← Back to Phalanx

Evidence

Early-Stage Parkinson’s Disease

A selection of the most compelling studies describing exercise — particularly high-intensity aerobic exercise — as a disease-modifying therapeutic in early-stage Parkinson’s Disease. Each summary below outlines the protocol, the adherence profile observed, and the motor-outcome findings reported.

Key Takeaways

  1. All five studies examining the effects of aerobic exercise engaged in for 6 months or longer by early-stage PD patients demonstrated statistically significant improvement or stabilization in the primary disease rating scale used in clinical trials to assess Parkinson’s Disease severity (MDS-UPDRS III). The one study that assessed PD-related brain-imaging biomarkers further found these improved or stabilized as well. This phenomenon was observed despite a meaningful percentage of patients meaningfully undershooting the prescribed exercise dose, non-adherent patients typically being included in the MDS-UPDRS III average for the aerobic-exercise group, and patients often failing to hit the intended average heart rate.

  2. A consistent dose-response relationship was seen between the amount of exercise and UPDRS-III improvement or stabilization and the intensity of exercise and UPDRS-III improvement or stabilization, within those studies where a dose-response relationship could be assessed.

  3. The greatest MDS-UPDRS III improvement was seen in the study (Yale) where patients engaged in the highest weekly volume of exercise, and a meaningful number of sessions had at least 20 minutes conducted at ≥ 80% of max heart rate.

  4. The Yale study found that the 12-month MDS-UPDRS III score was better than the score at 6 months, suggesting exercise’s benefits may continue to accrue over time.

Comparative Summary

Side-by-side: protocol, adherence, and outcome

All seven metrics for the five studies. Below desktop widths the table scrolls horizontally; metric labels stay pinned on the left.

MetricCYCLE II TrialSPARX TrialYale StudyCommunity Pedaling for PDPark-in-Shape Trial
Actual Minutes Completed (Avg.)

Exercise Arm: 92.4 minutes per week

High-Intensity Arm: 84 minutes per weekof work (avg. 2.8 sessions/week × 30 min)

Moderate-Intensity Arm: 96 minutes per weekof work (avg. 3.2 sessions/week × 30 min)

Notes: Patients could elect to do a 5–10 min warm-up and 5–10 min cool-down.

High-Intensity Cohort: 149 minutes of work per week(~3.33 classes/week, 86.5 average over 6 months; ~45 min of work per class — 30 min exercise circuit + 15 min boxing).

Exercise Cohort: 49%: ≥ 80 minutes of work (≥ 2 × 40 min); 46% 40–76 minutes of work.

Notes: Warm-up and cool-down not included above.

Exercise Arm: ~97 min per weekof cycling (median 2.9 sessions × 33:34 min, including warm-up and cool-down — pure work likely less).

Actual Heart Rate Achieved (Avg.)

Exercise Arm: 60–80% of max HR goal, actual is not reported in abstract.

High-Intensity Arm: 80.2% of max HR

Moderate-Intensity Arm: 65.9% of max HR

High-Intensity Arm: 59.4% of classes at > 80% max HR target, 85.6% of classes were performed at > 70% max HR target.

Notes: This data is artificially high — avg. heart rate was calculated based on the 20 minutes in the session which had the highest avg. heart rate.

Exercise Cohort: 69.3% of age-estimated maxHR, avg. cadence 74.9 ± 9.0 RPM (below the 80–90 RPM target).

Exercise Arm: 76.4% of max HR for those who adhered to the program (average UPDRS includes non-adherents).

Average Change in MDS-UPDRS III

(Lower UPDRS Score is better, negative change is improvement. UPDRS numbers are averages that include patients assigned to the exercise group and who did not adhere to the exercise program, or did less than prescribed.)

Aerobic Exercise Arm: +0.2 pts

Usual Care Controls: +3.9 pts

Notes: (p < 0.0001), 12-month follow-up. Author note: patients with cadence ≥ 75 RPM and higher power-to-weight ratio had greater improvements.

High-Intensity Arm: +0.3 pts(95% CI −1.7 to 2.3)

Moderate-Intensity Arm: +2.0 pts

Usual Care Controls: +3.2 pts

Notes: (p = 0.03 for HI vs. UC), (p = 0.11 MI vs. UC), 6-month follow-up.

All Patients at 6 months: −0.6 pts

6/10 Patients Continuing for 12 Months: −3.0 pts

Aerobic Exercise Cohort: −3.4 pts

Notes: (p = 0.001), 6-month follow-up.
Author note: Participants who attended ~74% or more of available PFP classes experienced the greatest improvement in MDS‑UPDRS III scores; of those who attended less than 74% of classes, cycling greater than or equal to 76 RPM led to improvement.

Aerobic Exercise: −0.5 pts (raw change from baseline)

Stretching Controls: +4.2 pts (raw change from baseline)

Notes: As the study was an RCT with two arms that differed in their characteristics, the authors sought to adjust for a number of those differing characteristics. After statistical adjustments were applied, the adjusted statistics were AE: +1.3 pts (SE 1.8) and SC: +5.6 pts (p = 0.002), 6-month follow-up.
Training Prescribed in the Study

135 min/week (3 × 45 min).

