📄 medicine

Time-to-Event Analyses of UPDRS and MDS-UPDRS Motor Examination and Functional Outcomes in Early Parkinson’s Disease Clinical Trials

This study analyzes five early Parkinson's disease trials to demonstrate that time-to-event analyses using large worsening thresholds on motor and functional scales are feasible for detecting meaningful treatment benefits within 12-month clinical studies before treatment initiation.

Original authors: Andrew McGarry, Christopher R Meyer, Karl Kieburtz, Jordan Dubow, Milton H Werner, Charles Venuto

Published 2026-06-28

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Andrew McGarry, Christopher R Meyer, Karl Kieburtz, Jordan Dubow, Milton H Werner, Charles Venuto

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to prove that a new shield works against a slow-moving storm. The storm is Parkinson's disease, and the people you are testing are just starting to feel the first few drops of rain.

The problem is that these people are still doing quite well. They can walk, talk, and function almost normally. Because they are doing so well, it's hard to tell if your new shield is actually stopping the rain or if the storm just hasn't gotten bad enough yet to show the difference. Plus, as soon as the rain gets heavy enough, doctors will naturally give them standard umbrellas (symptomatic medication), which muddies the water and makes it hard to see if your shield was the hero.

This paper is like a group of researchers looking at old weather reports from five different studies to see if they can find a better way to measure the storm before the umbrellas come out.

The Old Way vs. The New Idea

The Old Way (Fixed Time): Imagine checking the weather only once, exactly one year from now. You ask, "How much rain has fallen?" The problem is, some people might have gotten soaked and started using umbrellas halfway through the year, while others stayed dry. It's messy to compare them at the end of the year.

The New Idea (Time-to-Event): Instead of waiting for a specific date, the researchers suggest watching the clock. They ask: "How long does it take for the rain to get really heavy?" They define "really heavy" as a specific, noticeable change in how the person moves (the "exam score") or how they function in daily life (the "functional score").

What They Did

The team looked at data from five different studies involving people with early Parkinson's who hadn't started taking strong medication yet. They played a game of "thresholds":

The "Exam" Score (Part III): This is like a doctor checking how stiff or shaky a person's hands are. They looked for specific jumps in the score, like a 4-point, 5-point, 6-point, or even a 7-point worsening.
The "Function" Score (Part II): This is like asking the person, "How hard is it for you to button your shirt or walk to the store?" They looked for similar jumps in difficulty.

They tracked how many people hit these "worsening" marks within 12 months, and crucially, they did this before those people started taking their standard medication.

What They Found

1. Big changes happen, even without treatment.
Even though these people had early Parkinson's, a significant number of them (between 32% and 42%) showed a large jump in their exam scores (7 points or more) within a year, even without taking new drugs. This is good news for researchers because it means there are enough "events" to measure. If the changes were tiny and rare, you'd need to test thousands of people to see a difference. But because these "big jumps" happen often enough, you can design a study with a manageable number of people.

2. The "Exam" and the "Function" are linked.
The researchers wanted to know: "If someone's exam score gets worse by 7 points, does their daily life actually get harder?"
They found that yes, there is a strong connection. When a person's exam score got significantly worse, a large chunk of them also reported that their daily life had become noticeably harder.

The Analogy: Think of it like a car engine. If the engine starts making a loud, terrible noise (the exam score), the car usually starts driving poorly (the function score) at the same time. The study found that for the biggest engine noises, the car was almost always driving poorly too. This suggests that if a drug stops the engine noise, it's likely also helping the car drive better.

3. The "Umbrella" problem is manageable.
In these studies, some people did start taking medication because they felt they needed it. The researchers treated this as part of the "event." It's like saying, "The event happened when the rain got heavy enough to force you to open an umbrella." By counting this as a "worsening event," they could still measure how long it took for the storm to get bad enough to require help, without losing data because the person left the study.

The Bottom Line

The paper concludes that using this "Time-to-Event" method is a smart way to test new drugs for early Parkinson's.

It works: You can find enough people who get noticeably worse within a year to run a proper study.
It matters: When the physical symptoms get significantly worse, the person's daily life usually gets worse too. This means if a drug can delay that "big jump" in symptoms, it's likely doing something meaningful for the patient.
The Plan: The authors suggest that future studies could be designed to last about 12 months, looking specifically for these larger, meaningful changes (like a 6 or 7-point jump) rather than tiny, hard-to-measure shifts. This could help prove that a drug works before patients even need to start taking standard symptom-fighting medication.

In short, they found a way to measure the "storm" of Parkinson's more accurately by waiting for the "big splash" rather than just counting the drizzle, and they confirmed that when the splash is big, the person definitely feels the wetness.

Technical Summary: Time-to-Event Analyses of UPDRS and MDS-UPDRS Motor Examination and Functional Outcomes in Early Parkinson's Disease Clinical Trials

Problem Statement
Clinical trials for early Parkinson's Disease (PD) face a fundamental design challenge: while enrolling patients before the initiation of symptomatic treatment offers a biological window to test disease-modifying therapies, the mild phenotype and slow progression of early PD make demonstrating clinically meaningful benefit difficult. Traditional fixed-timepoint analyses of mean group score changes are confounded by the inevitable initiation of symptomatic treatment (levodopa, dopamine agonists, etc.) during the study period, which acts as an intercurrent event. Furthermore, there is no consensus on the magnitude of change in examination scores (e.g., UPDRS Part III) that constitutes a clinically meaningful treatment effect, nor is it guaranteed that such exam changes translate to functional improvements perceived by patients. The authors posit that Time-to-Event (TTE) outcomes, which define a specific worsening threshold as the primary endpoint, may offer a more robust approach to detect benefit before treatment initiation begins, balancing the sensitivity of exam-based metrics with the clinical relevance of functional decline.

