Relapse Prophylaxis With N-803 for AML and MDS Pts Following Allo HSCT: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Relapse Prophylaxis With N-803 for AML and MDS Pts Following Allo HSCT (NCT02989844), a na phase 2 study. The primary outcome is Incidence of Relapse.

2. Clinical Question

The question is about relapse incidence in a defined post-transplant setting, not a randomized estimate of the difference between N-803 and another treatment or no maintenance. Because all participants are assigned to the N-803 group, the observed experience can describe what occurred in that group, but cannot by itself isolate the effect of N-803 from the effects of transplantation, underlying disease, patient characteristics, follow-up, or other aspects of care.

The clinical context is AML and MDS after RIC allogeneic stem cell transplantation. The record describes N-803 as maintenance therapy and names relapse as the primary outcome. It also lists adverse events, acute and chronic graft-versus-host disease, minimal residual disease, overall survival, non-relapse mortality, and relapse among secondary outcomes. Numerical findings for these outcomes, apart from the safety summaries recorded in the registry, are not reported in the registry record.

3. Trial Design

This is a Phase II, single-group, multi-center trial of N-803 maintenance after allogeneic hematopoietic cell transplantation. The conditions are acute myelogenous leukemia and myelodysplastic syndrome, and the trial is listed as completed. Enrollment is reported as 20. The sponsor is Masonic Cancer Center, University of Minnesota.

The registry record gives the start date as 2017-04-12 and the primary completion date as 2022-03-19. These dates describe the trial record; they do not supply the timing or completeness of follow-up for each participant. The record does not report individual follow-up duration, the number of participants assessed for the primary endpoint, or the number with a relapse event.

Allocation is listed as not applicable and masking as none. The model is single group, with N-803 as the intervention. Therefore, the design does not include a randomized comparison or a concurrent control arm. Any observed relapse proportion or cumulative incidence would describe the N-803 group under the trial’s follow-up and analysis procedures; it would not be a direct estimate of a treatment contrast against a control group.

The primary endpoint is the incidence of relapse, expressed as a percentage of participants, with a 95% confidence interval and a timeframe of 24 months. Its description defines the assessment window from the first N-803 dose to 2 years after RIC allogeneic transplantation. The registry record does not explain how the timing of the first dose was handled relative to transplant, how participants without complete follow-up were treated in the analysis, or how death before relapse was addressed.

The record lists 9 outcome measures and 0 analyses. Its secondary outcomes include adverse events, acute graft-versus-host disease, chronic GVHD, minimal residual disease, overall survival, non-relapse mortality, and relapse. Acute graft-versus-host disease appears twice in the listed outcomes. The record does not report a statistical analysis plan, a sample-size justification, or details of a prespecified analysis population.

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 Incidence of Relapse, measured over 24 months.

The primary endpoint is the cumulative incidence of relapse from the first N-803 dose to 2 years after RIC allogeneic hematopoietic cell transplantation. It is expressed as a percentage of participants, and the endpoint description specifies a 95% confidence interval. The result itself, including the estimate and its interval, is not reported in the registry record.

Relapse is an important outcome to track after transplantation because it records recurrence of the underlying disease within the defined follow-up window. A cumulative-incidence estimate is intended to describe the chance of experiencing the event over time, rather than simply counting events without regard to when they occurred. The specified timeframe matters: an estimate for one follow-up window does not describe relapse risk outside that window.

General statistical context: when another event can prevent the event of interest from occurring or being observed, such as death before relapse, the analysis may need to account for that competing event. A cumulative-incidence approach can be used for this purpose, but the registry record does not state which estimator or competing-event method was used. It also does not report how censoring, missing follow-up, or participants without an observed relapse were handled.

The endpoint is framed as an incidence in the N-803 group, not as a comparison of relapse between treatment groups. Without a control group, it cannot quantify a difference attributable to N-803. A confidence interval would communicate uncertainty around the estimated incidence, but no interval bounds are reported, so the precision of the primary estimate cannot be evaluated from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a cumulative-incidence endpoint describe?

For an outcome recorded over follow-up, the timing of events matters. A simple proportion with an event may not convey when events occurred or how long participants were observed. A cumulative-incidence estimate describes the probability of experiencing the specified event by a point in time, under the analysis definitions used. Here, the endpoint is relapse incidence through 24 months, with the assessment window described as beginning at the first N-803 dose and ending 2 years after transplantation.

General statistical context: cumulative incidence differs from a rate based only on the number of events and total observation time. It also differs from a measure that treats competing events as if they were ordinary loss to follow-up. If a participant dies before relapse, relapse can no longer be observed for that participant. Depending on the scientific question, death may therefore need to be treated as a competing event rather than handled as routine censoring. The registry record does not report the method used to address death before relapse, censoring, or incomplete follow-up.

The endpoint description names a 95% confidence interval. A confidence interval is a range calculated using a statistical method to reflect uncertainty in an estimate. A narrower interval generally indicates greater precision under that method; a wider interval indicates greater uncertainty. The interval does not correct for bias in the study design, and it does not make a single-group estimate equivalent to a randomized treatment comparison. No confidence-interval bounds for the primary endpoint are reported in the registry record.

What can a single-group design estimate?

A single-group design can describe outcomes observed among participants receiving the intervention, provided the outcome definitions and analysis are specified. It does not directly estimate what would have happened to those same participants under a different treatment. A comparison against an external expectation or a historical group would require an appropriate reference and assumptions about comparability; no such comparator is identified in this record.

Because the trial has no concurrent control group, changes in outcomes cannot be attributed to N-803 by comparison within the randomized design. This is a design limitation, not a criticism of the descriptive purpose of a Phase II study. The key interpretation is to distinguish an outcome observed in the trial group from an effect caused by the intervention. The registry record reports the study model as single group and allocation as not applicable.

What does the endpoint’s confidence interval contribute?

In general, the point estimate represents the result under the chosen analysis, while the confidence interval communicates sampling uncertainty around it. A confidence interval does not state the probability that a particular fixed value is true, and it does not account automatically for missing data, misclassification, selection, or an unsuitable analysis method. Interpreting its width requires the actual estimate, interval bounds, event information, and method. Those primary-endpoint quantities are not reported in the registry record.

The record describes the endpoint as a percentage of participants and specifies 24 months. It does not state whether that percentage is a raw proportion, a time-to-event cumulative-incidence estimate, or another calculation. The outcome title and description use cumulative incidence, but the estimator and its assumptions are not reported. This distinction matters because participants can have different observation times and because a competing event can affect the probability of observing relapse.

How should event counts and percentages be read?

The safety summary lists participants at risk and affected for serious and other adverse events. These are reported group counts. A count of affected participants is not necessarily a count of all event episodes: a participant could experience more than one event, and the record does not describe how recurring events were tabulated. The record also does not specify whether the displayed percentages for serious-event terms represent participants, event reports, or another denominator. Therefore, the listed 5.0% values should be read as registry-reported percentages, not recalculated or assigned a denominator not stated in the record.

General statistical context: safety summaries can use different units of analysis. A participant-level summary counts people with at least one event in a category, while an event-level summary counts episodes. These quantities answer different questions. Event severity, seriousness, relatedness, and frequency are also distinct attributes. A serious event is not automatically caused by treatment, and an event’s appearance in a serious-event listing does not establish attribution.

What does over-dispersion mean in general?

General statistical context: over-dispersion occurs when observed variability is greater than a statistical model expects. If a model assumes more uniformity than the data support, standard errors can be too small and confidence intervals can be too narrow. Methods that allow additional variability may produce more realistic uncertainty. Whether over-dispersion is relevant depends on the outcome and the model; the registry record does not report a model or an over-dispersion assessment for this trial.

What do the planning assumptions mean?

The registry record describes a planning illustration based on an observed effect and specifies a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It states that the effect to detect is a stated fraction of the observed effect. For time-to-event calculations, the illustration concerns total events; other calculations concern participants per arm. The numerical planning outputs and the effect used are not reported in the registry record.

These assumptions describe hypothetical planning calculations, not the analysis of this single-group trial. A planning scenario is not an observed result and is not observed power. In general, sample-size planning asks how much information might be needed to detect a specified effect under stated assumptions. Its answer depends on those assumptions, and it cannot substitute for reporting the trial’s actual outcome estimate, uncertainty, follow-up, and analysis population.

