BACKGROUND: Converging lines of evidence suggest that microglia are relevant to Parkinson's disease pathogenesis, justifying exploration of therapeutic agents thought to attenuate pathogenic microglial function. We sought to test the safety and efficacy of NLY01-a brain-penetrant, pegylated, longer-lasting version of exenatide (a glucagon-like peptide-1 receptor agonist) that is believed to be anti-inflammatory via reduction of microglia activation-in Parkinson's disease. METHODS: We report a 36-week, randomised, double-blind, placebo-controlled study of NLY01 in participants with early untreated Parkinson's disease conducted at 58 movement disorder clinics in the USA. Participants meeting UK Brain Bank or Movement Disorder Society research criteria for Parkinson's disease were randomly allocated (1:1:1) to one of two active treatment groups (2·5 mg or 5·0 mg NLY01) or matching placebo, based on a central computer-generated randomisation scheme using permuted block randomisation with varying block sizes. All participants, investigators, coordinators, study staff, and sponsor personnel were masked to treatment assignments throughout the study. The primary efficacy endpoint for the primary analysis population (defined as all randomly assigned participants who received at least one dose of study drug) was change from baseline to week 36 in the sum of Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS) parts II and III. Safety was assessed in the safety population (all randomly allocated participants who received at least one dose of the study drug) with documentation of adverse events, vital signs, electrocardiograms, clinical laboratory assessments, physical examination, and scales for suicidality, sleepiness, impulsivity, and depression. This trial is complete and registered at ClinicalTrials.gov, NCT04154072. FINDINGS: The study took place between Jan 28, 2020, and Feb 16, 2023. 447 individuals were screened, of whom 255 eligible participants were randomly assigned (85 to each study group). One patient assigned to placebo did not receive study treatment and was not included in the primary analysis. At 36 weeks, 2·5 mg and 5·0 mg NLY01 did not differ from placebo with respect to change in sum scores on MDS-UPDRS parts II and III: difference versus placebo -0·39 (95% CI -2·96 to 2·18; p=0·77) for 2·5 mg and 0·36 (-2·28 to 3·00; p=0·79) for 5·0 mg. Treatment-emergent adverse events were similar across groups (reported in 71 [84%] of 85 patients on 2·5 mg NLY01, 79 [93%] of 85 on 5·0 mg, and 73 [87%] of 84 on placebo), with gastrointestinal disorders the most commonly observed class in active groups (52 [61%] for 2·5 mg, 64 [75%] for 5·0 mg, and 30 [36%] for placebo) and nausea the most common event overall (33 [39%] for 2·5 mg, 49 [58%] for 5·0 mg, and 16 [19%] for placebo). No deaths occurred during the study. INTERPRETATION: NLY01 at 2·5 and 5·0 mg was not associated with any improvement in Parkinson's disease motor or non-motor features compared with placebo. A subgroup analysis raised the possibility of motor benefit in younger participants. Further study is needed to determine whether these exploratory observations are replicable. FUNDING: D&D Pharmatech-Neuraly.