Flotetuzumab in Primary Induction Failure (PIF) or Early Relapse (ER) Acute Myeloid Leukemia (AML): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Flotetuzumab in Primary Induction Failure (PIF) or Early Relapse (ER) Acute Myeloid Leukemia (AML) (NCT02152956), a non-randomized phase 1, phase 2 study. The primary outcome is Efficacy Based on CR or CRh Rate.

2. Clinical Question

3. Trial Design

NCT02152956 is a Phase 1/2, first-in-human study of flotetuzumab. The registry describes it as open-label, multi-dose, single-arm, and multi-center. Allocation was non-randomized and masking was absent. These features mean that the study was designed to assess the treatment in cohorts, rather than compare randomized groups under blinded conditions.

The study had three segments: a Single Patient Dose Escalation Segment, a Multi-Patient Dose Escalation Segment, and a Maximum Tolerated Dose and Schedule (MTDS) Expansion Cohort Segment. The first two segments are described as complete, and the expansion segment as closed. The study description says that safety and the maximum tolerated dose and schedule had been characterized before the primary objective shifted to evaluating response in PIF/ER AML.

Several flotetuzumab regimens are listed, with dose levels and administration schedules differing across cohorts. The described approach included increasing doses during the first week of Cycle 1, then continuous infusion. Subsequent cycles were described as four weeks of continuous infusion, with treatment potentially continuing for up to 8 cycles. The detailed assignment rules for individual cohorts are not reported in the registry record.

Enrollment is listed as 244. The study started on 2014-06-09, its primary completion date is listed as 2022-07-05, and its status is terminated. A reason for termination is not reported in the registry record. The record lists 23 outcome measures and no posted analyses. Because the design is single-group and dose cohorts are not randomized controls, observed outcomes cannot by themselves establish a treatment effect relative to an alternative.

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.

Only the first 6 of 13 groups are shown.

Overall Study

Reasons for not completing this period:

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.

Only the first 6 of 14 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Efficacy Based on CR or CRh Rate, measured over up to 14 months.

The primary endpoint is the number of participants whose best response was CR or CRh, assessed over up to 14 months. The endpoint description includes morphologic CR (mCR), cytogenetic CR (CRc), molecular CR (CRm), and CRh within its definition of CR. It characterizes mCR by normal neutrophil and platelet counts, less than 5% blast cells in a bone marrow smear, and no extramedullary disease. CRc refers to CR with no evidence of cytogenetic abnormalities in bone marrow; CRm refers to CR with no evidence of molecular abnormalities; and CRh refers to CR with partial hematologic recovery.

Using a best-response endpoint captures whether a participant met the response definition at any point in the specified assessment period, rather than requiring a particular response at every assessment. As a general statistical point, binary endpoints such as this are interpretable as counts or proportions, but their precision depends on the number of participants assessed and the number meeting the definition. The registry record gives the endpoint definition and timeframe, but does not report the response count, proportion, confidence interval, or details of how missing response assessments were handled. Consequently, the posted record does not permit a numerical assessment of the primary endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The study’s design determines what its results could support. It is described as non-randomized, open-label, and single-group, with dose escalation followed by an expansion cohort. General statistical principle: randomization helps make treatment groups comparable in expectation, while masking can reduce some forms of bias in assessment and reporting. Neither feature is present as a comparison mechanism in this study. This does not make descriptive outcomes meaningless, but it means they should not be interpreted as a randomized estimate of benefit relative to another treatment.

What does a single-group response count tell us?

The primary endpoint is a binary outcome for each participant: the best response either meets the CR-or-CRh definition or it does not. A count records how many participants meet that definition; a proportion places that count in relation to an analyzed group size. General statistical principle: the proportion is an estimate for the population represented by the analyzed participants, and its uncertainty depends on the denominator and the observed responses. The registry record does not post the numerator, denominator, or proportion for this endpoint, so the response rate and its precision cannot be assessed.

The response definition combines several forms of CR and includes CRh. Such a composite definition can capture multiple clinically specified response states under one endpoint. The interpretation depends on the exact criteria and how responses were assessed. The registry description names Interworking Group AML response criteria and gives component definitions, but does not report further adjudication procedures, assessment schedules, or analysis details.

Why does the cohort structure matter?

Dose escalation studies commonly examine tolerability and dosing across successive cohorts before evaluating an expansion population. Here, the registry describes single-patient escalation, multi-patient escalation, and an MTDS expansion segment, and lists a range of doses and schedules. General statistical principle: results from cohorts with different doses or schedules are not automatically a valid dose comparison. Participants may enter at different stages, and cohort sizes and eligibility can differ. The record does not report the allocation rules or a statistical method for comparing doses. No randomized dose comparison should therefore be inferred.

The study description says that safety and the maximum tolerated dose and schedule had been characterized, and that the primary objective then became assessment of anti-neoplastic activity in PIF/ER AML. The statistical plan for that objective—including any target response threshold, hypothesis test, or confidence-interval method—is not reported in the registry record. The sample-size justification is also not reported in the registry record. Without those details, it is not possible to judge whether the planned enrollment was selected to answer a specific efficacy hypothesis.

