Effect of Triheptanoin on Caudate Atrophy and Motor Scores in Patients With Early-Stage Huntington Disease
A Phase II Study
Abstract
Background and Objectives
Brain energy deficiency occurs at the early stage of Huntington disease (HD). Triheptanoin, a drug that targets the Krebs cycle, can restore a normal brain energetic profile in patients with HD. In this study, we aimed at assessing its efficacy on clinical and neuroimaging structural measures in HD.
Methods
We conducted a 6-month bicentric (Paris, Leiden) double-blind randomized controlled trial followed by a 6-month open-label phase, between June 2015 and December 2019. We enrolled 107 patients at the early stage of HD—total motor score (TMS) of the UHDRS between 5 and 40. Participants received triheptanoin 1 g/kg/day or placebo (ratio 1/1). The primary outcome was the rate of caudate atrophy at 6 months using the caudate boundary shift integral (cBSI) method. Main secondary outcomes were cBSI at 12 months, TMS, and diffusion imaging at 6 and 12 months. Analysis was conducted using ANOVA, and data were presented with a 95% CI. To perform a 12-month comparison, we used the placebo arm of a 12-month randomized controlled trial conducted in parallel, using the double robust propensity score method.
Results
One hundred patients were randomized (mean age 49 years, 52% women). Fourteen patients withdrew from the study, including 10 because of gastrointestinal effects. There was no difference in cBSI at 6 months between groups (mean 0.026 [0.018–0.033] vs 0.023 [0.014–0.032]). TMS at 12 months was stable in patients treated with triheptanoin for 12 months (mean 0.6 [−1.1 to 2.1]), whereas it increased in patients initially on placebo (2.5 [1.2–3.8]). Compared with the external placebo control group, caudate atrophy decreased by approximately 50% (0.038 [0.028–0.048] vs 0.070 [0.057–0.082]) and TMS stabilized (0.66 [−1.07 to 2.48] vs 2.65 [1.38–3.89]) in patients treated with triheptanoin for 12 months.
Discussion
There was no effect of triheptanoin on caudate atrophy over 6 months. Compared with the external placebo group, triheptanoin was associated with motor stability and decreased caudate atrophy in patients treated for 12 months, but the post hoc nature of these findings is a major limitation.
Trial Registration Information
clinicaltrials.gov NCT02453061, May 25, 2015. First patient enrolled on June 29, 2015.
Classification of Evidence
This study provides Class I evidence that triheptanoin does not slow caudate atrophy compared with placebo over 6 months in patients with early HD.
Introduction
Huntington disease (HD) is an autosomal dominant disease characterized by movement disorders, behavioral disturbances, and cognitive decline. HD is caused by an abnormal expansion of CAG repeats in exon 1 of the HTT gene. HD is associated with neuronal loss predominantly in the caudate nucleus and putamen. However, many other cerebral regions are involved in disease progression, and white matter alterations seem to occur at early disease stages.1,2 The clinical features of HD usually emerge in adulthood between 30 and 50 years of age. Patients progress into motor dysfunction (chorea, dystonia, extrapyramidal rigidity, and akinesia), well captured by the Unified Huntington's Disease Rating Scale (UHDRS), cognitive difficulties (dysexecutive syndrome and frontal behavior), and psychiatric symptoms. There is currently no curative or preventive treatment.
In addition to neurologic symptoms, patients with HD often present with significant weight loss from the early stage of the disease. Because the brain has little energy storage besides glycogen, it depends largely on peripheral organs for energy supply.3 In situation of chronic brain energy deficiency, as documented in HD,4,5 there is an excessive demand to peripheral organs, associated with weight loss,6 to provide energy to the brain. Using untargeted metabolomics in early affected HD patients, we found that weight loss was associated with decreased levels of branched chain amino acids in the plasma, which we interpreted as a need for substrates for the Krebs cycle.10 Indeed, branched chain amino acids are normally spared for anabolic purposes, but when oxidized, they produce acetyl-CoA and propionyl-CoA, thereby fueling the Krebs cycle. We then showed by 31phosphorus brain NMR spectroscopy that the brain energetic response of patients with HD is abnormal during activation.11 Unlike healthy individuals, for whom the ratio of inorganic phosphate over phosphocreatine (Pi/PCr) increases in the visual cortex when stimulated, the Pi/PCr ratio did not increase in patients with HD during visual stimulation.11
To correct this brain energy deficiency, we used triheptanoin, a compound with anaplerotic properties that can replenish the pool of metabolic intermediates in the Krebs cycle.12 Triheptanoin is a medium-chain triglyceride that was initially developed for patients with long-chain fatty acid oxidation disorders (LC-FAODs)13 at a dose of 1 g/kg/day in adults, that is, approximately 30%–35% of daily caloric intake.14 Triheptanoin led to improved peripheral energy metabolism with reduced episodes of hypoglycemia, rhabdomyolysis, and, therefore, hospitalizations of patients with LC-FAOD.15,16 The drug is now approved in the United States for LC-FAOD. The first indication of a possible benefit of triheptanoin in the treatment of brain energy deficit came from pyruvate carboxylase (PC) deficiency. Triheptanoin led to immediate biochemical and clinical recovery of a PC-deficient patient with severe encephalopathy.17 Triheptanoin was then used in patients with Glut1 deficiency syndrome (Glut1-DS), a more common model of brain energy deficiency whereby glucose cannot properly enter glial cells. Treatment with triheptanoin was associated with a dramatic reduction of paroxysmal movement disorders18 that was sustainable over time.19 A phase 3 study did not confirm the benefit of triheptanoin in movement disorders in patients with Glut1-DS.20 However, the mentioned study was conducted without proper dietary management, so its results have to be interpreted with caution.21 In HD, we conducted a proof-of-concept study using our brain NMR spectroscopy protocol and showed that triheptanoin was able to restore the Pi/PCr ratio during brain activation after 1 month of treatment.17
In this study, we wished to evaluate the therapeutic benefit of triheptanoin on clinical and neuroimaging structural (caudate volume and white matter trophicity) measures. To increase patient's acceptability, we conducted a 6-month double-blind randomized controlled trial in early affected HD patients, followed by a 6-month open-label phase. The primary objective was to determine whether triheptanoin can decrease the rate of caudate atrophy over 6 months compared with placebo. To allow for a comparison over a period of 12 months, we used the placebo arm of a study conducted in parallel, with identical methods, in patients with HD with similar characteristics.
