A Study to Evaluate the Safety and Clinical Response of Subretinal Administration of CNTO 2476 in Participants With Geographic Atrophy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Safety and Clinical Response of Subretinal Administration of CNTO 2476 in Participants With Geographic Atrophy (NCT02659098), a na phase 2 study. The primary outcome is Change From Baseline in Best Corrected Visual Acuity (BCVA) Total Letters Score at Month 1- Treated Eye.

2. Clinical Question

3. Trial Design

The registry identifies this as a Phase 2 study of CNTO 2476 delivered subretinally in participants with visual acuity impairment associated with geographic atrophy. It records enrollment of 21 participants, a start date of 2015-12-04, primary completion on 2019-06-05, and a completed status. The listed interventions are CNTO 2476 3.0 x 10^5 cells and the Subretinal Delivery System. The treatment-group description says that participants in an open-label safety run-in phase received the cells in 50 microliter using a custom-designed Delivery System.

There is an important difference between the official title and the registry’s design fields. The official title describes a Phase 2b, multicenter, double-masked, randomized study, whereas the recorded allocation is NA, masking is NONE, and the analysis model is SINGLE GROUP. The registry also flags that this is not a randomized comparison and that arms are not a like-for-like control comparison. The design fields therefore do not support interpreting the record as a randomized controlled contrast. The registry record does not explain the discrepancy between the title and those fields.

The primary outcome is change from baseline in treated-eye best corrected visual acuity at Month 1. The listed secondary outcomes include visual-acuity changes and losses in treated and fellow eyes, changes in geographic atrophy lesion growth rate, and adverse-event counts. The record lists 26 outcome measures and zero analyses. A detailed statistical analysis plan, including a prespecified model or approach to uncertainty estimation, is not reported in the registry record.

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: The safety analysis set defined as participants who were enrolled and received surgery in the study.

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.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Best Corrected Visual Acuity (BCVA) Total Letters Score at Month 1- Treated Eye, measured over Baseline and Month 1.

The primary endpoint measures the change from baseline to Month 1 in best corrected visual acuity (BCVA) in the treated eye. BCVA is assessed with logMAR visual acuity testing charts. The endpoint description gives the scoring formula as 1.7 - 0.02*number of letters read. It also explains that each letter corresponds to 0.02 log units and that a line of five letters represents 0.1 log units. On this scale, a lower BCVA score indicates better vision, and a negative change score indicates improvement.

Using a continuous score retains information about the measured level of visual acuity and its change, rather than reducing the result to a simple event category. A mean change can describe the average direction and size of change among the analyzed participants. Its interpretation depends on seeing the estimate alongside variability and uncertainty, as well as knowing how many participants contributed endpoint data. Those result details are not reported in the registry record.

The registry labels the endpoint as a change in “total letters score” but gives logMAR as its unit and describes a logMAR formula. Readers should interpret the endpoint according to the scoring description and unit actually recorded, rather than assume that a higher letter count is the reported analysis scale. The result itself, baseline and Month 1 summaries, and the number analyzed are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical methods that can be described are limited to the outcome definition and design information in the registry. It identifies the primary outcome as a mean change in a continuous BCVA measure, expressed in logMAR, from baseline to Month 1 in the treated eye. It does not report the statistical analysis used to estimate that change, calculate uncertainty, or handle incomplete observations. General explanations below clarify what such statistics would mean; they do not claim that a particular unreported method was used.

What does a mean change describe?

In general, a mean change summarizes the average difference between a participant’s follow-up measurement and baseline measurement across the participants included in an analysis. It can indicate the average direction and magnitude of change on the recorded scale. For this endpoint, the registry states that a negative change indicates improvement. A mean is a group-level summary: it does not show how much individual participants varied, whether all participants changed in the same direction, or how representative the average is of the full enrolled group.

The record specifies standard deviation as the dispersion measure for the primary endpoint. In general, a standard deviation describes the spread of individual measurements around a mean. It is not the same as a confidence interval: the standard deviation characterizes variability among observations, whereas a confidence interval expresses uncertainty about an estimated quantity under the analysis assumptions. The registry does not report the primary mean or standard deviation, and it does not report a confidence interval.

How should the logMAR scale be read?

The endpoint description states that BCVA was measured using logarithm of the minimum angle of resolution charts and gives the formula 1.7 - 0.02*number of letters read. It says each letter represents 0.02 log units, with five letters per line corresponding to 0.1 log units. The stated interpretation is that lower BCVA scores indicate better vision and negative change indicates improvement. These directions matter: the sign of a change should be read using the registry’s scale definition rather than assuming that a larger numerical score is better.

The endpoint title refers to a “total letters score,” while the recorded unit is logMAR and the description gives a logMAR scoring formula. That wording warrants care when describing the outcome. The registry does not provide a result that would allow readers to check how the scale was applied in the reported analysis.

What can be inferred from the design?

