Pridopidine in early-stage manifest Huntington's disease: a phase 3 trial

Abstract

Huntington's disease (HD) is a rare, neurodegenerative disorder for which only symptomatic treatments are available. The PROOF-HD study was a randomized, double-blind, placebo-controlled phase 3 trial evaluating the efficacy and safety of pridopidine, a selective Sigma-1 receptor agonist, in HD. The primary and key secondary endpoints, change in total functional capacity (TFC) and composite Unified Huntington's Disease Rating Scale (cUHDRS) score at week 65, were not met in the overall population. The TFC least-squares mean difference between pridopidine and placebo was -0.18 (95% confidence interval -0.49 to 0.14; P = 0.26). The cUHDRS least-squares mean difference between pridopidine and placebo was -0.11 (95% confidence interval -0.40 to 0.18; P = 0.45). Sensitivity analysis in a subgroup of participants not treated with antidopaminergic medications at any time demonstrated a consistent pattern favoring pridopidine across multiple measures, including TFC and cUHDRS. Notably, pridopidine 45 mg twice daily demonstrated a favorable safety and tolerability profile. Taken together, pridopidine has the potential to address a critical unmet need in HD. ClinicalTrials.gov identifier: NCT04556656 .

Fig. 1. Participant disposition flowchart.

The flowchart illustrates participant disposition across study populations. ‘Off ADMs’ refers to participants who were not treated with ADMs at any point during the study. The ITT population includes all randomized participants. The mITT population consists of ITT participants who received at least one dose of the study drug and had valid TFC scores at baseline and at least one post-baseline visit. The SP population includes randomized participants who received at least one dose of the study drug. The PP population represents the combined analysis set of the W65PP and W78PP subsets. The W65PP and W78PP subsets include mITT participants with valid TFC scores at weeks 65 and 78, at least 80% compliance with the study drug and no major protocol deviations. W65PP, per protocol week 65 population; W78PP, per protocol week 78 population. ^aPatients who did not receive ADM treatment at any point during the study; ^bthe analysis set consisting of patients from both W65PP and W78PP.

Fig. 2. Primary and key secondary endpoints in the overall population.

a,b, TFC (primary endpoint) (a) and cUHDRS (key secondary endpoint) (b) are presented as LSMean changes from baseline to week 65 for the ITT and mITT populations. LSMean differences between the placebo and pridopidine treatment arms are shown at each time point. Sample sizes at week 65 were: placebo, n = 249 and pridopidine, n = 250 (ITT); and placebo, n = 247 and pridopidine, n = 242 (mITT). All replicates represent biologically independent participants. Statistical analyses were performed using a maximum likelihood-based mixed-effects model for repeated measures (MMRM). Data are presented as LSMean ± 95% CI.

Fig. 3. Sensitivity analysis of clinical outcomes in the off-ADM subgroup.

a, In the off-ADM mITT population, pridopidine treatment was associated with a slower rate of decline in cUHDRS scores compared with placebo. Maximal differences were observed between weeks 26 and 52, with nominal statistical significance at weeks 26 (P = 0.004), 39 (P = 0.014) and 52 (P = 0.035). b–f, Forest plots display treatment effects for cUHDRS and its four components—TFC, SWR, SDMT and TMS—in both the off-ADM mITT and PP populations. For the mITT group, sample sizes were as follows: week 26, n = 112 (placebo) and n = 95 (pridopidine) (b); week 39, n = 107 (placebo) and n = 90 (pridopidine) (c); week 52, n = 104 (placebo) and n = 91 (pridopidine) (d); week 65, n = 106 (placebo) and n = 91 (pridopidine) (e); and week 78, n = 71 (placebo) and n = 70 (pridopidine) (f). For the PP population, group sizes were: week 26, n = 99 (placebo) and n = 78 (pridopidine) (b); week 39, n = 99 (placebo) and n = 76 (pridopidine) (c); week 52, n = 97 (placebo) and n = 78 (pridopidine) (d); week 65, n = 97 (placebo) and n = 78 (pridopidine) (e); and week 78, n = 60 (placebo) and n = 60 (pridopidine) (f). g, Cognitive performance based on SWR showed numerical improvements with pridopidine at week 26 (P = 0.018; n = 112 (placebo) and n = 95 (pridopidine)), week 39 (P = 0.058; n = 107 (placebo) and n = 91 (pridopidine)) and week 52 (P = 0.042; n = 104 (placebo) and n = 92 (pridopidine)). h, Motor skills assessed by FT IOI Mean showed significant improvements at week 26 (P = 0.025; n = 112 (placebo) and n = 95 (pridopidine)), week 65 (P = 0.016; n = 104 (placebo) and n = 91 (pridopidine)) and week 78 (P = 0.028; n = 71 (placebo) and n = 69 (pridopidine)). All replicates represent independent participants. Statistical analyses were performed using a MMRM with two-sided nominal P values. The unit of analysis was the individual participant. Data are presented as LSMean, with error bars indicating 95% CI. *P < 0.05, **P < 0.01.

Fig. 4. Responder and threshold analyses of cUHDRS scores in participants off ADMs.

a, Post hoc responder analyses in the mITT population show that pridopidine led to higher CDF response rates than placebo at all time points, with significant differences at weeks 26 (AUC = 0.60, P = 0.013; KS test P = 0.0093) and 39 (AUC = 0.61, P = 0.0054; KS test P = 0.024). b, A threshold-based analysis revealed that a ≥5% improvement in cUHDRS was achieved in approximately twice as many participants receiving pridopidine compared with placebo at weeks 26 (P = 0.004) and 39 (P = 0.003). All statistical tests were two-sided and unadjusted for multiple comparisons. Error bars represent 95% CI. **P < 0.01.