Methods
Study Design
TRIHEP3 is a phase 2, multicenter double-blind, randomized, controlled 2-armed study evaluating the efficacy of triheptanoin vs placebo for a 6-month treatment period, followed by a 6-month open-labeled period with all patients treated with triheptanoin. This design was chosen for acceptability purposes because HD is a rare disease, and the demands of the study (e.g., drug that presents like an oil and requested dietary changes) may have proven too much for patients to accept, without the certainty of benefiting from the active product for at least part of it.
The study was conducted in 2 centers: La Pitié-Salpêtrière University Hospital in Paris, France, and Leiden University Medical Center in the Netherlands.
Standard Protocol Approvals, Registrations, and Patient Consents
The study protocol was approved by the CPP Ile-de-France VI (EudraCT 2014-005112-42) and the study registered with clinicaltrials.gov (NCT02453061). Written informed consent was obtained from all the participants.
Participants
Each center enrolled patients with HD at the early disease stage drawn from their respective cohorts.
Inclusion criteria were as follows: a positive genetic test with CAG repeat length ≥39 in the HTT gene; at least 18 years of age; coverage by social security; UHDRS total motor score (TMS) between 5 and 40; ability to undergo MRI scanning; body mass index between 18 and 30.
Exclusion criteria were as follows: known hypersensitivity to triheptanoin or to one of its excipients; additional major comorbidities; history of severe head injury; participation in another therapeutic trial (3-month exclusion period); pregnancy or breastfeeding; inability to understand information about the protocol; deprivation of liberty by judicial or administrative decision; legal protection or inability to consent; treatment with tetrabenazine; treatment with neuroleptic drugs other than olanzapine and aripiprazole at small doses (≤10 mg and ≤15 mg, respectively).
Randomization and Masking
The randomization was stratified by center, and the therapy was concealed by anonymous codes. The generation of randomization lists was realized using validated commercial software “Rancode” (Version 3.6, IDV, Gauting, Germany).
To minimize unmasking, triheptanoin and the placebo (safflower) oils were indistinguishable in appearance, taste, and packaging.
ECRIN sent the bottle assignment list to an unblinded representative at Ultragenyx. ECRIN was responsible for managing and maintaining the bottle assignment list. Central Coordination submitted a request for a bulk supply of the blinded drug for each site. Ultragenyx drug depot then shipped a bulk supply of the drug to each site, and the site stored it in the pharmacy. The data management unit was responsible for assigning the correct blinded kits to the patients from the blinded drug at each pharmacy and managing the inventory and resupply requests for the Paris and Leiden sites. Outcome assessors, statisticians, and all the persons responsible for the data analysis were blinded to treatment allocation.
Procedures
During the double-blind period, dosing of the study drugs was titrated for 1 week until patients reached 1 g/kg/d, which represented approximately 30%–35% of their total caloric intake. If the target dose was not reached, titration continued until the maximum tolerated dose was reached. Triheptanoin or placebo oil was mixed into food and administered orally in 3–4 doses per day. A trained dietitian determined the patient's caloric intake and adjusted their daily menus so that their diet remained isocaloric when triheptanoin or safflower oil was introduced. Patients were advised to ingest the study drugs over a period of at least 20 minutes to ease digestive tolerance. Proper metabolism of triheptanoin was monitored every 3 months using plasma acylcarnitine profiles and C5 ketone bodies and urine organic acids reviewed by an independent investigator (common for both Paris and Leiden) blind to the study drug.
Patients attended the clinic at baseline and months 3, 6, 9, and 12, with home nurse visits at months 1, 2, 4, 5, 7, 8, 10, and 11 to monitor compliance. Home nurses collected urine samples at months 1, 2, 4, 5, 8, and 10 for centralized organic acid analyses. Body weight was assessed at every visit. In-between visits, patients were also regularly monitored by trained dieticians over the phone. Blood sampling 75–90 minutes after administration of triheptanoin was conducted at baseline and months 3, 6, 9, and 12 to monitor proper metabolism of triheptanoin.
