviations: PSP progressive supranuclear palsy, SPECT single photon emission computed tomography, DAT dopamine transporter, ¹²³I-FP-CIT ioflupane iodine-123

Serum UA differed significantly across diagnostic groups (F(2, 90) = 7.17, p = 0.001, η² = 0.14), indicating that diagnosis accounted for 14% of the variance in UA levels (Fig. 1a). Post hoc Tukey comparisons showed that controls had higher UA levels compared to PSP (p = 0.027) and AD patients (p = 0.002), whereas PSP and AD did not differ (p = 0.596). Within the PSP group, no significant difference in UA levels was observed between mild and severe phenotypes (mild: M = 4.91 mg/dL; severe: M = 5.06 mg/dL; Welch’s t(24.22) =  − 0.33, p = 0.74).

Fig. 1.

a Boxplots show the distribution of mean serum UA concentrations (mg/dL) between groups. Black horizontal lines inside the boxes indicate the group median, the boxes represent the interquartile range (IQR), and the whiskers show the data range, excluding outliers. The red horizontal lines denote the group means. **p < 0.002. b Sex-specific differences in serum uric acid levels between PSP, AD, and healthy controls. Mean serum UA levels (± SE) stratified by diagnostic group and sex. c Correlation matrix of serum UA and striatal DAT binding in PSP patients by sex. Pairwise Pearson correlations between mean serum UA levels and striatal DAT binding ratios in four brain regions, stratified by sex. Only the lower triangle is shown to avoid redundancy. In female patients, serum UA showed weak positive correlations with all striatal regions (r = 0.16–0.20), while striatal regions were highly intercorrelated (r = 0.96–0.99). In male patients, serum UA showed weak negative correlations with striatal regions (r = − 0.06 to – 0.14), with moderate to strong intercorrelations among brain regions (r = 0.75–0.94). Color intensity represents correlation strength (blue = negative, red = positive)

To account for potential sex effects, a two-way ANOVA including diagnosis and sex was performed (Fig. 1b). Diagnosis remained a significant predictor of UA levels (F(2, 89) = 7.83, p < 0.001, partial η² = 0.15), and sex also significantly influenced UA (F(1, 89) = 9.38, p = 0.003, partial η² = 0.10). The diagnosis × sex interaction was not significant (p = 0.093). Nevertheless, exploratory simple-effects contrasts suggested a significant sex difference within PSP (male–female difference = 1.23 mg/dL, p = 0.0009), whereas sex differences were not significant within controls (p = 0.845) or AD (p = 0.101) (Fig. 1b, Supplementary Fig. 1a).

Next, we investigated whether the mean patients’ serum UA levels were associated with FP-CIT SPECT binding to striatal dopamine transporters. The mean longitudinal follow-up period for available serum UA measurements was 8.6 years (SD 6.8). Within the PSP group, correlations between mean serum UA levels and FP-CIT SPECT–derived striatal dopamine transporter availability were examined separately for males and females across the four regions for each patient. In female patients with PSP, we found weak positive correlations with ¹²³I-FP-CIT SPECT –derived SUV (r = 0.16–0.20, all p > 0.48, Fig. 1C), whereas in male patients with PSP, correlations were weakly negative (r =  − 0.06 to − 0.15, all p > 0.54). None of these associations reached statistical significance, and Fisher’s r-to-z tests indicated no significant sex differences (all p > 0.37). Overall, these findings demonstrate that within the PSP group, serum UA levels were not significantly associated with striatal SUV in either sex (Fig. 1C). Although the mixed-effects model did not yield formally significant interactions, visual inspection of the fitted slopes suggested a possible sex-specific trend in the relationship between mean patients’ serum UA and ¹²³I-FP-CIT SPECT –derived SUV in severe PSP cases. Moreover, in mild PSP, the association between UA and SUV was similar for men and women. In severe PSP, however, female patients tended to show a positive association between UA and SUV, whereas male patients showed a negative trend (Supplementary Fig. 1b).

To further explore whether longitudinal changes (rather than single values) may affect the SUV, we investigated the longitudinal relationship between serum UA and SUV using a two-step random coefficient growth curve model. To this end, we regressed regional SUV values on each participant’s UA intercept and slope (derived from individual UA time-series models) while controlling for sex, disease duration and number of collected values. Across all four striatal regions (caudate and putamen, more affected and less affected sides), neither UA intercept nor slope significantly predicted SUV (all p > 0.7). The effects of sex were also not significant. The model fit indices indicated a minimal explained variance (adjusted R² ≈ 0).

