Safety and tolerability of intravenous liposomal GM1 in patients with Parkinson disease: A single center open-label clinical phase I trial (NEON trial)

Study Design

In this single-centre, open-label phase I interventional trial we enrolled 12 patients with PD to assess the safety, tolerability, and preliminary efficacy of liposomal GM1 as add-on medication with weekly intravenous infusions. Three patients received weekly ascending doses for 14 weeks to establish the highest well-tolerated dose (dose escalation group [DE]). The highest tested weekly dose of the API was 720 mg based on previous use of GM1 in humans (11) and was reached in all three patients. DE patients and 9 additional patients (dose consolidation group [DC]) received thereafter weekly TLN with an API dose of 720 mg for 8 weeks between December 13, 2021 and June 20, 2022. One of these patients stopped participation prematurely. One month after the last administration of TLN, a final safety follow-up visit was performed. Antiparkinsonian medication was kept stable if possible. In two patients with bilateral subthalamic nucleus stimulation, stimulation remained on and parameters unchanged throughout the study.

An independent data safety monitoring board (DSMB) evaluated safety data when the 3 DE patients had reached a GM1 dose of 180 mg and after completion of the dose escalation. The trial was approved by the local ethics committee (Kantonale Ethikkommission Bern) and the competent authority October 30, 2021. The trial is registered at the US National Institute for Health (ClinicalTrials.gov NCT04976127). The trial was conducted according to the principles of the Declaration of Helsinki and Good Clinical Practice guidelines. Talineuren is provided by the manufacturer and sponsor of the study, InnoMedica Switzerland AG. Talineuren is produced under GMP conditions at InnoMedica’s Nanofactory in Marly (Freiburg, Switzerland).

Patients

Patients aged 40-80 years were eligible if they were diagnosed with PD according to British brain bank criteria (12), had a Hoehn and Yahr Stage 0 – 2.5 on medication (13) and stable PD treatment for at least 4 weeks. To exclude major cognitive deficits, a score >25 on the Montreal Cognitive Assessment (14) was required. Patients were carefully selected to exclude medical, psychological, and behavioural problems that may have interfered with their compliance for study participation. Patients provided written informed consent and fulfilled all eligibility criteria.

Dose escalation (DE)

In patients 1-3 dose escalation was started at 6 mg GM1 and then increased weekly to 12 mg, 60 mg, followed by increases of 60 mg/week up to 720 mg. TLN was provided as a concentrated liposomal suspension containing cholesterol, sphingomyelin, and the active pharmaceutical ingredient GM1 in 30ml vials (180 mg GM1) in a phosphate buffered solution. Talineuren was diluted in phosphate buffered saline with a final concentration of 6 mg/ml GM1. TLN was added to 250 ml NaCl 0.9% and administered with 250 ml/h with a perfusor. The speed of infusion could be increased up to 370 ml/h if tolerated but reduced if necessary. Three days after the infusion each weekly dose increase of the TLN had to be cleared by an internal safety monitoring committee (MS, RPC, MW) by assessing clinical symptoms and lab values (Appendix 1). The second and third patient could start with the lowest dose with a delay of 1 week after confirmation of safety in the first patient. This 1+2 schedule was maintained for each weekly increase to allow for modification of the increasing of doses if needed according to a predefined dose modification matrix. All 3 patients had two clinical and lab assessments each week during the 14 weeks of the dose escalation phase. The average of the three maximal tolerated doses was defined as maximal suitable dose (MSD) for the dose consolidation (DC) part of the trial.

Dose consolidation (DC)

After DE, the initial 3 and additional 9 patients received weekly infusions of TLN with an API dose of 720 mg and clinical and lab safety assessments over 8 weeks except for one patient who stopped participation early. A final safety assessment was performed one month after the last administration of TLN.

Pharmacokinetics

Pharmacokinetic parameters were determined for GM1 from individual concentration time profiles obtained after the first intravenous infusion administration of 720 mg GM1 in the 9 DC patients. Blood samples were drawn before the infusion started and after the start of the infusion (250 ml/h, 370 ml total volume) at 5 minutes, 1 hour, 4, 8, 24, 48, 72, and 96 hours. Samples were immediately centrifuged and frozen at -70°C. GM1 concentration was measured using liquid chromatography with tandem mass spectrometry.

