Table 3.
Outcome measures changes in placebo-treated groups with FXS in failed clinical trials–Sensitivity Analysis.
| Endpoint | of Studies | Heterogeneity (I2) | Heterogeneity p-Value | Meta-Analysis Type | Mean Score Change | 95% CI | Change p-Value | |
|---|---|---|---|---|---|---|---|---|
| Lower | Upper | |||||||
| Clinician Rated ‡ | ||||||||
| CGI-I (<12 years) | 3 | 0.97 | 0.00 | Random | −0.45 | −2.00 | 1.10 | 0.570 |
| CGI-I (>12 years) | 4 | 1.00 | 0.00 | Random | 0.46 | −1.83 | 2.76 | 0.693 |
| Caregiver Rated ‡ | ||||||||
| Vineland-II Comp (<12 years) | 2 | 0.00 | 0.47 | Fixed | 1.71 | 0.38 | 3.05 | 0.012 * |
| ABC-CFX Irr (<12 years) | 3 | 0.85 | 0.00 | Random | −3.49 | −8.32 | 1.35 | 0.157 |
| ABC-CFX SU/L (<12 years) | 2 | 0.97 | 0.00 | Random | −2.28 | −7.41 | 2.85 | 0.383 |
| ABC-CFX SA (<12 years) | 2 | 0.98 | 0.00 | Random | −1.44 | −4.07 | 1.20 | 0.284 |
| VAS Comp (<12 years) | 2 | 0.00 | 0.42 | Fixed | 1.54 | 0.90 | 2.17 | 0.000 * |
| Vineland-II Comp (>12 years) | 4 | 0.00 | 0.44 | Fixed | 1.74 | 0.08 | 3.39 | 0.040 * |
| ABC-CFX Total (>12 years) | 2 | 0.00 | 0.72 | Fixed | −8.60 | −12.45 | −4.76 | <0.001 * |
| ABC-CFX Irr (>12 years) ** | 4 | 0.91 | 0.00 | Random | −1.96 | −5.11 | 1.19 | 0.222 |
| ABC-CFX SU/L (>12 years) ** | 4 | 0.93 | 0.00 | Random | −1.90 | −4.51 | 0.71 | 0.154 |
| ABC-CFX Stereo (>12 years) | 2 | 0.00 | 0.69 | Fixed | −1.04 | −1.58 | −0.50 | <0.001 * |
| ABC-CFX Hyper (>12 years) | 2 | 0.00 | 0.84 | Fixed | −1.37 | −2.26 | −0.49 | 0.002 * |
| ABC-CFX SA (>12 years) | 3 | 0.79 | 0.01 | Random | −1.65 | −2.61 | −0.68 | 0.001 * |
Abbreviations: FXS, fragile X syndrome; CGI-I, Clinician Global Impression-Improvement; Vineland-II Comp, Vineland Adaptive Behavior Scale-2nd edition, Composite; ABC-CFX Irr, Aberrant Behavior Checklist-Community Irritability refactored for FXS; ABC-CFX SU/L, Socially Unresponsive/Lethargy; ABC-CFX Hyper, Hyperactivity; ABC-CFX Stereo, Stereotypy; ABC-CFX SA, Social Avoidance; ABC-CFX Total, Total Score; VAS Comp, Visual Analog Scale Composite. ‡ Meta-Analysis was conducted on subscales that were used in 2 or more papers, using a pre-post correlation of 0.5. Meta-analysis conducted using a pre-post correlation of 0.0 has almost identical values. * Indicates p-value < 0.05. ** The ABC-C subscale was used for lethargy and irritability in 2 studies, in contrast to the ABC-CFX subscale used in other studies.