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("N-803")
n      <- c(20)
events <- c(6)
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  N-803  20  6  30.0  11.9  54.3  14.5  51.9  14.6  52.2

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 9 outcome measures, including the primary relapse endpoint and several secondary outcomes: adverse events, acute graft-versus-host disease, chronic GVHD, minimal residual disease, overall survival, non-relapse mortality, and relapse. Acute graft-versus-host disease appears twice in the listed secondary outcomes. The record reports 0 analyses, so numerical analyses for these outcomes are not available in the registry entry.

When a study evaluates multiple outcomes or makes multiple statistical comparisons, the chance of at least one apparently notable result can increase. General statistical context: a prespecified hierarchy, adjustment procedure, or clearly identified primary analysis can help readers understand how multiplicity was handled. The registry record does not report a multiplicity strategy, a hierarchy among secondary outcomes, or which outcomes were treated as confirmatory versus exploratory. It is therefore not possible to determine from this record whether multiplicity was addressed.

The registry record also does not report missing-data procedures, the number of participants with complete follow-up, or the number included in each outcome analysis. It does not define an analysis population, such as an intention-to-treat or per-protocol population. The enrollment figure alone cannot establish how many participants contributed to the primary endpoint estimate or any secondary outcome.

Other analysis details are not reported in the registry record, including the statistical method for the primary endpoint, treatment of death before relapse, censoring rules, handling of withdrawals, and any sensitivity analyses. These omissions prevent assessment of how robust the planned or reported results would be to alternative assumptions. No protocol-based explanation should be inferred where the registry entry is silent.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: respiratory failure (5.0%), atrial fibrillation (5.0%), infections and infestations - other, (5.0%), immune system disorders - other, (5.0%), sepsis (5.0%).

The registry safety summary lists 20 participants at risk for serious adverse events and 7 affected. For other adverse events, it lists 20 at risk and 20 affected. It also records 3 deaths. These are the recorded group figures; the record does not give additional detail about the timing, duration, severity grading, or cause of the events or deaths.

The serious-event terms listed are respiratory failure, atrial fibrillation, infections and infestations - Other, immune system disorders - Other, sepsis, and fall. Each is accompanied by a registry-reported percentage of 5.0%. The record does not specify whether these categories overlap, whether each participant could contribute to more than one term, or whether the percentages refer to participants or event reports. They should not be added together or interpreted as mutually exclusive categories.

General statistical context: counts of participants with an event and counts of event episodes are different summaries. A participant may have multiple events, and an event may be recorded in more than one relevant category. Seriousness is also separate from causality. The registry summary does not report whether events were judged related to N-803, nor does it provide a comparison group that would support an estimate of excess risk attributable to treatment.

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

The registry record establishes that a completed Phase II, single-group trial evaluated N-803 maintenance after RIC allogeneic stem cell transplantation in AML and MDS. It identifies relapse incidence through 24 months as the primary endpoint and specifies that the endpoint should be reported as a percentage of participants with a 95% confidence interval. The primary endpoint estimate and interval are not reported in the registry record.

Accordingly, the record does not establish the observed relapse incidence or how precisely it was estimated. It also does not show a numerical result for the listed secondary outcomes. The safety summary contains counts and serious-event term percentages, but it does not provide enough detail to establish event attribution, timing, or comparative risk. A completed status indicates that the trial is recorded as completed; it is not itself an outcome result.

Even if a numerical relapse estimate were available, this design would describe the experience of participants in the N-803 group rather than provide a randomized comparison. It would not, on its own, distinguish the contribution of N-803 from the effects of transplantation, disease characteristics, participant selection, follow-up, or other care. The record identifies no control group and no randomized allocation.

General statistical context: an estimate and its confidence interval answer different but complementary questions. The estimate describes the observed outcome under the analysis; the interval conveys uncertainty under the method’s assumptions. Neither alone establishes causation, and neither resolves limitations from missing data or study design. Since the estimate and interval are not reported, readers cannot assess the magnitude or precision of the primary outcome from this entry.

