Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. For the Mixed Model for Repeated Measures (MMRM) analysis, the model adjusted for treatment arm, visit, visit-by-treatment arm interaction, baseline BCVA (continuous), baseline BCVA (<64 vs. ≥64 letters), prior intravitreal anti-VEGF therapy (yes vs. no), and region of enrollment. An unstructured covariance structure was used. Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were implicitly imputed by MMRM. Invalid BCVA values were excluded. 97.5% CI is a rounding of 97.52% CI.

ITT Population and Treatment-Naive Population

The Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Severity Scale (DRSS) classifies diabetic retinopathy into 12 severity steps ranging from absence of retinopathy to advanced proliferative diabetic retinopathy. Ocular imaging assessments were made independently by a central reading center. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world regions were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 97.5% confidence interval (CI) is a rounding of 97.52% CI.

ITT Population and Treatment-Naive Population. Only participants with non-missing, valid assessments at Baseline and Week 52 were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. For the Mixed Model for Repeated Measures (MMRM) analysis, the model adjusted for treatment arm, visit, visit-by-treatment arm interaction, baseline BCVA (continuous), baseline BCVA (<64 vs. ≥64 letters), prior intravitreal anti-VEGF therapy (yes vs. no), and region of enrollment. An unstructured covariance structure was used. Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were implicitly imputed by MMRM. Invalid BCVA values were excluded. 95% CI is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization.

Best-Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score attainable), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. For the Mixed Model for Repeated Measures (MMRM) analysis, the model adjusted for treatment group, visit, visit-by-treatment group interaction, baseline BCVA (continuous), baseline BCVA (<64 vs. ≥64 letters), and region of enrollment. An unstructured covariance structure was used. Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were implicitly imputed by MMRM. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization.

BCVA was measured on the ETDRS chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA from baseline indicates an improvement in visual acuity. For each participant, an average BCVA value was calculated across the three visits, and this averaged value was then used to determine if the endpoint was met. The results were summarized as the percentage of participants per treatment arm who met the endpoint. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world). Treatment policy strategy and hypothetical strategy were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. Only participants with at least one non-missing, valid assessment at Weeks 48, 52, or 56 were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

BCVA was measured on the ETDRS chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA from baseline indicates an improvement in visual acuity. For each participant, an average BCVA value was calculated across the three visits, and this averaged value was then used to determine if the endpoint was met. The results were summarized as the percentage of participants per treatment arm who met the endpoint. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world). Treatment policy strategy and hypothetical strategy were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. Only participants with at least one non-missing, valid assessment at Weeks 48, 52, or 56 were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. For each participant, an average BCVA value was calculated across the three visits, and this averaged value was then used to determine if the endpoint was met. The results were summarized as the percentage of participants per treatment arm who met the endpoint. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy and hypothetical strategy were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. Only participants with at least one non-missing, valid assessment at Weeks 48, 52, or 56 were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The weighted estimates of the percentage of participants avoiding a loss of letters in BCVA from baseline were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The weighted estimates of the percentage of participants avoiding a loss of letters in BCVA from baseline were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The weighted estimates of the percentage of participants avoiding a loss of letters in BCVA from baseline were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. For each participant, an average BCVA value was calculated across the three visits, and this averaged value was then used to determine if the endpoint was met. The results were summarized as the percentage of participants per treatment arm who met the endpoint. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy and hypothetical strategy were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. Only participants with at least one non-missing, valid assessment at Weeks 48, 52, or 56 were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The weighted estimates of the percentage of participants avoiding a loss of letters in BCVA from baseline were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The weighted estimates of the percentage of participants avoiding a loss of letters in BCVA from baseline were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The weighted estimates of the percentage of participants avoiding a loss of letters in BCVA from baseline were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

BCVA was measured on the ETDRS chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA from baseline indicates an improvement in visual acuity. For each participant, an average BCVA value was calculated across the three visits, and this averaged value was then used to determine if the endpoint was met. The results were summarized as the percentage of participants per treatment arm who met the endpoint. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world). Treatment policy strategy and hypothetical strategy were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population and Treatment-Naive Population. Only participants with at least one non-missing, valid assessment at Weeks 48, 52, or 56 were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

