A Randomized Trial of a Combination of Nintedanib/Placebo in Combination With Induction Chemotherapy for Patients With Refractory or First Relapse Acute Myeloid Leukemia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Randomized Trial of a Combination of Nintedanib/Placebo in Combination With Induction Chemotherapy for Patients With Refractory or First Relapse Acute Myeloid Leukemia (NCT02665143), a na phase 2 study. The primary outcome is Count of Patients With Treatment-Emergent Adverse Events (Phase 1).

2. Clinical Question

The registry title describes a randomized trial, and its brief summary frames the study around whether nintedanib plus induction chemotherapy could be an effective strategy. Those descriptions establish the clinical motivation, but they do not establish a treatment effect. The record’s listed primary outcome is a count of participants with treatment-emergent adverse events in phase I, and the posted group-level safety figures are separated into phase I and phase II. They are not results for nintedanib versus placebo.

The record names incidence of hematological improvement as a secondary outcome for phase II. Its result is not reported in the registry record. A comparative question about benefits or harms between nintedanib and placebo therefore cannot be answered from these posted figures. The available information supports a description of the study’s safety focus and the recorded adverse-event counts, while leaving efficacy and comparative safety unresolved.

3. Trial Design

The registry record identifies NCT02665143 as a phase II study sponsored by Yale University. Its title calls it a phase I-II randomized trial, and the study description refers to an initial phase I portion followed by a randomized phase II portion. The record lists an enrollment of 25 and a completed status. The recorded start date is 2016-07-21, and the primary completion date is 2021-03-08.

The treatment description combines AML induction chemotherapy with nintedanib or placebo. The regimen is described as idarubicin 12mg/m2/d on days 1 to 3 and cytarabine 0.667g/m2/d by continuous infusion on days 1 to 3. The phase II description says nintedanib or placebo would be added at day 8 and continued until the end of the cycle; the description is truncated after that point, so further treatment details are not reported in the registry record.

There is an important distinction between the planned design described in words and the comparison supported by the posted data. The registry lists masking as none and allocation as NA. Its safety groups are a phase I group of 6 and a phase II group of 19, both titled as receiving nintedanib and AML induction. They are not presented as nintedanib and placebo arms. The record therefore does not provide a like-for-like control comparison, despite the title and the planned phase II comparison described in the arm text.

The primary outcome is a phase I safety measure. The registry also names incidence of hematological improvement in phase II as a secondary outcome. Three outcome measures are listed, but only that secondary outcome is named in the available record summary. Detailed randomization procedures, allocation concealment, protocol deviations, and the full analysis plan 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.

Phase 1

Phase 2

Analysis population for the primary outcome, as defined in the registry: Phase 1 patients.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: There were 6 Phase 1 patients and 19 Phase 2 patients.

6. Primary Endpoint

The registry lists the primary outcome as Count of Patients With Treatment-Emergent Adverse Events (Phase 1), measured over 60 days.

The primary outcome is titled count of patients with treatment-emergent adverse events in phase I. It is expressed as a count of participants, with a timeframe of 60 days. The outcome description defines treatment-emergent adverse events as adverse events occurring within 60 days of treatment initiation and says that safety was to be evaluated according to NCI CTCAE V4 criteria. The stated population is phase I patients.

A participant count is a binary-outcome summary at the individual level: each participant either meets the stated event definition during the timeframe or does not. This kind of endpoint is used to describe the number of people experiencing an event, rather than the total number of event episodes. It can make safety findings easier to interpret at the participant level, but a count alone does not describe event severity, duration, recurrence, or causal attribution.

The description says safety would be assessed after inclusion of the first 6 patients to evaluate tolerance and adverse events and decide on any action. The registry record does not report a primary outcome result for this endpoint. The posted safety group data include figures for serious and other adverse events, but those figures should not be treated as a reported result for the specifically named primary endpoint unless the endpoint definition and corresponding analysis are explicitly presented. No such primary endpoint analysis is reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Clinical trial statistics depend on how treatment groups, endpoints, and analysis populations are defined. In this record, the title and description use randomized-trial language, but the allocation field is NA and the posted safety groups are organized by phase. That difference matters: a comparative treatment estimate requires clearly identified groups whose treatment assignments correspond to the question. Counts from differently labelled phases cannot be interpreted as a randomized comparison merely because the study description mentions a planned placebo comparison.

