Ropinirole hydrochloride for amyotrophic lateral sclerosis: A single-center, randomized feasibility, double-blind, placebo-controlled trial

1 INTRODUCTION

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease. The median survival time of patients with ALS is approximately 2 years, and muscle weakness eventually causes respiratory failure and death.¹ Currently, there are two Food and Drug Administration-approved drugs for the treatment of ALS: riluzole and edaravone. Riluzole extends survival by 2–3 months with no reported benefits to muscle strength, whereas edaravone improves short-term functional outcomes but has no statistically significant effect on survival.² Therefore, the need for effective ALS therapies remains unmet.

Drug development for diseases of the nervous system has the second-lowest success rate at the pivotal phase compared with other therapeutic areas.³ In particular, neurodegenerative diseases are heterogeneous and the majority of them is sporadic, limiting the translational potential of preclinical animal models. By exploiting induced pluripotent stem cell (iPSC)-derived motor neurons (MNs) generated from patients with ALS, we sought to enable large-scale drug screening and overcome the limitations of preclinical animal models for the development of drugs to treat highly heterogeneous neurodegenerative diseases.⁴

Ropinirole hydrochloride (ropinirole) was identified as a treatment candidate for ALS from 1232 FDA-approved drugs in a drug screening analysis conducted at Keio University. The analysis examined 3 FUS- and 2 TDP-43 (TARDBP)-ALS patient iPSC-derived lower MNs (LMNs) for the suppression of ALS-related phenotypes in vitro, such as MN death and damage, neurite regression, mislocalization of FUS and TDP-43, and stress granule formation. Furthermore, we established iPSCs derived from 32 sporadic ALS patients and induced LMNs. Among 32 LMNs, 22 of 24 LMNs with cell damage and neurite regression demonstrated increased apoptosis.

Finally, we confirmed that ropinirole hydrochloride improved multiple phenotypes of ALS in 16 of 22 LMNs (72·73%).⁵

Ropinirole is a commonly used drug for Parkinson’s disease (PD) that can permeate the blood-brain barrier. Of note, our in vitro model suggested that ropinirole was more potent than other ALS drugs and investigational drugs, including riluzole, edaravone, and ceftriaxone.⁵ Interestingly, other dopamine D2 receptor agonists such as bromocriptine and pramipexole

(dexpramipexole), have been reported to be candidates for therapeutics for ALS, ^6,7 but these drugs failed to show efficacy in clinical trials had.^8-11

Based on the safety profile of ropinirole for PD, we hypothesized that ropinirole would be well tolerated in patients with ALS. Here, we present the results of the Ropinirole Hydrochloride Remedy for Amyotrophic Lateral Sclerosis (ROPALS) single-centre, randomized, placebo-controlled trial, which evaluated the safety and efficacy of ropinirole with the following parameters: ALS Functional Rating Scale-Revised (ALSFRS-R) score,¹² composite functional endpoint, event-free survival, and time to ≤50% forced vital capacity (FVC), of the ropinirole hydrochloride extended-release tablet (Requip CR) in participants with ALS.^13,14

2 METHODS

3 RESULTS

3.1 characteristics of participants

From December 3, 2018, to September 12, 2019, we screened 29 participants, of whom 21 completed the run-in period and were randomized. Before drug administration, 1 participant withdrew due to protocol deviation. Among the remaining 20 participants, 13 received ropinirole, and 7 received placebo. Of the 18 participants who completed the last dose at week 24 in the double-blind period, 8 completed follow-up visits 24 days after the last dose at week 48 in the open-label active extension period (figure 1).

• Download figure
• Open in new tab

Figure 1. CONSORT Flow Diagram.

The diagram illustrates the screening, randomization, and follow-up processes in the ROPALS trial. Trial participants who completed the run-in period were randomized to receive either ropinirole or placebo at a 3:1 ratio. Due to dynamic allocation and allocation adjustment factors, 13 participants were assigned to receive ropinirole and 7 participants were assigned to receive placebo (approximately a 2:1 ratio). The in vitro evaluations of drug effects and exploration of new biomarkers using patient iPSC-derived motor neurons are ongoing.

