A Study of BI 765128 in Patients With an Eye Condition Called Diabetic Macular Ischemia Who Have Received Laser Treatment: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of BI 765128 in Patients With an Eye Condition Called Diabetic Macular Ischemia Who Have Received Laser Treatment (NCT04919499), a randomized phase 1, phase 2 study. The primary outcome is Single Rising Dose Part - Number of Subjects With Ocular Dose Limiting Events (DLEs) From Drug Administration Until Day 8 (7 Days After Treatment).

2. Clinical Question

The study addressed early safety and tolerability, rather than a definitive comparison of clinical benefit. Its official title describes a first-in-human trial. The registry identifies diabetic retinopathy as the condition and describes enrollment of 46 participants across the study. The primary endpoint concerns only the single rising dose part; it does not summarize the randomized multiple dose comparison.

The study also recorded secondary outcomes, including adverse events, changes in foveal avascular zone size and best corrected visual acuity. Numerical secondary results are available for best corrected visual acuity at visit 7. Results for the other listed secondary outcomes are not reported in the registry record.

3. Trial Design

The trial is listed as phase 1 and phase 2, completed, randomized, double-masked and sequential. Those overall registry labels need to be read alongside the two-part design. Part A was a non-randomized, uncontrolled single rising dose study. Participants received one low, medium or high intravitreal dose of BI 765128. The primary endpoint was assessed in this part from initial administration through day 8.

Part B used a randomized, double-masked, sham-controlled design. Participants received high-dose BI 765128 or sham injections at week 1, 4 and 8. The brief summary describes treatment once every month for 3 months. The study design therefore included an uncontrolled dose-escalation component and a separate randomized comparison; the primary endpoint belongs to the former, not the latter.

The single rising dose arms had 3 participants at low dose, 3 at medium dose and 6 at high dose in the posted primary endpoint results. In the multiple dose results for best corrected visual acuity, the sham group had 10 participants and the high-dose BI 765128 group had 20. These endpoint-specific group sizes should not be treated as a single analysis population: the parts, outcomes and analysis populations differ. The registry record does not report the rationale for the enrollment target or explain how those endpoint-specific counts relate to overall enrollment.

The planned study duration and visit schedule differed by part. The brief summary describes about 4 months in the study and about 8 site visits for part A, and about 5 months and about 7 site visits for part B. The registry record describes regular health checks and recording of unwanted effects. It does not report further operational details about allocation procedures or the sequence used to escalate doses.

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.

Before assignment: All subjects were screened for eligibility prior to participation in the trial. Subjects attended a specialist site which ensured that they (the subjects) strictly met all inclusion and none of the exclusion criteria. One eye was selected, according to inclusion /exclusion criteria, as the study eye to be treated.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Treated set (TS) - single rising dose part: all subjects who were treated with at least one dose of study drug.

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: Treated set (TS) - all subjects who were treated with at least one dose of study drug in the single rising dose part and all subjects who were randomized and treated with at least one dose of study drug (either treatment with BI 765128 or Sham) in the multiple dose part.

6. Primary Endpoint

The registry lists the primary outcome as Single Rising Dose Part - Number of Subjects With Ocular Dose Limiting Events (DLEs) From Drug Administration Until Day 8 (7 Days After Treatment), measured over From initial drug administration (day 1) until day 8..

The primary endpoint counts participants with ocular dose-limiting events (DLEs) during the period from initial drug administration on day 1 through day 8. The registry defines DLEs using several protocol criteria: specified levels and duration of sterile inflammation, visual loss of more than 15 letters at any given time-point, persistent intra-ocular pressure over 30 mmHg for 3 days, or signs of retinal vessel occlusion in a main branch or a vessel after its first bifurcation. The inflammation criteria also specify grading systems and thresholds.

This endpoint is used to identify early ocular events considered sufficiently important to limit dose escalation. Counting affected participants offers a direct description of how many people met at least one of the stated criteria in the short evaluation period. It is not a measure of average change in vision, and the count alone does not describe the timing, duration or severity of each event beyond the endpoint definition.

