Tretinoin and Arsenic Trioxide in Treating Patients With Untreated Acute Promyelocytic Leukemia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Tretinoin and Arsenic Trioxide in Treating Patients With Untreated Acute Promyelocytic Leukemia (NCT02339740), a na phase 3 study. The primary outcome is Event-free Survival (EFS) in Standard Risk Acute Promyelocytic Leukemia (APL) Patients.

2. Clinical Question

3. Trial Design

NCT02339740 is registered as a phase III study of patients with newly diagnosed acute promyelocytic leukemia. The registry lists a single-group model, allocation as not applicable, and no masking. It lists enrollment as 158 and identifies the Children's Oncology Group as sponsor. The study is active, not recruiting, in the registry record. The listed start date is 2015-07-21, and the primary completion date is 2020-10-31.

The treatment description names tretinoin, arsenic trioxide, and chemotherapy. The registry lists several drug interventions, including arsenic trioxide, cytarabine, dexamethasone, idarubicin, and mitoxantrone hydrochloride. It also lists laboratory biomarker analysis, quality-of-life assessment, and questionnaire administration. The arm description refers readers to a detailed description, but treatment schedules and the roles of each listed drug are not reported in the registry record.

The absence of a randomized comparator shapes what the study can answer. Outcomes can be described for the treated group, but a single-group record does not directly estimate a between-treatment contrast. Comparisons with results from other studies would require attention to differences in populations, follow-up, outcome definitions, and study conduct; those comparative data are not reported in the registry record.

The primary endpoint is specified for standard-risk patients, while the record also reports standard-risk and high-risk group counts for adverse-event information. The registry describes exclusions from the primary EFS analyses, but does not report a detailed analysis population flow or the numbers excluded for each reason. The record lists 13 outcome measures and 0 analyses. An observed primary-outcome estimate 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: Patients ineligible, did not start treatment, or RQPCR positive not confirmed are excluded from analyses of EFS

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 Event-free Survival (EFS) in Standard Risk Acute Promyelocytic Leukemia (APL) Patients, measured over Up to 24 months.

The primary endpoint is event-free survival (EFS) in standard-risk APL patients, with a timeframe of up to 24 months. EFS is defined from on study to the first of several events: failure to achieve hematological complete response before consolidation, persistence of molecular-positive disease after an MRD-positive consolidation course, relapse that is molecular, morphologic, or extramedullary, or death. This composite captures more than death alone, combining treatment-response and disease-status events with relapse and mortality.

The registry specifies a Kaplan-Meier estimate of 2-year EFS and 90% confidence limits using a log-minus-log transformation. In general, a time-to-event endpoint accounts for both whether an event occurred and when it occurred. Kaplan-Meier methods can use follow-up information from participants whose event has not been observed by the end of their available follow-up, under assumptions about censoring. The registry record does not report details of censoring or follow-up completeness.

The endpoint population excludes patients who were ineligible, did not start treatment, or had RQPCR-positive disease that was not confirmed. The numbers excluded for these reasons are not reported in the registry record. Nor is an observed EFS estimate or confidence interval reported, so the endpoint definition and planned estimation method can be described, but the study's measured EFS result cannot be interpreted from this record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a single-group design tell us?

The study is described as single-group, with allocation listed as not applicable and masking listed as none. In general, a single-group study can describe outcomes observed among participants receiving the study treatment. It does not create a randomized comparison with another treatment. Randomization in a comparative study is intended to make groups comparable, on average, with respect to measured and unmeasured factors at baseline. Without a randomized control group in this record, differences between the observed outcome and an outcome from another source could reflect differences in populations, follow-up, or other aspects of care rather than treatment alone.

Masking refers generally to whether participants, clinicians, or assessors are kept unaware of treatment assignments. The registry lists no masking, but does not report further details about who knew what or whether outcome assessment procedures included safeguards against bias. The registry record also does not report a statistical analysis plan or further design details that would clarify these procedures.

How is event-free survival estimated?

The stated primary analysis uses the Kaplan-Meier method to estimate 2-year EFS. In general, time-to-event analysis uses the time from a defined starting point until an event, while accommodating participants whose event time is not observed during available follow-up. Such participants may contribute information up to the time their follow-up ends; this is commonly represented through censoring. Valid interpretation depends on how follow-up and censoring are handled. The registry describes the endpoint origin as on study, but detailed rules for censoring and follow-up are not reported in the registry record.

EFS is a composite endpoint: the first qualifying event can be a response failure, persistent molecular-positive disease after the specified consolidation context, relapse, or death. Composite endpoints can capture several clinically relevant ways an outcome may occur, but their interpretation depends on knowing which components occurred and how often. The registry defines the components but does not report observed component counts or an EFS estimate. The endpoint population excludes patients who were ineligible, did not start treatment, or had RQPCR-positive disease that was not confirmed; the numbers in each excluded category are not reported in the registry record.

