Gene Therapy for Achromatopsia (CNGB3): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Gene Therapy for Achromatopsia (CNGB3) (NCT03001310), a non-randomized phase 1, phase 2 study. The primary outcome is Number of Participants Meeting the Primary Outcome Defined as Any of the Below Events Occurring During the 6 Weeks Following Administration, at Least Possibly Related to the Advanced Therapy Investigational Medicinal Products (ATIMP), Not Surgery Alone..

2. Clinical Question

3. Trial Design

This was an open-label, multicentre phase I/II trial of AAV - CNGB3 for adults and children with achromatopsia owing to CNGB3 defects. The registry describes its model as sequential and its allocation as nonrandomized. The stated intervention was a single subretinal administration, with low, intermediate, and high dose groups described. A further safety grouping is labelled “Other Dose”; the registry record does not explain that label further.

The record lists enrollment as 23 and status as completed. The recorded start date is 2017-01-16, and the primary completion date is 2019-10-25. These registry entries describe the study record; they do not establish that each enrolled participant completed every assessment or had complete outcome data.

Because participants were not randomized between dose groups and there was no listed control group, differences between groups cannot be read as controlled estimates of dose effects. The sequential dose-escalation structure is important context: the groups are part of a study evaluating an intervention across dose levels, not parallel randomized treatment arms. The registry record does not report the allocation sequence, the rules governing progression between dose levels, or the statistical analysis plan.

The listed secondary outcomes concern visual acuity, retinal function assessed by static perimetry, and quality of life assessed separately for children and adolescents and for adults. Five outcome measures are recorded, but the registry record lists zero analyses. Numerical results for these secondary outcomes, and the methods used to analyze them, are 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

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 Number of Participants Meeting the Primary Outcome Defined as Any of the Below Events Occurring During the 6 Weeks Following Administration, at Least Possibly Related to the Advanced Therapy Investigational Medicinal Products (ATIMP), Not Surgery Alone., measured over 6 weeks.

The primary endpoint is a participant count for a composite safety outcome over the 6 weeks following administration. A participant meets the definition if any listed event occurs and is judged at least possibly related to the investigational medicinal product, rather than surgery alone. The events include a reduction in visual acuity by 15 ETDRS letters or more that does not resolve to within 15 letters of baseline in a 4-week period once prophylactic treatment commences; severe unresponsive inflammation; infective endophthalmitis; ocular malignancy; or a grade III or above non-ocular SUSAR.

In general, a composite endpoint combines multiple events into one outcome. This can capture several kinds of safety concern in a single measure, but it also means that the overall count does not by itself show which component occurred or how serious each event was. The registry definition also includes an attribution condition: events must be at least possibly related to the investigational product, not surgery alone. That judgment is part of the endpoint definition, not proof that the product caused an event.

A defined follow-up window makes the period of observation explicit. The registry record does not report the primary endpoint count, component-specific counts, or details of how attribution judgments were made. Thus, the definition is available, but the primary endpoint result cannot be interpreted from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a nonrandomized, sequential trial. In general, random allocation helps make comparison groups similar on average with respect to factors measured and unmeasured at baseline. Without randomization, differences between dose groups may reflect participant characteristics, timing, clinical decisions, or other influences as well as dose. The registry’s design information therefore supports describing the groups, but not treating their outcomes as a controlled estimate of what would have happened under a different dose or no intervention.

What does the sequential dose design imply?

In general, a sequential dose-escalation study evaluates dose levels in an ordered study process rather than assigning participants concurrently at random to balanced groups. That can be appropriate to a study whose focus includes dose evaluation and safety, but it limits simple comparisons across groups. The registry record calls the model sequential and describes low, intermediate, and high dose groups. It does not report the escalation criteria, how dose decisions were made, or whether any additional rules affected who entered each group. Those details should not be inferred from the group labels.

How should counts be read?

A participant count records how many people in a group met a specified event definition. Counts need their denominators for context, and the adverse-event record lists both participants at risk and participants affected. The serious-event total is 2 of 23, while the group counts are 0 of 3, 1 of 12, 0 of 3, and 1 of 5. These are observations in the enrolled groups, not estimates from a randomized comparison. With counts of this kind, a small change in the observed number of events can materially alter the descriptive picture; general statistical uncertainty is therefore important, even when no interval estimate is posted.

In general, a confidence interval expresses the uncertainty around an estimated quantity under a specified statistical model. It is not a guarantee that the true value lies within the interval, and its interpretation depends on the design and assumptions used to calculate it. The registry record does not report confidence intervals for the event counts or a model-based comparison. No new interval should be inferred from the posted counts alone as a sponsor analysis.

What does the composite endpoint capture?

The primary endpoint combines several distinct events into a single participant-level outcome, within the 6 weeks following administration, and uses a relatedness threshold of at least possibly related to the investigational product, not surgery alone. In general, composite endpoints can make an endpoint responsive to a range of clinically important events. Their overall count can be harder to interpret if the component events differ in type or seriousness, so component-specific counts are useful. The registry record does not report the primary composite count or the counts for its individual components.

