A Study to Investigate Aqueous Humor and Multimodal Imaging Biomarkers in Treatment-Naïve Participants With Diabetic Macular Edema Treated With Faricimab: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Investigate Aqueous Humor and Multimodal Imaging Biomarkers in Treatment-Naïve Participants With Diabetic Macular Edema Treated With Faricimab (NCT04597918), a na phase 2 study. The primary outcome is Percentage of Participants With a ≥2-Step Improvement From Baseline on the Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Severity Scale (DRSS) in the Study Eye at Week 24.

2. Clinical Question

The study was designed to explore how clinical assessments, multimodal imaging assessments, aqueous humor biomarker patterns, and genetic polymorphisms relate to one another over time in participants treated with faricimab. That exploratory aim is broader than the primary endpoint, which is a specific change in retinal disease severity grading. A single-arm design can describe outcomes observed during the study, but it cannot isolate a treatment effect from other explanations for change over time. The record does not report biomarker association estimates or the primary endpoint result.

3. Trial Design

NCT04597918 is a prospective, multicenter, open-label, single-arm, interventional Phase IIb study sponsored by Hoffmann-La Roche. Its official title identifies it as an exploratory study of aqueous humor and multimodal imaging biomarkers in treatment-naïve patients with DME treated with faricimab. The registry lists the condition as diabetic macular edema, enrollment as 99, and study status as completed. The recorded study start is 2020-11-25 and primary completion is 2022-12-22.

All participants were assigned to the faricimab group; the record does not describe a separate control arm. The regimen was six 6 mg intravitreal injections, one every 28 days, starting at Day 1 and ending on the Day 140 visit. The protocol description also specifies a safety follow-up visit after at least 28 days and within <35 days following the last study treatment. The study was open-label, so masking is listed as none.

The primary outcome was the percentage of participants with at least a 2-step improvement from baseline on the ETDRS DRSS in the study eye at Week 24. The record also lists secondary outcomes for adjusted mean change in best-corrected visual acuity and central subfield thickness at Week 24, median time to first absence of DME during the study, and absence of intraretinal fluid or subretinal fluid at Week 24. The study’s broader purpose was to examine associations among clinical, imaging, aqueous humor, and genetic measures over time. Specific statistical procedures for those exploratory associations are not reported in the registry record.

Because this is a single-arm study, its design supports description of participants’ observed outcomes during follow-up, not a randomized treatment-versus-control contrast. The registry record does not report a statistical analysis plan, sample-size justification, or interim-analysis plan. Those omissions prevent assessment of how the planned analyses were selected beyond the endpoint description recorded in the registry.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants in the modified intent-to-treat (mITT) Population with missing baseline assessments were excluded from the analysis. The number analyzed indicates participants with assessments at both Baseline and Week 24.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: There were 2 analysis populations in this study, the modified intent-to-treat (mITT; all patients enrolled in the study that received any amount of study treatment) and safety-evaluable (SAF; all patients enrolled in the study who received at least 1 injection of study treatment) populations. All 99 participants were included in both the mITT and SAF populations.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With a ≥2-Step Improvement From Baseline on the Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Severity Scale (DRSS) in the Study Eye at Week 24, measured over Baseline and Week 24.

The primary endpoint is the percentage of participants whose study eye improved by at least 2 steps from baseline on the ETDRS Diabetic Retinopathy Severity Scale (DRSS) at Week 24. Trained personnel at study sites obtained ultra-wide field color fundus photographs, and a central reading center assessed the photographs. Baseline was defined as the participant’s last observation before study drug initiation. The endpoint therefore uses a change in a photo-based severity grade rather than a direct measure of visual function.

A step-based severity scale can express whether a participant’s graded retinal disease severity changed by a specified amount over follow-up. Focusing on a threshold makes the endpoint a binary outcome for each participant: the participant either meets the improvement threshold or does not. The reported group percentage would describe the share meeting that threshold, while the confidence interval would convey uncertainty around that percentage.