Noncompartmental pharmacokinetic analysis was performed using Phoenix WinNonlin version 8.0 [Pharsight Corporation, Mountain View, CA, USA] using the intravenous infusion dosing option. Pharmacokinetic variables were estimated from the plasma concentration versus time curves: tmax (time to reach the maximum plasma concentration read directly from the plasma concentration-time curve), Cmax (maximum plasma concentration read directly from the plasma concentration-time curve), AUC0-∞ (area under the plasma concentration-time curve from time point zero to infinity, estimated from AUC0-t + Ct/λz, where t is the last sampling time with a concentration above the limit of quantification and λz is the terminal elimination rate constant, estimated by log-linear least squares regression of the plasma concentration versus time data in the terminal phase; AUC0-∞ was calculated according to the linear trapezoidal with linear/log interpolation rule), t½: (apparent terminal half-life calculated as ln2/λz), CL: (apparent clearance calculated as dose divided by AUC0-∞), Vz (apparent volume of distribution calculated as dose divided by λz * AUC0-∞).

Outcome measures

The primary outcome was safety defined as the occurrence of adverse events. Safety assessments included a full medical history at baseline, a full general physical medical and neurological examination at baseline, a final assessment and last follow-up, and detailed unstructured interviews at each visit and if needed by phone. A narrative description of adverse events and their duration, intensity, seriousness, and relation to TLN were recorded by the site investigators. Adverse events were coded by an independent expert with the Medical Dictionary for Regulatory Activities (MedDRA®).

Adverse events were considered unrelated to TLN if a plausible other explanation for the observation was available. Otherwise, adverse events without plausible causal relation to TLN but lacking a different explanation were considered to be unlikely related to TLN. Safety data were weekly reviewed by an internal safety board (MS, RPC, MAW) during the dose escalation period. Any out-of-range laboratory result was counted as an adverse event.

Parkinsonian motor signs and non-motor symptoms were further explored weekly using the Movement Disorders Society-Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) (15) parts 1 (non- motor experiences of daily living), 2 (motor experiences of daily living), 3 (motor examination), and 4 (motor complications). The MDS-UPDRS-2 was assessed for best and worst condition in the preceding week in patients with motor fluctuations. The MDS-UPDRS-3 was assessed before and after each administration of TLN and also at baseline and final assessment in a levodopa challenge test (LCT). For the LCT patients paused their dopaminergic medications for at least 12 hours before the assessment “off” medication. The usual morning dose of levodopa equivalence plus 50 mg levodopa was then given as liquid formulation of levodopa/benserazide (in 1 patient levodopa/carbidopa) for the ensuing assessment “on” medication. Levodopa equivalent daily doses were noted weekly and calculated according to standard procedures (16). The number of PD-related non-motor symptoms was assessed with the Non-Motor Symptoms Questionnaire (NMSQuest) (17) at baseline and weekly during the course of the trial. Questionnaires at baseline and final assessment included the Parkinson’s disease Questionnaire-39 (PDQ-39) (18) for disease-related quality of life, the Epworth Sleepiness Scale (19), the Montreal Cognitive Assessment (MoCA (20) for mental performance, the Beck Depression Inventory (BDI) (21) to assess mood, and the Starkstein Apathy Scale (22).

Statistical analysis

After completion of data entry, data validation and cleaning were performed. Data analysis was started. All patients enrolled in this study received at least one dose of study medication and were considered in the safety and tolerability analysis.

PK analysis plasma concentrations and PK parameters were analyzed for the 9 patients of the consolidation group and reported in a descriptive fashion. Descriptive statistics include arithmetic mean, SD, minimum, median, maximum, geometric mean, and coefficient of variation of arithmetic and geometric means.

Other study outcomes including scores measuring the Parkinsonian motor signs and non-motor symptoms, disease related quality of life, levodopa equivalent daily dose, and mental performance were assessed in the 11 patients who received at least two full-dose infusions and completed at least the MDS-UPDRS at these two visits. For patients included in the escalation part we used as baseline value the values assessed at the start of the trial (i.e. before the start of the escalation part).

We used summary statistics to describe the outcomes values at baseline, after 8 weeks of treatment, and the observed changes. For each outcome, the significance of the change from baseline was assessed using a paired Wilcoxon signed-rank test. No correction for multiplicity was applied.

Few outcomes including the blood lab values of cholesterol (total, LDL, HDL, ratio total/HDL), triglycerides, and apolipoprotein B were defined, post-hoc, as additional outcomes. For these outcomes, we used summary statistics to describe the weekly assessed values. At each time point, the significance of the change from baseline was assessed using a paired Wilcoxon signed-rank test.

Trial population

There were no screening failures. Twelve patients (including three women) were recruited. At inclusion the median age was 65 (range 46-75) years old and had had motor parkinsonian signs for 7.9±5.2 (range 2-19, median 6.5) years. Three patients had only akinetic-rigid signs and nine patients also rest tremor. All patients had been clinically diagnosed with PD. For the patient characteristics see Table 1.