What a participant count represents

The primary endpoint is a count of participants with treatment-emergent adverse events during the stated 60-day timeframe. In general, a participant-level binary endpoint records whether each person met the event definition. It is distinct from a count of event episodes: one participant may experience more than one event, while a participant count generally counts that person once for the endpoint. The registry description identifies the event window and the safety grading criteria, but it does not report a result for this primary endpoint.

Counts are most informative when the denominator, eligibility rules, observation period, and handling of incomplete follow-up are clear. The safety groups have at-risk counts of 6 and 19, and affected-participant counts are posted for serious and other adverse events. However, the record does not provide the primary endpoint analysis or detail how missing event information was handled. A count should not be treated as a complete account of burden: it does not, by itself, indicate event duration, recurrence, timing, severity beyond the category shown, or whether an investigator considered an event related to study treatment.

Why the comparison structure matters

General statistical principle: random allocation can make treatment groups comparable, on average, with respect to both measured and unmeasured baseline factors. A treatment contrast can then estimate the effect of assignment, subject to appropriate conduct and analysis. But that interpretation requires the record to identify the groups and their assigned treatments. Here, the posted groups are labelled phase I and phase II under the same intervention description, not as active treatment and placebo. The registry does not provide a treatment-by-treatment outcome contrast. Consequently, the safety figures are descriptive within the listed groups and cannot estimate the comparative effect of nintedanib versus placebo.

The phase I safety description says the first 6 patients would be assessed for tolerance and adverse events, with decisions about action to follow. This describes a safety-monitoring purpose, but no further decision rules or stopping boundaries are reported in the registry record. It would be inappropriate to infer a dose-escalation method, interim-analysis plan, or stopping rule from that statement alone.

Uncertainty and confidence intervals

General statistical explanation: a confidence interval describes the range of effect estimates reasonably compatible with the observed data under a specified statistical model. Its width reflects uncertainty; intervals are typically wider when the available information is less precise. The record does not report a confidence interval or an effect estimate for the primary endpoint. The posted counts therefore cannot be used to describe the precision of a comparative treatment effect. No recalculated effect estimate is presented in the registry information summarized for this analysis.

General statistical principle: for binary outcomes, the choice of effect measure matters. A risk difference, risk ratio, or odds ratio answers a different mathematical question, and each requires a clearly defined comparison and an analysis population. Since the posted groups are not a like-for-like comparison, selecting an effect measure would not solve the underlying design and reporting problem.

Planning illustrations and power

The sample-size assumptions describe illustrations based on fractions of an observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. These are planning scenarios, not observed power and not evidence of the trial’s actual design rationale. The registry record does not report the trial’s own sample-size justification or establish that its planned comparison followed these assumptions. General explanation: power is a design-stage probability of detecting a specified effect under assumed conditions; it is not a measure of whether a result is true, and a post hoc calculation based on an observed estimate is not observed power.

For binary data, statistical methods can also be affected by sparse observations. General explanation: when event counts are limited, estimates can be unstable and uncertainty can be substantial. The posted phase I group is small, and the total enrollment is recorded as 25, but the registry does not report a comparative model or interval estimate for these adverse-event outcomes. No new inferential calculation should be read into the descriptive counts.

Multiplicity and analysis choices

General statistical explanation: multiplicity arises when several outcomes or comparisons are assessed, increasing the chance that at least one result appears statistically notable by chance if each is tested separately. A statistical plan may define a primary endpoint, specify which secondary analyses are confirmatory, and describe any adjustment or hierarchy. This record names the phase I adverse-event count as primary and hematological improvement as a secondary outcome, but the full analysis plan, hypothesis-testing strategy, and multiplicity approach are not reported in the registry record. The record lists three outcome measures, yet the remaining outcome details are not fully described in the summary.

Analysis populations also affect interpretation. General explanation: an intention-to-treat population usually retains participants in their assigned groups, while other populations may require treatment exposure or sufficient follow-up. The registry record does not define an analysis population for the named endpoint, describe exclusions, or explain how participants with incomplete follow-up were handled. These omissions limit reproducibility and make it difficult to determine precisely which participants contribute to each result.