BCVA was measured on the ETDRS chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA from baseline indicates an improvement in visual acuity. For each participant, an average BCVA value was calculated across the three visits, and this averaged value was then used to determine if the endpoint was met. The results were summarized as the percentage of participants per treatment arm who met the endpoint. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥69 vs. <69 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world). Treatment policy strategy and hypothetical strategy were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population and Treatment-Naive Population. Only participants with at least one non-missing, valid assessment at Weeks 48, 52, or 56 were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥69 vs. <69 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥69 vs. <69 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

BCVA was measured on the ETDRS chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA from baseline indicates an improvement in visual acuity. For each participant, an average BCVA value was calculated across the three visits, and this averaged value was then used to determine if the endpoint was met. The results were summarized as the percentage of participants per treatment arm who met the endpoint. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world). Treatment policy strategy and hypothetical strategy were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population and Treatment-Naive Population. Only participants with at least one non-missing, valid assessment at Weeks 48, 52, or 56 were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score), and a gain in BCVA letter score from baseline indicates an improvement invisual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Best Corrected Visual Acuity (BCVA) was measured on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart at a starting distance of 4 meters. The BCVA letter score ranges from 0 to 100 (best score attainable), and a gain in BCVA letter score from baseline indicates an improvement in visual acuity. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. Invalid BCVA values were excluded from analysis. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

The Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Severity Scale (DRSS) classifies diabetic retinopathy into 12 severity steps ranging from absence of retinopathy to advanced proliferative diabetic retinopathy. Ocular imaging assessments were made independently by a central reading center. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world regions were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. Only participants with non-missing, valid assessments at Baseline and each timepoint were included in the analysis.

The Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Severity Scale (DRSS) classifies diabetic retinopathy into 12 severity steps ranging from absence of retinopathy to advanced proliferative diabetic retinopathy. Ocular imaging assessments were made independently by a central reading center. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world regions were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. Only participants with non-missing, valid assessments at Baseline and each timepoint were included in the analysis.

The Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Severity Scale (DRSS) classifies diabetic retinopathy into 12 severity steps ranging from absence of retinopathy to advanced proliferative diabetic retinopathy. Ocular imaging assessments were made independently by a central reading center. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world regions were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. Only participants with non-missing, valid assessments at Baseline and each timepoint were included in the analysis.

The Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Severity Scale (DRSS) classifies diabetic retinopathy into 12 severity steps ranging from absence of retinopathy to advanced proliferative diabetic retinopathy. Ocular imaging assessments were made independently by a central reading center. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world regions were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. Only participants with non-missing, valid assessments at Baseline and each timepoint were included in the analysis.

The Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Severity Scale (DRSS) classifies diabetic retinopathy into 12 severity steps ranging from absence of retinopathy to advanced proliferative diabetic retinopathy. Ocular imaging assessments were made independently by a central reading center. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world regions were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. Only participants with non-missing, valid assessments at Baseline and each timepoint were included in the analysis.

The Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Severity Scale (DRSS) classifies diabetic retinopathy into 12 severity steps ranging from absence of retinopathy to advanced proliferative diabetic retinopathy. Ocular imaging assessments were made independently by a central reading center. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters) and region (U.S. and Canada vs. rest of the world; Asia and rest of the world regions were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. Only participants with non-missing, valid assessments at Baseline and each timepoint were included in the analysis.

The Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Severity Scale (DRSS) classifies diabetic retinopathy into 12 severity steps ranging from absence of retinopathy to advanced proliferative diabetic retinopathy (PDR). PDR was defined as an ETDRS DRSS score of ≥61 on the 7-field/4-wide field color fundus photographs assessment by a central reading center. The weighted percentages of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world regions were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% CI is a rounding of 95.04% CI.

ITT Population and Treatment-Naive Population. Only participants with non-missing, valid assessments at Baseline and Week 52 were included in the analysis.

The Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Severity Scale (DRSS) classifies diabetic retinopathy into 12 severity steps ranging from absence of retinopathy to advanced PDR. High-risk PDR was defined as an ETDRS DRSS score of ≥71 on the 7-field/4-wide field color fundus photographs assessment by a central reading center. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world regions were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% CI is a rounding of 95.04% CI.

ITT Population and Treatment-Naive Population. Only participants with non-missing, valid assessments at Baseline and Week 52 were included in the analysis.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. The number analyzed includes all participants in Arm B: Faricimab 6 mg PTI who had not discontinued the study prior to Week 52.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. The number analyzed includes all participants in Arm B: Faricimab 6 mg PTI who had not discontinued the study prior to Week 52.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. The number analyzed includes all participants in Arm B: Faricimab 6 mg PTI who had not discontinued the study prior to Week 96.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization. The number analyzed includes all participants in Arm B: Faricimab 6 mg PTI who had not discontinued the study prior to Week 96.

