Glucagon-like peptide-1 class drugs show clear protective effects in Parkinson ’ s and Alzheimer ’ s disease clinical trials: A revolution in the making?

Parkinson ’ s disease (PD) is a complex syndrome for which there is no disease-modifying treatment on the market. However, a group of drugs from the Glucagon-like peptide-1 (GLP-1) class have shown impressive improvements in clinical phase II trials. Exendin-4 (Bydureon), Liraglutide (Victoza, Saxenda) and Lixisenatide (Adlyxin), drugs that are on the market as treatments for diabetes, have shown clear effects in improving motor activity in patients with PD in phase II clinical trials. In addition, Liraglutide has shown improvement in cognition and brain shrinkage in a phase II trial in patients with Alzheimer disease (AD). Two phase III trials testing the GLP-1 drug semaglutide (Wegovy, Ozempic, Rybelsus) are ongoing. This perspective article will summarize the clinical re-sults obtained so far in this novel research area. We are at a crossroads where GLP-1 class drugs are emerging as a new treatment strategy for PD and for AD. Newer drugs that have been designed to enter the brain easier are being developed already show improved effects in preclinical studies compared with the older GLP-1 class drugs that had been developed to treat diabetes. The future looks bright for new treatments for AD and PD.


Introduction
Parkinson's and Alzheimer's disease (PD and AD) are complex neurodegenerative syndromes that have long resisted attempts to develop effective drug treatments.Mimetics of the peptide hormone Glucagon-Like Peptide − 1 (GLP-1) are on the market as treatments for diabetes.Bydureon), liraglutide (Victoza, Saxenda), and lixisenatide (Lyxumia, Adlyxin) have shown impressive protective effects in phase II clinical trials in PD patients, and liraglutide has shown clear effects in a phase II trial in AD patients (Athauda et al., 2017;Foltynie and Athauda, 2020;Hölscher, 2022b).These clinical results are a proof of concept that this approach is effective.In addition, a phase III trial testing exendin-4 in PD patients and two phase III trials testing the GLP-1 analogue semaglutide (Ozempic, Rybelsus, Wegovy) in AD patients are ongoing (Reich and Hölscher, 2022).These exciting developments demonstrate that a treatment for chronic neurodegenerative disorders could be very close to reality.
The reason for the failure of previous drug discovery projects is manifold, ranging from wrong hypotheses to misleading data from animal experiments.One key issue is that many approaches to treat AD or PD assume for each a monocausal origin of the disease.The aim is to 'fix' one biomarker that correlates with the disease and therefore must be the cause of the disease appears to be a common misconception that almost guarantees failure.We know that the pathologies that underlie PD and AD are complex and involve very disparate systems such as the chronic inflammation response, loss of energy utilization, loss of synaptic activity, impaired autophagy and responses to oxidative stress (Johnson et al., 2019;Reich and Hölscher, 2022).Any successful treatment will have to involve a multi-pronged approach that simultaneously improves many different physiological systems and underlying disease mechanisms.
One approach that has produced considerable success is that of the use of growth factors, or neurotrophics.Growth factors such as neuronal growth factor (NGF), brain-derived neurotrophic factor (BDNF) and glia-derived neurotrophic factor (GDNF) have shown convincing protective effects in preclinical studies of different neurodegenerative disorders (Allen et al., 2013;Nagahara and Tuszynski, 2011).The main stumbling block that prevented the translation into the clinic has been the fact that such growth factors do not cross the blood-brain barrier (BBB) (Covaceuszach et al., 2009;Whone et al., 2019).However, other growth factors exist that have neuroprotective effects and can cross the BBB.One example is the peptide hormone family that signals energy E-mail address: christian_holscher@mac.com.status and metabolism, including the incretin hormones Glucagon-Like Peptide-1 (GLP-1) and Glucose-dependent Insulinotropic Polypeptide (GIP) (Salameh et al., 2020;Yang et al., 2022).These peptide hormones play several roles in physiology, acting as a signal for the presence of energy-rich molecules such as glucose (Baggio and Drucker, 2007), as an anti-inflammatory cytokine modulating the inflammation response (Chowen et al., 1999;Iwai et al., 2006), and as a growth factor (Holscher, 2014;Perry and Greig, 2004).
Based on the impressive protective effects in different preclinical models and the approval for the use of some GLP-1 class drugs as a treatment for humans, numerous clinical trials have been conducted or are currently ongoing in AD and PD.The first clinical trial outcomes show clear neuroprotective and disease-modifying effects.