Methodology
The study conducted a secondary analysis of data from five early PD cohorts: four randomized, double-blind clinical trials (SURE-PD3, STEADY-PD III, FS-1, FS-TOO) and one observational study (PPMI). All participants had recent PD diagnoses and were not yet requiring symptomatic dopaminergic therapy at baseline.

Endpoints: The analysis utilized TTE methodology to measure the time from baseline to the earliest occurrence of:
1. Motor Worsening: Defined as a specific increase from baseline in clinician-rated UPDRS or MDS-UPDRS Part III (Motor Examination). Thresholds evaluated were $\ge$ 4, $\ge$ 5, $\ge$ 6, and $\ge$ 7 points.
2. Functional Worsening: Defined as increases in clinician-rated UPDRS Part II (ADL) or patient-reported MDS-UPDRS Part II (M-EDL). Thresholds evaluated were $\ge$ 2, $\ge$ 3, and $\ge$ 4 points.
3. Treatment Initiation: The start of symptomatic therapy (levodopa, agonists, amantadine, anticholinergics, or MAO-B inhibitors).
Composite Event: The primary event of interest was the earliest date of either motor worsening or treatment initiation.
Analysis: Kaplan-Meier methods estimated event-free survival. The authors specifically examined "co-attainment," determining the proportion of participants who crossed a specific Part III motor threshold who had also crossed a corresponding Part II functional threshold by the same time point.
Sample Size: Calculations were performed using the Schoenfeld formula to estimate the sample sizes required for Phase 2/3 trials to detect a 30% reduction in hazard (Hazard Ratio = 0.70) with 80% and 90% power, assuming 1:1 randomization and varying dropout rates (5–20%).

Key Results

Event Frequencies: Across all cohorts, the proportion of placebo participants reaching the composite endpoint increased as worsening thresholds were lowered. At the largest motor thresholds ( $\ge$ 7 points), 32%–42% of participants achieved untreated worsening. At the $\ge$ 4 point threshold, this rose to higher percentages, though the FS-1 and FS-TOO cohorts showed lower event rates primarily due to fewer treatment initiations.
Motor-Functional Correlation: A substantial proportion of participants who crossed Part III motor thresholds also experienced concordant Part II functional worsening. Higher motor thresholds were generally associated with greater functional worsening. For the most severe combination (Part III $\ge$ 7 and Part II $\ge$ 4), the proportion of motor attainees who also met the functional threshold ranged from 20% to 50% across cohorts. Even at the 7-point motor threshold, more than half of participants in most cohorts declined by at least 2 functional points.
Treatment Initiation vs. Worsening: Motor worsening accounted for the majority of events, while treatment initiation contributed a smaller share that declined as motor-worsening thresholds became less stringent.
Baseline Medication: The use of allowable symptomatic medications at baseline (e.g., MAO-B inhibitors in SURE-PD3, amantadine/anticholinergics in STEADY-PD III) did not significantly alter the time to reaching worsening thresholds or treatment initiation.
Active vs. Placebo: No statistically significant differences were observed between active and placebo arms in any of the clinical trials for Part III or Part II composite endpoints, consistent with the known futility or lack of efficacy of the tested agents (inosine, isradipine, creatine, minocycline, GPI-1485, CoQ10) in these settings.

Significance and Claims
The paper argues that TTE analyses focusing on larger magnitude worsening thresholds (specifically 6- or 7-point changes in Part III) are feasible and potentially meaningful for early PD trial design. The authors claim that:

Feasibility: A substantial percentage of untreated early PD patients experience large magnitude motor worsening (6–7 points) within 12 months, suggesting that studies powered for these thresholds are practical.
Clinical Meaningfulness: There is a demonstrable intersection between "how much" exam change occurs and "so what" functional change results. The data suggests that forestalling larger magnitude motor worsening (6–7 points) is likely to concurrently delay functional disability, providing a rationale for using these thresholds as primary outcomes to support clinical meaningfulness.
Design Implications: TTE designs allow for the specification of study periods long enough to observe sufficient events while minimizing the confounding impact of treatment initiation. The authors suggest that a 12-month study design powered for these larger thresholds could successfully demonstrate benefit on both phenotype and function.
Sample Size Viability: The sample size calculations indicate that Phase 2 or 3 studies using these TTE outcomes are plausible to conduct, provided dropout rates are managed.

The authors conclude that while TTE outcomes do not replace the need for fixed-timepoint analyses, they offer a promising framework to navigate the challenges of early PD trials, specifically by aligning exam-based deterioration with functional decline in a way that may be more acceptable to regulators and patients.

The Old Way vs. The New Idea

What They Did

What They Found

The Bottom Line

Technical Summary: Time-to-Event Analyses of UPDRS and MDS-UPDRS Motor Examination and Functional Outcomes in Early Parkinson's Disease Clinical Trials

More like this