The registry records a single-group model and says allocation is NA. In general, a within-group change is not equivalent to a randomized comparison. Participants may change over time for reasons other than the intervention, and a baseline measurement does not serve as a concurrent control group. The record does not describe a between-group analysis, and it explicitly notes that the arms are not a like-for-like control comparison. It would therefore be inappropriate to interpret the listed primary endpoint as a randomized treatment-versus-control effect.

Repeated measurements from the same participant are related. In general, an analysis of change can account for that relationship, but the precise method depends on the analysis plan and available data. The registry does not report whether the primary analysis used change scores, a repeated-measures model, or another approach. It also does not report covariate adjustment, assumptions, diagnostic checks, or an analysis population for the primary endpoint.

What do uncertainty and significance measures add?

In general, a confidence interval places an estimated effect in the context of its statistical uncertainty. A narrow interval indicates greater precision than a wide interval, conditional on the model and data; neither interval width nor statistical significance alone establishes clinical importance. A p-value, when reported, describes how compatible the data are with a specified null hypothesis under the analysis assumptions. It does not measure the probability that the null hypothesis is true or quantify the size of an effect. The registry record does not report a confidence interval, p-value, or hypothesis test for the primary outcome.

The sample-size material is described as a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. These are planning assumptions, not evidence of observed power. The illustration is not the trial’s own design, and the registry does not report the sample-size justification or design calculation used for this study. The numerical planning results and the observed primary effect are not reported in the registry record, so they cannot be interpreted.

What does the safety summary count?

The record reports participants at risk and participants affected for serious and other treatment-emergent adverse events. These counts are descriptive summaries. In general, a count of affected participants is distinct from a count of event episodes; the registry’s listed summary does not provide event-level timing, recurrence, severity grading, or causal assessment. No comparative safety estimate can be derived from a single-group record without a suitable comparison and the relevant analysis.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("CNTO 2476 3.0*10^5 Cells")
n   <- c(21)
m   <- c(0.132)
sp  <- c(0.2093)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  CNTO 2476 3.0*10^5 Cells  21  0.132  0.209  0.037  0.227

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 26 outcome measures, including treated-eye and fellow-eye visual-acuity outcomes, geographic atrophy lesion growth measures, and adverse-event outcomes. Considering multiple outcomes can raise multiplicity concerns: when many hypotheses are examined, the chance of finding an apparently notable result by chance can increase. In general, a prespecified primary outcome and a stated strategy for controlling or interpreting multiple comparisons help readers understand how findings should be weighed. A multiplicity strategy is not reported in the registry record.

The registry lists zero analyses. It does not report a hierarchy of secondary outcomes, an adjustment method, or which outcomes were formally tested. The listed outcomes therefore describe the measurement program, but they do not establish that every measure was analyzed or tested in a particular way. The record does not provide outcome-level estimates or uncertainty for the listed measures.

Missing-data procedures for the primary endpoint are not reported in the registry record. The endpoint information does not state how participants without a Month 1 measurement were handled, whether any values were imputed, or what assumptions were made about missingness. The number with available primary endpoint data is also not reported. In general, missingness can affect both the precision and interpretation of a change estimate, particularly if the probability of having a measurement is related to outcome status.

The registry defines the safety analysis set as participants who were enrolled and received surgery. That definition is specific to safety; the primary endpoint analysis population is not reported in the registry record. Enrollment is recorded as 21, but the record does not establish how many contributed to the Month 1 endpoint analysis or explain exclusions from that analysis. These gaps limit assessment of the outcome denominator and the potential effect of missing data.

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: haemorrhagic anaemia (4.8%), aortic valve stenosis (4.8%), atrial fibrillation (4.8%), osteoarthritis (4.8%), lung neoplasm malignant (4.8%).

The registry reports 5 participants affected by serious adverse events among 21 at risk, and 18 participants affected by other treatment-emergent adverse events among 21 at risk. It also reports 0 deaths. These are participant counts, not a comparison between treatment groups. Because the record describes a single group, the counts do not estimate how event frequency would differ under a control intervention.

The listed serious-event terms are Haemorrhagic Anaemia, Aortic Valve Stenosis, Atrial Fibrillation, Osteoarthritis, Lung Neoplasm Malignant, and Subarachnoid Haemorrhage. Each term is accompanied by 4.8% in the registry. These term-level percentages should be read as reported; the record does not provide additional event-level details that would explain recurrence, timing, or the denominator used for each term. The terms do not establish that the intervention caused the events.

The counts for serious and other treatment-emergent adverse events may overlap, so they should not be added to form a total. The registry does not report a control-group safety profile, a causal assessment for each event, or enough detail to compare risks across interventions. The figures describe safety reporting in the recorded group, not a comparative estimate.

