Low-Dose Benzodiazepines Linked With Worsened Disease Progression in Progressive Supranuclear Palsy
Even after adjusting for indications for benzodiazepine administration and baseline disease severity, results showed greater patient disease worsening while on such medications, regardless of duration and dose.
Anne-Marie Wills, MD
In a post-hoc secondary analysis of a phase 2/3 randomized controlled trial of davunetide for patients with progressive supranuclear palsy (PSP), findings showed significant differences in PSP Rating Scale (PSPRS) among those taking vs not taking benzodiazepines. Although low-dose benzodiazepines are commonly prescribed for this patient group, the results suggest an increased rate of worsening while on them.1
The study, originally published in 2014, randomly assigned patients with PSP 1:1 to either davunetide 30 mg twice daily or placebo for a 52-week treatment period, with change on the PSPRS and Schwab and England Activities of Daily Living scale as the primary end points. In the post-hoc analysis, least-square means (LSM) adjusted for missing values using model parameter estimates were computed for each medication at each time point and used to estimate within-medication mean PSPRS changes from baseline and between-group differences in change from baseline.
Led by Anne-Marie Wills, MD, director of the CurePSP Center of Care at Massachusetts General Hospital, 305 of the original 313-patient cohort was included in the new analysis. Of these patients, only 1 medication class, benzodiazepine derivatives (lorazepam, clonazepam, alprazolam, and diazepam), was associated with more rapid worsening of PSPRS scores after Bonferroni correction (parameter estimate, 6.98; 95% CI, 4.50-9.47; P <.001). Notably, benzodiazepine-related drugs, such as zolpidem, zopiclone, and eszopiclone, were not associated with worsening scores.
READ MORE: Case Study Shows ITPR1 Variant Associated With Spinocerebellar Ataxia Type 29
When plotting the difference of adjusted LSM of PSPRS scores between each time point, investigators observed a significant difference at weeks 39 (7.52 [95% CI, 6.68-8.37] vs 12.10 [95% CI, 9.43-14.75]) and 52 (10.02 [95% CI, 9.33-11.04] vs 16.69 [95% CI, 13.92-19.45] points; P <.001) between those not taking vs taking benzodiazepines at any point during the study. Notably, there was no significant differences observed between participants who started benzodiazepines before vs after the study began, arguing against faster PSP progressing leading to benzodiazepine prescription.
Investigators performed another analysis assessing change in PSPRS scores after adjusting for indications for benzodiazepine administration and baseline disease severity. This included conditions such as insomnia, anxiety, sleep disturbances, restless legs syndrome, and depression. All told, following adjustment, those taking benzodiazepines experienced a mean worsening of 17.1 (95% CI, 13.30-20.83) PSPRS points per year, vs 9.9 (95% CI, 8.55-11.23) points per year for those not taking benzodiazepines. Duration and dose of benzodiazepine exposure did not appear to be associated with the rate of change in PSPRS, possibly because of the limited sample size.
Despite the results, investigators wrote "We were unable to establish any causal relationship between benzodiazepines and worsening disease. In addition, infrequent study visits precluded our ability to evaluate acute symptomatic effects of benzodiazepines. Since the davunetide trial only enrolled patients with Richardson syndrome, this secondary analysis is not generalizable to other PSP variants."1
In the original study, davunetide failed to distinguish itself from placebo in change on PSPRS (median, 11.8 [95% CI, 10.5-13.0] vs 11.8 [95% CI, 10.5-13.0] P = .41) or SEADL (–0.20 [95% CI, –0.20 to –0.17] vs –0.20 [95% CI, –0.22 to –0.17; P = .92), the primary end points. Davunetide, an eightamino acid peptide that promotes microtubule stability and decreases tau phosphorylation, was considered not an effective treatment for PSP. In addition, 54 serious adverse events were reported in each of the treatment groups, including 11 deaths in the davunetide group and ten in the placebo group.2