A pilot study of exendin-4 (byetta, exenatide) in PD patients
A first study of the potential protective effects of GLP-1 type drugs was an open label trial in PD patients, testing the GLP-1 receptor agonist exendin-4 (NCT01174810).Unfortunately, the group at UCL conducting the study could not get the funding and support to obtain a placebo for this trial.44 patients were tested, with 20 receiving the drug.All patients continued to take their routine PD medication such as L-DOPA.The patients were already advanced in the disease progression, as the average time from diagnosis was 10 years.Patients were given the drug for 12 months.After that, exendin-4 was no longer given, and further motor tests were conducted 3 months and 12 months later to test if the drug induced improvements are still visible.When giving L-DOPA, the improvements in motor activity will fade once the drug is used up.Therefore, it is of interest to see if any new drug can induce permanent changes that are long-lasting (disease-modifying) and not just temporary.Patients were tested using the standard MDS-UPDRS part 3 (Movement Disorder Society-Unified Parkinson's disease rating scale) test battery which evaluates the patients' ability to initiate and control movements.The Neurologist that assessed the patients was blind to the treatment given to the patients.Another assessment was the Mattis Dementia Rating Scale 2, a test battery to analyze cognitive impairments often found in late-stage PD (Matteau et al., 2011).
The drug effects were clear-cut.As shown in Fig. 1, motor control as assessed by the MDS-UPDRS part 3 test battery in the off state (no L-Dopa drug was given before the test) did not change at all in the drug group over the course of 2 years, while the control group deteriorated steadily over time as expected for this disease.This is even more impressive as the drug group only received exendin-4 for 12 months, and the improvement remained stable for another 12 months after drug treatment had stopped.In contrast, the control group that only took standard medication such as L-DOPA deteriorated continuously.The difference between groups in the Mattis cognitive assessment tests is even clearer.The drug group did not deteriorate at all, and if anything, improved slightly.The control group, however, deteriorated to a point where the Mattis dementia score 2 cannot measure any more deterioration (Llebaria et al., 2008).The patients had reached 'rock-bottom' in this parameter, so to speak.
The drug was well received and no major side effects were observed.Initial fears that patients may lose weight when given the drug proved to be unfounded.The caveat in this study Is that there was no placebo group.The scores in the MDS-UPDRS part 3 motor test battery are notoriously sensitive to the placebo effect (de la Fuente-Fernandez and Stoessl, 2002).The improvement seen in the drug group can be attributed to some degree to that effect.However, the protective drug effect remained stable for 12 months after stopping drug administration.It is hard to imagine how a placebo effect, a temporary upregulation of motor performance due to expectation, can continue for that long.The drug group did not deteriorate for 24 months, while the control group deteriorated significantly.If the placebo effect was that powerful, any treatment could protect PD patients from disease progression.That is clearly not the case.L-DOPA is a powerful drug that can reduce PD motor symptoms to a great extent, but does not affect disease progression.If the protective effect observed in this study was entirely due to the placebo effect, then it is not clear why the control group that continued to take its standard PD medication did not show this effect.More importantly, the stark difference in the Mattis dementia rating scores cannot be explained by a placebo effect.Deterioration of cognitive performance in dementia patients has not been shown to be sensitive to the placebo effect to such a degree.It is inconceivable how the placebo effect can protect the brain from the neurodegenerative processes that take place in the brains of PD patients and that spread from primarily motor areas to more cognitive and memory-oriented areas (Aarsland et al., 2017).The only rational explanation is that the drug treatment protected the brain, stopped disease progression, and stabilized motor performance.