This was a small-scale phase 1/2a randomized control trial, and two more subjects were allocated to the placebo group than planned; therefore, the number of subjects in the placebo group was seven with an allocation ratio of 2:1. According to Kang et al, at the initial stages of a clinical study or for trials with small sample sizes, using simple randomization can cause an extreme imbalance among the treatment groups by random chance alone.²⁰

Table 1 summarizes the demographic and baseline characteristics of the 20 participants who were randomized and received treatments. All participants were diagnosed with sporadic ALS with no detectable known pathogenic mutations in ALS causative genes (supplementary method). Both treatment groups had similar yet high proportions of participants with ≤grade 2 ALS severity. The ropinirole group had a longer disease duration (median: 24 months) than the control group (median: 17 months). Nevertheless, the total ALSFRS-R scores at pre-observation and baseline were balanced in both groups.

View this table:

• View inline
• View popup
• Download powerpoint

Table 1. Demographic and Clinical Characteristics of the 20 Participants at Baseline.*

3.3 secondary outcomes

3.3.1 Functional outcomes

During the double-blind period, the mean ALSFRS-R slope per month before and after treatment was 1·042 (95% CI, 0·249 to 1·835) in the ropinirole group compared with 1·254 (95% CI, 0·165 to 2·344) in the placebo group (table S2). The difference between groups was −0·212 (95% CI, −1·577 to 1·153). The change in the ALSFRS-R score from day 1 to week 24 of the double-blind period was −5·36 (95% CI, −8·09 to −2·64) for the ropinirole group and −6·82 (95% CI, −10·54 to −3·10) for the placebo group (table S3). The between-group difference in the change in ALSFRS-R scores was 1·46 points (95% CI, −3·15 to 6·07) over 24 weeks of treatment (figure 2A). In addition, participants in the ropinirole group had higher daily physical activity, as measured by METs (figure 2B). The difference in METs per month between treatment groups at week 24 was 317·5 (95% CI, 66·3 to 568·7) (table S3). There was a persistent increase in between-group differences in the change in ALSFRS-R scores beyond 24 weeks. The change in the ALSFRS-R score during the entire treatment period was −7·64 (95% CI, −10·66 to −4·63) for the RR group and −17·51 (95% CI, −22·46 to −12·56) for the PR group (table S3). The treatment difference in the change in ALSFRS-R scores was 9·86 points (95% CI, 4·07 to 15·66) over 48 weeks, indicating that RR reduced functional decline (figure 2A). The ALSFRS-R score for each of the 12 items is also shown (figure S3). Furthermore, we applied a composite endpoint that combined multiple clinically relevant functional components, expressed as the mean z-score. The between-group differences in the mean z-score were 11·27 (95% CI, −5·85 to 28·39), 25·04 (95% CI, 4·34 to 45·74), and 15·27 (95% CI, −11·92 to 42·47) at the end of 24, 39, and 50 weeks, respectively (figure 2C). The results of individual components within the composite endpoint are shown in Figures S4–S7 and Tables S4–S7.

• Download figure
• Open in new tab

Figure 2. Effect of Ropinirole Treatment on ALSFRS-R score, Physical Activity and Z-score.

(A) Estimated change from baseline in ALSFRS-R total score every four weeks (FAS, entire period).

(B) Level of physical activity every four weeks (FAS, double-blind period).

(C) The composite z-score functional outcome based on a battery of functional and QoL endpoints (FAS, entire trial period; post hoc).

→ Z: The amount of change at each measurement time (weeks 13, 24, 39, 50) from the values at 12 weeks after interim registration in each item was calculated, and then converted to a z-score. Next, a comparison between groups was performed.

‡Differences >0 indicate a positive treatment effect of ropinirole.

I bars represent the 95%CI.

Abbreviations: ALSFRS-R = Revised ALS Functional Rating Scale, FAS = Full Analysis Set,

3.3.2 Survival outcomes

To examine the relevance of the ALSFRS-R benefit to survival, we used the CAFS measurement, which adjusts the ALSFRS-R score against mortality. The CAFS score favoured ropinirole only in the open-label extension period (difference in median CAFS scores, 6·0; 95% CI, −5 to 9) and the entire treatment period (difference in median CAFS scores, 9·0; 95% CI, 1 to 12) but not in the double-blind period (difference of median CAFS scores, 4·0; 95% CI, −6 to 10) (figure 3A). Related to respiratory outcome, 14·3% of participants in the placebo group reached ≤50%FVC in contrast to 0% in the ropinirole group in the double-blind period (95% CI, −51·3 to 11·4), while 50% of participants in the PR group reached ≤50%FVC beyond 40 weeks in contrast to 0% in the RR group (RR group: 7·7%; PR group: 42·9%; difference, −35·2; 95% CI, −67·9 to 2·1) (figure 3B). Notably, %FVC at baseline was not different between the RR and RP groups (95% CI, −4·85 to 30·8).