Safety

Dose escalation

Weekly dose escalation of TLN from 6 to 720 mg GM1 i.v. at a rate of 250 ml/h was well tolerated without major adverse effects. Mild neck and lumbar pain occurred for 15 minutes in one patient during the second (12 mg) and reappeared very mildly for a few minutes during the third (60 mg) infusion with TLN. Otherwise, dose escalation was unremarkable for all three patients and reached 720 mg with good tolerance. One patient reported transient beneficial effects of TLN treatment on sleep, mood, and general energy levels during dose escalation lasting longer with increasing doses. At the end of the dose escalation, the subjective beneficial effects of the weekly infusions lasted almost a week in this patient.

Pharmacokinetics

After first intravenous application of TLN with an API dose of 720 mg, repeated blood sampling showed the plasma peak of TLN in the sample taken 4 hours after start of the infusion in all but one patient in whom the peak was reached at 1 hour after start of the infusion. This patient was the patient who had the severe acute infusion reaction at the second administration. Pharmacokinetic variables are given in Table 2.

Dose consolidation

During the dose consolidation phase, acute infusion reactions occurred in 6 patients shortly after the start of the 2^nd (6 patients), 3^rd (1 patient), and 4^th (1 patient) administration of TLN (see below). One patient with a severe infusion reaction did not receive any further TLN treatment. During all following treatments with TLN, no acute infusion reactions occurred. The maximum dose of GM1 720 mg/week could be maintained throughout the study.

Adverse events

Overall, 304, mostly mild adverse events occurred. No serious adverse event was reported in this study. Twenty-three adverse events were affirmatively related to the study treatment. The causal relation to the study treatment was considered probable in 17 (all mild), possible in 100, and unlikely in 123 observations. Forty-one adverse events were unrelated to the study treatment.

Adverse events definitively or probably related to TLN can be separated into a group of immediate mild to severe infusion reactions that abated after halting the administration of TLN (see below), and a group of more diffuse, always mild reactions often occurring with some hours’ delay after the TLN infusion. The latter group comprises mild tension headache (n=3), nausea (n=1), and arterial hypertension during emotional tension (n=1) during the infusion, and dizziness (n=6), inner tension and tremulousness (n=2), and tension headache (n=4) occurring within hours after the infusion, spontaneously remitting within less than a day. One patient suffered from mildly depressed mood during the withdrawal of TLN after the dose escalation phase, partly because he missed the beneficial effects of the weekly infusions. This symptom remitted spontaneously before TLN was resumed in the consolidation phase of the study.

Acute infusion reactions occurred in one patient during dose escalation (2^nd and 3^rd administration), and in six patients during dose consolidation, always within minutes of the 2^nd administration of TLN. In two patients, an acute infusion reaction was observed at the 3^rd administration, in one very mildly during the 4^th administration. Symptoms occurred after as little as 3 mg of liposomal GM1 was administered and as early as within 1 minute after start of the infusion. Symptoms built up at variable speeds and with mild to severe intensity among different patients. All symptoms abated within minutes after stopping the infusion. Infusion could then be resumed at a low speed (10 ml/h) without recurrence of symptoms. In most cases, infusion speed could be increased swiftly up to the planned 250 ml/h except in one patient who required several hours for the 2^nd infusion. However, later the patient tolerated administrations of TLN at 250 ml/h without any adverse events. In one patient, the acute infusion reaction at the beginning of the 2^nd administration quickly led to severe back, neck, and chest pain. According to the study protocol, the patient had to be removed from the study and could not be re-exposed at a lower infusion rate despite the patients’ explicit request to continue study participation. The acute infusion reactions (Table 3) included lower back, neck, and chest pain as the most common symptom; lower back pain was present in all patients with an acute infusion reaction. Lower back pain was very mild in two patients during three administrations and did not require the infusion to be halted. One patient had itching during 2^nd, 3^rd, and 4^th administration of TLN, and urticaria on the trunk and the thighs (Table 3) during the 3^rd infusion. Tryptase was normal before start of TLN and within 30 minutes of the appearance of urticaria. Clemastine 2 mg was given i.v. and TLN could be resumed without worsening of symptoms. Urticaria disappeared within a day and did not recur during the following administrations of TLN. After the last occurrence of an acute infusion reaction, patients received 19 (n=1, dose escalation), 6 (n=4), and 4 (n=1) infusions without experiencing an acute infusion reaction. One patient was withdrawn from the study (Figure 1). In the remaining 5 patients overall 68 TLN infusions were given without an acute infusion reaction.