A placebo-controlled phase II study of exendin-4 (bydureon) in PD patients
The positive outcome of the pilot study made it possible for the C. Hölscher research group at UCLto secure funding for a follow-up study, testing exendin-4 in PD patients in a placebo-controlled phase II study (NCT01971242).Here, 62 patients were enrolled and randomly assigned, 32 to the drug and 30 to the placebo group.Patients were given the once-weekly formulation of exendin-4 (Bydureon).The disease progression in these patients when they enrolled was not as long as in the pilot study, as the average time period after diagnoses was around 6 years in this phase II trial.Patients received drug or placebo for 48 weeks, and another test was conducted 12 weeks after drug treatment had stopped (Athauda et al., 2017).Fig. 2 shows that the drug group performed better than the placebo group.Even 12 weeks after drug delivery had stopped, the improvement in the drug group was still significant compared to placebo.CSF samples were analyzed and showed that the drug can cross the BBB, and that 12 weeks after stopping drug treatment, there was no drug present.This demonstrates that the drug has disease-modifying effects, and the improvement remains after the drug has been removed from the body.The outcome confirms the clear findings of the pilot study.As the patients had not progressed as much in the phase II study than in the pilot study, the deterioration of the placebo group was not as strong, making the result look less impressive.This does not mean the drug effects were weak.Non-motor symptoms, such as clinically evaluated 'emotional well-being' and mood ratings improved in drug-treated patients.Interestingly, cognitive performance improved as well, mirroring the results of the pilot study (Athauda et al., 2018).Therefore, the results from the pilot study cannot be simply dismissed as a placebo effect.

A phase II study of liraglutide (victoza, saxenda) in PD patients
Exendin-4 was the first GLP-1 receptor agonist brought to the market to treat diabetes.Several other GLP-1 class drugs are on the market (Tan et al., 2022).It was soon followed by a popular and effective drug, the GLP-1 analogue liraglutide (Victoza, Saxenda) (Knudsen and Lau, 2019).In a recently completed randomized, double-blind, placebo-controlled phase II clinical trial it was shown that liraglutide can improve PD pathology (NCT02953665).PD patients received once-daily subcutaneous injections of liraglutide for 54 weeks in addition to standard clinical PD medication such as L-Dopa.The trial included 37 patients in the drug group and 18 in the placebo group.Despite the relatively low number of patients the results were very clear-cut.The drug group showed a significant improvement in MDS-UPDRS part 2 scores, see Fig. 3.This test battery evaluates everyday activities such as walking, tremor, talking, eating/swallowing, getting dressed, personal hygiene activities, and more.The phase II trial testing Bydureon already showed improvements in these scores, but the outcome of this trial testing liraglutide was more pronounced.Furthermore, significant improvement in the overall Global MDS-UPDRS motor-assessment score was found.In addition, non-motor symptom scores had improved in the liraglutide group and worsened in the placebo group over the course of the trial.The improvements were mainly found in cognitive and memory tests, confirming the results of the exendin-4 pilot study.In addition, the PDQ-39 (quality of life) scores had improved (p < 0.001).The main parameters that contributed to this improvement were the ability to move better and a reduced stigma of having PD.Patients noticed that they were able to control their movements much better, and that people in the streets were noticing their motor impairments a lot less.In the MDS-UPDRS part 3 motor assessment, both groups improved during the course of the study, not just the drug group.The improvements in the placebo group are most likely due to the placebo effect.For the MDS-UPDRS part 3 assessment, patients have to visit the hospital for the assessment by a Neurologist.This may induce a placebo effect in the patients triggered by the attention given to them.In contrast, the MDS-UPDRS part 2 assessment and the other assessments were not performed in the clinic and happen retrospectively via report by the patients or care givers.Therefore, no placebo effect can take place in these assessments, and a clear difference between drug group and placebo group was observed (Hogg et al., 2023).