• Download figure
• Open in new tab

Figure 3. Effect of Ropinirole Treatment on Survival.

(A) Median CAFS scores^¶ of participants in the double-blind period (n=20), open-label period (n=16), and entire trial period (n=20). Scores range from (1) (worst) to (20) (best).

(B) The proportion of participants whose %FVC was reduced to 50% over 48 weeks.

(C) The proportion of participants with an occurrence of disease progression events, including death, over 48 weeks.

^¶CAFS ranks patient clinical outcomes based on survival time and change in the ALSFRS-R score. Each patient’s outcome is compared with every other patient’s outcome, assigned a score, and the summed scores are ranked. The mean rank score for each treatment group is then calculated. A higher mean CAFS score indicates a better group outcome.

Abbreviations: CAFS = Combined Assessment of Function and Survival, %FVC = percent predicted Forced Vital Capacity, ALSFRS-R = Revised ALS Functional Rating Scale.

We also investigated the time to death or certain disease progression events. There were 7 out of 7 (100%) events in the PR group and 7 out of 13 (54%) events in the RR group, suggesting a twofold decrease in disease progression in the RR group (tables S9, S10A). The RR group had an extended time to the first disease progression event (median event-free survival: 50·3 weeks vs. 22·4 weeks in the PR group) (log-rank test: 95% CI, 4·3 to 37·4; Cox regression model analysis: hazard ratio = 0·25, 95% CI, 0·08 to 0·80) (figure 3C). We observed a higher proportion of participants who progressed to grade 4 ALS severity or above in the placebo group at the end of 24 weeks (71% for placebo and 53% for ropinirole groups) and 48 weeks (100% for PR and 58% for RR groups) (figure S8, table S9B). Moreover, we identified liquid biomarkers related to the pathomechanism of ALS (figure S9A–C).

4 DISCUSSION

In our previous in vitro drug efficacy study using patient-derived iPSC-MNs⁵, the effect of ropinirole on ALS was evident at 0·1, 1, and 10 µmol/L. The estimated ropinirole concentration at 2 mg of Requip CR is approximately 5 nmol/L and that at 16 mg is approximately 50 nmol/L in plasma and CSF.²¹ In other experiments, the efficacy of ropinirole was dose-dependent at ropinirole concentrations of 1 nmol/L and 10 µmol/L (figure S10), indicating that 2 mg of Requip CR is effective. However, it is desirable to administer up to 16 mg, which is the maximum approved dosage for use in PD in Japan.

Following the administration of 16 mg of ropinirole, the CSF concentration of ropinirole reached approximately 18 nmol/L (table S1) and remained at a steady state. This observation is consistent with the effective concentration of ropinirole in suppressing ALS-related damage in iPSC-derived MNs, which ranged from 1 nmol/L or more based on the present (figure S10) and previous results.⁵ Steady exposure to 18 nmol/L ropinirole was not associated with any serious AEs that led to trial discontinuation. This trial demonstrated that fixed-dose ropinirole (16 mg) had no specific safety concerns.

Overall, the incidence of AEs, most of which had been reported previously, were similar within both groups (table 2A, B). Although gastrointestinal disorders had a high incidence at 76·9% in the ropinirole group (14·3% in the placebo group), these symptoms were temporary mild nausea and diarrhoea, and there were no participants in which the continuation of oral administration was hindered (table 1). Therefore, ropinirole (2–16 mg) is tolerable for ALS patients.

