Recent Developments in the Drug Treatment of Parkinson’s Disease

Abstract

Parkinson’s disease (PD) is a common neurodegenerative disorder caused by a spread of misfolded α-synuclein and ascending neuronal degeneration mainly in dopaminergic neurons, leading to progressive nigrostriatal dysfunction and disruption of other pathways. It is clinically defined by its cardinal motor features of bradykinesia, rest tremor and rigidity, which are usually accompanied by a variety of non-motor symptoms. Here we provide a review on recent developments in the pharmacological treatment of PD with a focus on recently approved drugs, new modes of delivery and agents that are in advanced stages of clinical development. Pharmacological dopamine substitution remains the mainstay of symptomatic treatment approaches to control PD motor symptoms. While levodopa is still considered the gold-standard of symptomatic efficacy, its chronic use is associated with the development of response oscillations and drug-induced dyskinesias in a majority of patients. Thus, much effort has been put into developing more continuous ways of levodopa delivery. Novel formulations and modes of application include extended-release (ER) oral levodopa (IPX203), levodopa powder for inhalation, and levodopa infusion therapies through subcutaneous (foslevodopa/foscarbidopa, ND0612) or intrajejunal pumps (levodopa [entacapone] carbidopa intestinal gel). Additionally, different preparations for subcutaneous and sublingual administration of apomorphine, a dopamine agonist with equivalent efficacy to levodopa, are available for the treatment of PD motor fluctuations. Novel dopamine agonists have been developed and tavapadon, a selective dopamine D1/D5 receptor partial agonist, was shown to be efficacious as monotherapy in early PD and adjunct to levodopa in patients with motor fluctuations. In addition, new data on the early use of the COMT-inhibitor opicapone have emerged. This expanding drug armamentarium is complemented by an increasing number of drugs targeting non-dopaminergic pathways with increasing evidence for amantadine’s symptomatic efficacy in treating levodopa-induced dyskinesia and motor fluctuations. Beyond the approval of different botulinum toxin preparations for the treatment of sialorrhea, efforts to address the plethora of non-motor symptoms of the disease have not translated to novel PD-specific approvals over the last few years. PD symptoms can usually be satisfactorily controlled in the early to mid-stages, but the progressive course of the disease inevitably leads to increasing functional disability underlining the yet unmet need for disease-modifying therapies. While there is currently no conclusive evidence for disease-modifying efficacy on any of the numerous efforts, we summarize recent late-stage clinical trial evidence focusing on glucagon-like peptide 1 (GLP1) agonists, approaches targeting the glucocerebrosidase (GBA) pathway and inhibition of the leucine‐rich repeat kinase 2 (LRRK2), as well as α-synuclein based treatments.

Key Points

Dopamine replacement remains the mainstay treatment for the motor symptoms of Parkinson’s disease (PD) with levodopa still representing the gold standard of efficacy. Recent developments around levodopa include the introduction of novel formulations like an inhalation powder for intrapulmonary delivery for on-demand treatment of OFF-episodes and soluble formulations for continuous subcutaneous infusion to control motor fluctuations. In addition, new data from studies with opicapone support the early use of COMT-inhibitors to treat wearing-off fluctuations.

A novel D1 dopamine agonist (tavapadon) has completed late stage clinical development showing efficacy both as monotherapy in early PD and as adjunct to levodopa in patients with motor fluctuations.

A growing number of disease modification trials are testing promising new candidate agents to slow PD progression and recent trials with the α-synuclein antibody prasinezumab, while not meeting their primary endpoint, have provided some signals for efficacy.

Introduction

Parkinson’s disease (PD) is the second most common neurodegenerative disease with an estimated number of 11.8 million people affected worldwide in 2021 [1]. Based on recent epidemiological surveys it is the fastest growing neurodegenerative condition, and its prevalence has been projected to increase to 25.2 million people within the next 25 years [2]. While the clinical diagnosis of PD is based on the presence of defining motor features of bradykinesia, rest tremor and rigidity, additional motor symptoms like freezing of gait, postural instability, falls, dysarthria and dysphagia are key drivers of disability in advanced disease [3]. In addition, a plethora of non-motor symptoms (NMS) like sleep disorders, autonomic dysfunction, apathy and depression as well as cognitive dysfunction are major contributors to disease burden and poor quality of life [3–5].

Drug treatment of PD motor symptoms is centred around pharmacological dopamine substitution with levodopa still being considered the most efficacious agent, while dopamine agonists (DA agonists) have overall smaller effect size and greater risk of troublesome adversity including daytime sleepiness and impulse dyscontrol [6–9]. Levodopa’s long-term efficacy is, on the other hand, compromised by the development of response oscillations and drug-induced dyskinesias in a majority of patients due to the progressive degeneration of dopaminergic neurons and pulsatile administration of levodopa, and these limitations are the main reasons for continued efforts towards improvements of current dopaminergic strategies. This includes optimizing levodopa pharmacokinetics and delivery as well as the development of novel DA agonists and various dopaminergic and non-dopaminergic adjunct therapies. Additional drug combinations to treat the multiple NMS make the medical management increasingly complex as the disease progresses. Late stage PD is characterised by an increasing burden of drug refractory motor and non-motor symptoms with limited efficacy of available treatment options highlighting the pressing need for disease-modifying therapies.

Here we aim to provide an overview over recent developments in the drug treatment of PD with a focus on promising drug candidates for the symptomatic or disease-modifying treatment of PD excluding non-pharmacological approaches like surgical therapies such as deep brain stimulation, gene and cell therapy as well as diet- and exercise-based treatments.

Search Methods

For this narrative review, original papers, and reviews were screened on PubMed to identify drugs that were approved within the last 5 years or had reached advanced stages of clinical development up to March, 2026. Additionally, we screened abstracts presented in 2025 at 3 major international congresses (Movement Disorder Society [MDS] congress, the annual meeting of the American Academy of Neurology and the Alzheimer’s Disease and PD [AD/PD] conference) for PD clinical trial results and reviewed ongoing clinical trials in PD on ClinicalTrials.gov.

Drugs to Treat the Motor Symptoms of PD

The modern era of pharmacological treatments for the motor symptoms of PD started with the pioneering discoveries of the antagonizing effects of levodopa on the akinesia induced by reserpine in rabbits by the Swedish pharmacologist Arvid Carlsson [10] and of dopamine depletion in the striatum of brains from PD patients by the Austrian pharmacologist Oleh Hornykiewicz, who – together with the neurologist Birkmayer also reported on the striking effects of small doses of intravenous levodopa on PD motor symptoms [11]. The final breakthrough of levodopa substitution was achieved after the neurologist George Cotzias described the transforming clinical effects of oral levodopa [12]. Since then, pharmacological dopamine replacement has remained the central pillar of PD therapy and non-dopaminergic drugs like amantadine and anticholinergics play a lesser role. Amantadine, however, is still the only efficacious drug to treat levodopa-induced dyskinesias (LIDs) [13], and new non-dopaminergic approaches to treat PD motor symptoms are being explored.

Dopaminergic Drugs

Pharmacological dopamine substitution is the mainstay of medical management of PD motor symptoms and levodopa combined with a dopa decarboxylase inhibitor (DDCI) still remains the gold-standard of symptomatic efficacy [3, 6–9, 14].

While the routinely available dopaminergic drugs listed in table 1, alone or as part of a combination regimen, allow for effective control of motor symptoms in a majority of patients over many years, unmet needs remain regarding efficacy as well as safety, tolerability and prevention of long-term motor complications. Except for P2B001 (a combination preparation listed in Sect. 3.1.1.4), we did not identify any recently completed clinical trials on monoamine oxidase B (MAO-B) inhibitors and thus refrained from including a separate subchapter on this drug class.

Table 1.

Dopaminergic drugs to treat motor symptoms of PD

Drug	Mode of action	Indication	Phase of development
Levodopa formulations
Standard (IR) levodopa/DDCI	DA substitution	PD motor symptoms	In clinical use
Inhaled levodopa (Inbrija®) [32, 36]	DA substitution	On-demand treatment of OFF-episodes	In clinical use
CR levodopa/DDCI	DA substitution	PD motor symptoms and motor fluctuations	In clinical use
ER levodopa/DDCI  (IPX066, Rytary®) [18]  (IPX203, Crexont®) [22, 23]	DA substitution	PD motor symptoms and motor fluctuations	In clinical use (US only)
Levodopa/DDCI/entacapone	DA substitution and COMT inhibition	Motor fluctuations	In clinical use
LCIG [47]	DA substitution	Motor fluctuations	In clinical use
LECIG [51]	DA substitution	Motor fluctuations	In clinical use
Foslevodopa/foscarbidopa continuous subcutaneous infusion [70, 71]	DA substitution	Motor fluctuations	In clinical use
ND0612 continuous subcutaneous infusion	DA substitution	Motor fluctuations	Phase 3 completed [83, 84]
Levodopa delivery by intraoral micropump (DopaFuse®)	DA substitution	Motor fluctuations	Phase 2 completed [24]
Prefilled nasal levodopa (TR-012001)	DA substitution	On-demand treatment of OFF-episodes	Phase 2 completed [43]
Dopamine agonists
Pramipexole	D2 agonist	PD motor symptoms and motor fluctuations	In clinical use
Ropinirole	D2 agonist	PD motor symptoms and motor fluctuations	In clinical use
Piribedil	D2 agonist	PD motor symptoms	In clinical use
Rotigotine	D2 > D1 agonist	PD motor symptoms and motor fluctuations	In clinical use
Apomorphine intermittent injection	D1 and D2 agonist	On-demand treatment of OFF-episodes	In clinical use
Apomorphine sublingual film [110, 111]	D1 and D2 agonist	On-demand treatment of OFF-episodes	In clinical use
Apomorphine continuous subcutaneous infusion [65, 66]	D1 and D2 agonist	Motor fluctuations	In clinical use
Tavapadon	Selective D1 partial agonist	PD motor symptoms and motor fluctuations	Phase 3 completed [94–96]
ER pramipexole/ER rasagiline (P2B001)	D2 agonist combined with MAO-B inhibitor	PD motor symptoms	Phase 3 completed [116]
Rotigotine ER microspheres for intramuscular administration (LY03003)	D2 agonist	PD motor symptoms	Phase 3 completed [118]
Glovadalen	Positive allosteric D1 receptor modulator	Motor fluctuations	Phase 2 completed [99]
Mesdopetam	D3 receptor antagonist with agonist-like physicochemical properties	Levodopa-induced dyskinesia	Phase 2 completed [120], phase 3 trial planned [121]
MAO-B inhibitors
Selegiline	Central inhibition of enzymatic dopamine degradation	PD motor symptoms	In clinical use
Rasagiline		PD motor symptoms and motor fluctuations	In clinical use
Safinamide		Motor fluctuations	In clinical use
COMT inhibitors
Entacapone	Peripheral inhibition of enzymatic levodopa degradation	Motor fluctuations	In clinical use
Tolcapone		Motor fluctuations	In clinical use
Opicapone		Motor fluctuations	In clinical use

Established drugs are given based on previous reviews [7, 9, 14]

COMT catechol-O-methyltransferase, CR controlled-release, DA dopamine, DDCI dopa decarboxylase inhibitor, D1 (2/3) dopamine D1 (2/3) receptor, ER extended-release, IR immediate-release, LCIG levodopa carbidopa intestinal gel, LECIG levodopa entacapone carbidopa intestinal gel, MAO-B monoamine oxidase B, PD Parkinson’s disease

Novel Levodopa Formulations and Delivery

Extended-Release Levodopa

Efforts to increase levodopa’s half-life and thereby reduce motor fluctuations and prolong ON-time have started more than 5 decades ago with the introduction of sustained and controlled-release (CR) levodopa formulations [15–17], however the evidence base for the efficacy of the first-generation agents has remained slim [14]. IPX066 was the first oral extended-release (ER) levodopa/carbidopa (LD/CD) formulation, which demonstrated a clinically meaningful reduction in motor fluctuations in a randomised, double-blind trial with a reduction of total daily OFF-time of − 1.2 h vs immediate-release (IR) LD/CD. It was approved by the United States (US) Food and Drug administration (FDA) 10 years ago and is marketed as Rytary^® in the US [18].