TMS, Total Functional Capacity (TFC), and Problem Behaviors Assessment–Short (PBA-s) were estimated at baseline and months 3, 6, 9, and 12. A short battery of neurocognitive testing—Symbol Digit Modalities Test (SDMT), Stroop Test, Digit Span Forward/Backward, and Trail Making Test (TMT) A/B—was conducted at baseline, 6 months, and 12 months, as well as the Short Form 36 Health Survey Questionnaire (SF-36) to evaluate patient's quality of life.
Measurements of caudate volume (using BSI) were conducted at baseline, 6 months, and 12 months. Measurements of the Pi/PCr ratio were performed by 31phosphorus brain NMR spectroscopy before, during, and after a visual stimulation using a flashing checkerboard.17 In addition, a novel method of diffusion tensor imaging called fixel-based analysis (FBA), already developed in Paris for patients with HD,1 was used to measure white matter integrity in French patients from the study (n = 52).
Adverse events were recorded at all in-clinic visits, at home visits, and during telephone calls.
Participants could continue concomitant use of select therapies for management of HD symptoms. We did not enroll patients treated with tetrabenazine because it was shown to be a possible confounding factor in a clinical trial conducted in patients at the early stage of HD.22 The dose or regimen of these therapies and activities were recorded in patient diaries and had to remain constant during the first 12 months of the study.
Neuroimaging Methods
All magnetic resonance acquisitions were performed on a 3T scanner (Siemens Medical Solutions, Erlangen, Germany) with a standard Siemens transmit body coil and 32-channel receive head coil array. A 3-dimensional T1-weighted image (TR: 2,300 ms, TE: 2.98 ms, TI: 900 ms, field-of-view: 256 × 240 mm2, slice thickness: 1 mm) was acquired to allow for positioning of the volume of interest and for morphometry. The caudate boundary shift integral (cBSI) approach enabled effective evaluation of longitudinal caudate atrophy on the 3D T1-weighted images (eMethods).23
For NMR spectroscopy analyses, we targeted the visual cortex for the single-voxel 31phosphorus because it has a higher energy metabolism and is close to the scalp allowing an increased sensitivity to small surface coils. A 6-cm 31phosphorus transmit/receive surface coil (Rapid Biomedical GmbH, Rimpar, Germany) was used to collect free induction decays for 4 minutes at rest, 8 minutes during visual stimulation (activation) with 6-Hz red/black checkerboard flashes, and 8 minutes after stimulation (recovery). The delta of Pi/PCr between activation and recovery was calculated to determine the brain response to cortical activation (eMethods).11
Diffusion-weighted 2D spin-echo planar imaging (b value: 1,500 s/mm2 (30 directions), FOV: 256 × 256 mm2, TE: 93 ms, TR: 14,400 ms, slice thickness: 2 mm) was acquired to evaluate the brain microstructure. A fieldmap image was acquired for susceptibility correction. The diffusion data were processed using MRtrix3Tissue version 5.2.9,24 a fork of MRtrix3, and FSL version 6.01 tools. Noise in the data was first removed using principal component analysis,25 and Gibbs-ringing artifacts26 were removed based on local subvoxel shifts (eMethods).
We performed FBA to analyze multiple crossing fibers within a voxel (fixels) to extract complex white matter pathways, which are otherwise not properly estimated using the diffusion tensor model.27 The tissue-specific response function was estimated for the white matter, gray matter, and CSF using an unsupervised estimation that did not require segmentation of the T1 image.1 The fiber orientation distribution (FOD) was estimated from the average baseline response function of all participants using the single-shell 3-tissue constrained spherical deconvolution1 that reduced contribution of signal from non–white matter compartments, that is, the gray matter and CSF compartments. A template FOD was created from 10 patients from each treatment group. Each participant's normalized FOD, including that of follow-up, was then registered to the template and then segmented to generate an estimation of the intra-axonal compartment volume fiber density (FD). The voxels occupied by the fiber bundle (fiber cross section, FC) were estimated from the deformation fields from the registration of the FODs to the template space. The product of FD and FC was calculated to provide the fiber density and cross section (FDC) (eMethods).
Outcomes
The primary efficacy end point was the change from baseline to 6 months in caudate atrophy as measured by cBSI.
Secondary End Points
1.
The change from baseline to 12 months in caudate atrophy as measured by cBSI
2.
The change from baseline to 6 and 12 months in TMS of the UHDRS
3.
The change from baseline to 6 and 12 months in TFC
4.
The change from baseline to 6 and 12 months in PBA-s
5.
The change from baseline to 6 and 12 months in neurocognitive tests, that is, SDMT, Stroop Test, Digit Span, and TMT
6.
The change from baseline to 6 and 12 months in SF-36
7.
The change from baseline to 3, 6, and 12 months in the delta of Pi/PCr ratio when comparing visual stimulation with recovery
8.
The change from baseline to 6 and 12 months in fiber integrity as measured by FBA
9.
The safety and long-term tolerance of triheptanoin based on review of adverse events and changes in clinical laboratory results and physical examination/vital signs.