120 min/week of treadmill work in both arms (4 × 30 min, HI; 4 × 30 min, moderate).

135 minutes work per week: 3 Beat Parkinson’s Today HIIT classes/week for 6 months (work time: 2 × 15 min exercise / 30 sec rest + 15 min boxing = 45 min, plus 15 min warm-up + cool-down).

No fixed/week prescription — community program; ~65 classes offered over 6 months (~2.5/wk); per class: 40 min cycling + 5–10 min warm-up and 5–10 min cool-down.

90 minutes of work per week (3 × 30 min per week; with warm-up / cool-down for 15 minutes).

Heart Rate Prescribed

60–80% of age-predicted max HR (AAPMHR) at ≥ 75 RPM cadence.

80–85% of max HR (HI arm); 60–65% of max HR (moderate arm).

80% of max HR (high-intensity target, Beat Parkinson’s Today HIIT).

60–80% of age-predicted max HR (at 80–90 RPM cadence).

50–80% of HR reserve (Karvonen): lower bound ramps from 50% → 70% HRR over 6 mo; upper bound set at 80% HRR.

Change in Biomarkers (DAT, Neuromelanin)

Not measured.

Not measured.

Putamen DAT: +4.32% vs. −5.35% matched cohort (p = 0.004).

Substantia nigra DAT: +19.95% vs. −2.25% (p = 0.010).

Caudate DAT: +20.16% vs. −3.90% (p = 0.160, NS).

SNpc neuromelanin MRI contrast: +5.3% vs. −3.15% (p = 0.008).

Notes: Matched cohorts: Delva et al. (n = 27) for DAT; Xing et al. (n = 46) for neuromelanin.

Not measured.

Not measured.

% Non-Adherent Patients Still Included in Exercise Average

7% of aerobic arm non-adherent (overall adherence 93%).

~9% of HI arm (3 never started + 1 found it too burdensome ≈ 4/43) per Evidence page; not individually broken out in published abstract.

Not applicable (pilot, single-arm ITT); 4/10 did not continue to 12-month follow-up.

5% (2/41) attended < 1 class/week — per Evidence page (abstract reports avg. 65% attendance overall).

15% (10/65) did not complete the program; 22% (14/65) averaged < 2 sessions/week.

  1. Study 1: CYCLE II Trial: An Exercise Prescription to Slow the Progression of Parkinson’s Disease

    Authors & Journal
    Alberts et al. · Presented at MDS International Congress (2025)
    Institutions
    Cleveland Clinic · University of Utah
    Reference
    MDS Abstract

    This randomized controlled trial involved 256 individuals with mild-to-moderate PD assigned to either aerobic exercise — stationary cycling at ≥ 75 RPM at 60–80% of age-predicted maximum heart rate, three times per week for 45 minutes — or usual care.

    On average, patients assigned to aerobic exercise completed 92.4 minutes of exercise per week. Although this fell short of the 135 minutes prescribed and 7% of the aerobic-exercise group was considered non-adherent, the average MDS-UPDRS III score of the aerobic-exercise group effectively stabilized — an average increase of 0.2 points (clinically insignificant) over 12 months, compared to an average increase of 3.9 points over 12 months in the usual-care group. Notably, patients who maintained a cadence ≥ 75 RPM and increased their power-to-weight ratio (proxies for intensity) experienced the most benefit.

    The study authors concluded: “Long-term, high-intensity aerobic exercise with a focus on cadence significantly slowed motor sign progression in PD. These findings support the use of structured, rate-focused aerobic exercise as a viable disease-modifying intervention.”

  2. Study 2: SPARX Trial

    Authors & Journal
    Schenkman et al., JAMA Neurology (2018)
    Institutions
    University of Colorado Anschutz Medical Campus (lead) · Northwestern University · University of Pittsburgh · Rush University
    Reference
    PubMed

    This phase 2 multicenter randomized controlled trial enrolled 128 patients with de novo (no PD drug history) early-stage (Hoehn & Yahr Stages 1 & 2) Parkinson’s Disease. Participants were randomized to high-intensity treadmill exercise (4 days per week at 80–85% of maximum heart rate), moderate-intensity exercise (4 days per week at 60–65% of maximum heart rate), or usual-care control for 6 months. The high-intensity session included a warm-up, cool-down, and 30 minutes of work on the treadmill.

    On average, patients in the high-intensity group completed 2.8 sessions per week at an average HR of 80.2% of calculated max. That said, the average HR was determined from one monitored session per month in the last three months of the trial, so the HR achieved in unmonitored sessions may have been lower. The high-intensity group included 3 patients who did not start the exercise program and 1 who found exercise too burdensome.