IPX203

IPX203 is a novel combined IR/ER formulation of an oral multiparticulate capsule of LD/CD in a ratio of 4:1. The capsule contains granules of IR LD/CD with a disintegrant polymer and ER levodopa beads coated with 3 different layers allowing for sustained enteral adhesion, absorption and release where IR granules carry 100% of the carbidopa and 25% of the levodopa dose while 75% of the levodopa dose is within ER pellets [19]. Thereby, the formulation enables a rapid increase of LD/CD to peak plasma concentrations comparable to immediate-release LD/CD and thereafter holds a steady level for several hours [19–21]. In an open-label, rater-blinded, single-dose crossover study, IPX203 treatment led to levodopa plasma concentrations larger than 50% of the maximum concentration for 4.6 h compared to 3.9 h with IPX066 accompanied by an increase in ‘good ON-time’ measured with the Investigator Assessment of Subject's Motor State (5.3 h for IPX203 vs 4.5 h for IPX066) [21].

A phase 3 double-blind, double-dummy, randomised controlled trial (RCT) comparing IPX203 with immediate-release (IR) LD/CD showed a statistically significantly greater increase in daily ON-time of + 0.53 h, and this difference was achieved with a mean of 3 daily doses vs 5 in the IR LD/CD arm [22]. OFF-time was reduced to a similar degree (− 0.48 h), the proportion of participants with improvement according to the Patient Global Impression of Change (PGI-C) score also favoured IPX203, and post-hoc analysis showed a mean increase in ON-time without troublesome dyskinesias (‘good ON-time’) per dose of 1.6 h compared to IR-LD/CD [22]. An open-label extension study of this trial showed maintained efficacy at stable dosing over 9 months [23]. In the double-blind phase of the pivotal RCT, AEs and AE-related discontinuations were numerically more common with IPX203 compared to IR LD/CD treatment (42.2% vs 31.6%; 5.5% vs 1.2%) [22], with nausea (4.3%), anxiety (2.7%), and dizziness (2.3%) being the most prevalent AEs with IPX203. [22].

IPX203 was approved by the FDA in 2024 for the treatment of PD and is marketed in the US as Crexont^® at levodopa dose sizes of 140 mg, 210 mg, 280 mg and 350 mg each combined with carbidopa in a ratio of 4:1. A phase 3b, open-label, randomised trial of IPX 203 vs IR LD/CD has recently been initiated in the European Union (EU) with the primary objective to further explore the effect of different dose frequencies and interdose intervals on a primary outcome of ‘good ON-time’ (ADIP study; EUCT 2025-521772-57-00). Overall, this novel LD/CD formulation may offer further ON-time gains for patients not sufficiently controlled by other oral levodopa-based strategies.

Other Novel Oral Delivery Approaches

Additional strategies aiming to extend the duration of effect from oral doses of levodopa or provide more continuous oral delivery are in early stage clinical development. DopaFuse^®, is an oral minipump that is attached to a dental retainer and uses a propellant to enable continuous delivery of a highly viscous LD/CD formulation [24]. A recent, 2-week, open-label pilot-study in 16 PD patients showed reduced plasma level variability compared to oral IR LD/CD and that DopaFuse^® treatment over 11 days was associated with matching OFF-time reduction and ON-time gain of 1.7 h [24]. Further development data are not publicly available.

CLE-600 has been developed for nighttime dosing of levodopa to target nocturnal and early morning motor disability using a proprietary oral long acting release platform which enables extended drug delivery via a gastric retention system. A healthy volunteer pharmacokinetic study showed persisting levodopa plasma levels over 8 h [25] but no further development data have been reported.

CP-012 is a novel delayed pulsatile release formulation of levodopa also designed to target nocturnal and early morning PD symptoms with bedtime dosing. Results of a phase 1b pharmacoscintigraphic study in PD subjects have recently been published on the company website and provide proof of the pharmacokinetic principle with delayed levodopa release at the intended intervals post dosing [26]. Transition into phase 2 is currently planned.

Non-Oral On-Demand Formulations

Absorption of levodopa after oral dosing is subject to a variety of potentially interfering mechanisms including dysphagia, esophageal dysmotility, delayed gastric emptying due to gastroparesis or decreased intestinal absorption due to transport competition with dietary amino acids or decarboxylation by gut microbiota [27–29]. This may lead to considerable variability in plasma levels and clinical response including latency to onset of effect following individual oral doses. Soluble levodopa is available in several countries and used as an on-demand treatment to treat OFF-periods [30], but its efficacy is limited by problems of gastrointestinal transport and absorption as mentioned above. Several non-oral delivery systems of levodopa designed to bypass GI transport and absorption issues have recently become available for clinical routine use (Fig. 1).

Fig. 1.

Optimised dopaminergic drug delivery. Abbreviations: CR, controlled-release; ER, extended-release; IR, immediate-release; LCIG, levodopa carbidopa intestinal gel; LECIG, levodopa entacapone carbidopa intestinal gel. Created in https://BioRender.com

Inhaled Levodopa

CVT-301 (Inbrija^®) is a powder formulation of levodopa based on 5–10 μm diameter porous low-density particles suitable for aerosolizability and intrapulmonary delivery via a breath-actuated inhaler [31]. The marketed encapsulated formulation contains 42 mg of levodopa per capsule, and inhalation of two capsules delivers a fine particle dose corresponding to 50 mg of levodopa [32] leading to peak plasma concentrations within 30 min [31] and onset of clinical effects after about 15 min [33, 34].

A pivotal double-blind, placebo-controlled phase 3 RCT (SPAN-PD) compared two doses (60 mg and 84 mg) of inhaled levodopa to placebo in levodopa-treated PD patients with response fluctuations and at least 2 h of cumulative daily OFF-time [32]. Active drug or placebo were self-administered at the onset of OFF-periods and the primary endpoint was the change in UPDRS motor scores from pre-dose to 30 min post-dose assessed during an in-clinic OFF-period at week 12. There was a significant difference from placebo for the 84 mg dose of − 3.9 points (p < 0.01) in favour of active drug and the proportion of participants achieving a full ON-state after 60 min was 58% in the 84 mg treatment group compared to 36% in the placebo arm (odds ratio of 2.65 and 95% confidence interval of [1.48;4.76], p < 0.01) [32]. The clinical efficacy of this inhaled levodopa formulation is further supported by open-label longer-term trials suggesting that motor improvements are maintained over 12 months [35, 36]. A meta-analysis of published RCTs favoured inhaled levodopa over placebo regarding proportion of patients achieving a full ON-state within 60 min post dosing as well as magnitude of UPDRS motor score reduction and PGI-C self-rating improvements [37].

While initial studies and the pivotal SPAN-PD trial did not permit use of inhaled levodopa for the treatment of early-morning OFFs before intake of the first regular oral LD/CD dose [32–34] because of the absence of a DDCI in the inhaler formulation, a randomised, double-blind, 2-way crossover study did not reveal any additional safety signals with early-morning administration [38].

Other than typical dopaminergic AEs inhaled levodopa may cause coughing, throat irritation and sputum discoloration due to levodopa oxidation [32, 35, 36]. Discontinuation rates ranged from 15.0% to 30.8% of participants treated with inhaled levodopa in the SPAN-PD trial and its open-label 12-month extension study [32, 36]. A dedicated RCT focusing on pulmonary safety found no substantial differences regarding longitudinal change of pulmonary function with inhaled levodopa treatment compared to oral standard of care treatment [35].

Inbrija^® was approved by the FDA and the European Medical Agency (EMA) in 2018 and 2019 respectively [32] and has recently become available in several European countries including Germany, Spain, Portugal and Austria as an on-demand treatment to abort OFF-periods in PD patients with levodopa response fluctuations. Inhaled levodopa is the first on-demand levodopa delivery strategy that circumvenes potential intestinal absorption delay or failure and thus a useful addition to previous on-demand options like subcutaneous or sublingual apomorphine (see Sect. 3.1.1.4).

A differently formulated levodopa inhalation powder [39] administered with a disposable, preloaded inhaler [40] is in early phase clinical development [41, 42].

In addition to inhaled levodopa formulations, a prefilled nasal levodopa on-demand preparation (TR-012001) has shown peak plasma levels after 15 min post-dosing with variable motor improvement in subjects experiencing OFF-periods in a small phase 2 proof-of-concept study [43].

Levodopa Infusions

Continuous delivery of levodopa via waking-day intravenous infusions were first shown to smooth out response fluctuations in PD patients with sustained oral levodopa substitution in the 1980s in two small open-label proof-of-concept trials [44, 45] and similar results were subsequently reported for continuous duodenal delivery via nasoduodenal pumps [46]. Aside from the need for central lines for intravenous delivery, these approaches were unsuitable for clinical routine use due to the poor solubility of levodopa requiring large volumes of fluid around 2 L per infusion [45].

Levodopa Carbidopa Intestinal Gel Formulations

It took about 2 decades before an intestinal gel formulation for intrajejunal LD/CD delivery (LCIG) received market authorization in the US, EU, Japan and other parts of the world. A pivotal randomised, double-dummy, double-blind trial showed a significant OFF-time reduction of − 1.9 h vs placebo [47] while uncontrolled open-label real-world registry studies reported reductions in the order of 4 h vs baseline [48, 49]. In addition, a 12-week randomised trial of LCIG versus optimised medical therapy with oral drugs found a significant reduction in Unified Dyskinesia Rating Scale (UDysRS) scores with infusion therapy [50] supporting the role of continuous drug delivery in the management of LIDs.