Comparison With an External Parallel Placebo Group
In parallel to TRIHEP3, we conducted REV-HD, a phase 2, multicenter double-blind, randomized, controlled 2-armed study evaluating the efficacy of 80-mg resveratrol per day vs a placebo for a 12-month treatment period, in 102 early affected HD patients. The study was conducted in 5 university hospital centers in France (NCT02336633). Inclusion criteria were similar to TRIHEP3, especially CAG repeat length ≥39 in the HTT gene and TMS between 5 and 40. Exclusion criteria were similar to TRIHEP3, including treatment with tetrabenazine and neuroleptic drugs other than olanzapine and aripiprazole at small doses (≤10 mg and ≤15 mg, respectively). This study had similar design and clinical and radiologic assessments as TRIHEP3. Patients attended the clinic at baseline and months 3, 6, 9, and 12, with phone calls at months 1, 2, 4, 5, 7, 8, 10, and 11 to monitor compliance. TMS and TFC were estimated at baseline and months 3, 6, 9, and 12. MRI assessments with morphometry (i.e., caudate volume measured by cBSI) were conducted at baseline and 12 months. The primary efficacy end point was the change from baseline to 12 months in caudate atrophy as measured by cBSI. Secondary efficacy end points were (1) the change from baseline to 12 months in TMS and TFC and (2) the change from baseline to 12 months in diffusivity and/or fiber integrity as measured by DTI and FBA.
Statistical Analyses
We hypothesized that, in patients under placebo, the progression of the rate of caudate atrophy over 6 months would be reduced by 42%. Based on such hypothesis and data from the TRACK-HD study,28 50 patients under placebo and 50 patients treated with triheptanoin would be required to achieve 80% power at the 1-sided 0.025 α level. Therefore, a total of 100 patients were required.
Participants were classified into triheptanoin and placebo groups. Continuous variables are presented as means with SDs and medians with interquartile ranges (IQRs, described as 25th and 75th percentiles) for normal and skewed distributions, respectively. Categorical variables were reported as numbers with percentages.
All analyses were performed in all randomized patients (i.e., intention-to-treat). The 2-sided significance level was fixed at 5%. All tests were performed using SAS version 9.4 statistical software (SAS Institute Inc., Cary, NC).
The primary efficacy end point was the change from baseline to 6 months in caudate atrophy as measured by cBSI and analyzed between the 2 groups using ANOVA (SAS PROC MIXED) with Kenward-Roger adjusted degrees of freedom. Adjusted analysis was performed using ANCOVA with age, sex, and BMI as covariates. Before performing statistical tests, the assumptions of normality and homogeneity of variances of residuals were tested using the Shapiro-Wilk test and Levene test, respectively (eMethods). The 95% 2-sided CI for mean and mean difference estimates were computed using the bias-corrected and accelerated (BCa) bootstrap interval. Mean and mean difference estimates were presented together with 2-sided 95% BCa CI.
To avoid inflation of α risk due to multiplicity, no p values were presented if the primary efficacy end point did not reach the significance level. In this case, mean and mean difference estimates were presented with the 95% BCa CI for all secondary end points without statistical testing, and these analyses should be considered exploratory.
For DTI analyses, FSL randomize with 5,000 permutations was used to make voxel-wise comparisons of the DTI maps between and within patient groups with family-wise error–corrected significance at α = 0.05. The mean of each region that showed significant differences was extracted using the John Hopkins University white matter atlas. Furthermore, the threshold-free cluster connectivity–based fixel enhancement was used to test for significant voxel-wise differences of FD, FC, and FDC between and within patient groups with family-wise error–corrected significance at α = 0.05.
For the comparison between the triheptanoin group and the external placebo control group at 12 months, the propensity score model was developed using logistic regression. Two propensity score models were developed, inverse probability of treatment–weighted (IPTW) and double robust (DR) estimation. A linear regression adjustment model was performed for sensitivity analysis. These models included the following covariates: age, sex and BMI; the outcome regression model of DR and regression adjustment model further included baseline value of the outcome of interest. Standardized difference was used to assess any differences in baseline characteristics between treatment groups before and after propensity score weighting.
Data Availability
The study protocols (initial and amended) are available in eSAP 1 and eSAP 2 and the statistical analysis plan in eSAP 3. Anonymized data not published within this article will be made available by request from any qualified investigator.
Results
Participants
TRIHEP3 patients were recruited in Paris and Leiden from June 2015 to December 2017. The primary outcome analysis did not find a significant site by treatment interaction (eTable 1). A total of 107 patients were enrolled in the study, and 100 were randomized. Five of them withdrew consent before randomization, whereas 2 were excluded before randomization because of failure to undergo the MRI (claustrophobia). Eighty-six patients completed the 1-year TRIHEP3 study (Figure 1); 4 patients withdrew because of psychological issues or insufficient family support in Paris and Leiden and 10 patients because of gastrointestinal issues in Leiden. In addition, cBSI analyses could not be performed in 6 patients at 6 months and 5 patients at 12 months, because of extensive motion artifacts during scanning. No change in neuroleptics doses occurred during the trial. Baseline characteristics of the patients are summarized in Table 1. The gender imbalance between groups (more women in the triheptanoin arm) was considered during statistical analyses—that is, adjusted analyses were performed using ANCOVA with baseline value, age, sex, and BMI as covariates. There was no difference between groups for TMS (eTable 2).