ITT Population and Treatment-Naive Population. The number analyzed includes all participants in Arm B: Faricimab 6 mg PTI who had not discontinued the study prior to Week 52.

ITT Population and Treatment-Naive Population. The number analyzed includes all participants in Arm B: Faricimab 6 mg PTI who had not discontinued the study prior to Week 96.

Central subfield thickness (CST) was defined as the distance between the internal limiting membrane (ILM) and Bruch's membrane (BM) as assessed by a central reading center. For the Mixed Model for Repeated Measures (MMRM) analysis, the model adjusted for treatment group, visit, visit-by-treatment group interaction, baseline CST (continuous), baseline BCVA (<64 vs. ≥64 letters), prior intravitreal anti-VEGF therapy (yes vs. no), and region of enrollment (U.S. and Canada vs. the rest of the world; Asia and rest of the world regions were combined). An unstructured covariance structure was used. Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were implicitly imputed by MMRM. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population and Treatment-Naive Population

Central subfield thickness (CST) was defined as the distance between the internal limiting membrane (ILM) and Bruch's membrane (BM) as assessed by a central reading center. For the Mixed Model for Repeated Measures (MMRM) analysis, the model adjusted for treatment group, visit, visit-by-treatment group interaction, baseline CST (continuous), baseline BCVA (<64 vs. ≥64 letters), prior intravitreal anti-VEGF therapy (yes vs. no), and region of enrollment (U.S. and Canada vs. the rest of the world; Asia and rest of the world regions were combined). An unstructured covariance structure was used. Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were implicitly imputed by MMRM. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization.

Central subfield thickness (CST) was defined as the distance between the internal limiting membrane (ILM) and Bruch's membrane (BM) as assessed by a central reading center. For the Mixed Model for Repeated Measures (MMRM) analysis, the model adjusted for treatment group, visit, visit-by-treatment group interaction, baseline CST (continuous), baseline BCVA (<64 vs. ≥64 letters), and region of enrollment (U.S. and Canada vs. the rest of the world; Asia and rest of the world regions were combined). An unstructured covariance structure was used. Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were implicitly imputed by MMRM. 95% confidence interval (CI) is a rounding of 95.04% CI.

Treatment-Naive Population: all participants randomized in the study who had not received any intravitreal anti-VEGF agents in the study eye prior to randomization. Participants were grouped according to the treatment assigned at randomization.

Absence of diabetic macular edema was defined as achieving a central subfield thickness (CST) of <325 microns in the study eye. CST was defined as the distance between the internal limiting membrane and Bruch's membrane. For each participant, an average CST value was calculated across the three visits, and this averaged value was then used to determine if the endpoint was met. The results were summarized as the percentage of participants per treatment arm who met the endpoint. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world). Treatment policy strategy and hypothetical strategy were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population and Treatment-Naive Population. Only participants with at least one non-missing, valid assessment at Weeks 48, 52, or 56 were included in the analysis.

Absence of diabetic macular edema was defined as achieving a central subfield thickness of <325 microns in the study eye. Central subfield thickness was defined as the distance between the internal limiting membrane (ILM) and Bruch's membrane (BM) as assessed by a central reading center. The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world regions were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Retinal dryness was defined as achieving a central subfield thickness (ILM-BM) of <280 microns. Central subfield thickness was defined as the distance between the internal limiting membrane (ILM) and Bruch's membrane (BM) as assessed by a central reading center. The weighted estimates of the percentage of participants was based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world; Asia and rest of the world regions were combined). Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Intraretinal fluid was measured using optical coherence tomography (OCT) in the central subfield (center 1 mm). The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world); Asia and rest of the world regions were combined due to a small number of enrolled participants. Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Subretinal fluid was measured using optical coherence tomography (OCT) in the central subfield (center 1 mm). The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world); Asia and rest of the world regions were combined due to a small number of enrolled participants. Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

Intraretinal fluid and subretinal fluid were measured using optical coherence tomography (OCT) in the central subfield (center 1 mm). The weighted estimates of the percentage of participants were based on the Cochran-Mantel Haenszel (CMH) weights stratified by baseline BCVA (≥64 vs. <64 letters), prior IVT anti-VEGF therapy (yes vs. no), and region (U.S. and Canada vs. rest of the world); Asia and rest of the world regions were combined due to a small number of enrolled participants. Treatment policy strategy (i.e., all observed values used) and hypothetical strategy (i.e., all values censored after the occurrence of the intercurrent event) were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were not imputed. 95% confidence interval (CI) is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization. At each timepoint, only participants with non-missing, valid assessments were included in the analysis.