Levodopa entacapone carbidopa intestinal gel (LECIG) infusion is a modification of LCIG adding a catechyl-O-methyltransferase inhibitor (COMT-I) to the formulation. This increases levodopa bioavalibilty and enables smaller volumes that can be delivered by smaller pumps compared to the LCIG device. One randomised 2-day crossover trial [51] and multiple open-label series [52–55] suggest similar efficacy and safety to LCIG but head-to-head comparisons are lacking. LECIG is marketed in several EU and non-EU countries and is not as widely available as LCIG [56].

Levodopa Carbidopa Formulations for Subcutaneous Delivery

Despite of the remarkable efficacy of intestinal levodopa infusions, invasiveness due to a need for percutaneous endoscopic gastrostomy (PEG) and the complication risk associated with it as well as cosmetic and other limitations linked to external tubing are important limitations. Subcutaneous routes of delivery offer major advantages in this regard. They were first introduced for apomorphine—a drug with equivalent efficacy to levodopa—in the 1980s [57] and subcutaneous infusions of this dopaminergic agent have since been widely used in Europe, Australia and parts of Asia with multiple open label [58–65] and one randomised controlled trial [66] supporting their efficacy in reducing motor fluctuations.

The poor solubility of levodopa at physiological pH has long been a road-block for subcutaneous levodopa delivery but novel levodopa formulations with greatly enhanced solubility have now been successfully developed [67]. Foslevodopa/foscarbidopa (pLD/pCD) is a prodrug converted to LD/CD by alkaline phosphatases quickly after subcutaneous administration [68]. By replacing a hydroxyl group with a phosphonooxy group in levodopa and carbidopa the water solubility of the molecules is markedly increased at a physiological pH enabling foslevodopa concentrations of 240 mg/ml and offering the possibility to cover a wide range of daily dose requirements [67, 68]. Plasma levels of levodopa were shown to be stable with only minimal fluctuations in two phase 1 trials of pLD/pCD [68, 69].

The pivotal phase 3 study of pLD/pCD was designed as a double-blind, double-dummy, placebo-controlled 12-week RCT [70] with change from baseline to end of trial in normalised ‘good ON-time’ as primary endpoint. pLD/pCD was delivered as a 24-h infusion with up to 20% lower nighttime infusion rates vs daytime [67]. The trial which enrolled 174 PD patients with motor fluctuations and at least 2.5 h OFF-time per day found a significantly greater increase in ‘good ON-time’ of 1.75 h as compared to placebo with a corresponding reduction of − 1.97 h of OFF-time [70]. Percentages of patients reporting an OFF-status in their home diaries for the first half-hour period after awakening were also significantly smaller with active infusion (17% vs 63% on placebo). An open-label 52-week phase 3 trial in 244 PD patients with similar inclusion criteria found a sustained reduction in motor fluctuations after one year [71].

Both trials had high discontinuation rates ranging from 35.1% in the 12-week RCT to 44.8% in the 12-month open-label study [70, 71]. AEs were the main reason for discontinuation in 22% with infusion site reactions being the most common including skin erythema, nodules, cellulitis, oedema, pain and abscess, but hallucinations and psychosis were also reported in 15% of patients in the active arm of the double-blind RCT (vs 3% on placebo) [70] and in 17% of subjects in the open-label 52-week study [71].

pLD/pCD has received marketing authorization of the EMA in 2022 and FDA-approval in 2024 for the treatment of motor fluctuations associated with advanced PD. It is marketed in a dose of 240 mg foslevodopa and 12 mg foscarbidopa per ml and administered subcutaneously by a Vyafuser™ pump with flexibly adjustable flow rates [67, 72]. Although the maximum daily dose of pLD in the EU is set at 6000 mg and at 3525 mg in the US corresponding to approximately 4260 mg and 2500 mg of levodopa respectively [72], dose requirements in clinical practice are usually well below these limits [73–76]. High levodopa exposure over 24 h and especially during the night may increase risk for confusional states and hallucinosis especially in patients with cognitive impairment [77] and nocturnal flow rates in particular should be chosen as the best compromise between dopaminergic AE risk and clinical benefit based on clinical experience with 24-h LCIG infusion [78, 79].

ND0612 is a LD/CD formulation with increased levodopa aqueous solubility suitable for subcutaneous infusion delivered via 2 cannulas aiming to reduce skin burden [67]. Early-phase clinical trials have revealed sustained and stable LD/CD plasma concentrations in healthy participants and persons with PD [80]. A LD/CD ratio of 8 to 1 was identified as the most appropriate formulation to receive an optimal levodopa bioavailability and the final formulation is a liquid LD/CD solution of 60/7.5 mg per mL [80].

Efficacy of ND0612 in reducing motor fluctuations has been studied in several phase 2 trials [81–83] and a pivotal phase 3 randomised, double-dummy, double-blind, active-controlled trial [84]. The BouNDless trial enrolled 381 PD patients with motor fluctuations who experienced at least 2.5 h of OFF-time per day. A 12-week open-label run-in phase including an oral LD/CD optimization period followed by an ND0612 optimization period preceded the 12-week double-blind phase. The optimised ND0612 regime included at least an additional morning dose of oral IR LD/CD, and further oral doses were allowed if required, as maximum daily ND0612 infusion dose was restricted to a maximum of 720/90 mg of LD/CD. The primary outcome was change from double blind-baseline ‘good ON-time’ as recorded by home diaries, self-completed on 3 consecutive days before the study visits. The trial found a significantly greater gain in ‘good ON-time’ of 1.7 h and significantly greater reduction in OFF-time of − 1.4 h in the ND0612 treatment group compared to the oral LD/CD control group on placebo-infusions [84]. Additionally, ND0612 was shown to be superior to oral LD/CD regarding changes in the MDS-sponsored revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) part II (motor experiences of daily living) score, Patient Global Impression of Change (PGIC) score, and Clinical Global Impression of Improvement (CGI-I) score [84]. Long-term 3-year efficacy data from the BeyoND study showed OFF-time reductions of − 2.8 h and matching increases in ‘good ON-time’ [85].

Tolerability and safety were similar in both the 12-week RCT and the 12-months safety study (BeyoND) enrolling 214 patients with levodopa response fluctuations [83, 84]. Discontinuation rates ranged from 19.6% in the 6-week ND0612 open-label optimization period within the BouNDLess-study to 44% after 12 months within the BeyoND-study [83, 84]. AEs were the second most common reason for early termination after consent withdrawal [83, 84] and among these infusion site reactions were the most frequent (84% in the open-label optimization period of BouNDless), usually of mild-to-moderate severity and included mainly skin nodules, hematomas, infection, eschars and pain. Three-year outcomes from the extension phase of the BeyoND study show that the overall frequency of dopaminergic AEs were 5.3% for hallucinosis, 4.4% for dyskinesias and orthostatic hypotension and 2.6% for somnolence [85]. ND0612 has recently completed the EMA review with a positive opinion issued by EMA’s Committee for Medicinal Products for Human Use (CHMP) on February 26, 2026 [86].

DIZ102 is a subcutaneous levodopa formulation that exploits levodopa’s increased solubility and potentially favourable skin tolerability at acidic pH. This drug is still in its early stages of development with results of a phase 1 trial demonstrating feasibility of this approach and first exploratory efficacy signals [87, 88].

The availability of soluble levodopa/carbidopa formulations for continuous subcutaneous marks an important advance over the previous intestinal pump delivery systems in terms of invasiveness and patient comfort. Their efficacy appears equivalent to that of subcutaneous apomorphine infusions while AE-related discontinuation rates in the phase 3 trial of pLD/pCD were higher compared to those in the TOLEDO trial of apomorphine [66, 70, 89]. The lack of direct comparative studies and the much shorter real-world history of subcutaneous levodopa infusions presently do not allow to conclude on potential longer term tolerability differences vs subcutaneous apomorphine, particularly related to local skin reactions.

New Dopamine Agonists and Novel Formulations

Non-ergolinic DA agonists such as pramipexole, ropinirole, and rotigotine are efficacious in treating PD motor symptoms both as monotherapy as well as adjuncts to levodopa in patients with or without motor complications [7]. Their major limitations are related to an unfavourable safety profile regarding their potential to induce daytime somnolence and ICDs. The latter affect around 17% in cross-sectional studies up to 46% of exposed patients in a 5-year longitudinal study [90, 91] and are associated with long exposure and high doses of DA agonists [77]. A number of recent drug development efforts aim at improving the benefit-risk ratio of oral DA agonists (Fig. 2).

Fig. 2.

Pharmacodynamic principles of dopamine agonists. Drugs in current clinical development are written bold. Tavapadon is a selective dopamine D1/D5 receptor partial agonist, while Glovadalen is a positive allosteric modulator of the D1 receptor. Mesdopetam is an oral dopamine D3 receptor antagonist with agonist-like physicochemical properties. D1R, dopamine D1-like receptor; D2R, dopamine D2-like receptor. Created in https://BioRender.com

D1 Dopamine Agonists

D1-selective DA agonists were first developed in the 1990s based on the notion that they might offer similar efficacy as levodopa but reduce dyskinesia risk [92].

Tavapadon is a novel selective long-acting dopamine D1/D5 receptor partial agonist suitable for once daily oral administration [93]. Its development has progressed to phase 3 and results of 3 trials have recently become available [94–96].

TEMPO-1 and TEMPO-2 (NCT04223193) are placebo-controlled, double-blind, randomised trials investigating tavapadon as monotherapy in participants with early stage PD over 27 weeks. TEMPO-1 is a 3-arm trial of two doses of tavapadon (5 mg and 15 mg) vs placebo [95], and TEMPO-2 is a 2-arm parallel group trial of flexible dose tavapadon vs placebo currently published in abstract form only [94]. Both studies met their primary efficacy endpoint of change from baseline of MDS-UPDRS part II + III scores with a statistically significant difference in favour of tavapadon vs placebo: − 11.5 points for 5 mg, − 12.1 points for 15 mg, and − 9.1 points in the flexible-dose study [94].

The TEMPO-3 trial assessed tavapadon’s efficacy as an adjunctive treatment in levodopa-treated PD patients with motor fluctuations [96]. The primary endpoint of this randomised, double-blind, placebo-controlled parallel group study of flexibly dosed tavapadon was the change in ON-time without troublesome dyskinesia assessed by Hauser diaries between baseline visit and week 26. This study also met its primary endpoint with a significantly greater increase in ‘good ON-time’ versus placebo (+ 1.1 h) [96].

