Urate, a major antioxidant as well as the end product of purine metabolism in humans, has emerged as a biomarker of PD at multiple stages of the disease process.17
Specifically, prior to diagnosis lower serum urate indicates an increased risk of the disease becoming clinically manifest years later.4–7
Shortly after symptoms have developed, urate can serve as a prognostic biomarker with lower serum or CSF levels foretelling a greater likelihood of faster disability progression,18, 19
or of faster loss of dopamine transporter binding sites in striatum15
. The current findings demonstrate the potential diagnostic relevance of serum urate early in clinically manifest disease. They do so by identifying amongst those with mild parkinsonism a significant correlation between lower serum urate and reduced striatal uptake of a dopamine transporter ligand in a pattern indicative of idiopathic PD because of its characteristic deficit of nigrostriatal dopaminergic innervation. Conversely, higher -- but still normal – levels of urate were linked to SWEDDs and thus to a lesser likelihood of having true PD.
The correlation between urate and SWEDD status was robust in men but unclear in women. This apparent gender difference parallels similarly strong links between urate and PD risk or its progression18, 19
in men more than women. The gender difference in urate correlations may represent a true biological difference between men and women. Alternatively, the lack of significance for women might simply reflect the much smaller (here 6-fold compared to men) contribution of women to the highest two urate quintiles because of naturally lower levels of urate in women as well as the smaller number of women enrolling in PRECEPT.
Although higher serum urate amongst subjects recently diagnosed with ‘PD’ may be indicative of an alternative diagnosis, urate as biomarker of PD is not sufficiently specific on its own to distinguish PD and non-PD diagnoses. Nevertheless it may be combined with other diagnostic biomarkers to form a more useful composite indicator of non-PD diagnoses in early parkinsonism. In addition to SWEDD status, preserved olfaction has recently been substantiated as a potentially practical indicator of a non-PD diagnosis amongst newly diagnosed parkinsonian patients. It has been proposed that olfactory test performance, which has also been associated with SWEDDs in early parkinsonisim,20
in combination with distinctive PD motor features (e.g., asymmetric rest tremor21
) may serve as a more accessible and less expensive diagnostic biomarker compared to this functional neuroimaging technique. Thus it will be conceptually informative and possibly clinically useful to combine multiple readily available, simple tests – e.g., of serum urate and olfactory function – to determine how the diagnostic value (i.e., of higher urate and preserved olfaction) compares to that with either test result alone. Similarly, a crude composite index of suspicion for a non-PD diagnosis may aid in triaging subjects for more incisive diagnostic assessment with DAT neuroimaging.22
Amongst subjects diagnosed with early PD in the PRECEPT18
trials, those with higher levels of serum urate were found to decline clinically and radiographically at a slower rate than those with lower urate levels. The current findings suggest that at least some of urate’s capacity as a prognostic biomarker amongst those newly diagnosed with PD can be explained by the higher proportion of SWEDDs amongst ‘early PD’ subjects with higher urate levels. Because SWEDD subjects show little evidence of clinical and radiographic progression for at least 2–4 years from baseline,10, 11, 20, 23, 24
their greater representation amongst those with higher urate likely contributed to the relatively favorable prognosis observed amongst those with a PD diagnosis and higher urate levels.18, 19
However, the urate-SWEDD link could only partially account for the predictive association between higher urate at baseline and slower clinical progression because the association remains strong even after excluding all 90 subjects with a baseline SWEDD from a secondary analysis of the PRECEPT cohort18
(unpublished observations, AA, KM, A. Watts, SE, IS, MAS). Specifically, the HR for reaching the endpoint still declined with increasing serum urate in analyses including men and women combined (p= 0.002 for trend) with the an HR of 0.61 for those in the highest versus the lowest serum urate quintile (95% CI: 0.43 to 0.86; p=0.005). As before,18
the HR reduction was even greater and more significant in men and was absent in women. Similarly, the reported18
predictive association between higher baseline urate and slower decline in striatal β -CIT uptake persisted after excluding those with a baseline SWEDD from analysis, and though diminished it remained statistically significant amongst men.
Accuracy in the diagnosis of PD early in the course of parkinsonian conditions is of considerable importance for clinical research as well as for prognostic counseling.25
Although experience and specialization in movement disorders neurology appears to improve the positive predictive value of an idiopathic PD diagnosis to nearly 99% based on definitive pathological assessment,26
this high level of probability was achieved only after following patients who presented with parkinsonism for five years on average before diagnostic revision. Diagnostic accuracy is naturally lower early in the course of parkinsonian conditions when symptoms are fewer and subtler and when responsiveness to dopaminergic drug therapy – an important diagnostic indicator in PD – has not yet been assessed. However, this early so-called de novo
clinical phase of the disease has traditionally been the focus of therapeutic trials investigating potentially neuroprotective agents. Thus, it is not surprising that such clinical trials enroll a significant percentage of parkinsonian subjects who ultimately receive a diagnosis other than PD, for example, with 7% of the 800 subjects in the DATATOP trial who were initially thought to have PD upon enrollment being later deemed no longer likely to have the disease.14
In several more recent trials also investigating potentially disease-modifying agents in de novo
PD a similar proportion (~10%) of subjects have been identified as possibly not having PD, though in these studies the evidence was available at baseline when functional brain scans unexpectedly showed normal labeling of dopaminergic innervation in the striatum, i.e., SWEDDs. The demonstration that higher urate is a predictor of SWEDD status suggests that urate may be useful as a component of a composite diagnostic indicator early in the course of parkinsonian conditions. The findings broaden the diagnostic biomarker potential of urate in PD, and lend further support to the possibility of a role for urate in the pathophysiology of the disease and its future treatment.