| Triheptanoin | Placebo | |
|---|---|---|
| N = 50 | N = 50 | |
| Age, y | 50 (45; 58) | 48 (40; 58) |
| Sex, no. (%) | ||
| Male | 18 (36.0) | 30 (60.0) |
| Female | 32 (64.0) | 20 (40.0) |
| Body mass index, kg/m2 | 23.9 (21.7; 25.5) | 22.8 (21.4; 25.1) |
| TFC score at baseline | 12.0 (10.0; 13.0) | 11.0 (10.0; 13.0) |
| No. of CAG repeats on the pathologic allele | 43 (41; 44) | 44 (42; 45) |
| UHDRS TMS at baseline | 15.5 (12.0; 27.0) | 17.5 (13.0; 25.0) |
| Medications, n (%) | 21 (42.0) | 19 (38.0) |
| SSRI | 18 | 16 |
| Neuroleptics | 5 | 6 |
Abbreviations: TFC = total functional capacity; TMS = total motor score; UHDRS = Unified Huntington's Disease Rating Scale.
Efficacy
Primary End Point
There was no significant difference in caudate atrophy at 6 months between patients on triheptanoin or placebo (Table 2).
| Triheptanoin Mean (95% CI) | ANCOVAa Placebo Mean (95% CI) | Mean difference (95% CI) | p Value | |
|---|---|---|---|---|
| Caudate boundary shift at 6 mo | 0.026 (0.018 to 0.033)b | 0.023 (0.014 to 0.032)b | 0.002 (−0.011 to 0.013)b | 0.723 |
| Caudate boundary shift at 12 mo | 0.036 (0.027 to 0.045)b | 0.034 (0.023 to 0.045)b | 0.002 (−0.013 to 0.017)b |
Abbreviation: cBSI = caudate boundary shift integral.
a
Adjusted for age, BMI, and sex.
b
95% CIs using the bootstrap bias-corrected and accelerated method.
Secondary End Points
There was no difference in caudate atrophy at 12 months between groups—for example, difference in means 0.002 [−0.013 to 0.017] (Table 2)—nor between 6 and 12 months (eTable 3).
There was no difference in TMS from baseline to 6 months between patient groups (Table 3), but TMS was stable in patients treated with triheptanoin for 12 months, whereas it increased in patients treated for only 6 months (0.6 [−1.1 to 2.1] vs 2.5 [1.2 to 3.8], Table 3), an effect that seemed more pronounced between 6 and 12 months (−0.4 [−1.6 to 0.7] vs 1.7 [0.3 to 3.0], Table 3).
| Triheptanoin Mean (95% CI) | ANCOVAa Placebo Mean (95% CI) | Mean difference (95% CI) | |
|---|---|---|---|
| Absolute change (between baseline and mo 6) in TMS | 1.3 (0.0 to 2.7)b | 0.7 (−0.4 to 2.0)b | 0.6 (−1.0 to 2.5)b |
| Absolute change (between baseline and 1 y) in TMS | 0.6 (−1.1 to 2.1)b | 2.5 (1.2 to 3.8)b | −1.9 (−4.2 to −0.0)b |
| Absolute change (between mo 6 and 1 y) in TMS | −0.4 (−1.6 to 0.7)b | 1.7 (0.3 to 3.0)b | −2.1 (−4.1 to −0.3)b |
Abbreviation: TMS = total motor score.
a
Adjusted for baseline TMS, age, BMI, and sex.
b
95% CIs using bootstrap bias-corrected and accelerated method.
There was no difference in TFC, PBA-s, short cognitive assessments (SDMT, Stroop Test, Digit Span, TMT), and SF-36 between patient groups from baseline to 6 months or from baseline to 12 months (data not shown).
Measurements of the Pi/PCr ratio could not be performed because of a deleterious increase in the NMR signal in both centers—that is, in Paris, change in NMR magnet soon after the study start, and in Leiden, failure to obtain a sufficient brain activation after visual stimulation. No other neuroimaging measurements were altered by technical issues.
Diffusion imaging using FBA, a novel method to investigate the contribution of individual crossing fibers to white matter integrity, revealed reduction in FD (i.e., fiber trophicity) in the corpus callosum, corona radiate, and internal capsules of patients treated with triheptanoin, which were more pronounced at 6 months than at 12 months of treatment (Figure 2). Instead, FBA showed increased FC (i.e., total number of fibers in a voxel) in the corpus callosum at both 6 and 12 months in patients treated with triheptanoin since baseline while increased FC was only measured at 12 months in patients who started triheptanoin at 6 months (Figure 2).
Fixel-based analysis revealed microstructural alterations in both patient groups of the TRIHEP3 study. Compared with baseline, significant reduction in fixels was observed in the corpus callosum, corona radiate, and internal capsules, as early as 6 months (A), which persisted at 12 months (B), in patients treated since baseline. Of interest, in the patients treated from baseline, recovery in the FC started at 6 months (C) and continued at 12 months (D), while in patients treated after 6 months, recovery began at 12 months (E).
