Predicting progression in patients with Parkinson disease (PD) continues to be an elusive clinical goal. Overlapping etiologies with a variety of other Parkinsonian disorders such as progressive supranuclear palsy and multiple system atrophy, which have more rapid progression patterns, makes initial diagnosis more challenging. In addition, the tools to accurately measure rates of disease progression, including peripheral biomarkers, have not yet been identified.
According to David K. Simon, MD, PhD, director of the Parkinson’s Disease & Movement Disorders Center at Beth Israel Deaconess Medical Center in Boston, Massachusetts, numerous factors likely contribute to the rate of progression, with different combinations of factors showing relevance for individual patients. Dr Simon told Neurology Advisor that the possibilities include genetic factors (including numerous common variants that each play a small role in risk, sometimes through complex interactions with environmental factors), physical exercise, diet, exposure to pesticides, perhaps exposure to certain medications (eg, isradipine) or metabolic factors such as whether or not a person has diabetes or metabolic syndrome, or levels of uric acid. “But an important related problem is that we don’t know for sure if these factors truly do have any impact on rate of progression. And if they do, that relationship to progression is likely to be complex with interactions between different mechanisms,” he said.
Known Predictive Features
Age has consistently been predictive of PD progression, Dr Simon said, with younger-onset patients tending to progress more slowly compared with patients with onset at later ages. Other factors that show limited predictive value include PD subtype, with tremor-predominant patients tending to progress more slowly than akinetic-rigid patients. “But this too is limited, with some patients shifting categories,” he said. “Blood uric acid levels inversely correlate with the rate of progression in some studies. Presence of prominent cognitive difficulties early in the course is associated with a poorer prognosis, although in that case it may reflect in part limitations on our ability to treat cognitive deficits rather than the rate of loss of neurons. Men tend to progress a bit more quickly than women, and patients with metabolic syndrome on average progress more quickly than those who don’t have metabolic syndrome. But all of these factors and others are limited in their predictive value,” Dr Simon explained.
Jeanne C. Tourelle, DSc, director of Precision Medicine at GNS Healthcare and a research assistant professor at the Boston University School of Public Health in Massachusetts, explained that early and late PD have different progression rates. “I think it is very important to keep this in mind,” she told Neurology Advisor. “The rate of progression is not linear over the full course of the disease, and while the factors that influence progression may be similar, the effects are likely to change.”
Dr Simon, who is also a professor of neurology at Harvard Medical School in Boston, has contributed to several investigations into different potential biomarkers for disease progression. “Traditional biomarker research has focused on clinically defining subtypes of PD patients, and then looking for biomarkers that correlate with the clinically defined subtypes,” he said, discussing a recent study he coauthored on biomarker-driven phenotyping in PD.1 “Instead, here we’re proposing that we should define subtypes of PD patients based on pathophysiologically relevant biomarkers that will be more likely to predict responsiveness to therapies targeting those pathophysiological mechanisms, as compared to treatments aimed at clinically defined subtypes.”
He cited the recent SURE-PD3 phase 3 clinical trial by Paganoni and Schwarzschild,2 in which inosine was used to increase uric acid as a potential disease-modifying therapy. “This means that we are selecting a subset of patients based on a biomarker (uric acid levels) of relevance to the putative mechanism of a potential neuroprotective agent: those with the greatest chance of benefitting from inosine,” he said.
Biomarkers for disease progression have not been identified, and other methods have yielded inconsistent results. “Many groups have looked at [cerebrospinal fluid] and blood-based biomarkers to look for differences between PD patients and controls, and to look for markers that change over time as a marker of progression, but there is considerable ‘noise’ for any of the biomarkers reported thus far, and more studies are needed,” Dr Simon said.
Some studies have indicated a utility to imaging DaTscans,3 but as Dr Simon pointed out, “they do not have a high degree of accuracy in measuring progression.”
In a study of a novel model,4 Dr Latourelle and colleagues applied a Bayesian multivariate predictive inference platform to known predictors of PD progression, including age, sex, genetic features, and Movement Disorder Society-Unified Parkinson Disease Rating Scale (MDS-UPDRS) scores, using data from 2 large cohorts: the Parkinson’s Progression Markers Initiative for the first evaluation, and the Longitudinal and Biomarkers Study in Parkinson’s Disease study for replication.
“The goal is to develop an individual prediction of the rate of progression for each patient,” she explained. “The ensemble approach allows us to combine predictions from many models, any of which may be more or less appropriate to a given patient profile. Taking the median over all of these predictions allows us to identify individual predictions for each individual that are robust to the heterogeneity in the patient population.”
The value of individual factors in the study varied. In particular, DaTscans were not found to be informative to predicting the rate of progression, whereas cerebrospinal fluid-derived biomarkers, and specifically those for alpha-synuclien, contributed to the accuracy of the predictions. Women in both cohorts also demonstrated a slower rate of progression compared with men.
Genetic markers were more predictive of who would be affected by PD. “LRRK2 is definitely an important genetic factor for risk prediction, meaning the likelihood of being affected with PD. We also saw and replicated an association between LINGO2 and increased rate of progression, which is a slightly different question. We do replicate this association, but further investigation in additional cohorts would be warranted [to confirm] this relationship,” Dr Latourelle said.
Dr Simon pointed to other measures under development, such as an MRI-based “free water” imaging modality, with early data suggesting good potential as a biomarker of progression.5
Dr Latourelle observed that, “For a very simple model that doesn’t involve invasive tests (like genotyping and [cerebrospinal fluid] draws)…we are already able to make broad predictions, which may be useful at a population level for trials or study design, but these [methods] really would not be accurate enough to affect individual-level care. Once the technology catches up to allow quick and easy genetic, genomic, and other molecular tests to be done routinely, then I think progression will be able to be modeled much more accurately.”
- Espay AJ, Schwarzschild MA, Tanner CM, et al. Biomarker-driven phenotyping in Parkinson’s disease: A translational missing link in disease-modifying clinical trials. J Parkinsons Dis. 2015;5:681-697.
- Paganoni S, Schwarzschild MA. Urate as a marker of risk and progression of neurodegenerative disease. Neurotherapeutics. 2017;14:148-153.
- Sierra M, Martínez-Rodríguez I, Sánchez-Juan P, et al. Prospective clinical and DaT-SPECT imaging in premotor LRRK2 G2019S-associated Parkinson disease. Neurology. 2017;89:439-444.
- Latourelle JC, Beste MT, Hadzi TC, et al. Large-scale identification of clinical and genetic predictors of motor progression in patients with newly diagnosed Parkinson’s disease: a longitudinal cohort study and validation. Lancet Neurol. 2017;16(11):908-916.
- Burciu RG, Ofori E, Archer DB, et al. Progression marker of Parkinson’s disease: a 4-year multi-site imaging study. Brain. 2017;140:2183-2192.