• Download figure
• Open in new tab

Fig 1:

CONSORT diagram of dose escalation group (with prolongation) and dose consolidation group.

• Download figure
• Open in new tab

Fig 2.

Mean (±SD) GM1 plasma concentration-time profiles after intravenous infusion administration of 720 mg GM1 (linear).

• Download figure
• Open in new tab

Fig 3.

Observed adverse events (n = 304) during dose escalation and dose consolidation with Talineuren.

Adverse events possibly related to TLN (Table 4) comprised 46 mild out of normal range measures of cholesterol, triglycerides, or apolipoprotein B. These observations were considered possibly related to TLN because the study drug contains cholesterol and sphingomyelin. However, lipid assessments were performed in a non-fasting state for which no ranges of normality were available. The lipid levels of all study participants were fluctuating over the study period but did not consistently change except for HDL cholesterol that was slightly but significantly reduced on all visits compared to baseline (see Appendix 2). In one patient, one observation of a slightly elevated level of anti-GM1 IgM antibodies was clinically asymptomatic and followed by normal levels in consecutive assessments. Many adverse events possibly related to TLN could also be explained as motor (e.g. muscle cramps, tremor, dysarthria, dysphagia) or non-motor (e.g., fatigue, sleep disorders, constipation, dysphoria) signs and symptoms of PD. During the four weeks pause after the dose escalation phase, one patient reported 25 adverse events of worsening parkinsonian motor and non-motor problems beyond baseline level. This rebound remitted spontaneously to baseline for all reported aspects before TLN was resumed in the dose consolidation phase. Mild tension headache, dizziness, nausea, and neck and lumbar pain were noted as possible infusion reaction symptoms.

Adverse events unlikely related to TLN (table 4) comprised mainly lab results mildly outside of the normal range. Two measures of hyperkalaemia (5.5 and 5.8 mmol/l) were clinically asymptomatic and occurred without other lab abnormalities; an artefact due to erroneous sample handling was assumed. A self-remitting episode of acute vestibular vertigo was the only severe adverse event considered unlikely to be related to TLN.

Secondary outcome measures

At an early stage, 1 patient discontinued the study and was removed from the efficacy analysis, meaning that the present analysis was performed on 11 patients. At the final assessment, parkinsonian motor signs off medication improved 6.73±7.16 points (Pval = 0.01 ) on the MDS-UPDRS 3 assessed after 12 hours of withdrawal of dopaminergic medication (see Table 6). This corresponded with the observation that pausing medication for the final assessment was much better tolerated than at baseline. With medication (on state), no significant change in motor signs was found during the study. TLN treatment did not result in a consistent and clinically relevant immediate beneficial effect on parkinsonian motor signs (data not shown). Motor complications of dopaminergic treatment (MDS- UPDRS 4) did not change significantly during the study. There were fewer non-motor symptoms at follow-up compared to baseline as assessed with the NMS-Quest (8.64 vs. 4.63, Pval = 0.009). Non- motor parkinsonian symptoms (MDS-UPDRS 1) and activities of daily living in the best condition (MDS- UPDRS 2 best) were significantly improved at follow-up compared to baseline (mean difference from baseline -1.91, pval = 0.057 and -2.82, pval = 0.02; respectively).

The total MDS-UPDRS (including the MDS-UPDRS 2 in the worst condition and the MDS-UPDRS 3 off medication) improved by 11.1±10.47 from baseline to follow-up (pval = 0.006).

Quality of life related to PD as measured with the PDQ39-SI improved by 2.91±2.25 (pval = 0.005), mainly driven by an improvement of quality of life for activities of daily living. The other subdomains of the PDQ39 did not improve significantly. No significant change in mood (BDI), motivation (Starkstein Apathy Scale), overall cognition (MoCA), and dopaminergic medication (levodopa equivalent daily dose [LEDD]) was observed.

During dose-escalation, one patient reported a general effect of well-being and more energy that lasted longer with increasing doses up to 1 week with GM1 720 mg. All patients reported mild to moderate reduction of the previously perceived beneficial effects during the month between the last infusion and the follow-up safety assessment. No lasting worsening beyond baseline was observed after one month’s withdrawal of TLN. In the first three patients, total GM1 dose was resumed at 720 mg weekly after a month’s pause following dose escalation; within one or two infusions the previous level of well-being was re-attained. All patients emphatically requested to prolong TLN treatment at the end of the study. An amendment for study prolongation was therefore submitted.