Detailed safety data is available for the TEMPO-1 and TEMPO-3 trial. Discontinuation rates ranged from 24.3% in the 5 mg tavapadon treatment arm of the TEMPO-1 trial to 36.9% in the flexible-dose tavapadon treatment arm of the TEMPO-3 trial compared to 15.4% and 19.2% in the placebo arms [95]. AEs were the most common reason for discontinuation in both trials. Nausea, dizziness, and headache occurred in more than 10% (1.7% to 5.1% with placebo) of tavapadon-treated individuals of the TEMPO-1 trial and together accounted for more than half of AE-related dropouts [95]. Dyskinesia was reported as AE in 10% of the tavapadon-treated participants (vs 1.6% with placebo) of the TEMPO-3 trial [96]. Also orthostatic hypotension (6.0% vs 1.2%) and visual hallucinations (5.6% vs 1.2%) were reported more frequently with active treatment while rates of somnolence and fatigue were comparable to placebo . Incidence rates for ICDs with tavapadon treatment were 0.8% and 2.4% in the RCTs and 1.4% in a preliminary interim analysis of a 58-week, open-label study (TEMPO-4, NCT04760769) [95–97].

Glovadalen (UCB0022) is a positive allosteric modulator (PAM) of the D1 receptor that enhances D1 signaling in the presence of dopamine. Similar to tavapadon, the rationale for this approach is that D1-selectivity may avoid D2-related side effects while providing sufficient efficacy.

Results of a phase 1 trial have been published in abstract from only and report good tolerability [98], and the phase 2 trial (ATLANTIS, NCT06055985) performed in 207 participants with motor fluctuations met its primary endpoint with significantly greater change in the average number of OFF-hours per day from baseline to day 70 as compared to placebo [99].

The prime unmet needs regarding current D2-agonist treatment relate to their AE-profile, in particular regarding daytime somnolence, psychosis and impulse dyscontrol. While the above studies support the antiparkinsonian efficacy of drugs selectively targeting D1-receptors they do not support improved safety and tolerability. Longer term data will be required to assess ICD and dyskinesia risk.

Apomorphine

Apomorphine is a D1/D2 dopaminergic agonist with a receptor affinity profile of dopamine [100, 101]. It is the only dopaminergic drug with similar clinical efficacy for treating PD motor symptoms as the gold-standard agent levodopa [102, 103]. Due to an extensive first-pass effect apomorphine lacks adequate enteral bioavailability and subcutaneous on-demand injections or continuous infusions have been in clinical routine use in Europe for more than 30 years [57, 104, 105] but continuous subcutaneous apomorphine infusions were only recently granted FDA approval in 2025 [64, 66, 67].

Subcutaneous apomorphine delivery is usually well-tolerated, but injection and infusion site reactions are limiting AEs and ‘needle phobia’ may stand in the way of patient acceptance, prompting other modes of apomorphine administration, including nasal [106], transdermal [107], rectal [108], pulmonary [109], and sublingual routes [104].

A soluble bilayer film strip for sublingual administration containing apomorphine on one and a pH-buffer solution on the other layer has been approved by the FDA in 2020 and more recently by some European countries via a decentralised regulatory path [110].

The pivotal study assessing the efficacy and safety of this sublingual apomorphine film formulation was a randomised, double-blind, placebo-controlled trial in 109 levodopa-treated PD patients experiencing 2 or more hours of OFF-time per day and predictable morning OFF-phases [110]. Patients were randomised 1:1 to active drug (n = 54) or placebo (n = 55) and the primary endpoint was change in the MDS-UPDRS motor section (part III) from predose to 30 min postdose at week 12 assessed in a practically defined OFF-state. The trial found a significant difference of − 7.6 points favouring sublingual apomorphine over placebo at 30 min post-dosing and a statistically significant difference of − 3.4 points was already observed at 15 min [23] while open-label studies subsequently reported maintained efficacy over 48 weeks [111].

Thirty-seven percent of patients randomised to sublingual apomorphine in the double-blind 12-week study prematurely discontinued, most commonly due to AEs [110]. Oropharyngeal mucosal irritation occurred in 31% of patients treated with apomorphine and led to treatment discontinuation in 17%. 28% of participants in the active arm experienced nausea and 13% reported somnolence but except for one case with nausea and vomiting discontinuing these were mild to moderate in severity. The side effect profile was similar in 496 patients enrolled in an open-label long-term safety study for up to 3 years of which 120 (24%) completed the study and 34% discontinued because of AEs, the most common of which were nausea and oropharyngeal events [111].

Sublingual instead of subcutaneous delivery of apomorphine as an on-demand treatment to abort OFF-periods offers the advantage of non-invasiveness and extends options for subjects with ‘needle-phobia’ or those unable to handle a pen-injector. Local mucosal tolerability is a potential limitation and onset of effect after dosing is somewhat slower than that typically seen after subcutaneous injections of apomorphine. Clinical routine use of sublingual apomorphine should be initiated under supervision at a 10 mg dose and further titration to an optimal response dose (max. 30 mg) can be performed on an outpatient basis with or without prophylactic antiemetic treatment [112, 113]. Despite its efficacy sublingual apomorphine has not been commercially successful in the US market where it is currently no longer available.

Other dopamine agonist based approaches

P2B001 is a fixed low-dose, ER combination of 0.6 mg pramipexole and a novel ER formulation of 0.75 mg rasagiline in clinical development as a once daily monotherapy for early PD. By lowering the dose of pramipexole below the range of current use in PD this combination is intended to reduce dopaminergic AEs while maintaining efficacy [114–116]. P2B001 was shown to be superior to placebo in improving UPDRS total scores (− 4.7 points difference in adjusted mean change) in a phase 2b study [115].

A subsequent 12-week, double-blind phase 3 RCT in participants with early untreated PD demonstrated superior efficacy in reducing combined UPDRS Part II+III scores compared to either of its components alone as well as similar efficacy as conventionally dosed ER pramipexole (mean dose of 3.2 mg) [116]. Orthostatic hypotension (2.7% vs 12.2%), hallucinations (2.0% vs 4.1%) and somnolence (14.7% vs 31.3%) were more seldomly reported as AEs in the P2B001 group compared to conventionally dosed ER pramipexole, and overall results suggested improved safety and non-inferior efficacy. No ICDs were reported in this trial, which does not inform on potential ICD risk reductions of this formulation [116]. Since the publication of these results no information on the further development of P2B001 has been publicly provided.

LY03003 is a novel rotigotine ER microsphere formulation for once weekly intramuscular administration aiming for continuous drug delivery [117]. Results of a phase 3, randomised, placebo-controlled, double-blind study in participants with early PD diagnosed within the last 5 years were recently presented in abstract form [118]. After 24 weeks of maintenance treatment, the LY03003 treatment group showed significantly reduced UPDRS Part II+III scores versus placebo (− 11.8 vs − 5.6). Serious AEs occurred in 4.8% and 2.7% of subjects on active drug or placebo but are not detailed in the abstract [118]. Thorough peer reviewed results of this study have not yet been published, and the drug has not been approved in the US, Europe or Japan up to now.

Mesdopetam is an oral dopamine D3 receptor antagonist with agonist-like physicochemical properties currently in clinical development as a treatment of LIDs in PD patients [119]. A phase 2b, placebo-controlled, dose-finding study included 156 participants with PD 2 or more hours of dyskinesia per day [120]. The primary endpoint of change in ‘good ON-time’ from baseline to week 12 was not met, but secondary endpoint analyses demonstrated significant improvements in dyskinesia ratings (sum score of Unified Dyskinesia Rating Scale [UDysRS] subitems 1, 3 and 4) in the mesdopetam treatment groups compared to placebo.

The drug was well tolerated with similar rates to placebo of serious AEs (3.4% vs 7.7%) and AE-related discontinuations (8.6% vs 10.3%) [120]. Plans for a phase 3 trial in the same target population with change in UDysRS scores as endpoint were recently announced by the company [121]. Given the limited options to treat LIDs (see 3.2.1) the demonstration of antidyskinetic efficacy of a D3-dopamine antagonist without worsening of parkinsonism would be of great interest.

New Data on COMT-Inhibitors

The long-acting oral COMT inhibitor opicapone was approved for the treatment of PD as adjunct therapy to LD/DDCI in patients with wearing-off type motor fluctuations by the EMA in 2016 and in 2020 by the FDA based on efficacy and safety data from 2 pivotal trials (BIPARK I and II) [122, 123]. A pooled post-hoc analysis of these trials showed similar effect sizes in subjects with a short (≤ 1 year) or longer (> 2 years) history of motor fluctuations at trial start [124].

Three recent trials and one post-hoc analysis of the OL-extension phases of BIPARK I and BIPARK II have provided new information on the early use of opicapone as an adjunct to levodopa treatment. Ferreira and colleagues conducted a pharmacokinetic and clinical short term randomised trial in which patients with motor fluctuations were assigned to a regimen of five daily doses of 100 mg levodopa with DDCIs for 2 weeks (500 mg of levodopa per day) and subsequently randomised to one of two regimens where opicapone 50 mg was added and the daily levodopa dose was reduced by 100 mg—either by reducing dose frequency to 4 doses of 100 mg or by reducing two of the 5 doses to 50 mg. Following 2 weeks on these regimens, levodopa pharmakokinetic parameters and clinical response (total daily OFF- and ON-times) were assessed and compared to the corresponding outcomes after 2 weeks of the 500 mg/day regimen [125]. Results showed superiority of both levodopa dose reduction regimens with opicapone adjunct in terms of significant reductions of OFF-time and increases in ON-time vs the original 500 mg/day as well as significant increases in the minimum level of levodopa plasma concentrations and area under the curve (AUC) values.

These results triggered the randomised, open-label ADOPTION trial program (comprising a Korean and a European trial with identical design) comparing the effects of adding 50 mg of opicapone or 100 mg of levodopa to subjects with a motor fluctuation history of less than 2 years and a maximum of 600 mg of levodopa per day. More than 90% of patients in the Korean trial were on 3 daily doses of levodopa with a mean daily dose around 400 mg and average daily OFF-time of 3.4 h [126]. After 4 weeks, reductions in daily OFF-time were significantly greater after adding 50 mg opicapone vs adding 100 mg levodopa (− 62 vs − 17 min; mean difference − 45.4. min). The pooled analysis of both ADOPTION trials also showed a significant benefit of adding opicapone with a difference vs adding 100 mg of levodopa of 29 min (p = 0.022) [127].

A recent post-hoc analysis of the 52-week, open-label extension phases of BIPARK I and II compared the longer-term outcome of subjects with a fluctuation history of 2 years or less (n = 227) between those who had received opicapone vs placebo in the initial double-blind phase of the trial. While again showing significant improvement in OFF- and ON-time outcomes when switching from placebo to open-label opicapone treatment those initiated on opicapone at open-label start did not have the same degree of improvement as those with opicapone treatment at start of the double-blind period with a group difference in OFF- and ON-time of 30 (p = 0.2) and 44 min (p = 0.05) at end of open-label week 52 despite similar levodopa equivalent daily dosage in both groups. Importantly the 2 groups did not differ in time spent with troublesome dyskinesia at week 52 [128].

Ever since the Stride-PD trials of early adjunct treatment with entacapone in PD patients initiated on levodopa [129] there has been an unresolved debate on the potential benefit of LD/COMT-I combination in early PD. While the effects on the development of levodopa-related motor complications require long-term trials in de-novo subjects, the EPSILON study of early opicapone adjunct in levodopa-treated subjects without motor complications has recently provided evidence for beneficial effects on motor control without increasing motor complication risk over 76 weeks. The trial randomised 380 subjects on stable treatment with levodopa without motor complications to double-blind adjunct therapy with either 50 mg of opicapone or placebo for 24 weeks and offered transition into a 52-week extension on open-label treatment with 50 mg of opicapone for all participants [130]. The primary outcome of the double-blind phase was the change from baseline in the score of the MDS-UPDRS part III and showed a statistically significant difference in favour of opicapone of 2.2 points less worsening at month 6 (p = 0.01) without a difference in UPDRS IV motor complication scores. The results of the open-label extension showed also that longer exposure to opicapone did not increase motor complication rates and interestingly a larger percentage of patients randomised to opicapone in the double-blind phase (‘early starters’) remained free of motor complications (score of 0 on the MDS-UPDRS part IV) compared to those starting opicapone at open-label baseline (80.2% vs 69.7%, p = 0.1).

While the EPSILON study 1-year results of no increase in motor complications—in particular dyskinesia—rates after total exposure over 76 weeks are encouraging, they cannot exclude that such an increase would be seen with prolonged treatment over longer time periods. The STRIDE-PD trial randomised 747 patients with early PD to treatment with LD/CD plus entacapone vs LD/CD plus placebo for up to 4 years. The mean time to onset of dyskinesias with entacapone was 74 weeks roughly corresponding to the duration of the EPSILON study [129]. However, the levodopa equivalent daily dose at onset of dyskinesias in the entacapone group of STRIDE-PD was higher than in the EPSILON study, which may explain the different dyskinesia outcome.

Non-dopaminergic Drugs to Treat PD Motor Symptoms

Historically, the first drugs discovered to improve PD motor symptoms were non-dopaminergic agents. The founding-father of clinical neurology, Jean-Martin Charcot, started to prescribe hyoscyamine (since recognised as an anticholinergic agent) to his patients with ‘Paralysis Agitans’ in the 1860s and observed beneficial effects particularly on tremor. Synthetic anticholinergics subsequently became the principal drugs to treat PD in the first half of the twentieth century and were joined by amantadine as a second non-dopaminergic agent in the late 1960s [131, 132]. While non-dopaminergic agents thus have a long history in the treatment of PD motor symptoms, there have only been few new drug approvals in recent years. Nonetheless, new data have emerged around the efficacy of a novel formulation of amantadine, and some investigational new drugs are in early stage clinical development (Table 2).

Table 2.

Non-dopaminergic drugs to treat motor symptoms of PD

Drug	Mode of action	Indication	Phase of development
Anticholinergics	Antagonism at striatal interneuron cholinergic receptors	PD motor symptoms, mainly tremor	In clinical usea
Clozapine	Serotonergic receptor modulation (5HT2A/2C antagonism, 5HT1A agonism)	LIDs and PD tremor	In clinical usea
Istradefylline	A2A receptor antagonism	Motor fluctuations (adjunct to levodopa)	In clinical use (Japan, US)
Zonisamide	MAO-B inhibition plus blockade of T-type Ca-channnels	PD motor symptoms and motor fluctuations (adjunct to levodopa)	In clinical use (Japan only)
Amantadine	Antiglutamatergic (NMDA receptor anatagonism)	PD motor symptoms and LIDs	In clinical use
Amantadine ER (Gocovri®) [133, 134]	Antiglutamatergic (NMDA receptor anatagonism)	LIDs and motor fluctuations	In clinical use
NLX-112	Selective 5-HT1A receptor agonism	LIDs	Phase 2 completed [139]
CPL500036	PDE10A inhibition	LIDs	Phase 2 completed [142]
Solengepras	GPR6 inverse agonism	Motor fluctuations	Phase 2 completed [152, 153] phase 3 trial ongoingb

Established drugs are given based on previous reviews [7, 9, 14]

ER extended-release, GPR G-protein coupled receptor, LIDs levodopa-induced dyskinesias, MAO-B monoamine oxidase B, NMDA N-methyl-d-aspartate, PD Parkinson’s disease, PDE phosphodiesterase, 5-HT 5-hydroxytryptamine

^aSecond-line treatment

^bNCT06553027

Non-dopaminergic Drugs to Treat Levodopa-Induced Dyskinesias

Amantadine is a well-established antiparkinsonian drug with multimodal dopaminergic and non-dopaminergic properties including an antagonistic effect on N-methyl-d-aspartate (NMDA) glutamate receptors [13]. Multiple randomised trials from the 1970s and 1980s have shown its antiparkinsonian efficacy while its antidyskinetic efficacy in PD patients with LIDs was only demonstrated some 30 years after its introduction into PD therapy [13]. Despite a solid evidence base for its antidyskinetic efficacy this indication is not part of standard (IR) amantadine’s regulatory approval. FDA approval as an antidyskinetic agent was only granted in 2017 following two positive placebo-controlled trials of a novel ER formulation of amantadine (marketed as Gocovri^® in the US) [133, 134].

While all previous studies of amantadine as an antidyskinetic agent tested its symptomatic effect in PD patients who had established LIDs, the PREMANDYSK trial assessed if early adjunct therapy with amantadine has the potential to reduce dyskinesia risk in levodopa-treated early PD patients without LIDs [135]. The primary outcome of this trial was the percentage of patients with incident LIDs by month 18 and showed a statistically significant difference in favor of amantadine (11% vs 22% on placebo; p = 0.023). Following a 3-month delayed-start period, during which all participants were treated with amantadine and a subsequent double-blind 1-month washout period the difference in LID prevalence between those treated with adjunct amantadine vs those treated with placebo in the first 18 months was no longer statistically significant. The trial therefore provides no conclusive evidence that early adjunct therapy with amantadine provides true protective effects against the development of LIDs as opposed to its well-known symptomatic efficacy [135].

Conversion of levodopa to dopamine in serotonergic neurons followed by dysregulated dopamine release may play a role in the pathogenesis of LIDs [136]. Previous trials of serotonergic drugs in PD patients have failed due to worsening of parkinsonism [137], but serotonergic approaches are still being pursued.

NLX-112 (befiradol or F13640) is a centrally active, selective serotonin 5-HT1A receptor agonist that has shown antidyskinetic effects in a rat model, probably by inhibiting serotonergic neurons’ false dopamine conversion and release through activation of 5-HT1A autoreceptors [138]. NLX-112 has shown an acceptable safety profile in a small phase 2a, double-blind trial in 27 PD patients [139]. Mild to moderate AEs occurred primarily in the uptitration phase at a target dose of 1 mg of NLX-112 twice daily. The NLX-112 group showed significant reductions in UDysRS) average sum scores at predefined time points after a levodopa challenge test on day 28 (end of uptitration phase) and 42 (end of stable dosing phase) on treatment as compared to placebo treated patients. Importantly, PD motor symptoms did not worsen with NLX-112 treatment and UPDRS scores even showed small improvements [139].

CPL500036 selectively inhibits phosphodiesterase (PDE) 10A, an enzyme almost exclusively expressed in the medium spiny neurons of the striatum. Modulating cyclic-nucleotide levels in the striatum may balance striatal dysfunction by inhibiting the indirect and activating the direct basal ganglia pathway, which potentially translates to both antidyskinetic and antiparkinsonian effects [140, 141]. Results of a phase 2 trial (NCT05297201) in patients with LIDs have recently been published online by the manufacturing company and support both antidyskinetic efficacy and a favourable safety profile for further clinical development [142].

Non-dopaminergic Drugs to Treat Motor Fluctuations

The 2 pivotal trials of ER amantadine to treat LIDs (see Sect. 3.2.1, [133, 134]) also provided sufficient evidence for an FDA approval for the treatment of OFF-episodes in 2021. A pooled analysis of these two randomised-controlled trials demonstrated a placebo-adjusted reduction in OFF-time of 1 h from baseline to week 12 [143], and an open-label extension showed persistent improvement of MDS-UPDR part IV motor fluctuation subscore after 100 weeks of treatment [144, 145]. Whether similar efficacy in improving OFF-episodes can be achieved with amantadine IR tablets is currently investigated in the AMANT-OFF trial (NCT06817200).

Istradefylline is an adenosine A2A receptor antagonist which was shown to reduce OFF-time ranging from 12 to 114 min in several trials and has been in clinical use to treat motor fluctuations in Japan for many years and more recently (2019) also received FDA approval [14, 146, 147].

Zonisamide is an antiepileptic agent with multimodal pharmacodynamics including sodium and calcium channel blocking as well as dopaminergic properties (e.g. MAO-B inhibition) which was shown to reduce OFF-time by around 40 to 85 min [14, 148, 149]. The drug is marketed in Japan as an adjunct to levodopa to improve PD motor symptoms and motor fluctuations.

Solengepras (also known as CVN424) is a G-protein coupled receptor 6 (GPR6) inverse agonist primarily targeting D2 receptor expressing medium spiny neurons in the striatum and inhibiting the indirect basal ganglia pathway [150, 151].

A proof-of-concept phase 2 study compared two doses (50 and 150 mg) of this first-in-class drug to placebo in PD patients on dopaminergic treatment with an OFF-time of at least 2 h [152]. Solengepras was well-tolerated and did not show a typical dopaminergic AE profile beyond nausea. Further, OFF-time was significantly reduced by 1.3 h with the higher dose of solengepras in a secondary endpoint analysis [152].

A phase 3 trial (NCT06553027) in patients with motor complications is ongoing. Recently published results of another phase 2 trial (NCT06006247) investigating the drug in early untreated PD did not show a significant difference regarding MDS-UPDRS part II+III score [153].

Drugs to Treat Non-motor Symptoms of PD

NMS are an integral part of the spectrum of clinical features in PD and a key driver of long-term disability [3]. Despite their major impact on PD-related quality of life [4, 5], efficacious drugs to treat NMS in PD are still limited in number and effect size. Table 3 lists clinically useful pharmacological options to treat NMS in PD [154] and also drugs with novel evidence from advanced stage RCTs.

Table 3.