A phase II study of lixisenatide (lyxumia, adlyxin) in PD patients
The results of a phase II clinical trial has been reported, testing the GLP-1 receptor agonist lixisenatide (Christensen et al., 2009) in PD patients (NCT03439943).In a randomized, placebo-controlled, double-blind trial, 156 patients received drug or placebo for 12 months.Two months after drug treatment had stopped, patients were tested again to   3. Performance in the MDS-UPDRS part 2 motor assessment that measures everyday activities such as walking, chewing, getting dressed, talking, and other activities were significantly improved by liraglutide treatment The difference between groups at 54 weeks was p < 0.001.For details see (Hogg et al., 2023).
C. Hölscher evaluate if the drug effect wears off after discontinuation of treatment.A first report of the results states that the primary outcome change from baseline to end-point after 12 months in the MDS-UPDRS III motor tests was highly significant (p = 0.0068).Two months after treatment stopped, patients were tested again without drug and without L-Dopa, and the difference between groups was still significant (P < 0.05) (Meissner et al., 2024).The significant difference even after the drug had been eliminated from the body demonstrates a disease-modifying effect, very similar to the Bydureon trial results.The fact that L-Dopa was no longer in the system showed that the lixisenatide related improvement was not simply due to a facilitation of improved dopamine production by L-Dopa that may not be long-lasting.Other secondary read-outs were not significant.This is the third GLP-1 class drug that shows meaningful improvements in PD patients.

A phase II clinical trial testing liraglutide in patients with Alzheimer's disease
Due to the clear neuroprotective effects of liraglutide in preclinical trials (McClean et al., 2011), we were able to conduct a placebo-controlled double-blind phase II clinical trial testing liraglutide in 200 patients with AD (NCT01843075).Liraglutide or placebo was given daily for 12 months.The main measurements in this trial were cognition and memory (the ADASexec test battery) and changes in brain volume as measured in MRI brain scans (Femminella et al., 2020).We found that liraglutide significantly slowed down the deterioration in cognitive impairments, and found that brain temporal lobe volumes and parietal lobe volumes shrank less, and the total grey matter cortical volume shrank less in the liraglutide group compared to the placebo group as shown in MRI brain scans.This suggests that neuronal loss in the brain has been reduced by liraglutide (Edison et al., 2022(Edison et al., , 2023)).The effects of Liraglutide were relatively modest, but one has to keep in mind that this was the very first clinical phase II trial, testing a GLP-1 class drug in AD.The results are a proof of concept that GLP-1 receptor agonists can slow AD progression in a meaningful and significant way.The improvements in cognition in this trial confirm the cognitive improvements by exendin-4 and by liraglutide in the clinical trials in PD patients (see above).It is clear from these results that GLP-1 receptor agonists not only protect the brain from PD-related disease progression, but also from AD related disease progression.
Further clinical trials of GLP-1 receptor agonists in AD and PD patients are currently underway, a phase III clinical trial testing exendin-4 in PD patients (NCT04232969) is currently ongoing, and two phase III clinical trial testing semaglutide (Wegovy, Ozempic, Rybelsus) in AD patients are underway (NCT04777396 and NCT04777409).We have shown that semaglutide has neuroprotective effects in a 3xtg mouse model of AD (Wang et al., 2023).This demonstrates the increasing interest in this drug discovery area for novel treatments of AD and PD.