The NEI VFQ-25 captures a patient's perception of vision-related functioning and quality of life. The core measure includes 25 items that comprise 11 vision-related subscales and one item on general health. The composite score ranges from 0 to 100, with higher scores, or a positive change from baseline, indicating better vision-related functioning. For the Mixed Model for Repeated Measures (MMRM) analysis, the model adjusted for treatment arm, visit, visit-by-treatment arm interaction, baseline NEI VFQ-25 Composite Score (continuous), baseline BCVA (<64 vs. ≥64 letters), prior intravitreal anti-VEGF therapy (yes vs. no), and region of enrollment. An unstructured covariance structure was used. Treatment policy strategy and hypothetical strategy were applied to non-COVID-19 related and COVID-19 related intercurrent events, respectively. Missing data were implicitly imputed by MMRM. 95% CI is a rounding of 95.04% CI.

ITT Population: all participants who were randomized in the study, grouped according to the treatment assigned at randomization.

This analysis of adverse events (AEs) includes both ocular and non-ocular (systemic) AEs. Investigators sought information on AEs at each contact with the participants. All AEs were recorded and the investigator made an assessment of seriousness, severity, and causality of each AE. AEs of special interest included the following: Cases of potential drug-induced liver injury that include an elevated ALT or AST in combination with either an elevated bilirubin or clinical jaundice, as defined by Hy's Law; Suspected transmission of an infectious agent by the study drug; Sight-threatening AEs that cause a drop in visual acuity (VA) score ≥30 letters lasting more than 1 hour, require surgical or medical intervention to prevent permanent loss of sight, or are associated with severe intraocular inflammation.

The safety-evaluable population comprised all participants who received at least one injection of active study drug (faricimab or aflibercept) in the study eye.

This analysis of adverse events (AEs) only includes ocular AEs, which are categorized as having occurred either in the study eye or the fellow eye. Investigators sought information on AEs at each contact with the participants. All AEs were recorded and the investigator made an assessment of seriousness, severity, and causality of each AE. Ocular AEs of special interest included the following: Suspected transmission of an infectious agent by the study drug; Sight-threatening AEs that cause a drop in visual acuity (VA) score ≥30 letters lasting more than 1 hour, require surgical or medical intervention to prevent permanent loss of sight, or are associated with severe intraocular inflammation.

The safety-evaluable population comprised all participants who received at least one injection of active study drug (faricimab or aflibercept) in the study eye.

This analysis of adverse events (AEs) only includes non-ocular (systemic) AEs. Investigators sought information on adverse events (AEs) at each contact with the participants. All AEs were recorded and the investigator made an assessment of seriousness, severity, and causality of each AE. The non-ocular AE of special interest was: Cases of potential drug-induced liver injury that include an elevated ALT or AST in combination with either an elevated bilirubin or clinical jaundice, as defined by Hy's Law.

The safety-evaluable population comprised all participants who received at least one injection of active study drug (faricimab or aflibercept) in the study eye.

Faricimab concentration in plasma was determined using a validated immunoassay method.

This analysis only included participants in Arms A and B who received treatment with faricimab and with at least one plasma sample, provided sufficient dosing information (dose and dosing time) was available. The number of participants analyzed at a given timepoint includes those with an available plasma sample and dosing information at that timepoint.

Anti-drug antibodies (ADAs) against fariciamb were detected in plasma using a validated bridging enzyme-linked immunosorbent assay (ELISA). The percentage of participants with treatment-emergent ADA-positive samples includes post-baseline evaluable participants with at least one treatment-induced (defined as having an ADA-negative sample or missing sample at baseline and any positive post-baseline sample) or treatment-boosted (defined as having an ADA-positive sample at baseline and any positive post-baseline sample with a titer that is equal to or greater than 4-fold baseline titer) ADA-positive sample during the study treatment period.

The analysis population consisted of all participants receiving faricimab with at least one determinant post-baseline ADA assessment.