No DLEs were reported in the low-dose group, the medium-dose group or the high-dose group in this part. The corresponding treated-set group sizes were 3, 3 and 6 participants. Because the endpoint is based on small groups and a short observation window, zero observed events should be read as a description of those participants and that period, not as proof that such events cannot occur or that risks are absent. The registry record does not report a confidence interval for these event counts.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a binary participant-level endpoint: each treated participant either met the ocular dose-limiting event definition during the stated period or did not. The registry reports counts rather than a continuous measurement. For low-dose BI 765128, the count was 0 among 3 participants; for medium dose, 0 among 3; and for high dose, 0 among 6. These counts are the direct observed results. The registry record does not report an inferential comparison across the dose groups or a confidence interval for the primary outcome.

What does a confidence interval describe?

In general, a confidence interval expresses the statistical uncertainty around an estimate under the analysis assumptions. A wider interval indicates less precision than a narrower interval. For the secondary visual acuity comparison, the sponsor-reported adjusted mean difference was 0.1 letters, with a 95% confidence interval from -4.7 to 5.0. Because the interval includes values below and above zero, it is compatible with more than one direction of difference. It should not be read as a range containing all possible true values with certainty.

How should the secondary visual acuity analysis be read?

The registry describes a repeated-measures model (MMRM) that included measurements at baseline, day 29±3, day 57±3, day 85±7, day 113±7 and day 141±7. The reported visit 7 change is from day 141±7. A repeated-measures model is generally used to analyze measurements collected over time, accounting for their relationship within participants and using the scheduled measurement history. The registry record does not report the model’s full specification, covariance structure, or handling of missing observations, so those details cannot be assessed.

The sponsor-reported summary gives least-squares mean changes of 1.1 letters for sham and 1.3 letters for high-dose BI 765128, with standard errors of 1.9 and 1.3. Least-squares means are model-based estimates, rather than necessarily simple arithmetic averages. The adjusted mean difference and its confidence interval are the sponsor-reported comparison. A separate unadjusted recalculation gives a mean difference of 0.2 letters, a 95% confidence interval from -4.312 to 4.712, and p = 0.931. This recalculation uses an approximate standard deviation derived from the posted summary statistics; it is not a replacement for the adjusted model.

In general, a p-value describes how unusual data at least as incompatible with a specified null model would be if that model and the analysis assumptions held. It is not the probability that the null hypothesis is true, nor a measure of clinical importance. The recalculated p-value is unadjusted. A p-value for the sponsor-reported adjusted comparison is not reported in the registry record.

What do the planning illustrations mean?

The sample-size illustrations are planning calculations based on the observed visual acuity effect, not the trial’s original design and not observed power. The assumptions are a two-sided significance level of 0.05, equal allocation between arms and a normal approximation. For planning around 100% of the observed effect, the illustration gives 13,555 participants per arm for 80% power and 18,146 participants per arm for 90% power. For 75% of the observed effect, it gives 24,098 participants per arm for 80% power and 32,260 participants per arm for 90% power. For 50% of the observed effect, it gives 54,219 participants per arm for 80% power and 72,584 participants per arm for 90% power. These are hypothetical planning quantities; they are not observed enrollment or evidence that the completed study had a particular power.

In general, over-dispersion occurs when outcome variability is greater than assumed by a simple statistical model. It can make standard errors too small if it is ignored. The registry record does not identify a count-model analysis or report an over-dispersion assessment for the primary endpoint; it reports the observed participant counts. The available information does not justify adding a model-based inference to those counts.

