Frequently Asked Questions

In adults aged 65 years and older, valbenazine was associated with clinically meaningful and sustained improvement in tardive dyskinesia over 48 weeks. Mean change in AIMS total score was 34.5 b1 0.7 at Week 8 (n = 50), 38.6 b1 0.9 at Week 24 (n = 26), and 38.8 b1 0.9 at Week 48 (n = 28), which exceeded the estimated minimal clinically important difference of 32 points at all analyzed timepoints.

Improvement was evident by Week 8 in the elderly subgroup. By that time, mean AIMS total score had improved by 34.5 b1 0.7, 58.0% of participants had achieved at least 30% improvement from baseline, and 40.0% had achieved at least 50% improvement; response rates then increased further through Week 24 and were sustained through Week 48.

Among participants aged 65 years and older, the proportion with a clinically meaningful AIMS response of at least 30% improvement increased from 58.0% at Week 8 to 88.5% at Week 24 and 89.3% at Week 48. Using the more stringent protocol-defined threshold of at least 50% improvement, response rates increased from 40.0% at Week 8 to 65.4% at Week 24 and 82.1% at Week 48.

Yes. In the elderly subgroup, the proportion rated as much improved or very much improved increased over time on both clinician- and patient-reported global measures and was sustained through Week 48. At Week 48, 92.9% had a CGI-TD score of 2 or less, and 85.7% had a PGIC score of 2 or less.

At Week 48, efficacy outcomes were generally similar between age groups, with one significant difference favoring the elderly subgroup on clinician-rated global improvement. CGI-TD response occurred in 92.9% of participants aged 65 years and older versus 76.5% of those younger than 65 years (P = .050), while PGIC response in the elderly subgroup was 85.7% and was not significantly different from younger adults; the discussion also states there were no statistical age-group differences for AIMS outcomes at Week 48.

Week 48 results were numerically larger with 80 mg than with 40 mg in elderly participants, but the authors state that no conclusions can be drawn about dose effects from these analyses. Mean AIMS total score change at Week 48 in adults aged 65 years and older was 36.4 with 40 mg (n = 8) and 39.8 with 80 mg (n = 20), and a greater proportion met AIMS response thresholds with 80 mg than with 40 mg in both age groups.

Overall tolerability was broadly similar between older and younger participants for treatment-emergent adverse events and serious treatment-emergent adverse events, with no statistically significant age-group differences for those outcomes. However, discontinuation due to a treatment-emergent adverse event was higher in the elderly subgroup than in younger participants: 25.5% versus 13.3% (P = .023).

In participants aged 65 years and older, urinary tract infection and somnolence were among the common treatment-emergent adverse events, each occurring in 10.9% of patients. Fatigue occurred in 3.6%, and dizziness and falls each occurred in 5.5% of elderly participants.

No worsening signal was identified in these analyses. Psychiatric status remained stable throughout treatment in elderly participants on PANSS, CDSS, YMRS, and MADRS, and mean changes on BARS and SAS indicated no emergence or worsening of akathisia or drug-induced parkinsonism; Week 48 changes were minimal and did not differ significantly from younger participants.

This was a post hoc pooled analysis of 2 long-term phase 3 valbenazine studies: the KINECT 3 extension and KINECT 4. The pooled analysis included 304 participants, 55 of whom were aged 65 years and older, and evaluated outcomes through up to 48 weeks using descriptive analyses with no imputation for missing values; because this was post hoc, the elderly subgroup was relatively small, analyses were not corrected for multiple comparisons, and the authors state that dose-effect conclusions cannot be made.

The main limitations are that these were post hoc analyses, the elderly subgroup was small (55 participants versus 249 younger participants), and the study population may not represent the broader tardive dyskinesia population because participants had to be psychiatrically stable before entry. The authors also note that analyses were not corrected for multiple comparisons, no conclusions can be drawn about dose effects because KINECT 4 dosing was based on investigator judgment of tolerability and response, and selection bias from study dropouts is possible, especially given the higher rate of adverse-event discontinuation in elderly participants.