Drugs to treat non-motor symptoms of PD

Non-motor symptom	Drug	Mode of action	Efficacy and practice implicationsa/phase of development
Neuropsychiatric symptoms
Depression	Pramipexol	D2 agonist	Efficacious and clinically useful
	Venlafaxine	Selective serotonin-norepinephrine reuptake inhibitor	Efficacious and clinically useful
	Nortriptyline	Monoamine reuptake inhibitor	Likely efficacious and possibly useful
	Desipramine	Monoamine reuptake inhibitor	Likely efficacious and possibly useful
Apathy	Rivastigmine	Acetylcholinesterase inhibitor	Efficacious and possibly useful
	Piribedilb	D2 agonist	Likely efficacious and possibly useful
Dementia	Rivastigmine	Acetylcholinesterase inhibitor	Efficacious and clinically useful
Psychosis	Clozapine	Serotonergic receptor modulation (5HT2A/2C antagonism, 5HT1A agonism)	Efficacious and clinically useful
	Pimavanserin	Serotonin 5-HT2A inverse agonist	Efficacious and clinically useful
Sleep and wakefulness
Insomnia	Rotigotine	D2 > D1 agonist	Likely efficacious and possibly useful
	Apomorphine nighttime subcutaneous infusion	D1 and D2 agonist	Positive phase 4 trial positive [165]
Autonomic dysfunction
Orthostatic hypotension	Droxidopa	Norepinephrine precursor	Efficacious and possibly useful
Erectile dysfunction	Sildenafil	PDE5 inhibitor	Efficacious and clinically useful
Constipation	Probiotics and prebiotic fiber	Correction of gut dysbiosis	Efficacious and clinically useful
	Macrogol	Osmotic agent	Likely efficacious and possibly useful
	Lubiprostone	Intestinal chloride secretagogue	Likely efficacious and possibly useful
	ENT-01 (squalamine phosphate)	α-synuclein aggregation inhibitor at the GIT	Phase 2b trial positive [166]
Anorexia, nausea and vomiting associated with levodopa and/or dopamine agonist treatment	Domperidone	Peripheral D2 antagonist	Likely efficacious and possibly useful
Sialorrhea	Botulinum toxin B	Inhibitor of peripheral cholinergic transmission by cleavage of SNARE proteins	Efficacious and clinically useful
	Botulinum toxin A		Efficacious and clinically useful
	Glycopyrrolate	Muscarinic antagonist	Efficacious and possibly useful
Urinary frequency, urgency, and/or urge incontinence	Mirabegrone	β3 adrenoreceptor agonist	2 phase 2 trials positive [167, 168]
Other
Fatigue	Rasagiline	MAO-B inhibitor	Efficacious and possibly useful

^aEstablished drugs are included based on efficacy and practice implications of the MDS evidence-based medicine review on treatments for non-motor symptoms [154], and investigational drugs in active development with at least one phase 2 randomized controlled trial with a positive primary outcome are listed

^bRecommendation only for PD patients after subthalamic nucleus deep-brain stimulation implantation [154]

D1 (2) dopamine D1 (2) receptor, GIT gastrointestinal tract, MAO-B monoamine oxidase B, PDE5 phosphodiesterase 5, SNARE soluble N-ethylmaleimide-sensitive-factor attachment receptor, 5-HT 5-hydroxytryptamine

No new data from RCTs or any advanced-stage drug development programmes were identified for the treatment of anxiety, erectile dysfunction and fatigue in PD. Several development programmes for the treatment of excessive daytime sleepiness [155, 156], orthostatic hypotension [157], and cognitive impairment [158, 159] were discontinued.

Intravenous ketamine may be effective in treating major depression [160], and this NMDA receptor antagonist is under investigation in two phase 2 trials in participants with PD and depression (NCT04944017, NCT06231563), but results are not yet available. Psilocybin, which is found in psilocybe mushrooms and primarily acts as a serotonin 5-HT2A receptor agonist, has recently received increased interest as a treatment of major depression [161, 162] and an ongoing phase 2 RCT is investigating its possible efficacy also in patients with PD depression (NCT06455293).

A placebo-controlled phase 2 RCT testing pimavanserin in 117 PD patients with moderately severe ICDs has been completed but results have not yet been made public (PIMPARK study, NCT03947216).

There have been no novel approvals to treat cognitive impairment in PD and PD dementia within the last 5 years. A placebo-controlled RCT investigated TAK-071, a muscarinic acetylcholine M1 PAM, as a treatment to improve gait function in 54 participants with PD with cognitive impairment assessed a cognitive composite score as a secondary endpoint and found a significant improvement [163]. Intranasal insulin has been investigated in a phase 2 trial in 31 PD participants (NCT04251585) showing acceptable safety but with efficacy data regarding cognitive endpoints yet to be published [164], and a combination of intranasal insulin and glutathione is currently investigated to improve verbal fluency in a phase 2 trial in PD subjects (NCT05266417).

Sleep disturbances were significantly improved by nighttime apomorphine infusion in a double-blind, placebo-controlled, crossover RCT in 46 participants with PD and insomnia [165].

ENT-01 (squalamine phosphate) inhibits α-synuclein aggregation in the gastrointestinal tract and significantly increased the number of complete spontaneous bowel movements per week after 25 days on ENT-01 treatment in a double-blind, placebo-controlled, phase 2b RCT in 150 participants with PD and constipation however at the expense of gastrointestinal AEs in up to one third of participants [166].

Mirabegrone was shown to reduce overactive bladder (OAB) symptoms after 8 weeks [167] respectively 12 weeks [168] from baseline with an acceptable risk profile in two phase 2 trials both including more than 100 participants with PD.

Botulinumtoxin injections into the salivary glands are efficacious in reducing sialorrhea in PD and different preparations have received regulatory approval in the US, EU and most recently also in Japan [169–172].

Cannabinoid use is popular among PD patients and a placebo-controlled randomised withdrawal trial of the synthetic tetrahydrocannabinol analogue nabilone showed a positive effect on MDS-UDPRS part I sum scores probably driven by beneficial effects on sleeping problems and anxious mood [173]. Overall, there is still limited high-quality evidence to conclude on the efficacy of cannabinoids in the treatment of NMS in PD [174].

Disease-Modifying Treatments

Despite the rich portfolio of symptomatic approaches to treat the cardinal motor features of PD as well as levodopa-related motor complications there are still major unmet needs in the pharmacological management of progressive motor and non-motor symptoms in the advanced stages of disease [175]. A multitude of clinical trials testing a wide variety of different drugs over the past 2 decades have revealed no agent capable of slowing the progression of PD and there are multiple reasons why this might be the case [176, 177]. Tables 4, 5, 6 and 7 provide a non-exhaustive list of agents grouped by their molecular pathway targets. Although none of these approaches has so far provided conclusive evidence for disease-modifying efficacy, 3 target areas have progressed to phase 3 trials with signals for efficacy in some of these studies.

Table 4.

Recent and ongoing disease modification trials in Parkinson’s disease: GLP1 agonists

Drug class/substance	Trial	Phase	Target population (n)	Duration	Primary outcome	Efficacy results	Safety results
Exenatide	Aviles-Olmos et al., 2013 [188]	2	Moderately advanced PD with H&Y stage 2 to 2.5 (45)	12 months + 2-month washout	Change in MDS-UPDRS Part III OFF-state at 12 and 14 months	MDS-UPDRS III OFF scores at 12 / 14 months improved by 2.7 / 1.7 points in the exenatide group and worsened by 2.2 / 2.8 points in the placebo group (adjusted mean difference- 4·9 / 4.4 points; p=0·032 / p=0.044)	GI symptoms common (nausea, weight loss)
	EXENATIDE-PD Athauda et al., 2017 [185]	2	Moderately advanced PD with H&Y stage <3 (62)	48 weeks + 12-week washout	Change in MDS-UPDRS Part III OFF-state at 60 weeks	MDS-UPDRS III OFF scores improved by 1.0 points in the exenatide group and worsened by 2.1 points in the placebo group (adjusted mean difference 3·5 points; p=0·032), but multiple motor, non-motor and secondary outcomes negative	Injection site reactions and GI symptoms common (nausea, weight loss)
	Exenatide-PD3 Vijiaratnam et al., 2025 [187]	3	Moderately advanced PD with H&Y stage <3 (194)	96 weeks	Change in MDS-UPDRS Part III OFF-state at 96 weeks	Negative; MDS-UPDRS III OFF scores worsened by 5.7 points in the exenatide group and by 4.5 points in the placebo group (p=0.47) and no change between groups in multiple motor, non-motor and QoL related secondary outcomes	Acceptable safety profile; GI AEs more common in the exenatide group
Pegylated exenatide (NLY01)	McGarry et al., 2024 [183]	2	Early untreated PD (255)	36 weeks	Change in MDS-UPDRS Parts II and III	Negative; MDS-UPDRS III OFF-medication scores worsened by 5.2–5.9 points in all groups (p>0.7); secondary outcomes negative	GI AEs common (61–75%) particularly nausea (39–58%); overall well tolerated
Liraglutide	Wu et al., 2022a [186] NCT02953665	2	Treated PD (63)	54 weeks	Change from in MDS-UPDRS III, NMSS, and MDRS-2	No between group difference in change in MDS UPDRS III and MADRS-2, but NMSS scores improved by 6.6 points in the liraglutide group and worsened by 6.5 points in the placebo group. Also, secondary outcomes positive (change in MDS-UPDRS total, MDS-UPDRS II and PDQ-39 scores)	Injection site reactions and GI symptoms common
Lixisenatide	LIXIPARK Meissner et al., 2024 [184]	2	Early PD <3 years disease duration (156)	12 months (+2-month lixisenatide washout)	Change in MDS-UPDRS Part III ON-state at 12 months	MDS-UPDRS III on medication scores unchanged in the lixisenatide group and worsened by 3 points in the placebo group (p=0·007); MDS-UPDRS Part III OFF scores 3 points higher after 14 months in the placebo group, other secondary outcomes negative	Nausea (46%), vomiting (13%); no major safety concerns

This table provides a non-exhaustive list of agents investigated for disease-modification in PD grouped by their molecular pathway targets

AE adverse event, GI gastrointestinal, H&Y Hoehn and Yahr, MDRS-2 Mattis Dementia Rating Scale, MDS-UPDRS Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale, NMSS Non-Motor Symptoms Scale, PD Parkinson’s disease, PDQ-39 Parkinson's Disease Questionnaire, QoL quality of life

^aUp to now results of this trial have only been reported in abstract form

Table 5.