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("Nintedanib and AML Induction")
n      <- c(6)
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  Nintedanib and AML Induction  6  6  100.0  54.1  100.0  61.0  100.0  59.0  99.6

11. Multiplicity, Missing Data and Analysis Populations

The registry record identifies a primary safety endpoint for phase I and names incidence of hematological improvement as a phase II secondary outcome. It lists three outcome measures, but the available outcome description does not give complete detail for all of them. The statistical plan, including any hierarchy among outcomes, planned hypothesis tests, or adjustment for multiple comparisons, is not reported in the registry record.

Multiplicity is a general concern when investigators assess several endpoints or comparisons. Testing multiple hypotheses without a prespecified strategy can increase the chance of a seemingly notable finding arising by chance. A trial may address this through a hierarchy, adjustment procedure, or a clear distinction between confirmatory and exploratory analyses. None of these approaches can be attributed to this study from the registry information available.

Missing-data rules and analysis populations are also not reported in the registry record. The record does not explain whether all enrolled participants were included in each outcome analysis, how incomplete follow-up was treated, or whether any participants were excluded from particular analyses. The endpoint description specifies a 60-day period, but does not state how an incomplete observation period was handled.

The record reports 0 analyses. It therefore does not present statistical analyses that would clarify endpoint denominators, uncertainty, or the handling of missing outcomes. The absence of these details does not imply that no analysis was conducted outside the registry; it means the registry record does not document one. These gaps limit assessment of the planned inferential approach and the reproducibility of any conclusions.

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: febrile neutropenia (31.6%), alanine aminotransferase increased (16.7%), thromboembolic event (16.7%), aspartate aminotransferase increased (5.3%), heart failure (5.3%).

The serious adverse-event figures are participant counts in groups labelled phase I and phase II. The phase I group lists 2 of 6 participants with serious adverse events; the phase II group lists 8 of 19. The other-adverse-event counts are 6 of 6 and 18 of 19, respectively. The registry lists 0 deaths in both groups. These are descriptive counts and should be read using the group labels and denominators recorded.

The serious adverse-event terms listed with percentages are febrile neutropenia at 31.6%, alanine aminotransferase increased at 16.7%, thromboembolic event at 16.7%, aspartate aminotransferase increased at 5.3%, heart failure at 5.3%, and pericardial effusion at 5.3%. The available record summary does not identify the denominator or group assignment for each term. The percentages therefore should not be interpreted as treatment-arm rates.

A serious adverse event is not necessarily caused by study treatment. The registry figures do not establish attribution, comparative risk, or severity beyond the recorded serious-event classification. They also do not show how many episodes each participant experienced or how long events lasted. Since the groups are separated by phase rather than by active treatment and placebo, their counts do not constitute a randomized safety comparison.

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

The registry record gives a limited descriptive account of safety in a study of nintedanib with AML induction chemotherapy. It records serious and other adverse-event counts in a phase I group and a phase II group, along with 0 deaths in each. It also lists serious adverse-event terms and associated percentages. These observations describe what the record reports for those groups; they do not establish that treatment caused an event.

The record does not report the result for the named primary endpoint, which is the count of phase I participants with treatment-emergent adverse events within 60 days. Nor does it report the phase II hematological-improvement result. No comparative outcome analysis is reported, and the posted groups do not provide a nintedanib-versus-placebo contrast. The title and planned-comparison language cannot replace that missing comparison.

General statistical principle: a treatment effect is a contrast between outcomes under different treatment assignments, interpreted in light of the study design and uncertainty. A within-group event count alone does not estimate that contrast. It cannot determine whether the addition of nintedanib altered adverse-event frequency, hematological improvement, or another clinical outcome relative to chemotherapy plus placebo.

The enrollment of 25 and the small phase-specific groups also mean that the descriptive record should not be treated as a precise estimate of event frequency in a broader population. The registry does not report confidence intervals or a statistical model for the safety data. The size of uncertainty around any underlying rate therefore cannot be assessed from the posted counts alone.