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("Single Rising Dose Part: Low-dose BI 765128", "Single Rising Dose Part: Medium-dose BI 765128", "Single Rising Dose Part: High-dose BI 765128")
n      <- c(3, 3, 6)
events <- c(0, 0, 0)
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  Single Rising Dose Part: Low-dose BI 765128  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Single Rising Dose Part: Medium-dose BI 765128  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Single Rising Dose Part: High-dose BI 765128  6  0  0.0  0.0  45.9  0.0  39.0  0.4  41.0

11. Multiplicity, Missing Data and Analysis Populations

The primary endpoint analysis population is defined as the treated set in the single rising dose part: all participants who received at least one dose of study drug. The posted primary results give counts for the three dose groups. The best corrected visual acuity analysis is from the multiple dose part, with posted group sizes of 10 for sham and 20 for high-dose BI 765128. The registry record does not report a fuller explanation of the relationship between these endpoint-specific populations and overall enrollment.

The registry describes an MMRM for the visual acuity outcome, including baseline and repeated follow-up measurements through day 141±7. It does not report the model specification in enough detail to evaluate its assumptions, or describe how missing measurements were handled. The number of missing observations, reasons for missingness, and any sensitivity analyses are not reported in the registry record. The primary endpoint’s missing-data handling is also not reported in the registry record.

Multiplicity is relevant because the study lists several secondary outcomes, including measurements at multiple visits. In general, testing multiple outcomes or time points can increase the chance of at least one apparently notable result even when there is no underlying effect. Statistical plans may address this through a prespecified hierarchy or adjustment, but neither a multiplicity procedure nor a hierarchy of secondary hypotheses is reported in the registry record. The sponsor-reported visual acuity result has no p-value listed; the recalculated p-value is unadjusted.

The registry record does not report interim analyses, a sample-size justification, or a detailed statistical analysis plan. Those omissions limit assessment of whether the analysis populations, missing-data methods and multiplicity approach were prespecified. They do not alter the descriptive counts and estimates that are posted, but they constrain how confidently inferential claims can be made from the available record.

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: cerebrovascular accident (33.3%), chest pain (10.0%), herpes zoster (10.0%), lumbar spinal stenosis (4.3%), malignant melanoma (4.3%).

For serious adverse events, the registry reports 1 affected participant among 3 in the low-dose single rising dose group, 0 among 3 in the medium-dose group and 0 among 6 in the high-dose group. In the multiple dose part, it reports 2 among 10 for sham and 3 among 23 for high-dose BI 765128. These are participant counts within each group; they should not be added across parts or interpreted as a direct comparison between the uncontrolled dose groups and randomized groups.

The registry also reports other adverse events: 1 among 3 in the low-dose group, 1 among 3 in the medium-dose group, 3 among 6 in the high-dose single-dose group, 5 among 10 in the sham group and 3 among 23 in the multiple-dose high-dose group. The record’s listed serious-event terms include cerebrovascular accident (33.3%), chest pain (10.0%), herpes zoster (10.0%), lumbar spinal stenosis (4.3%), malignant melanoma (4.3%) and syncope (4.3%). The registry record does not provide enough context in these figures to infer a causal relationship or compare event rates reliably.

Safety counts are descriptive. Small groups and differences between study parts limit precision and comparability. The counts do not by themselves establish whether an event was caused by study treatment.

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

The primary result means that no ocular dose-limiting events meeting the registry’s definition were recorded in the low-, medium- or high-dose single rising dose groups during the period through day 8. It provides an early, short-window description of observed tolerability in the treated participants. It does not show that the risk of such events is zero, address events outside the evaluation period, or establish long-term safety.

The study’s two parts answer different questions. Part A used single rising doses without a control group and generated the primary endpoint. Part B randomized participants to repeated high-dose BI 765128 or sham and included secondary outcomes such as visual acuity. The visual acuity estimate is imprecise, with a sponsor-reported 95% confidence interval that spans values below and above zero. The results do not establish a precise difference between the groups, and the record does not include numerical results for several other listed secondary outcomes.

These findings should not be interpreted as proof that BI 765128 improves vision or changes the foveal avascular zone. Nor do the zero primary events establish that the treatment is free of ocular risks. The primary outcome is a count of specific events in small groups during a limited period; it is not a comprehensive assessment of all possible benefits and harms.

