A new study found that dementia in patients with Parkinson's disease may occur less frequently or develop over a longer period than previously reported.

Researchers analyzed data from two large prospective studies: the Parkinson's Progression Markers Initiative (PPMI) and a University of Pennsylvania (Penn) cohort. The PPMI cohort included 417 de novo, untreated participants with Parkinson's disease (PD; mean age = 61.6 years, 65% male), while the Penn cohort was comprised of 389 participants with established PD (mean age = 69.3 years, 67% male).

Key Findings

The study, published in Neurology, found that in the PPMI cohort, 8.5% of participants (34/401) were diagnosed with dementia over the entire follow-up period. The estimated probability of dementia at 10 years disease duration was 9% (site investigator diagnosis), 15% (Montreal Cognitive Assessment [MoCA] score <21), or 12% (Movement Disorder Society—Unified Parkinson's Disease Rating Scale [MDS-UPDRS] Part I cognition score ≥3). In the Penn cohort, 47.3% of participants (184/389) were eventually diagnosed with dementia; in this group, the median time to dementia was 15 years (95% confidence interval [CI] = 13-15 years), and the estimated probability of dementia was 27% at 10 years, 50% at 15 years, and 74% at 20 years.

Combined results showed an estimated risk of dementia by disease duration of 3% to 12% at year 5, 9% to 27% at year 10, 50% at year 15, 74% at year 20, and 90% from year 25 onward.

Methods

The PPMI study enrolled participants from 24 sites internationally between 2010 and 2013. Inclusion criteria for PD participants were:

• Age ≥ 30 years
• Untreated disease and within 2 years of diagnosis
• Hoehn and Yahr stage <3
• At least two of the following symptoms: resting tremor, bradykinesia, or rigidity (must have either resting tremor or bradykinesia); or a single asymmetric resting tremor or asymmetric bradykinesia
• Dopamine transporter imaging demonstrating dopaminergic deficit.

The Penn cohort recruited participants from a tertiary movement disorders center between 2006 and 2021. Participants underwent annual or biennial cognitive assessments.

Dementia diagnoses were made by site investigators in PPMI and through expert consensus in the Penn cohort. Survival curves were fitted for time from PD diagnosis to stable dementia diagnosis.

Additional Findings

Age at disease diagnosis (Penn cohort): <56 years: 36/89 (40.4%) diagnosed with dementia, median time = 19.4 years (95% CI = 19.4-23.7 years); 56-70 years: 95/213 (44.6%) diagnosed with dementia, median time = 14.6 years (95% CI = 13.4-15.2 years); 70 years: 53/87 (60.9%) diagnosed with dementia, median time = 9.2 years (95% CI = 6.7-11.6 years).

Sex differences (Penn cohort): females: 49/128 (38.3%) diagnosed with dementia, median time = 19.4 years (95% CI = 16.1-19.4 years); males: 135/261 (51.7%) diagnosed with dementia, median time = 13.3 years (95% CI = 13.3-14.6 years).

Education level (Penn cohort): <13 years: 30/51 (58.8%) diagnosed with dementia, median time = 11.6 years (95% CI = 6.7-18.0 years); ≥13 years: 154/338 (45.6%) diagnosed with dementia, median time = 15.2 years (95% CI = 14.6-16.1 years).

Comparison of PD participants and healthy controls (PPMI cohort); for participants with baseline MoCA ≥27:

• PD: 19/266 (7.1%) diagnosed with dementia by investigator
• Healthy controls (HC): 2/180 (1.1%) diagnosed with dementia by investigator
• PD: 18/280 (6.4%) reached MoCA <21
• HC: 4/192 (2.1%) reached MoCA <21
• PD: 28/280 (10.0%) reached MDS-UPDRS item 1.1 score ≥3
• HC: 1/192 (0.5%) reached MDS-UPDRS item 1.1 score ≥3.

Worst-case sensitivity analysis (PPMI cohort):

• 145/401 (36%) participants withdrew before receiving a dementia diagnosis or study completion
• Under worst-case assumption: 165/401 (41%) would be diagnosed with dementia within 10 years of PD diagnosis
• Median time to dementia diagnosis under worst-case scenario: 11.4 years (95% CI = 10.3-12.8 years).

Baseline Characteristics of Participants

PPMI cohort (n=417):

• Mean education: 15.6 years (standard deviation [SD] = 3.0)
• Mean MDS-UPDRS III score: 20.9 (SD = 8.9)
• Median MoCA score: 28
• 397 (95%) had sporadic PD, 13 (3%) had PD with GBA1 pathogenic sequence variants, 6 (1%) had PD with LRRK2 G2019S variation, and 1 (0.2%) had PD with GBA1 + LRRK2 mutation.

Penn cohort (n=389):

• Mean education: 16.0 years (SD = 2.5)
• Mean MDS-UPDRS III score: 28.2 (SD = 13.2)
• Median MoCA score: 26
• Median levodopa equivalent daily dose: 600 mg.

Cognitive assessment tools used in the PPMI cohort were the MoCA, Hopkins Verbal Learning Test-Revised, Benton Judgment of Line Orientation, Symbol-Digit Modalities Test, Letter-Number Sequencing, and category (animal) fluency; in the Penn cohort, they were MoCA, Dementia Rating Scale-2, and eight other domain-specific cognitive tests.

The study's strengths included its large sample sizes, current data, serial assessments, and comprehensive cognitive evaluations; limitations included potential selection bias, high education levels of participants, and predominantly White cohorts.

The study authors concluded, "Results from two large, prospective studies suggest that dementia in PD occurs less frequently, or later in the disease course, than previous research studies have reported."

The study disclosed that financial support was received from the Michael J. Fox Foundation for Parkinson’s Research and various pharmaceutical companies, including Abbvie, Biogen, and Eli Lilly. Several authors also reported consulting fees and research support from organizations such as the NIH and industry sponsors. Full disclosures can be found in the published study.