← Back

PD MED

Long-term Effectiveness of Adjuvant Treatment With Catechol-O-Methyltransferase or Monoamine Oxidase B Inhibitors Compared With Dopamine Agonists Among Patients With Parkinson Disease Uncontrolled by Levodopa Therapy

Year of Publication: 2022

Authors: Richard Gray, Smitaa Patel, Natalie Ives, ..., for the PD MED Collaborative Group

Journal: JAMA Neurology

Citation: Gray R, Patel S, Ives N, et al. Long-term Effectiveness of Adjuvant Treatment... JAMA Neurol. 2022;79(2):131-140. doi:10.1001/jamaneurol.2021.4736

Link: https://doi.org/10.1001/jamaneurol.2021.4736

PDF: https://jamanetwork.com/journals/jamaneu...04362.66983.pdf

Clinical Question

Among patients with PD and motor complications uncontrolled by levodopa, which adjuvant class (dopamine agonist, MAO-B inhibitor, or COMT inhibitor) offers better long-term quality of life and functional outcomes?

Bottom Line

MAO-B inhibitors were superior to COMT inhibitors in patient-rated mobility and quality of life. Dopamine agonists and MAO-B inhibitors showed similar effectiveness, suggesting that MAO-B inhibitors may be underused as adjuvant therapy.

        Major Points
        500 patients with PD and uncontrolled motor complications were randomized 1:1:1 to dopamine agonist, MAO-B, or COMT inhibitors — all active treatment (no placebo/control arm)
Primary outcome was PDQ-39 mobility domain; secondary outcomes included other PDQ-39 domains, EQ-5D-3L, dementia, death, and institutionalization
Dopamine agonist vs DRI (combined MAO-B + COMT): PDQ-39 mobility difference +2.4 (95% CI -1.3 to 6.0, P=0.20) — not significant
MAO-B vs COMT: PDQ-39 mobility +4.2 (95% CI 0.4-7.9, P=0.03); EQ-5D-3L utility +0.05 (95% CI 0.003-0.09, P=0.04)
Dementia, mortality, and institutionalization were nonsignificantly lower with MAO-B vs COMT (all P=0.07-0.18)

      

Design

Study Type: Pragmatic semifactorial (2×1) randomized clinical trial

Randomization: 1

Blinding: Open-label; patients and clinicians not blinded (all arms received active treatment)

Enrollment Period: February 23, 2001 to December 15, 2009

Follow-up Duration: Median 4.5 years (range 0–13.3 years)

Centers: 64

Countries: UK, Czech Republic, Russia

Sample Size: 500

Analysis: Intention-to-treat; mixed-effects repeated measures for continuous outcomes, log-rank for time-to-event, negative binomial regression for admissions

Inclusion Criteria

Idiopathic Parkinson disease (UK Brain Bank criteria)
Motor complications uncontrolled by levodopa
Need for adjuvant therapy with uncertainty about drug class
No dementia

Exclusion Criteria

Dementia
Inability to provide informed consent
Inability to complete questionnaires

Baseline Characteristics

Overall Cohort (n=500):

Age: 73.0 ± 8.2 years
Sex - Male: 314 (62.8%)
Sex - Female: 186 (37.2%)

PDQ-39 Mobility at Baseline (mean, SD):

Dopamine Agonist: 49.0 (29.3)
MAO-B Inhibitor: 50.1 (29.1)
COMT Inhibitor: 50.1 (28.3)

EQ-5D-3L Utility at Baseline (mean, SD):

Dopamine Agonist: 0.53 (0.28)
MAO-B Inhibitor: 0.54 (0.27)
COMT Inhibitor: 0.51 (0.28)

PDQ-39 Summary Index at Baseline (mean, SD):

Dopamine Agonist: 31.6 (16.7)
MAO-B Inhibitor: 29.7 (14.2)
COMT Inhibitor: 31.9 (15.0)

Note: Per-arm demographic breakdown not reported in the primary paper; groups reported as balanced within each randomization stratum