Extended Data Fig. 1. Sensitivity Analysis in mITT Participants Off ADMs Showing the Effect of Pridopidine on TFC, cUHDRS, Cognitive Function (SWR) and Motor Performance (Q-Motor Assessments).

Sensitivity analyses of the change in (a) TFC show that pridopidine treatment showed a numerical trend favoring pridopidine, slowing TFC decline as compared with placebo through Week 78 with maximal TFC difference at Week 52 (difference: 0.26 vs. placebo; n = 104 placebo, n = 91 pridopidine). (b) Skilled Motor Performance (Pronation-Supination Inter-Tap Interval [Pro-Sup ITI] Mean): Improvements were significant at Week 26 (LSMean difference = -38.06 msec; p = 0.007; n = 112 placebo, n = 95 pridopidine) and Week 65 (LSMean difference = -23.92 msec; p = 0.04), with numerical trends favoring pridopidine at other timepoints. (c, d) Pridopidine treatment mitigated declines in motor function in mITT participants off ADMs. Negative values indicate improvement. Specifically: (c) Motor Variability (Finger Tapping IOI Standard Deviation [SD]): Significant reductions in variability were observed at Week 26 (LSMean difference = -0.43; p = 0.035; n = 112 placebo, n = 95 pridopidine), with trends favoring pridopidine at later timepoints. (d) Pronation-Supination Inter-Onset Interval (Pro-Sup IOI) Mean: Significant improvements were observed at Week 26 (LSMean difference = -17.2 msec; p = 0.044; n = 112 placebo, n = 95 pridopidine). Each panel represents mean change from baseline with 95% confidence intervals. *p < 0.05; **p < 0.01; ***p < 0.001. All statistical tests are based on mixed-effects models for repeated measures (MMRM) applied to the mITT population off ADMs. Negative values indicate improvement for all motor outcomes.

Extended Data Fig. 2. Sensitivity Analysis in PP Participants Off ADMs Showing the Effect of Pridopidine on TFC, cUHDRS, Cognitive Function (SWR) and Motor Performance (Q-Motor Assessments).

(a) Pridopidine showed a 0.37-point difference in TFC versus placebo at Week 78 (placebo: n = 160; pridopidine: n = 158). (b) For the cUHDRS, pridopidine slowed decline compared with placebo, with nominal statistical significance observed through Week 52 (Week 26: p = 0.006, placebo: n = 99, pridopidine: n = 78; Week 39: p = 0.003, n = 99, n = 76; Week 52: p = 0.035, n = 97, n = 78) (*p < 0.05; **p < 0.01). (c) Pridopidine demonstrated improvements in cognitive function (SWR) with significance reached at Week 26 (p = 0.034, n = 99, n = 78), Week 39 (p = 0.023, n = 99, n = 77), and Week 52 (p = 0.019, n = 97, n = 79) (*p < 0.05; **p < 0.01). (d–g) For motor outcomes, negative values indicate improvement. (d) Motor Skills (Finger Tapping Inter-Onset Interval [FT IOI]): Pridopidine resulted in significant improvements at Week 26 (p < 0.0001, n = 99, n = 78), with sustained differences through Week 78 (n = 76, n = 74; *p < 0.05; **p < 0.01; ****p < 0.0001). (e) Motor Variability (FT IOI Standard Deviation [SD]): Significant reductions were observed at Week 26 (p < 0.01; n = 99, n = 78) (**p < 0.01). (f) Skilled Motor Performance (Pronation-Supination Inter-Tap Interval [Pro-Sup ITI]): Improvements were significant at Week 26 (***p < 0.001), with trends favoring pridopidine through Week 78 (*p < 0.05; **p < 0.01; n = 99, n = 78). (g) Pronation-Supination Inter-Onset Interval (Pro-Sup IOI): Significant improvements were observed at Week 26 (**p < 0.01; n = 99, n = 78) and Week 78 (*p < 0.05; n = 76, n = 74). All data represent mean change from baseline in the PP population off-ADMs. Statistical significance was assessed using mixed-effects models for repeated measures (MMRM). P values are nominal and unadjusted for multiplicity. Error bars represent 95% confidence intervals (CIs).

Extended Data Fig. 3. Post Hoc Cumulative Distribution Function Analysis Based on Percent Change from Baseline in Composite Unified Huntington’s Disease Rating Scale Through Week 78 in Participants Off Antidopaminergic Medications (PP).

(a) This post hoc analysis presents the cumulative distribution function (CDF) of percent change from baseline in cUHDRS scores through Week 78 in the PP population of participants not receiving antidopaminergic medications (ADMs). Pridopidine consistently demonstrated higher response rates compared with placebo, with significant differences observed at Week 26 (AUC = 0.63, p = 0.003; KS Test p = 0.008; placebo n = 99, pridopidine n = 78) and Week 39 (AUC = 0.63, p = 0.001; KS Test p = 0.02; n = 99, n = 76). Trends favoring pridopidine persisted at Week 65 (AUC = 0.56; n = 97, n = 78) and Week 78 (AUC = 0.61; n = 60, n = 60). These results highlight pridopidine’s potential to improve or maintain function in participants off ADMs, with pronounced effects observed during the first 39 weeks. (b) A threshold-based analysis revealed that a ≥ 5% improvement in cUHDRS was achieved in approximately twice as many participants receiving pridopidine compared with placebo at Week 26 (p = 0.006) and Week 39 (p = 0.004) (**p < 0.01; *p < 0.05). All statistical tests were two-sided and unadjusted for multiple comparisons. Error bars represent 95% confidence intervals (CIs). AUC = area under the curve; KS test = Kolmogorov–Smirnov test.