Recent and ongoing disease modification trials in parkinson’s disease: pathway-specific approaches—GBA pathway modulation and inhibition of LRRK2

Drug class/substance	Trial	Phase	Target population (n)	Duration	Primary outcome	Efficacy results	Safety results
GCase enhancers
Ambroxol	ASPro-PD (NCT05778617)	3	GBA1 variant positive and negative PD (1:1; 330)	60 weeks	Change in MDS-UPDRS I to III score	Trial ongoing	-
	GREAT (NCT05830396)	3	GBA1 associated PD (80)	104 weeks	Change in MDS-UPDRS III	Trial ongoing
	AMBITIOUS (Colucci et al., 2023 [219]; NCT05287503)	2	GBA1 associated PD with disease durations >5 years and H&Y stage ≤4 in ON-phase (60)	52 weeks	Changes in the MoCA score and frequency of MCI / dementia	Trial ongoing
	Silveira et al., 2025 [220]	2	PD >1 year before mild to moderate dementia	52 weeks	Safety, ADAS-Cog-13 and CGIC	No group difference in primary outcomes	High dose ambroxol was safe and well-tolerated, with GI symptoms as most frequent AEs (23%)
BIA28-6156	ACTIVATE (NCT05819359)	2	GBA1 associated PD with disease durations <5 years and H&Y stage <3 (237)	78 weeks	Time to change; ≥2-point increase in MDS-UPDRS II and ≥0-point increase in MDS-UPDRS Part III	Trial ongoing
GCase substrate inhibition
Venglustat	Giladi et al., 2023 [195]	2	GBA1 associated PD with H&Y stage ≤2 (221)	52 weeks	Change in MDS-UPDRS Parts II and III OFF-state	No difference in MDS-UPDRS II and III increment in the two groups	GI AEs most common; overall well tolerated
LRRK2 inhibitors
BIIB122 (DNL151)	BEACON (NCT06602193)	2a	LRRK2 PD (50)	12 weeks	Safety	Trial ongoing
	LUMA (NCT05348785)	2b	PD with disease duration <2 years and H&Y stage 1–2 (650)	48 – 144 weeks	Time to worsening on combined MDS-UPDRS 2 and 3	Trial ongoing
NEU-411	NEULARK (NCT06680830)	2	Early PD with H&Y stage 1–2.5 (150)	52 weeks	Change from baseline in the Roche digital biomarker score	Trial ongoing

This table provides a non-exhaustive list of agents investigated for disease-modification in PD grouped by their molecular pathway targets

ADAS-Cog-13 Alzheimer Disease Assessment Scale-cognitive subscale, version 13, AE adverse event, GBA1 GBA1 gene, GCase glucocerebrosidase, GI gastrointestinal, H&Y Hoehn and Yahr, LRRK2 leucine‐rich repeat kinase 2, MCI mild cognitive impairment, MDS-UPDRS Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale, MoCA Montreal Cognitive Assessment, PD Parkinson’s disease, QoL quality of life

Table 6.

Recent and ongoing disease modification trials in Parkinson’s disease: α-synuclein based approaches

Drug class/substance	Trial	Phase	Target population (n)	Duration	Primary outcome	Efficacy results	Safety results
Reducing α-synuclein synthesis (small molecule translational inhibition)
Buntanetap	Fang et al., 2023 [221]	1/2	PD with H&Y< 4 (58), AD with CDR 0.5 or 1 (17)	25 +/- 2 days	Safety; Cmax as secondary endpoint		Well tolerated
	NCT05357989	3	PD with H&Y< 4 (523)	6 months	Change From Baseline to Month 6 in MDS-UPDRS Part II in OFF-state	Recently completed, no results published (abstract/press release [222, 223]: negative primary endpoint, but no cognitive deterioration in verum groups and positive signals in subgroup with mild dementia (MMSE 20–26) in terms of motor and non-motor deterioration)
Reducing α-synuclein aggregation (tyrosine kinase inhibition)
Nilotinib	PD-Nilotinib Pagan et al., 2020 [224]	2	Stable treated PD (75)	12 months plus 3-month washout period.	Safety and target engagement	Nilotinib altered exploratory CSF biomarkers, including brain dopamine turnover, oligomeric α-synuclein, and hyperphosphorylated tau	Reasonable safety, asymptomatic, dose-dependent elevations of amylase and/or lipase
	NILO-PD Simuni et al., 2021 [203]	2a	Stable treated PD (76)	6 + 2 months	Safety and tolerability; change in MDS-UPDRS as secondary endpoint (2 months washout)	Negative (low CSF penetration, lack of biomarkers effect, and MDS-UPDRS change trending in the negative direction
Vodobatinib	PROSEEK (NCT03655236)	2	Early untreated PD H&Y stage ≤ 2 (513)	40 weeks	Change in MDS-UPDRS III	Not available
Radotinib	NCT04691661	2	Early untreated PD H&Y stage ≤ 2.5 (40)	12 months	Safety; change in MDS-UPDRS as secondary endpoint	Trial ongoing
Risvodetinib	NCT05424276 (Werner et al., 2025a [204])	2	Early untreated PD (137)	12 weeks	Safety; change in MDS-UPDRS as secondary endpoint	Safe and well tolerated, functional assessments as secondary outcomes trended toward clinical benefit
Reducing α-synuclein aggregation (small molecule aggregation inhibitor)
Minzasolmin	ORCHESTRA (NCT04658186; Carson et al., 2025a [205, 206])	2	Early untreated PD, H&Y stage ≤2.5 (450)	12 to 18 months	Change in MDS UPDRS I-III	No significant difference from placebo in MDS-UPDRS change from baseline	Well tolerated
Passive α-synuclein Immunization (monoclonal α-synuclein antibody)
Prasinezumab	PASADENA (Pagano et al., 2022 [207])	2	Early PD with H&Y stage ≤2 (316)	52 weeks	Change from baseline to week 52 in the MDS-UPDRS total score	No between group difference in MDS-UPDRS change; and no difference in DAT-SPECT signal decrease over 52 weeks (secondary endpoint)	Infusion reactions as most frequent AEs
	PADOVA (NCT04777331; Nikolcheva et al., 2025a [211, 212])	2b	Early PD on stable symptomatic monotherapy with OFF-medication state H&Y stage ≤2 (586)	76 weeks	Time to ≥ 5-point increase in MDS-UPDRS Part III score	Numerically longer but statistically non-significant delay in time to motor progression in favour of prasinezumab
	PARAISO (NCT07174310)	3	PD on monotherapy with OFF-medication state H&Y stage ≤2 (900)	104 weeks	Time to ≥ 7-point in MDS-UPDRS Part III score	Trial ongoing
Cinpanemab	SPARK (Lang et al., 2022 [208])	2	Early PD with H&Y stage ≤2 (357)	52 and 72 weeks	Change from baseline to week 52 (and 72 for the active-treatment dose-blinded extension phase) in the MDS-UPDRS total score	No between group difference in MDS-UPDRS change; and no difference in DAT-SPECT signal decrease over 52 weeks (secondary endpoint)
Exidavnemab	EXIST (NCT06671938)	2a	PD with H&Y stage ≤2.5 (24) and MSA with H&Y stage ≤3 (12)	26 weeks	Safety, tolerability, and pharmacokinetics of multiple ascending dosing of exidavnemab	Trial ongoing
Active α-synuclein Immunization
PD01A	Volc et al., 2020 [215]	1	Early treated PD (32)	128 weeks	Safety and immunogenicity	Safe and well tolerated with fatigue, headache, and myalgia as AEs; substantial humoral immune response shown
PD03A	Poewe et al., 2021 [216]	1	Early treated PD (36)	52 weeks	Safety and immunogenicity	Safe and well tolerated with mild injections site reactions; substantial humoral immune response shown
UB-312	Yu et al., 2022 [213]; Eijsvogel et al., 2024 [214]	1	PD on stable treatment (50)	44 weeks	Safety, tolerability, and immunogenicity	Safe and well tolerated with mild injections site reactions
ACI-7104	VacSYn (NCT06015841)	2	Early treated PD (150)	74 weeks	Safety and target engagement	Trial ongoing

This table provides a non-exhaustive list of agents investigated for disease-modification in PD grouped by their molecular pathway targets

^aUp to now results of these trials have only been reported in abstract form

AD Alzheimer’s disease, AE adverse event, CDR Clinical Dementia Rating scale, CSF cerebrospinal fluid, DAT-SPECT dopamine transporter single photon emission computed tomography, H&Y Hoehn and Yahr, MDS-UPDRS Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale, PD Parkinson’s disease

Table 7.

Recent and ongoing disease modification trials in Parkinson’s disease: other approaches

Drug class/substance	Trial	Phase	Target population (n)	Duration	Primary outcome	Efficacy results	Safety results
Nicotinamide	NOPARK (NCT03568968)	3	Early PD (400)	52 weeks	MDS-UPDRS Total Score (sum of parts I, II, and III)	Recently completed, no results reported	–
BHV-8000	NCT06976268	2/3	Early untreated PD (550)	48 weeks	Time to prespecified worsening on MDS-UPDRS Part II	Trial ongoing
Telmisartan, terazosin, and ursodeoxycholic acid (Multi-Arm, Multi-Stage)	EJS ACT-PD (NCT07207057) [225]	3	Treated PD without dementia or orthostatic hypotension (1200)	36 months	MDS-UPDRS I and II combined	Trial ongoing

This table provides a non-exhaustive list of agents investigated for disease-modification in PD grouped by their molecular pathway targets

MDS-UPDRS Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale, PD Parkinson’s disease

GLP1 Agonists

Consistent findings of an increased risk of PD in individuals with type 2 diabetes [178] and risk reduction in those treated with certain antidiabetic drug classes including dipeptidyl peptidase-4 (DPP4) inhibitors and glucagon-like peptide 1 (GLP1) receptor agonists [179–181] have moved insulin signaling into the focus of PD research as a target for disease-modifying intervention. Experimental studies had shown a role of insulin in neuronal survival and synaptic maintenance and there is evidence for insulin resistance in PD brains [182]. To date 6 randomised placebo-controlled trials have investigated the efficacy and safety of GLP1 agonists in early to moderately advanced PD (Table 4; [183–188]).

An initial proof-of-concept phase 2 trial provided first disease-modifying efficacy signals of exenatide (see Table 4) but was limited by its single-blind design [188]. The following phase 2 placebo-controlled trial of exenatide included 60 patients with moderately advanced PD and found significantly less worsening of motor symptoms as assessed by MDS-UPDRS part III scores in the practically defined OFF-condition (− 3.5 points difference vs placebo) at week 48 [185]. Similarly, positive results have been reported by Meissner and colleagues in a 12-month placebo-controlled trial of lixisenatide in 156 early PD patients on stable symptomatic therapy [184]. Using MDS-UPDRS part III scores in the ON-condition at month 12 as primary outcome the trial found a statistically significant difference in favour of lixisenatide of 3 score points, but there was no significant difference on any of the secondary efficacy outcomes including MDS-UPDRS part II scores. Nausea and vomiting were the most common AEs affecting 46% and 13% of lixisenatide treated patients, respectively and 8% had weight loss [184].

While these 2 trials have generated considerable optimism about a future role of GLP1 agonists as disease-modifying both in the PD research and patient community, a recent placebo controlled phase 3 trial of exenatide in 215 subjects with early PD on symptomatic therapy failed to meet its primary endpoint and found no significant difference in MDS-UPDRS part III OFF-scores at 96 weeks [187]. Another placebo-controlled trial of 2 doses of a pegylated formulation of exenatide (NLY01) included 255 early PD patients on symptomatic therapy and found no signals of efficacy on motor or non-motor symptoms after 36 weeks of treatment [183]. Differences between the GLP1 receptor agonists and brain penetrance as well as heterogenous response to these drugs may account for the inconsistency of clinical outcomes in PD [189].