To validate these findings and account for estimation uncertainty in Stage 1 modelling, we employed a Bayesian, two-stage measurement error model. The model converged well (all R̂ = 1.00, ESS > 4000). The posterior median estimate for the standardized UA slope effect was − 0.03 (95% CrI [− 0.41, 0.35]), and for the UA slope × sex interaction, − 0.02 (95% CrI [− 0.71, 0.68]). The main effect of sex was 0.01 (95% CrI [− 0.54, 0.57]). Region effects showed higher ¹²³I-FP-CIT SPECT binding values in the caudate than in the putamen (caudate_less_affected: β = 0.26 [0.21, 0.31]; putamen_more_affected: β =  − 0.33 [− 0.38, − 0.29]; putamen_less_affected: β =  − 0.04 [− 0.09, 0.01]). The residual SD of the model was 0.22 (95% CrI [0.21, 0.23]). The 95% credible intervals for both UA-related predictors overlapped with zero, and the slope effect, in particular, was characterized by substantial uncertainty in prediction. Crucially, the results of the measurement error–corrected model were fully concordant with the initial two-stage analysis: the diagnostic group (PSP) remained the only reliable predictor of striatal dopaminergic degeneration, whereas the UA intercept and slope effects were negligible and non-significant. This consistency reinforces the robustness of our findings regarding the first-stage estimation variability (Supplementary Table 1).

Discussion

In this study, we investigated the relationship between serum uric acid (UA) levels and striatal ¹²³I-FP-CIT SPECT–derived SUV across a cohort of patients with progressive supranuclear palsy (PSP). To our knowledge, this is the first study to integrate repeated UA sampling with dopaminergic imaging in a substantial PSP cohort. Our findings provide new insights into sex- and disease-specific interactions between UA and striatal dopamine transporter availability in patients with PSP. Firstly, we noted that mean serum UA levels were significantly lower in PSP than in healthy controls. This effect was driven almost entirely by female patients, who showed markedly reduced UA levels relative to male patients with PSP and female controls. Secondly, serum UA levels were not significantly associated with FP-CIT SPECT–derived striatal dopamine transporter availability in patients with PSP. Finally, longitudinal UA changes were not related to striatal dopamine transporter (DAT) availability.

Multiple meta-analyses and cohort studies have shown that serum uric acid levels are significantly lower in PD patients compared to healthy controls (Wen et al. 2017; Shen and Ji 2013; Annanmaki et al. 2007). Many of those studies associated lower UA levels with increased PD risk, faster disease progression, and greater severity, including both motor and non-motor symptoms (Wen et al. 2017; Danau et al. 2022; Koros et al. 2023; van Wamelen et al. 2020). Interestingly, in PD this reduction was evident in both sexes, with men showing a slightly greater reduction (− 0.66 vs. − 0.53 compared to controls) (Wen et al. 2017; Shen and Ji 2013). This contrasts our observations, in which female patients with PSP showed significantly lower UA levels than both controls and male patients with PSP. Given the non-significant sex-specific changes of UA in controls and AD, this pattern may suggest that sex-related biological factors differentially influence UA regulation across neurodegenerative disorders, with a potentially heightened relevance in PSP. Notably, although males with AD demonstrated numerically higher UA levels than females (mean difference ≈ 0.6 mg/dL, compared to ≈ 1.2 mg/dL in PSP), this difference did not reach statistical significance in AD. Thus, while a sex effect in AD cannot be excluded, it appeared less pronounced than in PSP. Overall, UA levels were significantly lower in AD then in PSP, indicating disease-specific patterns. Estrogen is known to promote uricosuria and is associated with lower UA levels in women, and sex-dependent variation in renal urate transporters such as SLC2A9 and ABCG2 further contributes to this dimorphism (Hak and Choi 2008; Brandstatter et al. 2008; Merriman 2015). Potential differences between PD and PSP, and AD may reflect distinct underlying pathogenetic mechanisms, including mainly presynaptic vs. postsynaptic dopaminergic degeneration and divergent proteinopathies (alpha-synucleinopathy vs. 4R-tauopathy vs. mixed 3R/4R-tauopathy with β-amyloid accumulation) (Kovacs 2019; Goedert et al. 2010). However, further studies are needed to clarify the mechanistic basis of these observations, particularly across broader groups of Parkinsonian syndromes.

In our analysis neither mean UA levels nor their temporal fluctuations (i.e., trajectories of change) exhibited any discernible association with striatal DAT availability in patients with PSP. Likewise, neither baseline UA nor its rate of change predicted FP-CIT SPECT–derived DAT measures in any striatal region. The models explained virtually no variance in DAT availability. This result was corroborated by the Bayesian measurement-error–corrected analysis, despite being more robust compared to frequentist statistics. Within the PSP cohort, UA levels did not differ between mild and severe phenotypes. Although these comparisons did not reach statistical significance, exploratory analyses in female patients hinted at stage-dependent divergence, with trends toward inverse associations in milder disease and positive relationships in more advanced stages.