The endpoint description specifies a two-sided 95% Clopper-Pearson exact confidence interval. However, the registry record does not report the number meeting the threshold, the number analyzed, the resulting percentage, or the interval limits. Thus, the definition and assessment method are available, but the magnitude and precision of the primary endpoint result cannot be evaluated from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the primary percentage represent?

For each participant included in the primary analysis, the endpoint classification depends on whether the study eye had at least a 2-step improvement from baseline on the ETDRS DRSS at Week 24. The group percentage is the share of analyzed participants who meet that criterion. This is a binary outcome because each participant’s assessment is classified as meeting or not meeting the threshold. The registry record does not give the numerator or denominator used to calculate the percentage, or the resulting estimate.

The assessment was based on ultra-wide field color fundus photography. Trained personnel at study sites took the photographs, and a central reading center performed the analysis. Central reading can apply a common assessment process across images, although the registry record does not describe reading procedures in further detail. Baseline is defined as the last observation before study drug initiation, making the endpoint a comparison between that baseline assessment and the Week 24 assessment.

What does a confidence interval add?

A confidence interval expresses uncertainty around an estimated percentage. It gives a range of values compatible with the observed data under the method’s assumptions; it is not a guarantee that the true value lies within the interval. The registry specifies a two-sided 95% Clopper-Pearson exact confidence interval for the primary percentage. The exact method is designed for binomial proportions and does not rely on the same large-sample normal approximation used by some simpler interval methods. This can be useful when the event count or sample size is limited, because the method respects the bounded range of a proportion.

The method cannot replace the missing result. Without the observed number meeting the endpoint and the number analyzed, the interval cannot be reconstructed or interpreted. The width of an interval is one way to assess precision: all else equal, less information tends to produce less precise estimates. But no statement about this study’s actual precision is justified without its reported interval.

How should the analysis population be read?

The registry describes the primary analysis population as the modified intent-to-treat population and states that participants with missing baseline assessments were excluded. It further defines the number analyzed as participants with assessments at both Baseline and Week 24. These criteria mean that the primary endpoint denominator was intended to reflect participants with the necessary observations, rather than automatically including everyone listed in the enrollment figure.

The enrollment figure and the primary analysis denominator are not interchangeable. The record does not report the number analyzed, the number missing a Week 24 assessment, or the reasons assessments were unavailable. It also does not report whether any other missing-data methods were used. The stated exclusion and assessment requirements identify some aspects of the analysis population, but they do not establish how all missing observations were handled.

What can a single-arm design estimate?

A single-arm study can describe outcomes observed among participants receiving the study intervention. It has no concurrent randomized control group, and the registry lists allocation as not applicable. Consequently, a baseline-to-follow-up change is not a randomized contrast. General influences such as time, changes in measurement, or factors associated with who remains assessed can also relate to observed change; the registry record does not quantify these possibilities.

Open-label status means masking is listed as none. The central reading center’s role is stated for the fundus photographs, but the registry does not report masking details for other assessments. A single-arm design can be appropriate for exploratory description and association questions, but its results do not establish how outcomes would differ under another intervention or no intervention.

What do the planning assumptions mean?

The planning illustrations are described as calculations based on the observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. Those assumptions describe hypothetical planning scenarios; they do not describe the actual single-arm allocation in this study. The registry record does not report a trial-specific sample-size justification, and the illustration assumptions should not be read as the trial’s statistical plan.

Sample-size illustrations are not observed power. They ask what sample size a hypothetical design might require under specified assumptions. Their relevance depends on the effect estimate and assumptions used, and they do not establish that the observed study result is reliable or that an effect exists. No numerical illustration results are reported in the registry record for this analysis, so no particular planning scenario can be interpreted quantitatively.

