(347) Estimation of the minimal clinically important difference (MCID) and longitudinal change in the tardive dyskinesia impact scale (TDIS), a validated, tardive dyskinesia-specific, patient-reported outcome measure

Background: : Tardive dyskinesia (TD) is a potentially disabling involuntary movement disorder that can impair physical and socioemotional well-being. The 11-item Tardive Dyskinesia Impact Scale (TDIS) is a validated TD-specific, patient-reported outcome measure evaluating TD symptom impact over the previous 7 days evaluated in KINECT®3, KINECT®4, and KINECT-PRO.

Objective: : We aimed to estimate the TDIS minimal clinically important difference (MCID, also known as meaningful score difference) and evaluate change in TDIS total and individual item scores longitudinally with valbenazine.

Methods: : Anchor-based and distribution-based approaches used the Patient Global Impression of Change (PGIC) (range 1-7, higher score=worsening) to estimate TDIS MCID using KINECT3 and KINECT4 trial data. The anchor was a PGIC score=3 (“minimally improved”) at Week 6. TDIS score changes from baseline (CFB) to Week 6 in KINECT3 and Week 8 in KINECT4 were computed from patients who received valbenazine (40-80 mg) or placebo and had PGIC of 3. To assess valbenazine’s effect on TDIS, mean CFB to Week 48 in TDIS total (range=0-44) and individual item scores (range=0-4) were evaluated; higher scores indicate worse functioning.

Results: : Patients reporting minimal improvement (PGIC=3) had TDIS improvements of 3.8-3.9 points by Week 6 and Week 8, establishing an MCID for TDIS of 4 points from anchor- and distribution-based results. By Week 48, CFB in TDIS total score and individual items confirmed long-term efficacy of valbenazine with consistent improvement in TDIS scores for both physical and socio-emotional domains. When applying short-term MCID to longer-term studies, CFB to Week 48 in TDIS total score in both KINECT3 (−8.4) and KINECT4 (−11.0) far exceeded the MCID. The most impacted items at baseline were self-consciousness (mean score=1.99), embarrassment (1.89), and mouth noises (1.80), which were also the most improved at Week 48 (−1.23, −1.20, −1.05, respectively).

Conclusions: : A 4-point MCID was considered clinically meaningful for interpreting the change in TDIS total scores; TDIS total score improved through Week 48 in both trials and the improvement was substantially greater than the MCID. All TDIS items improved over 48 weeks, particularly socioemotional impacts, which supports use of TDIS to measure longitudinal change in TD burden. TDIS can help quantify impact of TD symptoms and provide a benchmark for assessment of treatment effects.

Sponsorship:: Neurocrine Biosciences, Inc.