Arms

Field	Dopamine Agonist	MAO-B Inhibitor	COMT Inhibitor
Intervention	Open-label dopamine agonist added to levodopa (ropinirole 43.2%, pramipexole 35.0%, other 21.9% of intended initial agents; n=144 randomized)	Open-label MAO-B inhibitor added to levodopa (oral selegiline 51.6%, sublingual selegiline 13.8%, rasagiline 29.5%, unknown 5.1% of intended initial agents; n=146 randomized)	Open-label COMT inhibitor added to levodopa (entacapone 90.8%, entacapone-levodopa conjugate [Stalevo] 6.8%, other/unknown 2.4%; n=210 randomized)
Duration	Up to 13.3 years follow-up	Up to 13.3 years follow-up	Up to 13.3 years follow-up

Outcomes

Outcome	Type	Control	HR / OR / RR	P-value
PDQ-39 mobility domain score (0–100 scale, higher = worse) over median 4.5-year follow-up. Two prespecified comparisons.	Primary			DA vs DRI: 0.20; MAO-B vs COMT: 0.03
EQ-5D-3L utility score (MAO-B vs COMT)	Secondary			0.04
PDQ-39 Summary Index (MAO-B vs COMT)	Secondary			0.07
PDQ-39 ADL (MAO-B vs COMT)	Secondary			0.03
PDQ-39 Emotional Well-Being (MAO-B vs COMT)	Secondary			0.009
PDQ-39 Social Support (MAO-B vs COMT)	Secondary			0.01
Dementia onset (MAO-B vs COMT)	Secondary		RR 0.70 (95% CI 0.47-1.03)	0.07
Mortality (MAO-B vs COMT)	Secondary		RR 0.76 (95% CI 0.56-1.03)	0.07
Institutionalization (MAO-B vs COMT)	Secondary		RR 0.74 (95% CI 0.48-1.15)	0.18
AE-related treatment withdrawals (predominant reason for discontinuation in all arms)	Adverse	Dopamine Agonist: 66/144 withdrew for AEs — mainly psychiatric (mental problems e.g. psychosis or confusion) in 45 participants · MAO-B Inhibitor: 58/146 withdrew for AEs — mainly psychiatric (mental problems) in 24 participants · COMT Inhibitor (entacapone): 72/210 withdrew for AEs — mainly gastrointestinal (e.g. diarrhea) in 29 participants
Serious AEs	Adverse	9 unexpected serious AEs reported among 9 participants; none considered unexpected on review
Hospitalizations (HES data, England/Wales)	Adverse	70% overall hospitalization rate; nonelective admissions similar between dopamine agonist and DRI groups (mean 2.25 vs 2.34/patient, P=0.96); MAO-B group had nonsignificantly fewer nonelective admissions than COMT (mean 1.99 vs 2.29, P=0.18) and shorter total admission duration (24.3 vs 33.6 days, P=0.10)

Subgroup Analysis

MAO-B outperformed COMT in PDQ-39 mobility, ADL, emotional well-being, and social support domains. Dopamine agonist vs MAO-B (exploratory): no differences in PDQ-39 mobility (+0.2, P=0.93) or summary index (+0.1, P=0.94). Treatment efficacy did not vary by baseline age, PD duration, Hoehn and Yahr stage, or previous therapy. Older age (≥70) was associated with higher treatment discontinuation (61.4% vs 49.7%, P=0.02).

Criticisms

Open-label design introduces potential performance and reporting bias (though authors argue any a priori bias would have favored dopamine agonists, opposite to findings)
Suboptimal adherence — only ~50% still on randomized drug class at 5 years, which likely attenuated observed treatment differences
COMT group used almost exclusively entacapone (90.8%); results may not generalize to tolcapone or opicapone
Between-group differences for MAO-B vs COMT were of borderline significance (P=0.03-0.07 range) and rely on consistency across domains rather than a single highly significant result

Funding

Health Technology Assessment Programme of the UK National Institute for Health Research (project number 98/03/02); UK Department of Health through March 2012 (University of Birmingham Clinical Trials Unit, supporting the salaries of Mss Ives and Patel); UK Medical Research Council (Mr R. Gray); and Parkinson's UK (Dr McIntosh)

Based on: PD MED (JAMA Neurology, 2022)