Levodopa-induced dyskinesia: do current clinical trials hold hope for future treatment?

KEYWORDS:

• AV-101
• befiradol
• buspirone
• CPL500036
• DSP-9632P
• dyskinesia
• JM-010
• ketamine
• lenrispodun
• levodopa
• mesdopetam
• Parkinson’s disease
• Tianqi Pingchan granule

1. Introduction

Whereas dopamine replacement therapy with levodopa remains the bedrock of Parkinson’s disease (PD) treatment, its use is almost invariably marred by the occurrence of abnormal involuntary movements, i.e. levodopa-induced dyskinesia [1]. Although dyskinesia is a frequent problem, treatment options remain relatively few, with only amantadine and clozapine being regarded as clinically efficacious by the International Parkinson and Movement Disorder Society [2]. However, not all patients are responsive or able to tolerate amantadine, while the use of clozapine is accompanied by life-long blood sampling to prevent potentially fatal agranulocytosis. As such, levodopa-induced dyskinesia very much remains an unmet medical need; hence, the on-going efforts to discover new therapies to alleviate this treatment-related complication.

In this article, we will briefly review some of the active clinical trials whose primary endpoints relate to levodopa-induced dyskinesia, followed by our modest point of view on whether these trials will result in novel therapies to be offered to patients with PD, or if they represent incremental steps in our search for a treatment that will suppress dyskinesia.

2. Molecules under active investigation in clinical trials for levodopa-induced dyskinesia

Searching through the clinicaltrials.gov and clinicaltrialsregister.eu databases (current as of 9 November 2023), we noted that there were relatively few active trials, with the following molecules being tested: ketamine (Phase II, NCT04912115, currently suspended and unclear future perspectives), JM-010 (Phase II, NCT04377945 and NCT03956979), mesdopetam (Phase III being planned, after a positive results in a Phase IIb trial, NCT04435431), DSP-9632P (Phase I recently completed, NCT05435729), Tianqi Pingchan granule (Phase ‘not applicable,’ unknown status, NCT04173832 and NCT04173845), AV-101 (Phase II, NCT04147949, not yet recruiting, and future plans for recruitment unclear), buspirone (Phase III, NCT02617017, recently completed), CPL500036 (Phase II, NCT05297201), befiradol (Phase II, NCT05148884, Phase IIb being planned), lenrispodun (Phase II, NCT05766813).

In terms of mechanisms of action, buspirone is a partial agonist at serotonin (5-HT) type 1A (5-HT_1A) receptors, but also has affinity for other targets, notably as a dopamine D₂ receptor antagonist [3]. DSP-9632P (transdermal tandospirone) is also a 5-HT_1A partial agonist [4], while befiradol is a potent and highly selective 5-HT_1A agonist [5]. JM-010 is a combination of buspirone and zolmitriptan, a 5-HT type 1B and 1D (5-HT_1B/1D) agonist [6]. For its part, ketamine presumably acts via a preferential N-methyl-D-aspartate (NMDA) antagonist mechanism, but the compound also exhibits affinity for several other targets, notably cholinergic receptors, 5-HT type 2A (5-HT_2A) and dopamine D₂ receptors, which makes its primary mechanism of action unclear [7]. AV-101 (L-4-chlorokynurenine) acts as an antagonist at the glycine-binding site on NMDA receptors [8]. Whereas the precise mechanism of Tianqi Pingchan granule is uncertain, it regulates the expression of the G protein-coupled receptor kinase 6 (GRK6) [9]. Mesdopetam is a dopamine D₃/D₂ antagonist with agonist-like binding properties [10]. CPL500036 is a selective phosphodiesterase 10A (PDE10A) inhibitor [11], while lenrispodun is a selective phosphodiesterase 1B (PDE1B) inhibitor [12]. It is worth mentioning that for some of these molecules, notably those acting at 5-HT_1A receptors, there is a large body of literature that support their investigation [13], whereas the same cannot be said for other compounds, e.g. Tianqi Pingchan granule or lenrispodun.