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

The registry record documents a study of CNTO 2476 delivered using a custom-designed system and identifies a Month 1 treated-eye BCVA change as the primary endpoint. It also provides the direction of the logMAR scale, a definition of the safety analysis set, and counts for serious and other treatment-emergent adverse events. These details clarify what the study set out to measure and how some reported quantities should be read.

The record does not report the primary endpoint estimate, its standard deviation, an uncertainty interval, or the number analyzed. As a result, it does not show the size or precision of the Month 1 mean change. The endpoint definition alone cannot indicate whether an observed change was small or large, consistent or variable, or statistically distinguishable from no change.

The record’s single-group design also matters. A change from baseline is descriptive of change over time in the measured group; it is not, by itself, a comparison with what would have happened without the intervention. The registry explicitly cautions that this is not a randomized comparison and that arms are not a like-for-like control comparison. The randomized and double-masked wording in the official title does not resolve the different design fields recorded in the registry.

The safety counts show that adverse events were recorded among participants in the group, including serious events. They do not establish that the intervention caused any particular event, nor do they quantify a difference from an alternative treatment. The record gives no comparative safety analysis.

Accordingly, the registry supports a careful description of the study’s endpoint, design fields, and reported safety counts. It does not support a quantitative conclusion about the primary visual-acuity result or a causal treatment effect. The posted results should be interpreted in light of the missing endpoint estimate and the absence of a like-for-like control comparison.

14. Why This Trial Matters Statistically

This trial is statistically notable because its official title and its recorded design fields point in different directions. The official title describes a multicenter, double-masked, randomized study, while the registry fields state allocation NA, masking NONE, and a single-group model. The registry also warns against treating arms as a like-for-like control comparison. This distinction is central: randomization and a concurrent comparator can support a different type of inference from a single-group change over time. The record does not explain the discrepancy, so the design fields and explicit caution deserve attention when interpreting results.

The primary endpoint is a continuous visual-acuity measure with a stated direction: lower logMAR scores indicate better vision, and negative change indicates improvement. This makes the scale interpretable, but interpretation still depends on the actual estimate, variability, and precision. Those primary numerical results are not reported in the registry record. A clearly defined outcome is not a substitute for a reported result.

The record also illustrates why safety summaries need careful framing. Participants with serious or other treatment-emergent adverse events are counted in a single group, and listed serious-event terms are accompanied by percentages. Without a control group or event-level context, these figures describe what was recorded but cannot establish comparative risk or causation.

Finally, the registry lists multiple outcome measures but no analyses. This leaves unresolved which secondary results were analyzed, how missing endpoint measurements were handled, and whether multiplicity was addressed. These omissions affect how confidently readers can connect the study’s stated objectives with its statistical conclusions. The record is useful for identifying the design and questions, while its limits prevent a complete assessment of the primary result.

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 limitation for statistical interpretation is that the registry record does not report the numerical result for the primary endpoint. It gives no mean change, standard deviation, confidence interval, p-value, or number analyzed at Month 1. Without these, neither the magnitude nor precision of the endpoint can be assessed.

The design information is internally difficult to reconcile. The official title describes a double-masked, randomized study, but the registry records allocation as NA, masking as NONE, and a single-group model. It also states that this is not a randomized comparison. The reason for the inconsistency is not reported in the registry record, so the title cannot be used to assume a controlled analysis.

The record does not report the analysis method for the primary outcome, the analysis population for that outcome, or procedures for missing measurements. It also does not provide details about adjustment, model assumptions, or how repeated measurements were handled. The registry lists 26 outcome measures and zero analyses, but does not describe a multiplicity strategy or clarify which secondary outcomes were formally analyzed.

For safety, the available information consists of participant counts, deaths, and selected serious-event terms with percentages. The record does not report comparative event rates, detailed event timing, or causal assessments. These limitations mean the counts cannot be interpreted as evidence of relative safety or attribution to the intervention.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study to Evaluate the Safety and Clinical Response of Subretinal Administration of CNTO 2476 in Participants With Geographic Atrophy — NCT02659098.
• Registry record last updated 2025-04-29; status last verified 2025-04.
• Results first posted 2020-06-23.
• 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 describes a completed Phase 2 study of CNTO 2476 delivered using a Subretinal Delivery System in participants with visual acuity impairment associated with geographic atrophy secondary to age-related macular degeneration. Enrollment is recorded as 21. The study started on 2015-12-04, and primary completion is recorded as 2019-06-05.

The primary outcome is change from baseline to Month 1 in treated-eye BCVA, reported as a mean continuous measure in logMAR. The registry says lower scores indicate better vision and negative change indicates improvement, but the primary numerical result and its precision are not reported in the registry record.

The recorded design is single group, with allocation NA and masking NONE, despite different wording in the official title. The record reports 5 participants with serious adverse events, 18 with other treatment-emergent adverse events, and 0 deaths. It lists 26 outcome measures and zero analyses. Missing-data methods, multiplicity procedures, and the primary endpoint analysis method are not reported in the registry record.