This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Reported estimates are kept separate from the interpretive commentary that follows them.
1. Trial at a Glance
BRIGHTER was a randomized phase 3 trial asking whether intravitreal ranibizumab 0.5 mg, given alone or combined with laser photocoagulation, improves best-corrected visual acuity (BCVA) at Month 6 more than laser monotherapy in patients with visual impairment due to BRVO.
| Feature | BRIGHTER |
|---|---|
| Phase | Phase 3 |
| Condition | Visual impairment due to branch retinal vein occlusion |
| Design | Randomized, parallel-group, open-label (no masking), active-controlled |
| Arms | 3: ranibizumab 0.5 mg; ranibizumab 0.5 mg + laser; laser monotherapy |
| Enrollment | 455 |
| Primary endpoint | Mean change in BCVA (ETDRS letters) from baseline to Month 6 |
| Primary analysis | ANOVA; difference in least-squares (LS) means versus laser monotherapy |
| ClinicalTrials.gov | NCT01599650 |
| Lead sponsor | Novartis Pharmaceuticals (industry) |
2. Clinical Question
The central question was whether anti-VEGF treatment with ranibizumab, either alone or together with laser, produces a larger mean gain in visual acuity at Month 6 than laser photocoagulation alone in patients whose vision is impaired by BRVO.
Population
Patients with visual impairment due to branch retinal vein occlusion.
Interventions
Ranibizumab 0.5 mg monotherapy, or ranibizumab 0.5 mg combined with laser.
Comparator
Laser monotherapy.
Primary question
Is the mean BCVA change from baseline to Month 6 greater with each ranibizumab-containing arm than with laser alone?
3. Trial Design
Ranibizumab 0.5 mg
- Intravitreal ranibizumab 0.5 mg
- No protocol laser
Ranibizumab 0.5 mg + laser
- Intravitreal ranibizumab 0.5 mg
- Laser photocoagulation
Laser monotherapy
- Laser photocoagulation as control
- Safety reported separately for laser patients who later received ranibizumab 0.5 mg (n = 63) and those who did not (n = 25)
4. Endpoint
| Endpoint | Registry definition | Time frame |
|---|---|---|
| Mean change in visual acuity: BCVA change at Month 6 compared to baseline | BCVA letters measured with an ETDRS-like chart, participant seated, at a testing distance of 4 meters. ETDRS range is 0 to 100 letters. The 95% CI and p-value for the mean change were calculated from a t distribution/t-test and assessed by an ANOVA model. | Baseline, 6 Months |
BCVA is a continuous outcome measured in letters, where higher values mean better vision. A positive change from baseline is an improvement. The between-arm effect measure is therefore a difference in mean letter change, and a positive value favours the ranibizumab arm.
5. Primary Results: Ranibizumab Monotherapy vs Laser
Both posted comparisons used an ANOVA model on patients with both a baseline and a Month 6 BCVA measurement, and each tested superiority against laser monotherapy.
Difference in LS mean BCVA change at Month 6
95% CI: 7.3 to 12.8 · P < 0.0001 · two-sided, superiority
Ranibizumab 0.5 mg monotherapy vs laser monotherapy (ANOVA)
What the estimate means. The model-estimated mean gain in BCVA from baseline to Month 6 was 10.0 letters larger in the ranibizumab monotherapy arm than in the laser arm. Least-squares means are group means adjusted for the other terms in the ANOVA model.
What it does not mean. It is not the change within the ranibizumab arm, and it does not say that every ranibizumab-treated patient gained 10 letters more than they would have with laser. Individual responses vary around the group means.
Precision. The 95% CI of 7.3 to 12.8 letters excludes zero comfortably. Even the lower limit represents a between-group difference of more than 7 letters on the ETDRS scale.
The p-value. P < 0.0001 means the observed difference would be very unlikely if the true mean difference were zero. It says nothing about whether 10 letters is clinically important. That judgment rests on the estimate and CI.
Cautions. The analysis included only patients with both baseline and Month 6 measurements, which is a completers-type population rather than a full intention-to-treat set. Patients who dropped out before Month 6 are excluded, which may bias the comparison if dropout was related to treatment or visual outcome. The trial was also open-label, although BCVA is a standardized chart-based measurement.
6. Primary Results: Ranibizumab Plus Laser vs Laser
Difference in LS mean BCVA change at Month 6
95% CI: 5.8 to 11.6 · P < 0.0001 · two-sided, superiority
Ranibizumab 0.5 mg + laser vs laser monotherapy (ANOVA)
What the estimate means. Patients receiving ranibizumab plus laser had an estimated mean BCVA gain at Month 6 that was 8.7 letters larger than patients receiving laser alone.
What it does not mean. The point estimate is lower than the 10.0 letters for monotherapy, but that does not show that adding laser reduces the benefit of ranibizumab. The two comparisons share the same laser control group and their CIs overlap substantially. The registry does not post a direct comparison of ranibizumab monotherapy with ranibizumab plus laser.
Precision. The CI of 5.8 to 11.6 letters lies entirely above zero and is similar in width to the monotherapy comparison.
The p-value and multiplicity. With two active arms each compared with one control, there are two primary hypothesis tests. The registry does not describe how type I error was controlled across them (for example by a hierarchical order or a Bonferroni-type split). Both p-values are below 0.0001, however, so the conclusion is unlikely to depend on the choice of adjustment.
| Comparison | Effect measure | Estimate (letters) | 95% CI | P-value |
|---|---|---|---|---|
| Ranibizumab monotherapy vs laser monotherapy | Difference in LS means | 10 | 7.3 to 12.8 | <0.0001 |
| Ranibizumab + laser vs laser monotherapy | Difference in LS means | 8.7 | 5.8 to 11.6 | <0.0001 |
The ClinicalTrials.gov record does not report baseline characteristics, secondary endpoint analyses, or subgroup results for BRIGHTER in a form that supports further between-arm inference.
7. Safety: Serious Adverse Events
| Reporting group | Participants with SAEs / at risk |
|---|---|
| Ranibizumab 0.5 mg | 30 / 180 |
| Ranibizumab 0.5 mg + laser | 33 / 183 |
| Laser monotherapy, with later ranibizumab 0.5 mg | 10 / 63 |
| Laser monotherapy, without ranibizumab | 3 / 25 |
SAE counts were similar in the two ranibizumab-containing arms. Splitting the laser arm by later ranibizumab use leaves small reporting groups, and their exposure periods differ, so between-group comparisons of SAE proportions are imprecise and not straightforward to interpret. The registry posts no formal statistical test of SAE rates.
8. Statistical Methodology
Analysis of variance (ANOVA)
BCVA change from baseline to Month 6 was analysed with an ANOVA model, and the difference in least-squares means was estimated for each ranibizumab arm versus laser. In a one-factor ANOVA, treatment is the categorical predictor and the model estimates a mean for each arm. Additional factors, when included, give adjusted (least-squares) means.
where αk is the effect of arm k. The treatment contrast is αranibizumab − αlaser, estimated with a pooled residual variance across all three arms.
t-based confidence intervals
The registry states that the 95% CIs and p-values were based on a t distribution. For each contrast the interval takes the form estimate ± t0.975, df × SE, where the standard error comes from the pooled ANOVA residual variance. Pooling across three arms gives more degrees of freedom than a two-sample t-test and can produce slightly narrower intervals.
Analysis population
The posted analysis population was defined as patients with both a baseline and a Month 6 data point. This is an observed-case analysis. It does not impute missing Month 6 values and does not include every randomized patient.
Multiple comparisons
Two primary contrasts were tested against a shared control. The registry does not describe the procedure used to control familywise type I error across these contrasts.
9. Statistical Methods Explained
Why was an ANOVA used for this endpoint?
BCVA change is a continuous outcome and there were three arms. ANOVA estimates all the group means in one model with a common variance, and specific contrasts, such as each ranibizumab arm versus laser, are then extracted from that model.
What is a "least-squares mean"?
An LS mean is the model's estimate of an arm's average outcome, adjusted for any other factors in the model and evaluated at balanced values of those factors. If treatment is the only factor, the LS mean equals the raw arm mean.
What does a difference of 10.0 letters represent?
On an ETDRS-like chart, letters are read line by line. A 10.0-letter difference means the ranibizumab arm's average gain exceeded the laser arm's average gain by 10.0 letters at Month 6. It is a group-level contrast, not a guaranteed individual gain.
Why does the completers-only population matter?
Restricting the analysis to patients with Month 6 data assumes that those who are missing are similar to those observed. If patients doing poorly on laser were more likely to leave early, for example, the completers analysis could understate or overstate the true difference. Intention-to-treat analyses with principled handling of missing data avoid relying on that assumption.
Does an open-label design undermine a BCVA endpoint?
BCVA is measured under a standardized protocol, but without masking, patients and assessors knew the treatment assignment, which can subtly influence effort during chart reading. Masked BCVA assessors are a common safeguard, and the registry lists no masking for this trial.
10. Limitations and Interpretation Issues
- Open-label design: no masking of participants, care providers or assessors, which can introduce performance or detection bias.
- Observed-case population: the primary analysis used patients with baseline and Month 6 data, not all randomized patients.
- Multiplicity: two primary comparisons against one control group, with no adjustment method described in the registry.
- No direct combination vs monotherapy test: the registry posts only comparisons against laser, so it does not answer whether adding laser to ranibizumab changes outcomes.
- Short primary horizon: the primary endpoint is at Month 6, and later access to ranibizumab in the laser arm complicates any longer-term between-arm comparison.
- Unreported characteristics: the ClinicalTrials.gov record does not provide baseline characteristics or subgroup results for assessing balance or consistency.
11. Why This Trial Matters Statistically
BRIGHTER is a compact example of a multi-arm trial analysed with a single linear model.
| Concept | How it appears in BRIGHTER |
|---|---|
| ANOVA | One model estimates all three arm means with a pooled variance |
| Least-squares means | Adjusted mean BCVA change per arm and their differences |
| Multi-arm trial with shared control | Two ranibizumab arms each compared with laser monotherapy |
| Confidence intervals | t-based 95% CIs for each contrast |
| P-values | Superiority tests with p < 0.0001, interpreted alongside effect size |
| Analysis population | Patients with baseline and Month 6 data (observed cases) |
| Open-label design | Shows why measurement standardization and masking matter |
12. Related Tutorials
Learn more about the methods used in this trial:
13. Related Calculators
14. Sources
- ClinicalTrials.gov: NCT01599650 — Efficacy and Safety of Ranibizumab With or Without Laser in Comparison to Laser in Branch Retinal Vein Occlusion.
- PubMed: PMID 34934034.
- PubMed: PMID 28807635.
- PubMed: PMID 27039022.
Explore more trial analyses
Each trial page connects its endpoints and methods to statistical tutorials and calculators.
15. Record Summary
BRIGHTER randomized 455 patients with BRVO-related visual impairment to ranibizumab 0.5 mg, ranibizumab 0.5 mg plus laser, or laser monotherapy. In the ANOVA-based primary analysis at Month 6, the difference in LS mean BCVA change versus laser was 10.0 letters (95% CI 7.3 to 12.8) for ranibizumab monotherapy and 8.7 letters (95% CI 5.8 to 11.6) for ranibizumab plus laser, both with p < 0.0001. Interpreting these results well means considering the effect size in letters, the interval precision, the observed-case analysis population, the open-label design, and the two-comparison structure of the primary analysis.