What can the adverse-event summaries establish?

The adverse-event figures are descriptive. In general, a descriptive summary reports observed events without resolving whether the intervention caused them. Attribution is particularly difficult to establish from an uncontrolled study, because events may have other explanations and there is no randomized comparison to estimate a counterfactual event rate. The primary endpoint’s relatedness rule defines what is counted for that endpoint; it does not mean that every event in the separate adverse-event summary was judged related. The record does not report enough detail to reconcile those two presentations.

The posted term percentages should be read as registry descriptors, not as a complete estimate of risk. The record lists Uveitis at 20.0% and Ectopic pregnancy at 8.3% among top serious terms, but does not report the denominators or calculation method for these percentages in the registry record. No conversion to counts, group rates, or comparative risks is warranted.

What statistical analyses are reported?

The record lists zero analyses. It does not report inferential tests, adjusted estimates, confidence intervals, covariate adjustment, or a statistical analysis plan. This absence means that a reader cannot assess whether particular planned analyses were carried out, how multiple outcomes were handled, or which analysis population was used. It does not justify assuming that no analyses occurred outside the registry; it means they are not reported in the registry record.

In general, statistical significance and clinical importance are different concepts. A statistical test assesses compatibility between observed data and a model-based null hypothesis, while the size and relevance of an outcome require interpretation in context. No p-values or estimates of visual-function change are reported in the registry record, so neither statistical significance nor magnitude of those outcomes can be evaluated from these data.

Any sample-size figures associated with planning are illustrations based on an observed effect, not the trial’s own design justification or evidence of observed power. In general, power is a prospective property of a specified design and effect assumption; it should not be presented as a measurement of how much evidence an observed study produced. The registry record does not report a sample-size justification, and the planning illustrations do not fill that gap.

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("Low Dose AAV - CNGB3", "Intermediate Dose AAV-CNGB3", "Other Dose - AAV - CNGB3", "High Dose AAV - CNGB3")
n      <- c(3, 12, 3, 5)
events <- c(0, 1, 0, 2)
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  Low Dose AAV - CNGB3  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Intermediate Dose AAV-CNGB3  12  1  8.3  0.2  38.5  1.5  35.4  1.9  36.0
ARM  Other Dose - AAV - CNGB3  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  High Dose AAV - CNGB3  5  2  40.0  5.3  85.3  11.8  76.9  11.8  77.7

11. Multiplicity, Missing Data and Analysis Populations

The registry lists secondary outcome areas covering visual acuity, retinal function assessed by static perimetry, and quality of life in children and adolescents and in adults. It records five outcome measures and zero analyses. Numerical results for these outcomes, and the tests or models used to analyze them, are not reported in the registry record.

In general, evaluating multiple outcomes creates a multiplicity issue: when many hypotheses are tested, the chance of at least one apparently positive finding can increase if each test is interpreted separately. Statistical approaches can address this, for example by defining a primary analysis in advance or adjusting inference across a family of tests. The registry record does not report whether any multiplicity strategy was planned or used. With no reported analysis results, it is not possible to assess how this issue affected interpretation.

The record does not report missing-data counts, reasons for missingness, outcome-specific analysis populations, imputation methods, or sensitivity analyses. Enrollment and the safety denominator do not establish completeness for every visual-function or quality-of-life assessment. In general, missing outcome data can affect estimates when the likelihood of missingness is related to participants’ characteristics or outcomes. Without the number and pattern of missing assessments, the extent or direction of any such effect cannot be judged.

The record also does not report whether all enrolled participants were included in every outcome analysis, or how withdrawals and incomplete follow-up were handled. These details are necessary to evaluate the analysis population and the robustness of results, but they are not reported in the registry 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: uveitis (20.0%), ectopic pregnancy (8.3%).

The adverse-event summary records serious events for 0 of 3 participants in the low-dose group, 1 of 12 in the intermediate-dose group, 0 of 3 in the “Other Dose” group, and 1 of 5 in the high-dose group. The total row records 2 of 23 with serious events. Deaths are recorded as 0 in every listed group and in the total row. These figures describe the registry’s reported counts; they do not establish whether events were caused by AAV - CNGB3.

The record also lists other adverse events for all participants at risk in each group and in the total row. The registry does not give enough detail to interpret the full definition, seriousness, timing, or treatment relationship of this category. Serious and other event categories should not be combined, and the figures should not be assumed to represent the primary endpoint: that endpoint has a specific 6-week window and an attribution criterion.

Uveitis at 20.0% and Ectopic pregnancy at 8.3% are listed as top serious-event terms. The registry record does not state the denominators or calculation method associated with these percentages. The small, nonrandomized dose groups also do not support reliable causal comparisons across dose levels. These counts are safety descriptions, not medical guidance or proof of benefit or harm.