More generally, a trial result is conditional on its participants, outcome definitions, follow-up and analysis. The registry record describes people with diabetic macular ischemia who had previously received laser treatment, and the posted primary endpoint is limited to the single rising dose part. Its findings should not be generalized beyond that study context without further evidence. The record does not report enough information to assess every analysis assumption or the handling of missing data and multiplicity.

14. Why This Trial Matters Statistically

This study is statistically notable because it separates early dose exploration from a later randomized sham-controlled component. The primary endpoint is drawn from the uncontrolled single rising dose part, while the posted visual acuity comparison comes from the multiple dose part. Keeping those analyses distinct prevents the randomized comparison from being incorrectly treated as the source of the primary safety count.

The primary result illustrates an important limitation of zero-event observations: a count of zero is informative about what was observed, but its precision depends on the size of the group and the observation period. With only 3, 3 and 6 participants in the respective groups, the data offer limited information about uncommon events. A zero count is not the same as evidence that an event cannot occur.

The secondary visual acuity result also shows why estimates should be read with their uncertainty. The sponsor-reported mean difference is accompanied by a confidence interval spanning both directions. Reporting the estimate without its interval would obscure how little precision the comparison offers. The approximate unadjusted recalculation is a different analysis from the sponsor-reported adjusted model and should not be substituted for it.

Finally, the record lists several secondary outcomes but does not report numerical results for all of them, nor does it report a multiplicity strategy or detailed missing-data methods. These gaps matter because they restrict assessment of the evidential strength of secondary findings. The study can inform how early outcomes were recorded in this trial, while the registry record does not support broad claims about efficacy or comprehensive safety.

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 primary endpoint has a short follow-up window, ending on day 8, and is based on small treated groups. It captures protocol-defined ocular events rather than every potential adverse effect. The registry record does not provide a confidence interval for the primary event counts, so their precision cannot be assessed from a reported interval.

The primary endpoint comes from an uncontrolled dose-escalation part. Its counts describe what occurred in the observed dose groups but cannot establish a comparative effect against sham or another control. The randomized, sham-controlled part is separate and contributes the posted visual acuity comparison, not the primary endpoint result.

The visual acuity comparison has a wide sponsor-reported confidence interval, and numerical results for the listed foveal avascular zone and other secondary outcomes are not reported in the registry record. The record does not describe the full MMRM specification, missing-data methods, sensitivity analyses or multiplicity procedure. The recalculated unadjusted result is approximate and should not be confused with the sponsor’s adjusted analysis.

Serious adverse event counts are reported for groups in both parts, but the parts differ in design and treatment schedule. The counts alone cannot establish causality or support reliable comparisons across those groups. The registry record also does not report the rationale for the enrollment target or detailed statistical planning. These limitations restrict interpretation to the posted descriptive results and the specific analyses reported.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of BI 765128 in Patients With an Eye Condition Called Diabetic Macular Ischemia Who Have Received Laser Treatment — NCT04919499.
• Registry record last updated 2024-11-21; status last verified 2024-11.
• Results first posted 2024-08-27.
• 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

The registry record describes a completed phase 1 and phase 2 study of BI 765128 in people with diabetic retinopathy and diabetic macular ischemia after panretinal photocoagulation. It included an uncontrolled single rising dose part and a randomized, double-masked, sham-controlled multiple dose part.

The primary endpoint counted ocular dose-limiting events through day 8 in the single rising dose part. No events were reported in the low-, medium- or high-dose groups. The secondary visual acuity result at visit 7 was an adjusted mean difference of 0.1 letters, with a 95% confidence interval from -4.7 to 5.0. Numerical results for several other listed secondary outcomes are not reported in the registry record.

The posted findings are descriptive and limited by small groups, short primary-endpoint follow-up, incomplete reporting of analysis details and the distinction between study parts. They do not establish a treatment effect or rule out uncommon or later events.