The registry also lists 0 analyses. This limits what can be learned about the details of statistical implementation from the record. The endpoint description names the measure and confidence-interval method, but the statistical plan, any interim analyses, and procedures for multiple outcomes are not reported in the registry record.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Faricimab")
n      <- c(64)
events <- c(32)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Faricimab  64  32  50.0  37.2  62.8  38.1  61.9  38.1  61.9

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and several secondary outcomes: adjusted mean change in best-corrected visual acuity, adjusted mean change in central subfield thickness, median time to first absence of DME, and percentages with absence of intraretinal fluid or subretinal fluid. Examining multiple outcomes raises a general multiplicity issue: the more outcomes assessed, the more opportunities there may be for an apparently notable result by chance. Whether and how that issue was addressed depends on a prespecified analysis plan. A multiplicity adjustment or testing hierarchy is not reported in the registry record.

The endpoint description states that participants in the mITT population with missing baseline assessments were excluded. It also says the number analyzed indicates participants with assessments at both Baseline and Week 24. The actual number analyzed is not reported in the registry record. The number and reasons for missing Week 24 assessments, and any imputation or sensitivity analyses, are also not reported. It is therefore not possible to assess how missing outcome data may have affected the primary estimate.

The record does not report a statistical analysis plan, interim analyses, or a sample-size justification. It reports 0 analyses, while the primary endpoint description specifies a percentage and a two-sided 95% Clopper-Pearson exact confidence interval. That method detail alone does not establish how secondary outcomes or exploratory biomarker associations were analyzed. In particular, the registry record does not report whether the primary endpoint was treated as confirmatory or how the broader set of exploratory questions was handled.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: localised infection (2.0%), sepsis (2.0%), covid-19 (1.0%), cellulitis (1.0%), escherichia urinary tract infection (1.0%).

For the faricimab group, the registry records 99 participants at risk for serious adverse events and 8 affected. It records 99 at risk for other adverse events and 21 affected. The record also lists 1 death. These are descriptive figures; they do not provide a randomized safety comparison. The registry does not report, in the registry record, a causal attribution for the death or the adverse events.

The listed serious adverse event terms include Localised infection at 2.0%, Sepsis at 2.0%, COVID-19 at 1.0%, Cellulitis at 1.0%, Escherichia urinary tract infection at 1.0%, and Osteomyelitis at 1.0%. The registry lists these as serious event terms and percentages. They should be read as entries in the safety record, not as evidence that faricimab caused those events. The record does not provide a control group against which to compare these frequencies.

Counts of participants affected and percentages attached to event terms may describe different aspects of safety reporting. Event terms can overlap within participants, so the listed percentages should not be added to infer a total. The registry record does not provide enough detail to infer event timing, duration, recurrence, or relationship to study treatment. These figures summarize what is recorded, not a complete causal assessment.

13. What the Result Does, and Does Not, Mean

The registry record establishes the study’s question, design, endpoint definition, and selected safety figures. It does not report the primary endpoint estimate or its confidence interval. The percentage achieving at least a 2-step improvement on the ETDRS DRSS at Week 24 is therefore unknown from the registry record. The same limitation applies to the listed secondary outcome estimates and the biomarker association results.

If the primary result were reported, it would describe the proportion of analyzed participants who crossed the defined improvement threshold during follow-up. Its confidence interval would indicate the estimate’s statistical precision under the stated method. Neither a percentage nor its interval, on its own, would show what would have occurred without faricimab. That causal question requires a suitable comparison, and this study has no concurrent randomized comparator.

The study’s exploratory purpose is to examine relationships over time among clinical assessments, imaging, aqueous humor biomarkers, and genetic polymorphisms. Associations can identify patterns for further study, but an association does not by itself establish that one measure caused another or that it predicts outcomes in other populations. The registry record does not report the association analyses, so their size and uncertainty cannot be assessed.

The safety counts provide a description of events recorded in the faricimab group. They do not establish whether individual events were caused by treatment, and without a control group they cannot quantify a comparative safety difference. The death is recorded, but the information available does not state its cause or relationship to treatment.

Overall, this registry record supports understanding of the study’s design and planned outcome definitions, not a conclusion about the magnitude of the primary outcome or a treatment effect. The absence of reported endpoint estimates, analyzed denominators, and confidence limits is central to interpretation. Completed study status and a detailed endpoint definition do not fill those reporting gaps.

