Randomized Trial Highlights Uncertainties of Ambroxol to Improve Cognition in Parkinson Disease Dementia

Stephen H. Pasternak, MD, PhD

Results from a recently published, 52-week, phase 2, double-blind, placebo-controlled study showed that ambroxol, a chaperone for β-glucocerebrosidase, was safe in patients with Parkinson disease dementia (PDD); however, its effects on cognition were not confirmed. Positive data revealed stabilization of neuropsychiatric symptoms and plasma glial fibrillary acidic protein (GFAP) levels, a marker of neurodegeneration.¹

After excluding the low-dose group due to recruitment challenges, the study featured a cohort of patients with PDD who were randomly assigned 1:1 to either high-dose ambroxol at 1050 mg (n = 22) or placebo (n = 24) for a 52-week period, with an additional 26-week open-label extension. All told, change in Alzheimer Disease Assessment Scale–cognitive subscale 13 (ADAS-Cog13) and Clinician Global Impression of Change (CGIC), the study’s primary outcomes, were not different between the groups at week 26 or week 52.

In the study, investigators observed no statistical between-group differences on secondary outcomes as well, which included Clinical Dementia Rating Scale, Parkinson’s Disease Cognitive Rating Scale, Trail Making Test, and Stroop test, among many others. The study authors did observe a trend in total Neuropsychiatry Inventory (NPI) score (t = –1.86; P = .07), where the placebo group worsened from baseline to week 52 while those in the ambroxol group remained stable. NPI score among GBA1 carriers revealed that 2 patients taking placebo changed by 0 and +10 points, whereas patients taking high-dose ambroxol showed score reductions of –1, –8, –12, and –15 points at week 52.

Led by senior author Stephen H. Pasternak, MD, PhD, the study aimed at testing the safety and tolerability of ambroxol, as well as its effects in slowing the progression of cognitive deficits in PDD. Ambroxol, an over-the-counter expectorant with documented safety profile, has been identified as a high throughput screen as a pharmacological chaperone of GCase and an inhibitor that binds and increases levels of GCase, while dissociating from GCase in the acidic environment of the lysosome. Prior to this study, ambroxol has shown an ability to increase GCase and reduce alpha-synuclein in mice and nonhuman primate brains.

Ambroxol was shown to be safe in patients with PDD, as those on the drug had a similar rate of adverse events (AEs) than those in the placebo group. Overall, 8 patients on ambroxol (2 low-dose; 6 high-dose) and 3 patients taking placebo (12.5%) withdrew because of AEs. There were 7 hospitalizations, considered a serious AE, found in the ambroxol group. These included one patient on low-dose admitted twice for delirium, and other participants admitted to urinary tract infection (low-dose), worsening hallucinations and recurrent falls (high dose), infection after elective surgery (low dose), rigidity and worsening hallucinations (high dose), and severe leg weakness (high dose).

READ MORE: FDA Approves Updated Label for Alzheimer Therapy Donanemab to Lower ARIA-E Risk

For gastrointestinal disorders, 12% of ambroxol-patients reported these vs 5% of those taking placebo. In addition, there were more psychiatric AEs in the placebo group (23%) compared with the ambroxol group (16.6%), as well as more injuries/falls (29.7% vs 22.8%).

The plasma biomarker analysis comprised 15 patients on high dose ambroxol and 20 on placebo, all who had baseline and week 52 samples. When excluding those who had a clinically significant AE or were not taking treatment within 1 month of the end of trial (6 taking ambroxol and 3 taking placebo excluded), the placebo group demonstrated an increase in GFAP from baseline over the 52 weeks (mean: 135.07 pg/mL to 167.90 pg/mL) whereas those on ambroxol high dose did not (mean: 127.95 pg/mL to 118.05 pg/mL).

At week 26, GCase levels were significantly higher in the ambroxol group vs placebo (12.45 ± 1.97 vs 8.50 ± 1.96 nmol/h/mg; 91% CI, 11.54–13.36 vs 7.65–9.34; P = .05), and remained elevated but not statistically different at week 52 (11.45 ± 1.59 vs 8.59 ± 1.56 nmol/h/mg; 95% CI, 10.61–12.28 vs 7.86–9.31). Considerable inter-patient variability was observed, including at baseline.

"In conclusion, results of this randomized clinical trial reveal that ambroxol was deemed safe and well tolerated in PDD. Sufficient drug levels were achieved, and target engagement was demonstrated based on increased white blood cell GCase activity levels,” Pasternak et al wrote. “Future studies should attempt to reduce the heterogeneity in study population, focus on recruiting more GBA1 carriers, and assess GFAP as a potential biomarker to be used in PD clinical trials."

REFERENCE
1. Silveira CRA, Coleman KKL, Borron K, et al. Ambroxol as a treatment for Parkinson disease dementia: a randomized clinical trial. JAMA Neurol. Published online June 30, 2025. doi:10.1001/jamaneurol.2025.1687