    Despite the non-adherence and fewer-than-prescribed weekly sessions on average, the high-intensity group’s average Unified Parkinson’s Disease Rating Scale (UPDRS) motor score changed by only +0.3 points — equivalent to disease stabilization — while a number of patients saw outright improvement, three of them by approximately 10 points or more. This was a statistically significant difference from the +3.2-point worsening in the usual-care group (p = 0.03). The moderate-intensity group worsened by +2.0 points, which did not reach significance relative to usual care (p = 0.11). This Phase II futility trial paved the way for the ongoing Phase III SPARX3 trial.

  3. Study 3: Intense Exercise Increases Dopamine Transporter and Neuromelanin Concentrations in the Substantia Nigra in Parkinson’s Disease

    Authors & Journal
    de Laat et al., npj Parkinson’s Disease (2024)
    Institutions
    Yale University
    Reference
    PubMed

    This pilot study followed 10 patients diagnosed with PD within the last 4 years, all at Hoehn & Yahr Stage 1 or 2, who engaged in the “Beat Parkinson’s Today” high-intensity interval training program three times per week over the course of 6 months. In two-thirds of all Beat Parkinson’s Today classes, patients achieved an average heart rate > 80% of target heart rate.

    A brain PET to assess DAT concentrations in the caudate, putamen, and substantia nigra, and a brain MRI to assess neuromelanin in the substantia nigra pars compacta, were performed at the beginning and end of those 6 months. These brain regions are all heavily implicated in PD. The ten patients in this Yale cohort were compared to a similar PD cohort.

    The study found that the Yale cohort experienced a statistically significant increase in DAT concentrations in the putamen (4.23% increase vs. 5.35% decrease in the matched cohort, p = 0.004) and the substantia nigra (19.95% increase vs. 2.25% decrease in the matched cohort, p = 0.010). DAT concentrations also increased in the caudate, though this did not reach significance (20.16% increase vs. 3.90% decrease in the matched cohort, p = 0.160). The study also found that the Yale cohort experienced a statistically significant 5.3% increase in MRI contrast ratio (marker of neuromelanin) in the substantia nigra pars compacta relative to the 3.15% decrease reported in the matched PD cohort (p = 0.008).

    Finally, the study authors performed a motor function exam at baseline, 6 months, and 12 months on the six patients who engaged in the program for a full year. Their average motor score at 12 months was 23.7 ± 6.6 — a 3.0-point improvement from the baseline average of 26.7 ± 7.2 (lower score is better).

  4. Study 4: Community-Based High-Intensity Cycling Improves Disease Symptoms in Individuals with Parkinson’s Disease: A Six-Month Pragmatic Observational Study

    Authors & Journal
    Rosenfeldt et al. (incl. Alberts), Health & Social Care in the Community (2022)
    Institutions
    Cleveland Clinic
    Reference
    Wiley Online Library

    41 PD patients participating in a community-based Pedaling for Parkinson’s cycling program were followed for 6 months. The program entailed a 5–10-minute warm-up, 40 minutes of cycling at a target of 60–80% of age-predicted heart rate and 80–90 RPM, and a 5–10-minute cool-down.

    20 of the 41 participants (49%) attended an average of 2 or more classes per week, and 19 of 41 (46%) attended 1 to 1.9 sessions per week. The remaining 2 participants (5%) attended less than 1 class per week. The average percentage of age-estimated maximum heart rate for the cohort was 69.3%.

    On average, patients experienced a significant decrease (symptom improvement) in their MDS-UPDRS III score (37.2 ± 11.7 → 33.8 ± 11.7, p = 0.001) and significant improvement in their immediate recall score over the 6-month exercise period (42.3 ± 12.4 → 47.1 ± 12.7, p = 0.02). Participants who attended ~74% or more of available PFP classes experienced the greatest improvement in MDS-UPDRS III scores; among those who attended less than 74% of classes, cycling at ≥ 76 RPM led to improvement.

  5. Study 5: Park-in-Shape Trial

    Authors & Journal
    van der Kolk et al., The Lancet Neurology (2019)
    Institutions
    Radboud University Medical Center (Nijmegen, Netherlands)
    Reference
    PubMed

    65 early-stage PD patients (Hoehn & Yahr Stages 1 & 2) were randomized to an aerobic-exercise program of 3 sessions per week for 6 months, each consisting of 30 minutes of cycling flanked by 15 minutes of warm-up and cool-down. A further 65 patients were randomized to a stretching control arm.

    14 patients (22%) in the aerobic-exercise group averaged less than two sessions per week, with 10 of those patients (15% of the total) dropping out of the exercise program entirely. The study assessed each patient’s MDS-UPDRS III at baseline and at 6 months.

    Within the aerobic-exercise group, the median was 2.9 sessions per week, the median time per session was 33:34 minutes, and the average heart rate as a percentage of max was 76.4% for patients who completed the protocol. Only 16 patients (25%) averaged ≥ 3 exercise sessions per week. 35 patients (54%) averaged 2–3 exercise sessions per week.

    Despite the high rate of non-completion, the relatively low number of weekly sessions, and the relatively short session time, the aerobic-exercise group nonetheless averaged a +1.3-point change in MDS-UPDRS III score over six months (still below clinical significance), compared to the +5.6-point change in the control group (p = 0.002).