Answers are grounded in this paper and cite their sources.
Background: The Rett Syndrome Behavior Questionnaire (RSBQ), a caregiver-reported measure that assesses behavioral manifestations of Rett syndrome (RTT), is used as a primary endpoint in RTT clinical trials, including the pivotal trofinetide studies. However, a minimal clinically important difference (MCID) for the RSBQ mean total score has not been established, limiting patient-level interpretation of change. Methods: MCID on the RSBQ was established using distribution- and anchor-based methods from data in RETT-002 (phase 2, 6-week, double-blind, randomized, placebo-controlled; N = 82) and LAVENDER (phase 3, 12-week, double-blind, randomized, placebo-controlled; N = 187) trofinetide trials. While distribution-based estimates were calculated as 0.4–0.5 × pooled standard deviation (SD) across treatment arms, anchor-based MCIDs were derived using between-patient and within-patient score changes on the Clinical Global Impressions-Improvement (CGI-I) scale by applying the upper bound of the 95% limits of agreement. The percentage of responders at week 12 was evaluated using MCID-based (who fit the responder criteria using both distribution and anchor-based methods) and threshold-based (≥15% improvement) mean total RSBQ score change from baseline. Adjusted multivariable logistic regressions assessed the association between treatment (trofinetide vs. placebo) and MCID-responder status. Sensitivity analysis examined patients who were both RSBQ responders and CGI-I responders (i.e., any improvement). Results: Pooled SDs yielded MCID ranges of ∼4.6–5.9 points (baseline) and 3.2–4.4 points (change from baseline) using distribution methods while anchor-based estimates ranged from ∼3–9 points. Together, these findings supported an overall clinically meaningful improvement range of 3–9 points on the RSBQ. A ≥6-point improvement in mean total RSBQ score, representing the midpoint of the MCID range across distribution- and anchor-based methods, was selected as the conservative responder threshold. At week 12 in LAVENDER, trofinetide versus placebo had higher percentages of MCID-based (42.11% vs. 22.35%; p < 0.05) and threshold-based (39.47% vs. 18.82%; p < 0.05) responders, respectively. Trofinetide was associated with greater than two-fold higher odds of response versus placebo (adjusted OR: 2.54; 95% CI: 1.24–5.17; p < 0.05). Sensitivity analyses also favored trofinetide versus placebo (21.05% vs. 5.88%; p < 0.05). Conclusions: Overall, a 3- to 9-point improvement on the mean total score, which may reflect meaningful changes in day-to-day symptoms such as mood, anxiety, and repetitive behaviors, was estimated to be the MCID range for RSBQ among individuals with RTT. In clinical trials, trofinetide versus placebo had a higher likelihood of achieving MCID- and threshold-based clinically meaningful improvements.
No takes yet. Share an insight, caveat, or question.
Rajagopalan et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: