Table 1 Analysis populations

Randomized population

All randomized participants as defined by the trial protocol

Cross-sectional analysis population

Participants who have completed the NVCI item, NVS, and all items of the NAVQ-P at Month 2. This population was used for all cross-sectional analyses performed at Month 2. Month 2 was selected since it was expected to have the greatest distribution of scores across the sample.

Test-retest analysis population

The test-retest analysis population consisted of participants who were determined as stable between Months 2 and 3 (primary test-retest analysis population), or, stable between Week 2 and Month 1 timepoints (secondary test-retest analysis population). Stability between these timepoints was defined using two different criteria:

1. No change in PGI-S

2. < 0.14 logMAR change in DCNVA

Test-retest reliability analysis was conducted separately for each of these populations and stability definitions. The threshold for DCNVA logMAR change was informed by previous literature [28, 29].

Interpretability analysis populations

For interpretability analyses, ‘stable’, ’improved’, and ‘worsened’ groupings were defined from the randomised population using multiple anchors including the PGI-S, PGI-C, and DCNVA. All anchor-based analyses were performed examining changes between baseline and Months 1, 2, and 3, with the change to Month 3 considered the primary analysis. The anchor groups defining change were as follows:

PGI-S (1 grade):

•Improved ( > = 1-point PGI-S improvement)

•Stable (0-point PGI-S change)

•Worsened ( > = 1-point PGI-S worsening)

PGI-S (2 grade):

•Improved ( > = 2-point PGI-S improvement)

•Stable (< 2-point PGI-S change)

•Worsened ( > = 2-point PGI-S worsening)

PGI-C:

•Improved (A little better/Much better)

•Stable (No change)

•Worsened (A little worse/Much worse)

DCNVA:

•Improved ( > = 0.14 logMAR decrease)

•Stable (< 0.14 absolute logMAR change)

•Worsened ( > = 0.14 logMAR increase)