Regarding the potential of buspirone as an anti-dyskinetic agent, one must note that the drug has already entered a few clinical trials before NCT02617017, with variable effects on dyskinesia and a tendency to worsen parkinsonian disability [14]. Regarding JM-010, the concerns pertaining to the possible deleterious effect of buspirone on parkinsonism remain, while the effects of zolmitriptan are unclear, but a pre-clinical study suggests that caution might also be warranted. Thus, the 5-HT_1B/1D agonist SKF-99101-H hindered the anti-parkinsonian action of levodopa, in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-lesioned primate [15]. However, in the case of JM-010, it is possible, and that would be the ideal scenario, that additive effects of 5-HT_1A partial agonism and 5-HT_1B/1D agonism will allow to administer doses of each molecule that would be low enough to dissociate an anti-dyskinetic effect from a possible deleterious effect on parkinsonism, time will tell. The possibility of an exacerbation of parkinsonian disability with DSP-9632P is not insignificant, as tandospirone, while reducing dyskinesia in some patients, also hampered the anti-parkinsonian efficacy of levodopa, in a pilot clinical trial [16]. Whether befiradol will exacerbate parkinsonism remains to be seen, but unlike buspirone and tandospirone, befiradol is a full 5-HT_1A agonist, with high selectivity for its target and did not worsen parkinsonism in the 6-hydroxydopamine (6-OHDA)-lesioned rat [17] or the MPTP-lesioned primate [18,19].

Regarding the future perspectives of ketamine, we note that it previously reduced dyskinesia in a case-series [20], however with the NCT04912115 study being currently suspended, it is unclear whether further development will be pursued. Hurdles to the widespread use of ketamine for the treatment of dyskinesia in PD involve intra-venous (i.v.) administration (although an oral formulation is available, the studies in PD were conducted with i.v. injection paradigms) and a possible dissociative effect [21].

Mesdopetam is the first D₃ antagonist with agonist-like binding properties to be tested in clinical trials for PD. Mesdopetam exhibits low micromolar/sub-micromolar affinity for other targets, notably dopamine D₂, sigma-1, 5-HT_1A and 5-HT_2A receptors [10]. There is extensive pre-clinical literature suggesting that antagonising D₃ receptors alleviates dyskinesia, generally without hindering the effect of levodopa on parkinsonism [22]. It is noteworthy that mesdopetam failed to meet the Phase IIb study primary endpoint, which was to increase the number of daily on-time without troublesome dyskinesia, yet showed positive effects in the full analysis set population, notably a reduction of Unified Dyskinesia Rating Scale scores [23] and a Phase III is being planned.

AV-101 alleviated dyskinesia in the MPTP-lesioned primate [24] and a Phase II clinical trial with dyskinesia as primary endpoint is yet to begin recruiting. Whether the development of AV-101 for dyskinesia will be pursued is unclear, as the company seems to be developing the drug for pulmonary arterial hypertension, with a Phase III on-going (NCT05557942). Of note, in a pilot clinical trial, D-serine, which is an agonist at the glycine-binding site on NMDA receptors, increased on-time without troublesome dyskinesia [25], which is difficult to reconcile with the mechanism of action of AV-101 which, as mentioned above, antagonises the glycine-binding site on NMDA receptors.

CPL500036 and lenrispodun are phosphodiesterase inhibitors. Whereas there are pre-clinical data in parkinsonian primates and rats supporting a potential anti-dyskinetic effect with PDE10A inhibitors, data regarding PDE1B inhibition are rather scant in the peer-reviewed literature [26], and it is difficult to speculate on the efficacy or tolerability of these agents in the clinic. The results of the on-going clinical trials will provide invaluable data pertaining to the possible use of phosphodiesterase inhibition as a therapeutic strategy for levodopa-induced dyskinesia.

Lastly, as the status of the clinical studies with Tianqi Pingchan granule is unclear, we will refrain from commenting on the possible future role of this agent in the treatment of dyskinesia.