Comparison With the External Placebo Control Group at 12 Months
We first validated that there was no difference at baseline for age, sex, BMI, number of CAG repeats on the pathologic allele, TFC, TMS, and medications between the treated arm of TRIHEP3 (n = 50) and the placebo arm of REV-HD (n = 51) (eTable 4). In addition, we applied a propensity score method to further reduce any potential imbalance between groups—which included age, sex, BMI, number of CAG repeats on the pathologic allele, TFC, TMS, and medications—as demonstrated by almost null calculated standardized differences for all these variables after propensity score–based adjustment (eTable 5). There was no difference at baseline for age, sex, BMI, number of CAG repeats on the pathologic allele, TFC, TMS, and medications between the placebo arm of TRIHEP3 and the placebo arm of REV-HD (eTable 6), either.
Compared with the external placebo control group (n = 48), caudate atrophy decreased by approximately 50% in patients treated with triheptanoin (n = 41) (0.038 [0.028–0.048] vs 0.070 [0.059–0.082], eTable). Furthermore, TMS was stable in patients treated with triheptanoin for 12 months, whereas it increased in the placebo group (0.66 [−1.07 to 2.48] vs 2.65 [1.38 to 3.89], eTable 8). In patients at the early stage of HD, a change in TMS of 2.1 is considered clinically meaningful.29
Safety, Tolerance, and Metabolism
Overall, triheptanoin was well tolerated and body weight remained very stable over 1 year. In total, 330 nonserious adverse events occurred in 95 participants, including 10 events definitely related, 47 probably related, and 100 possibly related. Most adverse events were gastrointestinal issues (diarrhea, stomachache, flatulence, or bloating), which were observed occasionally, especially if the oil was improperly ingested (e.g., too quickly or not mixed with semisolid or solid foods). These events were observed with both triheptanoin and placebo (safflower) oils and usually resolved with dietary management. Four serious adverse events occurred but all unrelated to the study drug. Altogether, only 14 patients withdrew during the study because dietary restrictions were overall well tolerated.
Urine organic acid profiles confirmed excellent compliance with the detection of 6-hydroxyheptanoic acid in all patients treated with triheptanoin. Biochemical plasma analyses showed proper metabolism of triheptanoin with increased propionyl-carnitine (0.49 ± 0.17 μmol/L vs 1.93 ± 0.86 μmol/L) and C5 ketone bodies—3-hydroxypentanoate (0.34 ± 0.20 μmol/L vs 16.18 ± 15.49 μmol/L) and 3-ketopentanoate (0.27 ± 0.11 μmol/L vs 8.64 ± 6.18 μmol/L)—in patients treated with triheptanoin.
This study provides Class I evidence that triheptanoin does not slow caudate atrophy compared with placebo over 6 months in patients with early HD.
Discussion
The 6-month randomized, double-blind, placebo-controlled TRIHEP3 study did not meet its primary end point, that is, decreased rate of caudate atrophy in patients with HD treated with triheptanoin. However, motor functions were stable in patients treated with triheptanoin for 12 months. This observation led us to compare clinical and neuroimaging measures of patients treated with triheptanoin for 12 months in TRIHEP3 with those of an external placebo control group of patients with HD. These analyses showed approximately a 50% reduction in the rate of caudate atrophy and confirmed motor stability in patients treated with triheptanoin.
While there was no difference in TMS between TRIHEP3 patient groups during the first 6 months of the study, we observed different time-dependent profiles from 6 to 12 months in favor of patients treated with triheptanoin from baseline. This resulted in stable TMS at 12 months in patients treated with triheptanoin for 1 year compared with patients treated for 6 months only. Because triheptanoin targets brain energy deficiency in HD, a short-term symptomatic effect on motor disorders was unlikely. Furthermore, the delayed clinical effect we observed could be underlined by delayed biological effects of triheptanoin, as supported by alterations on FBA metrics at 6 months that improved at 12 months. Triheptanoin may thus act as a disease-modifying drug, capable of slowing HD progression after a sufficient treatment period.
The TRIHEP3 study only had a controlled arm for 6 months. To better assess the treatment effects of triheptanoin at 12 months, we used an external placebo comparative group from the REV-HD study, which was conducted simultaneously to TRIHEP3, with the same number of site visits and the same outcome measures. The placebo arms of both studies were comparable at baseline, as well as the treated arm of TRIHEP3 and the placebo arm of REV-HD. Still, we were aware that inclusion of patients in one study or in the other could be influenced by factors that can bias the comparisons. Thus, even if the characteristics of the patients seemed similar, we used propensity score methods to further reduce the possibility of bias. Unlike patients treated with triheptanoin, patients on placebo displayed clinically meaningful disease progression,29 as expected in HD (i.e., mean of 2.5 TMS increase per year).18,19 This clinical stability under triheptanoin was associated with a 50% decrease in the rate of caudate atrophy over 1 year compared with placebo.
Triheptanoin was overall well tolerated. Despite dietary constraints, adding to the many constraints of living with HD, only few patients withdrew because of digestive side effects. This is especially relevant because, despite significant hope in antisense oligonucleotides designed to reduce the production of the huntingtin protein and slow HD progression (NCT03761849), the study did not meet its primary end point22 and was discontinued because of concerns about toxicity.23 More recently, negative but promising results have been reported with the anti-inflammatory drug laquinimod (LEGATO-HD study, NCT02215616).30 Laquinimod failed to show significant clinical improvement but reduced caudate volume loss after 1 year. Similarly, triheptanoin represents a new therapeutic strategy in HD, which may be envisaged as an add-on treatment to other disease-modifying therapies currently in clinical trials, such as gene therapy, small-molecule splicing modulators, or allele-selective antisense oligonucleotides.31 Still, owing to the post hoc nature of our findings, our encouraging results need to be confirmed in a multicentric phase 3 study with a longer double-blind phase, using TMS and caudate atrophy as end points.