The appropriate conclusion is limited to what the record documents: the trial design, its stated endpoint, the listed outcomes, and the available safety summaries. The record does not support a conclusion that N-803 did or did not prevent relapse, nor does it support a comparative claim about treatment benefit or harm.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of why design and result reporting must be considered together. The registry record specifies a clinically relevant time-to-event outcome after transplantation and identifies a confidence interval as part of the primary endpoint reporting. But without the endpoint estimate, interval, event information, and analysis method, readers cannot evaluate either the observed relapse experience or its precision.

The single-group design also clarifies the difference between describing outcomes and estimating treatment effects. A group’s observed outcome can be useful for characterizing experience under a treatment approach, but a causal contrast requires a suitable comparison and assumptions that support attribution. The record lists no control group or randomized allocation. A reader should therefore avoid treating an eventual single-group incidence as a direct estimate of what N-803 changes relative to another option.

The trial also illustrates the role of transparent analysis details. For a cumulative-incidence outcome, readers need to know how follow-up, censoring, and competing events were handled. For safety summaries, they need to understand whether counts refer to participants or event episodes and how seriousness and attribution were assessed. These details are not reported in the registry record.

The listed secondary outcomes cover several different domains, including disease recurrence, survival, graft-versus-host disease, minimal residual disease, and adverse events. Without reported analyses and information on multiplicity, the status of findings across those outcomes cannot be assessed. The record’s planning description should likewise be distinguished from trial findings: planning assumptions do not replace observed estimates or establish observed power.

For biostatisticians and clinical researchers, the central lesson is practical: an endpoint definition, study design, analysis method, and numerical result are all needed to interpret evidence. In this record, the first two are described more clearly than the latter two. That reporting gap limits interpretation without changing what the design itself can and cannot answer.

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.

Several trial-specific details needed to interpret the primary endpoint are not reported in the registry record. These include the number of relapse events, the numerical cumulative-incidence estimate, the 95% confidence interval bounds, the estimator used, and how death before relapse was handled. The record gives an endpoint timeframe of 24 months and describes the window as running from the first N-803 dose to 2 years after transplantation, but does not explain the timing of treatment initiation relative to transplant or how follow-up was managed.

The record does not report participant-level follow-up, withdrawals, loss to follow-up, missing outcome data, or the number included in each analysis. It also does not identify an analysis population or provide a statistical analysis plan. These gaps make it difficult to judge how the endpoint was operationalized and whether the analysis population matched the enrolled group.

The single-group design has no concurrent control and no randomized allocation. Any outcome in the N-803 group therefore cannot by itself distinguish an intervention effect from differences in disease, transplant course, participant characteristics, or other care. The registry record also does not report a sample-size justification or the numerical outputs of the described planning illustrations.

For safety, the record lists affected and at-risk counts, deaths, and serious-event term percentages, but does not explain event-level versus participant-level tabulation, overlap among categories, timing, severity grading, relatedness, or causes of death. These limitations apply to the information in the registry record and should not be filled in by assumption.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Relapse Prophylaxis With N-803 for AML and MDS Pts Following Allo HSCT — NCT02989844.
• Registry record last updated 2023-11-03; status last verified 2023-10.
• Results first posted 2023-11-03.
• 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 NCT02989844, “Relapse Prophylaxis With N-803 for AML and MDS Pts Following Allo HSCT,” as a completed Phase II, multi-center, single-group trial. It enrolled 20 participants with AML or MDS following RIC allogeneic stem cell transplantation. N-803 was the intervention; allocation is listed as not applicable and masking as none. The sponsor is Masonic Cancer Center, University of Minnesota.

The trial began on 2017-04-12, and the primary completion date is 2022-03-19. The primary endpoint is relapse incidence through 24 months, expressed as a percentage of participants with a 95% confidence interval. The numerical result and interval are not reported in the registry record. The record lists 9 outcome measures and 0 analyses.

Safety summaries include 7 affected among 20 at risk for serious adverse events, 20 affected among 20 at risk for other adverse events, and 3 deaths. The record does not report the primary endpoint analysis method, missing-data procedures, or a comparator group.