Interestingly, the incidence of ropinirole-induced gastrointestinal disorders in PD patients and healthy individuals in Japan is approximately 5-12% (pharmaceutical interview form for Requip CR tablets, version 6) and 0% (table S10), respectively, which is different from the results in ALS patients in this study. Therefore, gastrointestinal symptoms are not considered to be a major adverse effect of ropinirole, and the reason why many gastrointestinal symptoms appeared in ALS patients in this clinical trial is unknown. Regarding safety concerns, gastrointestinal adverse effects for ALS patients might complicate care because ALS patients have poor BMI, malnutrition, and cachexia; however, the ropinirole group did not differ significantly from the placebo group in weight drop or in the ALSAQ-40 (eating and drinking) score (figures S6 and S7). Thus, we thought that gastrointestinal symptoms did not significantly affect safety or tolerability in this trial. Because the study was primarily designed to assess safety and tolerability, we chose to present our clinical findings regarding the secondary outcomes as reference for further study. Moreover, regarding the assurance of blinding, the ALSFRS-R score includes both subjective and objective assessments. The investigators did indeed evaluate the patients objectively, assessing at least half of the ALSFRS-R items (i.e., speech, handwriting, cutting food and handling utensils, walking, climbing stairs and respiratory insufficiency). In our posthoc analysis of the transition of each of the 12 items in the ALSFRS-R, no obvious differences were noted between the results of subjective and objective assessment items (figure S3). Therefore, blinding was not broken, and outcomes were not influenced throughout the study. The blinding was kept throughout, and it was plausible that patients and investigators would not know which group had active or placebo treatment; therefore, outcomes were not influenced during the study.

The estimated effect size of ropinirole observed for a change in the ALSFRS-R score ranged from 1·46 points to 9·86 points over 48 weeks, equivalent to a 21%–60% slower rate of functional decline. Neurologists generally perceive a treatment difference of >20% as somewhat clinically meaningful.²² Accordingly, the FDA and PMDA-approved ALS drug edaravone reduced the rate of functional decline by 33% over 24 weeks.²³ Moreover, clinical evaluation items such as the CAFS score, muscle strength, physical activity, pulmonary function, and composite z-score might be useful as evaluation indexes for the efficacy of ropinirole. The CAFS endpoint is advantageous because it evaluates functional and survival outcomes concomitantly without relying on statistical assumptions. CAFS is well adjusted for missing values due to death or drop-outs.¹³ However, CAFS measurements may be less relevant because there was only one death in this study, indicating that mortality did not confound the survival analysis per se. Ropinirole, a non-ergot dopamine agonist, was previously approved for the treatment of PD. The full mechanism of action of ropinirole in ALS is not yet understood. Based on our previous reports and according to its structural characteristics, its effects might be dopamine D2 receptor (D2R)-dependent and D2R-independent.^{5, 21} With regard to the dopamine D2R-dependent anti-ALS mechanism of ropinirole, we confirmed the protein expression of dopamine D2R on anterior horn cells in a healthy human spinal cord and human iPSC-derived LMNs (figures S11 and S12), which is consistent with the transcriptome data.²¹ Furthermore, D2R couples with Gi proteins to inhibit adenyryl cyclase, resulting in decreased intracellular cAMP levels. As a result, neuronal hyperexcitability, which is thought to be the cause of MN toxicity, can be inhibited.^24,25 Recently, it was shown that D2R agonists other than ropinirole, such as bromocriptine and sumanirole, suppressed the hyperexcitability of human iPSC-derived MNs in vitro.²⁴ This suppression of neuronal hyperexcitation might protect ALS MNs, which is relevant to the action of riluzole. Furthermore, previous reports indicated that autophagy may be activated by dopamine D2R and D3R agonists via a Beclin-1-dependent pathway.^{26, 27} Thus, ropinirole-induced dopamine D2R activation might induce autophagy, supporting the degradation and disassembly of abnormal RNA-protein complexes in the MNs of patients with ALS.

With regard to dopamine D2R-independent mechanisms, it is notable that ropinirole is a lipophilic cation that readily localizes to the mitochondrial inner membrane due to its tertiary amine moiety and possesses antioxidant properties related to its oxindole structure.²¹ Consistent with this, urine 8-hydroxy-2’-deoxyguanosine, a surrogate marker of oxidative stress, was not increased in the ropinirole group throughout the entire study period (figure S9A). These findings suggest that ropinirole may scavenge mitochondrial reactive oxygen species and protect cells against mitochondrial damage in ALS. In addition, ropinirole suppressed increases in ALS pathomechanism-related markers, such as serum ferritin^{28, 29} and high-sensitivity C-reactive protein,³⁰ throughout the entire study period (figure S9A). Interestingly, bromocriptine mesylate, an ergot alkaloid and agonist of the same dopamine receptor, namely, D2R, is expected to have an effect on dopamine D2R-independent and neuronal apoptosis inhibitory protein (NAIP), an oxidative stress-induced cell death suppressor -mediated anti-ALS mechanism.^6,31 Collectively, the precise mechanism of action of ropinirole in ALS warrants further investigation.