What does the confidence interval represent?

The planned EFS estimate has 90% log-minus-log transformed confidence limits. In general, a confidence interval describes sampling uncertainty around an estimated quantity under the chosen model and method. It is not a guarantee that the true value lies within the particular interval, and it does not account automatically for every possible source of bias or uncertainty. A log-minus-log transformation is a method for constructing confidence limits for a survival estimate while respecting its bounded scale. The registry record does not report the estimate or its limits, so their width and location cannot be assessed.

The confidence level is stated as 90%, not as a statement that a treatment has a particular probability of succeeding. It also does not indicate whether differences from a comparator would be statistically significant, since a comparator and a comparative estimate are not reported. Interpretation should focus on the estimate, interval, population, and endpoint definition together when those results are available.

Why does the analysis population matter?

The registry explicitly excludes ineligible patients, those who did not start treatment, and patients with unconfirmed RQPCR-positive results from EFS analyses. In general, inclusion and exclusion rules determine which participants contribute to a particular estimate and can affect how broadly that estimate applies. Excluding people after enrollment or based on information related to outcome can also affect interpretation. The record does not report the counts excluded, the timing of those decisions, or a detailed analysis population flow. It is therefore not possible to assess the impact of these exclusions from the registry record alone.

What can the adverse-event counts establish?

The record gives affected counts and at-risk counts by risk group for serious events and other events. In general, counts are descriptive and should be interpreted with their stated group denominators and event definitions. They do not by themselves establish causation, comparative safety, or the timing and duration of events. The registry lists serious-event terms with percentages, but does not report additional details in the registry record about how the percentages were derived, whether people could contribute more than one event, or how events were adjudicated. No risk comparison between treatments is possible from a single-group record.

10. Exact Estimates for Each Arm

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

R code and output

arm    <- c("Standard Risk")
n      <- c(98)
events <- c(96)
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  Standard Risk  98  96  98.0  92.8  99.8  92.9  99.4  92.9  99.4

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 13 outcome measures and 0 analyses. It does not report a multiplicity plan, such as which outcomes were designated primary or how statistical testing across multiple outcomes would be handled. In general, when many outcomes or hypotheses are examined, the chance of at least one apparently notable result can increase. Prespecified priorities and appropriate methods can help readers distinguish confirmatory analyses from exploratory ones. The registry record does not provide enough information to determine how multiplicity was addressed.

The primary endpoint description gives specific exclusions from the EFS analysis population: patients who were ineligible, did not start treatment, or had RQPCR-positive disease that was not confirmed. However, missing-data procedures, handling of incomplete follow-up, censoring details, and the numbers excluded for each reason are not reported in the registry record. In general, incomplete follow-up can affect a time-to-event estimate, and interpretation depends on the reasons follow-up ended and the assumptions used by the analysis.

The record does not report a statistical analysis plan, a detailed approach to missing outcome data, or any sensitivity analyses for the primary endpoint. It also does not report whether subgroup analyses beyond the named risk groups were planned. These omissions limit assessment of whether the reported endpoint definition, analysis population, and statistical methods were fixed before outcomes were examined. An observed primary endpoint result is not reported in the registry record, so the effects of these issues on a numerical estimate cannot be assessed.

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: psychosis (1.8%), retinoic acid syndrome (1.8%), death nos (1.0%), multi-organ failure (1.0%), infections and infestations - other, specify (1.0%).

The serious-event counts are reported separately for standard-risk and high-risk groups. Standard risk has 101 participants listed at risk and 3 affected for serious events; high risk has 57 at risk and 2 affected. For other events, the corresponding figures are 89 affected among 101 at risk for standard risk and 50 among 57 at risk for high risk. The registry also lists 1 death for standard risk and 0 for high risk.

These are descriptive counts from the registry record. They are not treatment-versus-control comparisons, and the record does not establish that an event was caused by treatment. Counts alone also do not convey event timing, duration, recurrence, or severity beyond the distinction between serious and other events. The criteria used to define serious events are not reported in the registry record.

The listed serious-event terms include psychosis and retinoic acid syndrome, each reported at 1.8%; death NOS, multi-organ failure, infections and infestations - Other, specify, and depressed level of consciousness are each reported at 1.0%. These term percentages are presented without additional detail about their denominators or overlap in the registry record. They should not be treated as comparative rates or as evidence that any event was caused by the study treatment.

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

The record describes a phase III, single-group study of tretinoin and arsenic trioxide with chemotherapy in newly diagnosed APL. Its primary endpoint is 2-year EFS in standard-risk patients, defined using response failure, persistent molecular-positive disease after a specified consolidation context, relapse, or death. The registry specifies Kaplan-Meier estimation and 90% log-minus-log transformed confidence limits. These details explain what the primary measure is intended to capture and how it is planned to be estimated.

