Table 2.
Power of Detecting Significant Treatment Effect in DR for a Sample Size of 100 per Arm at Two-Sided Alpha = 0.05 and the Sample Size Needed to Achieve 80% Power Based on the Simulated Data Whose Distributions of DR Outcome are Displayed in Figure 4
| Endpoints | RBZ vs Sham | RBZ vs Active Control | ||
| Power (100 per Arm) | Sample Size per Arm | Power (100 per Arm) | Sample Size per Arm | |
| % ≥3-step progression | 25% | ~500 | 7% | >10,000 |
| % ≥2-step progression | 68% | ~160 | 7% | >8000 |
| % ≥3-step improvement | 99% | ~65 | 9% | ~1800 |
| % ≥2-step improvement | >99% | ~25 | 19% | ~680 |
| % ≥3-step progression, no change, ≥3-step improvement | 99% | ~55 | 10% | ~1600 |
| % ≥2-step progression, no change, ≥2-step improvement | >99% | ~25 | 20% | ~650 |
| Change in ETDRS DRSS | >99% | ~25 | 51% | ~200 |
Abbreviations: DR, diabetic retinopathy; DRSS, Diabetic Retinopathy Severity Scale; ETDRS, Early Treatment Diabetic Retinopathy Study; RBZ, ranibizumab.