Novel dual GLP-1/GIP receptor agonists that can cross the blood-brain barrier easier
All GLP-1 receptor agonists tested so far have shown good effects in patients with AD or PD.However, these drugs have been designed to treat type 2 diabetes (Müller et al., 2022).The main target in this drug development is to enhance the survival time in the blood to maximize efficacy in treating diabetes (Dong et al., 2022;Lau et al., 2015).However, a long half-life in the blood means less of the drug will cross the blood-brain barrier (BBB) into the CNS.While exendin-4 has only a half-life of 2.4 h in the blood, semaglutide has a 165 h half-life (Dong et al., 2022), which means that less of the drug will enter the brain via the BBB (Gabery et al., 2020;Salameh et al., 2020;Salinas et al., 2018;Shi et al., 2017;Zhang et al., 2019Zhang et al., , 2021)).There is a correlation between the ability of a drug to cross the BBB and its efficacy in protecting the brain in AD or PD animal models (Holscher, 2014;Liu et al., 2015;Lv et al., 2021;Maskery et al., 2020;Yang et al., 2022;Zhang et al., 2020).
Importantly, NLY01, a modified version of exendin-4 that has a 40 kDa pegylation added to increase the survival time in the blood (Yun et al., 2018), but which does not enter the brain (Lv et al., 2021), did not show any effects in a phase II clinical trial in PD patients (NCT04154072) (McGarry et al., 2024).Since exendin-4 on its own can enter the brain easily (Salameh et al., 2020) and shows good effects in PD (Athauda et al., 2022), this demonstrates the need for drugs to enter the brain in order to be effective.

What is the underlying mechanism of GLP-1 drug action?
It appears that a lot of people in the field are unsure what to make of the results of these clinical studies, even though quite a lot of research on the underlying mechanisms has been conducted in this area.Since the role and function of GLP-1 signalling in the brain is still unknown to most Neuroscientists and Neurologists, the main question often is how does GLP-1 achieve this impressive result, what is the underlying mechanism?We actually know a lot about the molecular processes that are activated by GLP-1 in neurons.Cell culture and animal studies demonstrate that when activating the GLP-1 or GIP receptors in the brain, energy utilization in neurons is normalized, cell repair and gene expression returns to normal, synaptic activity is normalized, the chronic inflammation response in the brain is reduced, and more (Hölscher, 2022a;Reich and Hölscher, 2022).This is due to the loss of growth factor signalling, in particular insulin.Insulin is an important growth factor in the brain, and previous studies have shown that insulin signalling is impaired in the brains of PD patients (Bassil et al., 2022;Cheong et al., 2020;Morris et al., 2011).GLP-1 is a growth factor as well and can compensate for the loss of insulin signalling, and can re-sensitize insulin signalling in the brain (Long-Smith et al., 2013;Yassine et al., 2022).In diabetes patients, GLP-1 mimetics can improve insulin sensitization (Doyle and Egan, 2003;Müller et al., 2019;Yang et al., 2022) and in patients with AD or PD, they can, too (Athauda et al., 2022;Cheong et al., 2020;Long-Smith et al., 2013;Talbot, 2014;Talbot and Wang, 2023).There are several other mechanisms, such as the reduction of the chronic inflammation response found in the brains of PD patients (Hirsch and Hunot, 2009;Orr et al., 2002).GLP-1 mimetics such as exendin-4, liraglutide and lixisenatide have shown good effects in reducing the inflammation response (Kim et al., 2009;Liu et al., 2015;Parthsarathy and Holscher, 2013;Solmaz et al., 2015;Yang et al., 2022).Other mechanisms of action include improvement of mitochondrial activity, synaptic activity, normalization of autophagy to remove misfolded proteins, and reduction of oxidative stress, see (Hölscher, 2022a(Hölscher, , 2022b;;Reich and Hölscher, 2022) for comprehensive reviews.Therefore, it is important to note that there is no single mechanism of action that underlies the improvements observed in the brain.Fig. 4 gives an overview of the main processes that underlie the neuroprotective effects.
Are the same mechanisms of action seen in animal models responsible for the improvements observed in PD patients in the clinical trials of GLP-1 type drugs, too?In order to address this, blood samples from the patients in the phase 2 trial had been taken.In these samples, exosomes can be found that originate from the CNS.Exosomes are vesicles that are released from cells for different purposes, eg.removal of cell junk such as misfolded proteins, signalling to other cells, or even transport of mRNA to target cells (Dutta et al., 2023;Meldolesi, 2021;Vandendriessche et al., 2020).In an analysis of exosome content, patient samples showed that the exosomes from the drug group contained biomarkers of improved insulin signalling in the brain, suggesting that the drug normalized insulin signalling (Athauda et al., 2019;Mustapic et al., 2019).Therefore, normalizing insulin signalling may be one of the key mechanisms of stopping PD progression.