In practical terms, the record supports reporting the observed safety counts and the study’s stated safety objective. It does not support a conclusion about comparative benefit, comparative harm, treatment effectiveness, or the probability that any particular event was caused by study treatment. Those questions remain unanswered by the results documented in the registry record.

14. Why This Trial Matters Statistically

This trial is statistically notable because its description and its posted data do not align as a direct treatment comparison. The title calls it randomized, and the phase II description says nintedanib would be compared with placebo alongside chemotherapy. Yet allocation is recorded as NA, and the available safety groups are labelled by phase rather than by assigned treatment. Recognizing that distinction prevents phase-specific safety counts from being misread as an estimate of a nintedanib-versus-placebo effect.

The named primary endpoint is a participant count for treatment-emergent adverse events in phase I, within a 60-day timeframe. That endpoint answers a focused safety question, but its result is not reported in the registry record. The posted serious and other adverse-event counts do not substitute for a result explicitly tied to the primary endpoint’s definition. Clear alignment between an endpoint definition, its population, its denominator, and its reported analysis is essential for reproducible interpretation.

The study also illustrates why outcome reporting and statistical planning matter. The registry names a phase II hematological-improvement outcome but does not report its result, and it does not document the full analysis plan, multiplicity strategy, or missing-data rules. Without these details, readers cannot determine how outcomes were prioritized or how uncertainty and incomplete observations were handled.

Finally, planning illustrations based on observed effects should be distinguished from the trial’s actual design and from observed power. The registry does not report a trial-specific sample-size justification. Keeping design assumptions separate from results avoids overstating what a small, descriptive record can establish. The central statistical lesson is that the strength of an interpretation depends not only on the recorded counts, but also on whether the comparison and analysis match the clinical question.

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 principal limitation is that the record does not provide a randomized treatment comparison in its posted results. Although its title and phase II description refer to a planned nintedanib-versus-placebo comparison, allocation is listed as NA and the two reported safety groups are labelled by phase. The available counts therefore cannot be interpreted as an active-treatment versus placebo contrast.

The named primary outcome is a phase I count of participants with treatment-emergent adverse events within 60 days, but its result is not reported in the registry record. The posted adverse-event summaries include serious and other events, but the record does not establish that these are the reported primary endpoint analysis. It also does not provide confidence intervals or a comparative statistical model.

The group sizes are limited, with 6 participants in phase I and 19 in phase II. The record does not provide enough detail to assess the precision of the adverse-event estimates, and it does not report the denominator or group assignment for each listed serious adverse-event term and percentage. Causal attribution and event-level details are also not stated in the available summary.

Other trial details relevant to interpretation are not reported in the registry record, including the full statistical analysis plan, analysis populations, missing-data handling, multiplicity approach, and trial-specific sample-size justification. The treatment description is truncated after saying that nintedanib or placebo would continue until the end of the cycle. These limitations constrain interpretation to a descriptive account of the recorded groups, rather than conclusions about comparative safety or efficacy.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Randomized Trial of a Combination of Nintedanib/Placebo in Combination With Induction Chemotherapy for Patients With Refractory or First Relapse Acute Myeloid Leukemia — NCT02665143.
• Registry record last updated 2022-04-08; status last verified 2022-04.
• Results first posted 2022-04-08.
• 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

NCT02665143 is listed as a completed phase II study sponsored by Yale University, with an enrollment of 25. Its title describes a phase I-II randomized trial in refractory or first-relapse acute myeloid leukemia. The recorded start date is 2016-07-21, and the primary completion date is 2021-03-08.

The intervention description combines nintedanib with AML induction chemotherapy and describes a planned phase II comparison with chemotherapy plus placebo. The primary outcome is a phase I count of participants with treatment-emergent adverse events within 60 days. Hematological improvement is named as a phase II secondary outcome.

Posted safety groups are labelled phase I and phase II, with 6 and 19 participants at risk, respectively. The registry record reports serious and other adverse-event counts, 0 deaths in each group, and several serious adverse-event terms with percentages. It does not report a result for the named primary endpoint, the secondary outcome, or a like-for-like randomized treatment comparison. Allocation, detailed statistical methods, and analysis populations are not reported in the registry record.