Pathway-Specific Approaches in Genetic PD

GBA1-variant associated PD is the most common genetic PD subtype and overall has an earlier age of onset, faster progression and greater risk for cognitive decline as well as generally greater NMS burden [190]. The key pathophysiological downstream effect of GBA mutations is decreased glucocerebrosidase (GCase) activity and lysosomal dysfunction and experimental studies have shown a reciprocal relationship of GCase activity and α-synuclein aggregation [191, 192].

A number of approaches targeting deficient GCase activity in GBA-PD have reached the stage of clinical phase 2/3 trials including orally active GCase chaperones to enhance enzymatic activity, substrate reduction therapies modelling current treatment of Gaucher’s disease and GBA gene therapy (Table 5). So far, a small open-label trial of the GCase enhancer ambroxol has provided evidence for CSF penetration and target engagement as well as safety of high oral doses [193] and a large phase 3 randomised placebo-controlled trial is currently recruiting in the UK (NCT05778617). BIA-28-6156 is an allosteric modulator of GCase enhancing its activity and is currently studied in a phase 2/3 placebo-controlled trial [194]. Venglustat, a brain penetrant small molecule inhibitor of glycosylceramide synthase has failed to show an effect on progression of MDS-UPDRS part II+III scores after 52 weeks of treatment in a double-blind, placebo-controlled trial despite evidence for target engagement, i.e. decreased CSF levels of glycosylceramide [195].

Leucine‐rich repeat kinase 2 (LRRK2) mutations were first identified in 2004 to cause autosomal dominant PD [196, 197]. The G2019S mutation is the most common variant in populations of European ancestry with an overall prevalence of 4% in familial cases, but much higher rates in the Ashkenazi Jewish population (10% of sporadic and 28% of familial cases; [198, 199]). Disease-causing LRRK2 variants are characterised by a gain of kinase function including phosphorylation of Ras-associated binding (Rab) proteins.

Currently, most targeted treatments in development for LRRK2-associated PD are LRRK2 kinase inhibitors, other approaches include inhibition of GTPase activity as an indirect means of controlling LRRK2 activity, and the development of ‘LRRK2 degraders’ using antisense oligonucleotide and the proteolysis targeting chimeras (PROTAC) technologies [200]. The LRRK2 inhibitor DNL201 (BIIB122) has shown target engagement and satisfactory safety in phase 1/2 studies [201, 202] and is currently tested in a phase 2 trial enrolling PD patients with LRRK2 mutations (BEACON; NCT06602193). LUMA is a phase 2b trial and aims to investigate the efficacy of BIIB 122 in PD patient regardless of their LRRK2 status (NCT05348785). Six hundred and fifty participants with early stage PD have been randomised to either 225 mg of oral BIIB122 or placebo once daily for 48 to 144 weeks. Progression is determined using the MDS-UPDRS parts II and III and results are expected in 2026/27. Another LRRK2 inhibitor NEU-411 is currently investigated in a phase 2 trial (NEULARK, NCT06680830, see Table 5)

Targeting α-Synuclein Pathology

Based on the robust evidence for a central role of intraneuronal misfolding, oligomerization and aggregation of α-synuclein as well as cell-to-cell transmission of pathological α-synuclein species these mechanisms have become key targets for drug development programmes focused on disease modification in PD. Approaches range from reducing α-synuclein synthesis by antisense oligonucleotides, small molecule inhibitors of α-synuclein misfolding and polymerization, c-Abl kinase inhibitors and—most prominently—passive and active immunotherapies (Fig. 3).

Fig. 3.

α-synuclein targeting disease-modifying trials in PD (simplified scheme). * Buntanetap does not only suppress the translation of the mRNA of α-synuclein, but also other neurotoxic aggregating proteins. c-ABL1 Abelson murine leukaemia viral oncogene homologue 1, DNA deoxyribonucleic acid, mRNA messenger ribonucleic acid, PD Parkinson’s disease. Created in https://BioRender.com

C-ABL Inhibitors

So far trials with Abelson murine leukaemia viral oncogene homologue 1 (c-ABL) kinase inhibitors have not been successful with two randomised placebo-controlled trials with nilitonib and vodobatinib both failing to detect efficacy signals on the change from baseline of MDS-UPDRS part III scores [203]. A phase 2a placebo-controlled trial in 137 early and untreated PD subjects has recently reported satisfactory safety, clinical (MDS-UPDRS part II) and biomarker (quantitative a-synuclein immunohistochemistry) outcomes trending in favour of risvodetinib but with mostly statistically non-significant differences from placebo [204]. A trial of the small molecule inhibitor of α-synuclein aggregation minzasolmin also failed to detect differences from placebo on the primary outcome of change from baseline in MDS-UPDRS scores [205, 206].

Passive Immunotherapy

Two placebo-controlled phase 2 trials of monoclonal anti-α-synuclein antibodies (prasinezumab and cinpanemab) have also failed to meet their primary endpoints of change from baseline to week 52 or 76 in combined MDS-UPDRS part I + II + III scores [207, 208]. The prasinezumab trial (PASADENA), however, showed significantly slower decline in MDS-UPDRS part III scores (a key secondary endpoint) at week 52 as well as consistent results in exploratory digital motor measures [207]. A post-hoc subgroup analysis of that trial suggested benefit of prasinezumab in subjects with faster progression [209] and a comparison of MDS-UPDRS progression during the 4-year open-label extension phase with historical controls showed divergent trajectories in favour of prasinezumab from year 2 of treatment onwards [210].

Most recently another controlled trial of prasinezumab (PADOVA), which had enrolled subjects with more advanced disease compared to the initial trial and randomised subjects to 4-weekly iv injections of 1500 mg prasinezumab or placebo for at least 76 weeks, used time to ≥ 5-point worsening on MDS-UPDRS part III scores as primary outcome measure. To date results have only been reported in abstract form [211, 212] and showed numerically longer delays in favour of prasinezumab (61.1 vs 49.7 weeks) but the HR of 0.86 missed statistical significance (p = 0.066) and only became significant in supplementary adjusted analysis. A subgroup analysis of subject treated with levodopa at baseline (approximately 75% of subjects in the trial) did show significantly longer delays to pre-defined motor worsening in prasinezumab vs placebo-treated subjects (64.4 vs 48.6 weeks; p = 0.043) [211]. A phase 3 trial of prasinezumab has been initiated with a primary outcome of time to ≥ 7-point increase in the MDS-UPDRS motor examination (PARAISO, NCT07174310).

The monoclonal anti-synuclein antibody exidavnemab is currently investigated in a phase 2a trial (NCT06671938, see Table 6)

Active Immunotherapy

Active immunization strategies are designed to provoke an immune response against oligomeric or insoluble forms of α-synuclein. They generally consist of polypeptides mimicking epitopes of α-synuclein and are conjugated to helper peptides. Different compounds such as UB-312 [213, 214], PD01A [215], and PD03A [216] have been developed and tested in Phase 1 trials. Intramuscular injection was generally well tolerated with mostly mild and transient adverse events, such as injection site rejections, and caused a measurable immune response. ACI-7104 is an optimised formulation of the PD0 vaccines and is the first compound tested in a larger phase 2 trial (VacSYn; NCT06015841) with results expected in 2028.

Conclusions

PD stands out among the neurodegenerative diseases by the availability of highly efficacious symptomatic drug therapies to improve and maintain motor function.

Levodopa substitution is still the most effective approach to treat PD motor symptoms, but standard oral drug delivery is associated with plasma level oscillations that give rise to the development of motor fluctuations and dyskinesias. Novel levodopa formulations and delivery modes address these issues and include novel oral ER formulations as well as non-oral approaches like an inhalation powder for intrapulmonary delivery for on-demand treatment of OFF-episodes and soluble formulations for continuous subcutaneous infusion to control motor fluctuations.

Despite all recent progress in dopaminergic drug delivery optimising striatal dopamine substitution through non-invasive oral dopamine replacement strategies remains a priority. Novel dopamine agonist drugs include the D1/D5 agonist tavapadon, which has successfully completed phase 3 development as well as a variety of candidates in earlier development stages.

Amantadine remains the only non-dopaminergic drug combining anitiparkinsonian and antidyskinetic efficacy. With the exception of istradefylline (US, Japan) and zonisamide (Japan) no other non-dopaminergic drugs have been approved for clinical use in PD.

Late stage PD is characterised by increasing disability from drug-refractory motor and non-motor symptoms including gait and balance problems, impaired speech and swallowing, autonomic dysfunction, and cognitive decline for all of which there is currently little prospect of highly efficacious drug therapies. Targeting distinct pathogenetic pathways that drive disease progression has reached the stage of active drug development programs for disease modification with drugs targeting α-synuclein, GBA-1, LRRK-2 and GLP1 receptors comprising the largest group of recent and ongoing disease modification trials in PD. Several of these have progressed to phase 3 clinical trials which are either ongoing or have failed their primary outcomes. Recent efforts towards biological definitions of PD [217, 218] are expected to facilitate early and even pre-symptomatic diagnosis and are an important step on the path towards disease prevention trials. More than 200 years after James Parkinson’s famous essay on the Shaking Palsy it seems appropriate to echo his visionary statement made in the last chapter of his monograph of 1817: ‘…there appears to be sufficient reason for hoping that some remedial process may ere long be discovered by which, at least, the progress of the disease may be stopped’.

Funding

Open access funding provided by University of Innsbruck and Medical University of Innsbruck. No funding was received for the preparation of this manuscript.

Declarations

Conflicts of Interest

SL is supported by a grant from the MJFF grant ID MJFF- 009277, outside the submitted work. PM has no conflicts of interest regarding the submitted work. WP has received lecture fees and honoraria for consultancy in relation to clinical drug development programs from AbbVie, AC Immune, Alterity, BIAL, Boehringer, Britannia, Lilly, Eisai, Lundbeck, Roche, Takeda, Britannia, Eisai, Roche, Stada, and Zambon; grant support from The Michael J. Fox Foundation and the EU FP7 & Horizon 2020 programs; and safety monitoring board membership for UCB. He has leadership roles in the Movement Disorder Society, Austrian Society of Neurology, and Austrian PD Society.

Ethics Approval

Not applicable for this review manuscript.

Consent to participate

Not applicable.

Consent to publish

Not applicable.

Authors’ Contributions

SL, PM, and WP jointly wrote the first draft of the manuscript, SL created the illustrations. All authors contributed to revising the manuscript. All authors have read and approve the final version of the manuscript and agree to be accountable for the work.

Data availability

Data sharing not applicable to this article as no datasets were generated or analysed during the current review.

Code availability

Not applicable.