Conclusion
GLP-1 class drugs that are on the market to treat diabetes showed clear neuroprotective effects in first clinical trials in AD and PD patients.This is a clear proof of concept that this strategy is viable, and that extensive drug discovery research is warranted in this area.Drugs designed to enter the brain may be more effective than the older drugs that had been designed to treat diabetes.
Currently, there are no drugs available to treat PD that can modify disease progression.The research in this area seems to focus mainly on the theory of misfolded proteins and on the protein alpha-synuclein that appears to aggregate in some, but not all patients with PD.
All clinical trials that tested the effects of antibodies directed against alpha-synuclein showed a complete lack of improvements in PD patients.The antibody Prasinezumab had been tested (Roche and Prothena) and showed no effect (Pagano et al., 2022).Cinpanemab (Biogen) had been shelved in 2021, due to lack of efficacy (Lang et al., 2022).AbbVie discontinued their drug development program of the antibody ABBV-0805 for the same reason (AbbVie, 2022).Other trials in this space are ongoing (Teng et al., 2021), though it is unlikely that this approach will bear fruit.
In contrast, the first clinical trials of GLP-1 receptor agonists that originally have been developed to treat diabetes have already shown significant protective effects in patients with AD or PD.Three different GLP-1 receptor agonists (Exendin-4, Liraglutide, Lixisenatide) in three separate clinical trials showed clear and meaningful improvements in PD, even after washout, and Liraglutide showed improvements in AD patients, too.This is a clear proof of concept that the GLP-1 strategy is effective and treatment with such drugs is making a clear difference in disease progression.This is only the beginning, and drugs are being developed that can cross the BBB better and show improved protective effects compared to older drugs that had been developed to treat diabetes.We are entering a new era of drug discovery that already has shown success in the clinic and could bring novel drug treatments to the market that are effective and potentially stop disease progression.The future looks bright for patients with AD or PD.The second messenger pathways control energy utilization, the chronic inflammation response in the brain, mitochondria function and mitogenesis, gene expression, autophagy, inhibition of apoptosis, modulation of ion channels and of synaptic activity, cell growth and repair, and the cellular response to oxidative stress (Doyle and Egan, 2007;Müller et al., 2019;Sharma et al., 2013).GLP-1 and GIP receptors can form dimers which show enhanced cAMP production and can activate other G-protein linked enzymes such as PLC (Baggio and Drucker, 2007;Müller et al., 2019;Wellman and Abizaid, 2015;Zhang and Holscher, 2019).For further details see (Hölscher, 2020;Reich and Hölscher, 2022;Sharma et al., 2013).

Funding
There is no funding associated with this article.

Declaration of competing interest
CH is a named inventor on patents that cover the use of GLP-1/GIP receptor agonists in treating neurodegenerative disorders.He is the CSO of the company Kariya Pharmaceuticals.

Fig. 1 .
Fig. 1.Scores in MDS-UPDRS part 3 motor test battery (left), and scores in the Mattis dementia rating scale 2 cognitive test battery (right).The drug group scored significantly better compared to controls.Compared to the control group of patients, patients previously exposed to exenatide had an advantage of 5.6 points (95% CI, 2.2-9.0;p = 0.002) using blinded video rating of the MDS-UPDRS part 3 motor subscale.There was also a difference of 5.3 points; (95% CI, 9.3-1.4;p = 0.006) between groups on the Mattis Dementia Rating scale.Figure modified from (Aviles-Olmos et al., 2013, 2014).

Fig.
Fig.3.Performance in the MDS-UPDRS part 2 motor assessment that measures everyday activities such as walking, chewing, getting dressed, talking, and other activities were significantly improved by liraglutide treatment The difference between groups at 54 weeks was p < 0.001.For details see(Hogg et al., 2023).