14. Why This Trial Matters Statistically

This study is statistically notable for combining a clinical endpoint with an exploratory program involving multimodal imaging, aqueous humor biomarkers, and genetic polymorphisms. Such designs can generate questions about how different types of measurements relate over time. Interpreting those relationships depends on clear definitions of outcomes, analysis populations, and statistical methods, particularly when many measures may be examined.

The primary endpoint illustrates the distinction between a well-defined measure and a reported result. The registry specifies the DRSS threshold, study eye, assessment timing, photographic assessment process, and confidence-interval method. Those details help identify what the endpoint means. But without its estimate, denominator, and interval, readers cannot evaluate the observed frequency or precision. Transparent outcome reporting is necessary to move from a protocol-level question to an interpretable statistical finding.

The single-arm design also makes the estimand important. A percentage of participants meeting a threshold can describe what happened in the observed group. It cannot, without a comparator, estimate a randomized difference between faricimab and another condition. This distinction matters when interpreting changes from baseline: a within-group change and a treatment-versus-control effect answer different questions.

Finally, the registry highlights why analysis details matter in exploratory research. The record lists multiple secondary outcomes and a broader set of biomarker and imaging aims, but does not report a statistical plan or multiplicity approach. Reporting those methods and the associated estimates would help readers distinguish prespecified questions from exploratory findings and judge the uncertainty around each result. In this record, these statistical details frame the questions the study set out to address, while the numerical outcome evidence remains unreported.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The most direct limitation is that the registry record does not report the primary endpoint result. It gives neither the percentage with at least a 2-step improvement nor the number analyzed and confidence-interval limits. The endpoint is therefore defined but cannot be quantitatively interpreted from this record. The listed secondary outcomes also lack reported estimates in the information available for this analysis.

The single-arm, open-label design has no randomized comparator. As a result, the study can describe outcomes in participants receiving faricimab but cannot separate treatment-related change from other influences using a concurrent comparison. The record’s exploratory focus on associations likewise does not by itself establish causal relationships. Masking is listed as none, while masking details for the central reading process beyond its role in photograph analysis are not reported.

The registry describes exclusion of participants with missing baseline assessments and says the primary analysis count includes participants with assessments at both Baseline and Week 24. It does not report the analyzed count, missing-assessment patterns, reasons for missingness, or any additional missing-data procedures. Those omissions limit assessment of how well the analyzed group represents the enrolled group.

The statistical analysis plan, sample-size justification, interim analyses, and multiplicity procedures are not reported in the registry record. The record lists 0 analyses. The stated confidence-interval method is useful information, but it does not resolve those gaps. Safety figures are descriptive and do not establish causality; the registry information does not identify the cause of the death or the relationship of reported adverse events to treatment.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study to Investigate Aqueous Humor and Multimodal Imaging Biomarkers in Treatment-Naïve Participants With Diabetic Macular Edema Treated With Faricimab — NCT04597918.
• Registry record last updated 2026-06-15; status last verified 2026-05.
• Results first posted 2023-12-01.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT04597918 is a completed, prospective, multicenter, open-label, single-arm Phase IIb study of faricimab in treatment-naïve participants with diabetic macular edema. The registry lists enrollment of 99. The recorded study start is 2020-11-25, and primary completion is 2022-12-22.

The regimen consisted of six 6 mg intravitreal injections, one every 28 days, from Day 1 through the Day 140 visit. The primary endpoint was the percentage with at least a 2-step improvement from baseline on the ETDRS DRSS in the study eye at Week 24, with a two-sided 95% Clopper-Pearson exact confidence interval specified. The endpoint estimate, analyzed denominator, and interval are not reported in the registry record.

The registry also lists secondary visual, anatomical, fluid-status, and exploratory biomarker aims. It records 8 participants affected by serious adverse events, 21 by other adverse events, and 1 death in the faricimab group. These safety figures are descriptive and do not establish causality or provide a randomized comparison.