News
Article
Author(s):
Leveraging the Enroll-HD dataset, recent data revealed consistent links between antidopaminergic medication use and worsened cognitive and motor outcomes in Huntington disease.
Michael Hayden, PhD, MBChB, FRCP
A recently published, 2-year causal analysis supported and extended previously reported findings that suggest antidopaminergic medications (ADM) worsen outcome measures in patients with Huntington disease (HD). To further validate these data, investigators called for future clinical trials that include a balanced randomization of patients with ADMs and a statistical plan that evaluates outcomes both with and without ADMs.1
Published in Movement Disorders, the causal analysis emulated a randomized controlled trial (RCT) using the most recent Enroll-HD dataset, the largest observational study of HD, to assess the impact of antipsychotics and VMAT2 inhibitors on motor, cognitive, and functional outcomes. The final analyzed dataset comprised 1173 participants, 380 exposed to either of these medications, and 793 unexposed.
For the study, investigators adopted and built a new-user design that required 3 contiguous annual visits. Similar to an RCT, the on-ADM participants were required to be off ADM medications for the first visit and then on-ADMs for all follow-up visits. Off-ADM participants had to be ADM naïve during all visits, with no exposure within or beyond the analysis window. In terms of outcome measures, the authors pulled data from 12 assessments within the Unified Huntington’s Disease Rating Scale (UHDRS), as well as several subscales of the TMS, which is included in the composite UHDRS (cUHDRS).
Senior author Michael Hayden, PhD, MBChB, FRCP, a professor in the Department of Medical Genetics, University of British Columbia, and colleagues used a doubly robust targeted maximum likelihood estimation (TMLE) model to control for 27 covariates. In the first analysis, the on-ADM group exhibited significantly lower mean outcomes at 2 years compared with the off-ADM group across medication classes, except for cUHDRS in the VMAT2 group, which was not significant. Notably, motor estimates revealed consistently higher bradykinesia means in the exposed group for both ADM and antipsychotic medications, as well as larger means for gait balance and hand movement.
READ MORE: Huntington Drug PTC518 Meets Primary End Point in Phase 2 Study
The new-user design was adjusted for a large number of baseline covariates using state-of-the-art statistical methods for causal analysis. Overall, the analytical platform captured baseline characteristics before the initiation of ADM use which helped control for possible confounding factors, that is, disease severity, prior medication use, psychiatric comorbidities, age, sex, and genetic predispositions, ensuring a more accurate estimation of treatment effects.
In the second analysis, the study authors estimated the average treatment effect (ATE) for dose groups of combinations of medications. To maximize the sample size, combined dose groups were formed by aggregating the groups among certain medications. Overall, there were 5 formed groups, including all ADMs, antipsychotics, newer antipsychotics, older antipsychotics, and VMAT2 inhibitors. The same unexposed off-ADM group, TMLE, and covariates were used, with each dose group compared to the unexposed group to yield two ATEs per medication group: low versus unexposed and high versus unexposed.
Data from the second analysis further showed that those on high doses of ADMs and antipsychotics had consistently smaller mean clinical outcomes for cUHDRS, Symbol Digit Modalities Test, and SWR compared with the off-ADM group. Overall, patients on high-dose antipsychotics demonstrated lower SWR and TFC scores than those off-ADM, while VMAT2 inhibitors impacted SWR and SDMT across doses. In addition, higher doses of ADMs led to worsened clinical measures of cognition and in cUHDRS, which were statistically reliable. Investigators observed no differences in any of the outcomes based on antidepressant usage.