Data Access, Responsibility, and Analysis
Prof. Fanny Mochel had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Glossary
- BCa
- bias-corrected and accelerated
- cBSI
- caudate boundary shift integral
- DR
- double robust
- FBA
- fixel-based analysis
- FC
- fiber cross section
- FD
- fiber density
- FDC
- fiber density and cross section
- FOD
- fiber orientation distribution
- HD
- Huntington disease
- IQR
- interquartile range
- LC-FAOD
- long-chain fatty acid oxidation disorder
- PBA-s
- Problem Behaviors Assessment–Short
- PC
- pyruvate carboxylase
- SDMT
- Symbol Digit Modalities Test
- TFC
- total functional capacity
- TMS
- total motor score
- TMT
- Trail Making Test
- UHDRS
- Unified Huntington's Disease Rating Scale
Acknowledgment
The authors thank all the patients who have participated in the studies and Bernardo Blanco Sanchez for technical assistance.
Supplementary Materials
eMethods
- Download
- 113.09 KB
eSAP_1
- Download
- 1.14 MB
eSAP_2
- Download
- 1.14 MB
eSAP_3
- Download
- 624.80 KB
eTables
- Download
- 176.13 KB
References
1.
Adanyeguh IM, Branzoli F, Delorme C, et al. Multiparametric characterization of white matter alterations in early stage Huntington disease. Sci Rep. 2021;11(1):13101.
2.
Zhang J, Gregory S, Scahill RI, et al. In vivo characterization of white matter pathology in premanifest huntington's disease. Ann Neurol. 2018;84(4):497-504.
3.
Bélanger M, Allaman I, Magistretti PJ. Brain energy metabolism: focus on astrocyte-neuron metabolic cooperation. Cell Metab. 2011;14(6):724-738.
4.
Grafton ST, Mazziotta JC, Pahl JJ, et al. Serial changes of cerebral glucose metabolism and caudate size in persons at risk for Huntington's disease. Arch Neurol. 1992;49(11):1161-1167.
5.
Kuwert T, Lange HW, Boecker H, et al. Striatal glucose consumption in chorea-free subjects at risk of Huntington's disease. J Neurol. 1993;241(1):31-36.
6.
Djoussé L, Knowlton B, Cupples LA, Marder K, Shoulson I, Myers RH. Weight loss in early stage of Huntington's disease. Neurology. 2002;59(9):1325-1330.
7.
Powers WJ, Videen TO, Markham J, et al. Selective defect of in vivo glycolysis in early Huntington's disease striatum. Proc Natl Acad Sci U S A. 2007;104(8):2945-2949.
8.
Buck E, Zügel M, Schumann U, et al. High-resolution respirometry of fine-needle muscle biopsies in pre-manifest Huntington's disease expansion mutation carriers shows normal mitochondrial respiratory function. PLoS One. 2017;12(4):e0175248.
9.
Josefsen K, Nielsen SMB, Campos A, et al. Reduced gluconeogenesis and lactate clearance in Huntington's disease. Neurobiol Dis. 2010;40(3):656-662.
10.
Mochel F, Charles P, Seguin F, et al. Early energy deficit in Huntington disease: identification of a plasma biomarker traceable during disease progression. PLoS One. 2007;2(7):e647.
11.
Adanyeguh IM, Rinaldi D, Henry PG, et al. Triheptanoin improves brain energy metabolism in patients with Huntington disease. Neurology. 2015;84(5):490-495.
12.
Roe CR, Sweetman L, Roe DS, David F, Brunengraber H. Treatment of cardiomyopathy and rhabdomyolysis in long-chain fat oxidation disorders using an anaplerotic odd-chain triglyceride. J Clin Invest. 2002;110(2):259-269.
13.
Vockley J, Burton B, Berry G, et al. Effects of triheptanoin (UX007) in patients with long-chain fatty acid oxidation disorders: results from an open-label, long-term extension study. J Inherit Metab Dis. 2021;44(1):253-263.
14.
Roe CR, Brunengraber H. Anaplerotic treatment of long-chain fat oxidation disorders with triheptanoin: review of 15 years experience. Mol Genet Metab. 2015;116(4):260-268.
15.
Vockley J, Burton BK, Berry G, et al. Triheptanoin for the treatment of long-chain fatty acid oxidation disorders: final results of an open-label, long-term extension study. J Inherit Metab Dis. 2023;46(5):943-955.
16.
Guffon N, Mochel F, Schiff M, De Lonlay P, Douillard C, Vianey-Saban C. Clinical outcomes in a series of 18 patients with long chain fatty acids oxidation disorders treated with triheptanoin for a median duration of 22 months. Mol Genet Metab. 2021;132(4):227-233.
17.
Mochel F, DeLonlay P, Touati G, et al. Pyruvate carboxylase deficiency: clinical and biochemical response to anaplerotic diet therapy. Mol Genet Metab. 2005;84(4):305-312.
18.