Interpretations of efficacy analyses in this feasibility study were limited by the small sample size of 20 participants. This was further compounded by the unexpectedly higher rate of discontinuation in this study than the historical rate in clinical trials of ALS (∼20%)³²;46·2% (6/13) of the participants in the ropinirole group and 85·7% (6/7) in the placebo group discontinued the trial (table S11), and the ratio of patients in the ropinirole versus placebo group changed, particularly in the open-label extension period.

These discontinuation rates, particularly during the open-label extension phase, were attributable, at least in part, to the COVID-19 outbreak, which accounted for 23% and 29% of the participants in the ropinirole and placebo groups, respectively. Therefore, it may have had some undesirable influence on the objective interpretation of the results. The number of participants in the placebo group who discontinued the trial due to a worsening condition (47%) was higher than that in the ropinirole group (23%). We did not apply any imputation to the missing data but used the MMRM to infer missing data. In a recent randomized and controlled trial of rasagiline for ALS, 60% of the placebo group withdrew from the trial. As a result, the investigators increased the power of the study by utilizing placebo (n=8) and historical placebo controls (n=177).³³ In the future, the addition or adoption of historical placebo controls may be a plausible approach to enrich the placebo cohort, particularly for trials with an extended period.

In this trial, the placebo group had more bulbar onset, more female patients and a lower BMI at baseline, which could be the relevant prognostic factors for ALS.^34-36 These are major weaknesses of this trial. In addition, there are some points to be discussed for further study. It is important to note that the slope for the ALSFRS-R decline (−3·4 points) was consistent between the ropinirole and placebo groups during the 12-week run-in period. Additionally, the onset of disease is self-reported and can be influenced by the sensitivity and character of an individual patient; therefore, it is difficult to confirm the accuracy of the period of disease duration. We considered the latest disease progression rate to be more important and emphasized this progression rate with the ALSFRS-R during the 12-week pre-observation period. These observations ruled out potential baseline imbalances related to differences in disease duration. Moreover, previous observational studies showed that the relationship between bulbar onset and survival prognosis were controversial after adjustments for the other clinical features of ALS. ^34-37 In practice, both the bulbar-onset group and the nonbulbar-onset group showed almost identical transitions in ALSFRS-R scores. In addition, both the ropinirole and placebo groups were divided into the bulbar-onset group and the nonbulbar-onset group (figure S13). When we compared the bulbar-onset and the nonbulbar-onset groups, no significant difference was observed at any time point. Therefore, in this clinical trial, the transition in the ALSFRS-R was not affected by the higher percentage of bulbar-onset patients in the placebo group.

Although we observed clinically meaningful differences between the treatment groups for functional and survival outcomes, these differences were primarily observed in the open-label phase. During the double-blind phase, we observed statistically significant treatment differences only for physical activity. Indeed, the rate of decline in the ALSFRS-R score in the placebo group was not improved even during the open-label extension period when the active drug was started. Interestingly, a similar phenomenon was observed in another trial.³⁸ Although the underlying mechanism of this observation remains unclear, we postulate that taking ropinirole earlier and for longer is necessary to demonstrate its efficacy in participants with ALS. Furthermore, the patient population was skewed towards early-stage ALS (the total ALSFRS-R score at baseline was 39·5±3·0); thus, we also have no conclusive empirical data to demonstrate whether ropinirole was effective in patients with advanced ALS. Taken together, the uncontrolled data with a limited sample size reported in this feasibility study should be interpreted with caution and must be further validated with, ideally, advanced trials and trial designs. Therefore, the efficacy of ropinirole in ALS deserves further investigation in larger, multinational, randomized controlled trials.³⁹

Although this trial is a randomized feasibility trial, the results of this trial encourage further study using ropinirole in ALS worldwide. To the best of our knowledge, the ROPALS trial is the representative touchstone of iPSC-based drug repurposing-enabled trials to define the feasibility of iPSC models in predicting clinical outcomes and replacing failure-prone preclinical transgenic mouse models of ALS.

6 APPENDIX

Data Sharing

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.