An observed primary EFS estimate and confidence interval are not reported in the registry record. As a result, the record does not establish the observed level of EFS or the precision of that estimate. It also does not report component-specific outcome counts that would show which types of event contributed to the endpoint.

The single-group design has no randomized comparator. The outcomes described for participants in this study therefore cannot, on their own, show how outcomes would differ under another treatment. The registry summary gives a rationale involving reduction or removal of anthracycline chemotherapy and possible long-term side effects, but those statements describe the rationale, not findings from the primary endpoint. No comparative estimate of long-term side effects is reported in the registry record.

The safety counts and listed event terms provide descriptive information by risk group. They do not establish causality or comparative risk, and the record does not give detail on timing, event recurrence, or attribution. More broadly, missing analysis details, including the statistical plan and missing-data procedures, constrain assessment of how the stated endpoint would be analyzed. The most defensible reading is limited to the design, endpoint definition, exclusions, and descriptive safety information recorded; treatment benefit or comparative safety cannot be inferred from these registry entries alone.

14. Why This Trial Matters Statistically

This study raises statistical questions that are important in interpreting time-to-event outcomes in a single-group setting. EFS combines several event types, and the definition makes clear that the endpoint is not limited to death or relapse. Component events can have different meanings, so knowing the composite estimate alone would not necessarily reveal which events drove it. The registry defines the components, but an observed estimate and component counts are not reported in the registry record.

The planned use of Kaplan-Meier estimation and confidence limits makes precision central to interpretation. A point estimate without an interval can conceal substantial uncertainty; an interval helps readers judge how precisely an outcome has been estimated. Here, the registry gives the method and confidence level but no observed estimate or interval. The study record therefore supports discussion of the planned analysis, but not a numerical conclusion about the primary outcome.

The study also illustrates the limits of causal interpretation without a concurrent randomized control group. A single-group outcome can describe what was observed under the study conditions, but it does not supply the counterfactual outcome that would have occurred under another treatment. This distinction matters particularly when the rationale concerns reducing or removing a class of chemotherapy and possible long-term effects: rationale and observed comparative evidence are different kinds of information.

Finally, explicit analysis-population exclusions and descriptive safety counts show why definitions and denominators matter. Readers need to know who contributed to each endpoint, how incomplete follow-up was handled, and how adverse events were classified. Several of those details are not reported in the registry record. Recognizing these limits is part of interpreting the evidence precisely, rather than treating a study label or an outcome definition as a 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 interpreting the main outcome is that an observed EFS estimate and its confidence interval are not reported in the registry record. Although the endpoint definition, timeframe, and planned estimation method are described, there is no numerical primary outcome to assess. The record also does not report the component-event counts, so the relative contribution of response failure, persistent molecular-positive disease, relapse, and death cannot be characterized.

The study is single-group, with allocation listed as not applicable and no masking. There is no randomized comparator in the record, which limits interpretation of treatment effects and comparative safety. The registry does not report detailed treatment schedules, the statistical analysis plan, censoring rules, missing-data procedures, or detailed follow-up information. It also does not report the number of patients excluded from EFS analyses for each stated reason.

Safety information is descriptive. Serious and other event counts are reported by risk group, alongside a list of serious-event terms and percentages, but the registry record does not provide detail on event timing, recurrence, attribution, or the denominators behind the term percentages. The group counts do not support a treatment comparison or a causal conclusion.

The record lists 13 outcome measures and 0 analyses, and does not report how multiplicity was addressed. This limits assessment of the statistical context for outcomes beyond the stated primary endpoint. These limitations concern what can be concluded from the registry record; they do not establish that unreported analyses or details were absent from other study materials.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Tretinoin and Arsenic Trioxide in Treating Patients With Untreated Acute Promyelocytic Leukemia — NCT02339740.
• Registry record last updated 2026-02-18; status last verified 2026-02.
• Results first posted 2021-12-16.
• 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

NCT02339740 is a phase III, single-group study sponsored by the Children's Oncology Group. The registry lists enrollment as 158, no masking, allocation as not applicable, and a treatment description naming tretinoin, arsenic trioxide, and chemotherapy. Its status is active, not recruiting. The listed start date is 2015-07-21, and the primary completion date is 2020-10-31.

The primary outcome is EFS in standard-risk APL patients, up to 24 months. The registry defines its component events and specifies a Kaplan-Meier estimate of 2-year EFS with 90% log-minus-log transformed confidence limits. The stated EFS analysis excludes ineligible patients, those who did not start treatment, and patients with unconfirmed RQPCR-positive disease. An observed primary estimate, confidence interval, and detailed analysis plan are not reported in the registry record.

Descriptive safety counts are listed for standard-risk and high-risk groups. Because the study has no randomized comparator, those counts do not provide comparative treatment effects or establish causality.