Mochel F, Hainque E, Gras D, et al. Triheptanoin dramatically reduces paroxysmal motor disorder in patients with GLUT1 deficiency. J Neurol Neurosurg Psychiatry. 2016;87(5):550-553.
19.
Hainque E, Gras D, Meneret A, et al. Long-term follow-up in an open-label trial of triheptanoin in GLUT1 deficiency syndrome: a sustained dramatic effect. J Neurol Neurosurg Psychiatry. 2019;90(11):1291-1293.
20.
De Giorgis V, Bhatia KP, Boespflug-Tanguy O, et al. Triheptanoin did not show benefit versus placebo for the treatment of paroxysmal movement disorders in Glut1 deficiency syndrome: results of a randomized phase 3 study. Mov Disord. 2024;39(8):1386-1396.
21.
Roze E, Méneret A, Hainque E, Mochel F. Comment on De Giorgis et al. “Randomized phase 3 study of triheptanoin for glut1 deficiency syndrome–associated paroxysmal movement disorders. Mov Disord. 2024;39(10):1898-1899.
22.
Dickey AS, La Spada AR. Therapy development in Huntington disease: from current strategies to emerging opportunities. Am J Med Genet A. 2018;176(4):842-861.
23.
Hobbs NZ, Henley SMD, Wild EJ, et al. Automated quantification of caudate atrophy by local registration of serial MRI: evaluation and application in Huntington's disease. Neuroimage. 2009;47(4):1659-1665.
25.
Veraart J, Novikov DS, Christiaens D, Ades-Aron B, Sijbers J, Fieremans E. Denoising of diffusion MRI using random matrix theory. Neuroimage. 2016;142:394-406.
26.
Kellner E, Dhital B, Kiselev VG, Reisert M. Gibbs-ringing artifact removal based on local subvoxel-shifts. Magn Reson Med. 2016;76(5):1574-1581.
27.
Raffelt DA, Tournier JD, Smith RE, et al. Investigating white matter fibre density and morphology using fixel-based analysis. Neuroimage. 2017;144(pt A):58-73.
28.
Tabrizi SJ, Scahill RI, Owen G, et al. Predictors of phenotypic progression and disease onset in premanifest and early-stage Huntington's disease in the TRACK-HD study: analysis of 36-month observational data. Lancet Neurol. 2013;12(7):637-649.
29.
Hamilton JL, Mills JA, Stebbins GT, et al. Defining clinical meaningfulness in huntington's disease. Mov Disord. 2023;38(6):1036-1043.
30.
Reilmann R, Anderson KE, Feigin A, et al. Safety and efficacy of laquinimod for Huntington's disease (LEGATO-HD): a multicentre, randomised, double-blind, placebo-controlled, phase 2 study. Lancet Neurol. 2024;23(3):243-255.
31.
Estevez-Fraga C, Tabrizi SJ, Wild EJ. Huntington's disease clinical trials corner: March 2024. J Huntingtons Dis. 2024;13(1):1-14.
Information & Authors
Information
Published In
Copyright
© 2024 American Academy of Neurology.
Publication History
Received: April 11, 2024
Accepted: November 14, 2024
Published online: December 30, 2024
Published in issue: January 28, 2025
Disclosure
F. Mochel received consultancy fees from Minoryx therapeutics and Ultragenyx pharmaceuticals. A. Méneret received speaker honoraria from AbbVie and travel funding from Merz-Pharma and Elivie. I. Adanyeguh, C. Giron, E. Hainque, M.P. Luton, M. Atencio, R. Debs, M. Jacobs, F. C.M. Veldkamp, M. Barbier, E.M. Coppen, K.F. van der Zwaan, A. Diallo, C. Ottolenghi, E. Vicaut, and R.A. Roos report no disclosures. A. Durr received research support for her institution from NIH, ANR, Biogen, Wavelife, Askbio, Servier, and Triplets Therapeutics; and served on advisory boards of Roche and UCB. Go to Neurology.org/N for full disclosures.
Study Funding
The study was funded by Ultragenyx Pharmaceuticals.
Authors
Author Contributions
F. Mochel: drafting/revision of the manuscript for content, including medical writing for content; study concept or design; analysis or interpretation of data. A. Méneret: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data. I.M. Adanyeguh: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data. C. Giron: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data. E. Hainque: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data. M-P. Luton: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data. M. Atencio: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data. R. Debs: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data. M. Jacobs: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data. F.C.M. Veldkamp: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data. M. Barbier: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data. E.M. Coppen: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data. K.F. van der Zwaan: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data. A. Diallo: drafting/revision of the manuscript for content, including medical writing for content; analysis or interpretation of data. C. Ottolenghi: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data. E. Vicaut: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data. R.A. Roos: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design. A. Durr: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design.
Metrics & Citations
Metrics
Citation information is sourced from Crossref Cited-by service.
Citations
Download Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Select your manager software from the list below and click Download.
Cited By
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginPurchase Options
The neurology.org payment platform is currently offline. Our technical team is working as quickly as possible to restore service.
If you need immediate support or to place an order, please call or email customer service:
- 1-800-638-3030 for U.S. customers - 8:30 - 7 pm ET (M-F)
- 1-301-223-2300 for customers outside the U.S. - 8:30 - 7 pm ET (M-F)
- [email protected]
We appreciate your patience during this time and apologize for any inconvenience.