3. Expert opinion

Here, we have briefly listed the agents under investigation for the treatment of levodopa-induced dyskinesia, along with their mechanism(s) of action. We would like to draw a few conclusions. The first one is that there is a paucity of different mechanisms of action begin explored, with an emphasis on activation of 5-HT_1A receptors. As we have pointed out above, it is unclear if the 5-HT_1A partial agonists buspirone, DSP-9632P and, by extension, JM-010 will be either efficacious or well tolerated, as they have previously exacerbated parkinsonism in earlier clinical studies. The 5-HT_1A full agonist befiradol is different, however, in terms of potency and has so far been well tolerated in parkinsonian rodents, primates and PD patients. If befiradol was to fail in the upcoming Phase IIb trial in which it will be assessed, it would shed doubt on the future of 5-HT_1A agonism as a strategy to alleviate levodopa-induced dyskinesia. Outside of 5-HT_1A receptors, we are excited about the upcoming Phase III trial with mesdopetam, a D₃/D₂ dopamine antagonist with agonist-like binding properties. If positive, mesdopetam could become the first D₃ antagonist approved for the treatment of levodopa-induced dyskinesia. We nevertheless remain cautious, as it failed to meet its primary endpoint in a Phase IIb study. It is unclear whether the development of ketamine and AV-101 will be pursued for the treatment of dyskinesia, but we have highlighted some concerns in the previous section. Lastly, whereas promising, it may be too early to speculate on the possible anti-dyskinetic effect of CPL500036 and lenrispodun in the clinic, although an important body of pre-clinical literature on PDE10A inhibitors would tend to support optimism regarding CPL500036.

The second point we would like to draw is that, while dyskinesia is a complex phenomenon involving a wealth of neurochemical anomalies [27], with few exceptions, most molecules being investigated harbour high selectivity for their primary target. This is in contrast with amantadine [28] and clozapine [29], which are both multi-target drugs. To this day, it is unclear if the best way to tackle dyskinesia is through a highly selective approach focussed on a selected target, or if aiming to act at several pharmacological targets would yield superior results.

The third point we would like to highlight is that sometimes, promising molecules are the unfortunate victims of unwarranted circumstances. For instance, the development of the metabotropic glutamate 5 (mGlu₅) negative allosteric modulator dipraglurant for levodopa-induced dyskinesia seems to have been ceased following the termination of a Phase IIb/III with dipraglurant (NCT04857359) due to slow enrollment. Development of molecules may also be discontinued for business or undisclosed reasons.

In summary, there are several active clinical trials that seek to alleviate levodopa-induced dyskinesia. Throughout the text, we have tried to show cautious optimism as to whether one of these candidates will be successful and will eventually be granted approval by regulatory authorities. It is noteworthy that, in the event that none of them were to be approved, the knowledge gained through these clinical studies would still be invaluable and will help moving the field of dyskinesia research forward, to one day put an end to dyskinesia. As such, to conclude and to answer the question in the title of this article, do current clinical trials hold hope for future treatment, yes we do believe so and are hopeful for the future.

Declaration of interest

P Huot has received payments from AbbVie, adMare BioInnovations, ConSynance Therapeutics, Neurodiem, Sanford Burnham Prebys, Sunovion, ConSynance Therapeutics and Throughline Strategy. P Huot has also had research support from Parkinson Canada, Parkinson Québec, Fonds de Recherche Québec – Santé, the Weston Brain Institute, the Michael J Fox Foundation for Parkinson’s Research, the Natural Sciences and Engineering Research Council of Canada, the Canadian Institutes of Health Research, the New Frontiers in Research Fund and Healthy Brains for Healthy Lives.

The authors have no other relevant affiliations or financial involvement with any organisation or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Authors contributions

J Al-Kassmy, M Alsalmi, M Palayew, W Kang wrote the manuscript. P Huot reviewed and commented on the manuscript.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.