In contrast, prior reports in PD have suggested that lower serum UA levels correlate with reduced striatal DAT availability (Baik et al. 2020) potentially reflecting more severe dopaminergic neuron degeneration (Kovacs et al. 2008). This association is observed across PD motor subtypes, with the lowest UA and DAT levels in patients with postural instability and gait disorder compared to tremor-dominant subtypes (Wen et al. 2017; Moccia et al. 2015). Further, some studies provided evidence for the association between UA and disease progression or severity, potentially with a stronger effect in men (Shen and Ji 2013; Lee et al. 2020, 2018). However, the absence of any association in PSP in our study suggests that peripheral UA dynamics may be insufficiently sensitive to central dopaminergic degeneration in this disorder. This interpretation aligns with findings from Hasimoglu et al., who likewise reported lack of relationship between longitudinal UA changes and dopaminergic or clinical progression in PD (Hasimoglu et al. 2020), and is further compatible with emerging genetic and longitudinal evidence suggesting that urate may function more as a peripheral correlate than as a mechanistically relevant neuroprotectant (Wang and Tang 2024; Fazlollahi et al. 2022; Zhang et al. 2024b). It should be borne in mind that the present study assessed serum UA levels exclusively, which may not accurately reflect CSF or CNS concentrations and thus may be of limited utility as a biomarker of central neurodegeneration. Peripheral UA alterations may instead be governed primarily by systemic and sex-dependent factors that transcend diagnostic categories, rather than indexing disease-specific neurobiological processes.

Evidence from other neurodegenerative conditions suggests stage-dependent variation in uric acid levels, as well as potential discrepancies between serum and CSF measurements (Maetzler et al. 2011). In Parkinson’s disease, elevated uric acid has been linked to earlier disease phases and, in certain cohorts, more favorable outcomes, whereas lower concentrations tend to predominate in advanced stages. Such patterns have been interpreted as indicative of a shift from antioxidant buffering toward oxidative stress–driven or secondary reactive mechanisms as neurodegeneration progresses (Ikeda et al. 2011). Nonetheless, these observations remain heterogeneous and underscore the need for longitudinal studies assessing uric acid dynamics in both serum and CSF.

Several limitations should be considered when interpreting our findings. First, we were unable to account for several confounding factors that may influence serum uric acid levels, such as dietary assessment, including daily alcohol consumption and BMI, which could have introduced some residual variability. Second, PSP diagnoses were based on established clinical criteria and confirmed by experienced movement disorder specialists but not neuropathologically confirmed. Third, as this was a single-center study using a specific ¹²³I-FP-CIT SPECT acquisition and analysis protocol, the generalizability of our imaging findings to other centers may be limited, despite the advantage of reduced methodological variability. Fourth, although clinically substantial, statistically rather moderate sample size within PSP subgroups—especially in sex-stratified analyses—may have limited our ability to detect subtle associations or interaction effects. Lastly, due to the retrospective design, standardized clinical severity scales such as the PSP Rating Scale were not systematically available at the time of serum uric acid sampling and [¹²³I]FP-CIT SPECT acquisition, limiting correlations with clinical disease severity.

In conclusion, our findings suggest that peripheral uric acid levels may provide only limited insight into central dopaminergic degeneration in PSP. Although UA was reduced—particularly in female patients—neither average concentrations nor longitudinal changes showed measurable associations with striatal DAT availability. Together with prior PD-based evidence, Mendelian randomization analyses and the absence of disease-specific effects across additional neurological comparison groups included in this study, these results may indicate that urate might rather reflect peripheral physiological processes than disease-specific neurodegeneration. While these observations do not rule out context-dependent, disease-stage related or sex-modulated roles of UA, they further highlight the need for multimodal biomarker approaches and for further investigation into the biological mechanisms underlying UA-related sex differences in PSP in larger cohorts.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

All authors contributed to the study conception and design. Data curation and dataset generation were performed by D.B. and T.A. Data acquisition was performed by C.B., E.B.-S., T.N., and T.T.-W. Imaging analysis and interpretation were performed by T.T.-W. and T.T. Statistical analysis was performed by I.M. and D.B. The first draft of the manuscript was written by D.B. and I.M., and all authors contributed to manuscript revision and approved the final version. I.M. and T.T.-W. contributed equally to this work.

Funding

Open access funding provided by Medical University of Vienna.

Data availability

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request due to ethical restrictions and data protection regulations.

Declarations

Conflict of interest

The authors declare that they have no competing interests.

Ethics approval

This study was approved by the local ethics committee (EK Nr. 1265/2020) and was conducted in accordance with the principles of the Declaration of Helsinki and its later amendments.

Consent to participate

This research was conducted retrospectively using data obtained for clinical purposes. In view of the retrospective nature of the study and the use of fully anonymized data, informed consent to participate was not required.

Consent to publish

Not applicable.