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

The record documents a completed, open-label, nonrandomized phase I/II study of subretinal AAV - CNGB3 in adults and children with achromatopsia owing to CNGB3 defects. It describes a sequential dose-escalation approach and records enrollment of 23. The primary endpoint was defined as a count of participants experiencing any of a set of events in the 6 weeks following administration, subject to an attribution criterion. The registry does not report the result for that endpoint.

The posted adverse-event data give descriptive counts, including 2 of 23 participants with serious events and 0 deaths recorded. They do not, on their own, establish that the intervention caused or prevented any event. The dose groups were not randomized, so differences between them cannot be interpreted as causal dose effects. The figures also do not provide the timing, duration, or detailed clinical course of each event.

The trial’s listed secondary outcomes include visual acuity, static perimetry, and quality of life. Their numerical results are not reported in the registry record. As a result, this record does not support conclusions about the magnitude or precision of changes in visual or quality-of-life outcomes. It also does not establish whether any observed changes, if assessed elsewhere, would be attributable to treatment rather than other influences.

In general, a trial’s design and endpoint definitions describe what it set out to assess; they are not themselves evidence of an outcome. A completed status does not supply missing numerical results, and a safety count is not a substitute for an efficacy estimate. The registry record is informative about the study structure and the event counts it posts, while leaving the primary endpoint result, secondary outcome findings, and statistical analyses unreported.

Accordingly, the defensible interpretation is limited to the design description and reported adverse-event counts. The record does not establish that treatment works or does not work, and it does not provide a basis for medical decisions.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of why design details matter when interpreting early-phase dose studies. Its open-label, nonrandomized, sequential structure is central to reading the group counts: those groups are not randomized controls, so between-group contrasts can reflect more than dose. Descriptive safety counts remain useful for documenting what was recorded, but they cannot replace controlled estimates of causal effects.

The primary endpoint is defined in detail and uses a fixed 6-week period, a group of specified safety events, and an attribution rule that excludes surgery alone. These features show how an endpoint definition determines what a reported count would mean. At the same time, the registry record does not report the count meeting that definition. The distinction between an endpoint definition and its result is essential: knowing what was measured is not the same as knowing the observed outcome.

The adverse-event summary illustrates another statistical caution. Small group counts can look very different when viewed as raw counts, but raw differences across nonrandomized groups do not establish a dose-response relation. The record’s event totals are best read as descriptive observations, with uncertainty and study design limiting stronger interpretations.

Finally, the registry lists several visual-function and quality-of-life outcome areas but reports zero analyses and no numerical outcome findings. This makes transparent reporting of analysis populations, missing data, uncertainty estimates, and multiplicity methods especially relevant to interpretation. The case underscores that a completed trial record may document design and safety counts while leaving readers unable to assess the precision or statistical evidence for the intended visual and quality-of-life outcomes.

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 main limitation for interpretation is that the registry record does not report a numerical result for the primary endpoint, despite defining its components and follow-up period. It also does not report numerical results for the listed visual acuity, static perimetry, or quality-of-life outcomes. The record lists zero analyses, leaving no reported inferential estimates or analysis details to evaluate.

The study was open-label and nonrandomized, with sequential dose groups. Consequently, the groups are not like-for-like controls, and their outcomes should not be used to infer causal dose effects. The registry record does not provide details about the escalation criteria or group assignment process. It also does not clarify the “Other Dose” group label beyond the name recorded.

The adverse-event information is limited to participant counts, denominators, deaths, and listed serious-event terms. It does not report event timing, duration, recurrence, severity beyond the serious classification, relatedness for the separate adverse-event rows, or component-level results for the primary endpoint. The denominators and calculation method for the listed term percentages are not reported in the registry record.

Missing-data handling, outcome-specific analysis populations, statistical methods, confidence intervals, p-values, multiplicity procedures, and sensitivity analyses are not reported in the registry record. The recorded enrollment does not show whether all participants contributed data to each outcome. These limitations prevent assessment of precision, robustness, or the consequences of incomplete follow-up from the registry information alone.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Gene Therapy for Achromatopsia (CNGB3) — NCT03001310.
• Registry record last updated 2023-03-08; status last verified 2023-02.
• Results first posted 2023-03-08.
• 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 NCT03001310, “Gene Therapy for Achromatopsia (CNGB3),” as a completed, open-label, multicentre phase I/II trial. It was nonrandomized and sequential, with subretinal administration of a single low, intermediate, or high dose of AAV - CNGB3. The study concerned adults and children with achromatopsia owing to CNGB3 defects. Enrollment is recorded as 23; the start date is 2017-01-16 and the primary completion date is 2019-10-25.

The primary endpoint was a participant count for specified safety events during the 6 weeks following administration, subject to an investigational-product relatedness criterion. The primary endpoint result is not reported in the registry record. The adverse-event summary records 2 of 23 participants with serious events and 0 deaths. Numerical results for the listed visual-function and quality-of-life outcomes, statistical analyses, missing-data methods, and the statistical analysis plan are not reported in the registry record.