How should uncertainty and analysis methods be read?

General statistical principle: a confidence interval communicates uncertainty around an estimate under stated assumptions. It is not a guarantee that the true value lies in the interval, nor does it address all biases in the study design. No confidence interval is posted for the primary endpoint. No p-value or analysis method for that endpoint is reported in the registry record. The record lists no analyses, so no sponsor analysis or inferential result can be weighed against a recalculation.

For a binary outcome, choices about the analysis population and missing assessments affect the denominator and interpretation. General statistical principle: excluding participants without an evaluable response can yield a different proportion from counting them as non-responders or using another prespecified missing-data rule. The registry record does not report which analysis population was used, whether response data were missing, or how missingness was handled. It also does not report whether endpoint assessment was independently reviewed.

Safety results require different caution from an efficacy comparison. The record gives numbers at risk and affected within named cohorts, plus death counts. General statistical principle: a count of participants with an event is descriptive, and comparisons of event rates require attention to follow-up time, cohort composition, and consistent event ascertainment. The record does not provide exposure-adjusted rates or a controlled comparison. A reported death count is not, by itself, a count of deaths attributed to treatment.

Overall, the statistical information posted supports a description of study structure and selected safety counts, but not a quantitative inference about the primary response endpoint. The detailed efficacy analysis plan, missing-data procedures, and analysis populations are not reported in the registry 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("Cohort 0-a", "Cohort 0-b", "Cohort 0-c", "Cohort 0-d", "Cohort 1", "Cohort 2")
n      <- c(1, 4, 5, 4, 3, 4)
events <- c(0, 0, 0, 0, 0, 0)
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  Cohort 0-a  1  0  0.0  0.0  97.5  0.0  79.3  1.3  84.2
ARM  Cohort 0-b  4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2
ARM  Cohort 0-c  5  0  0.0  0.0  52.2  0.0  43.4  0.4  45.9
ARM  Cohort 0-d  4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2
ARM  Cohort 1  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Cohort 2  4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2

11. Multiplicity, Missing Data and Analysis Populations

The record lists 23 outcome measures, including the primary CR-or-CRh endpoint and secondary outcomes for overall complete response, CR, CRh, overall response, hematopoietic stem cell transplant (HSCT), dose-limiting toxicity, adverse events, and serious adverse events. When a study evaluates multiple outcomes, general statistical principle is that the chance of at least one apparently notable finding can increase if each is tested independently. A prespecified approach to multiplicity can help control that risk. The registry record does not report a multiplicity adjustment, a hierarchy of hypotheses, or which secondary outcomes were formally tested.

The record lists no analyses and does not report a statistical analysis plan. It is therefore not possible to determine whether results were intended to be confirmatory or descriptive, how outcomes were prioritized, or whether any adjustment was planned. The primary endpoint result itself is not reported in the registry record.

Missing-data handling and analysis populations are also not described. The record does not state the number of participants assessed for the primary endpoint, how participants who discontinued treatment were handled, or how missing response assessments were treated. General statistical principle: these choices can affect a binary response proportion, particularly when the probability of missing information is related to prognosis or treatment experience. The record also does not report sensitivity analyses addressing alternative assumptions about missing data. These omissions limit evaluation of both the robustness and the intended inferential meaning of any response estimate.

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: asthenia (33.3%), neutropenic sepsis (33.3%), infusion related reaction (33.3%), febrile neutropenia (25.0%), fatigue (25.0%).

The registry record reports serious adverse event counts by cohort, including the number at risk and the number affected. For example, the MTD Expansion cohort lists 185 participants at risk and 100 affected by serious adverse events; MTD Expansion With Ruxolitinib lists 12 at risk and 6 affected. These are cohort descriptions, not randomized comparisons. The record also lists other adverse event counts and deaths by cohort. A death count should not be interpreted as a treatment-related death count: attribution is not specified by those figures.

Listed serious adverse event terms include Asthenia, Neutropenic sepsis, and Infusion related reaction, each at 33.3%; Febrile neutropenia, Fatigue, and Oedema peripheral are each listed at 25.0%. The denominator, method of deriving these percentages, and whether terms may overlap within participants are not reported in the registry record. The percentages therefore should not be combined or treated as mutually exclusive categories.

General statistical principle: adverse event counts describe what was recorded among participants observed, but interpretation also depends on exposure duration, event definitions, follow-up, and attribution. Those details are not reported in the registry record. The safety figures alone cannot establish comparative risk or causation.

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

The record establishes that this was a first-in-human Phase 1/2 study with a dose-escalation and expansion structure, and that its later primary objective was to assess CR or CRh in PIF/ER AML. It also describes the endpoint criteria, treatment schedules, and selected cohort-level safety counts. These are useful features for understanding what the study set out to examine and the type of information recorded.

However, the central efficacy result is not reported. There is no posted count of participants achieving CR or CRh, no denominator for that endpoint, and no estimate of uncertainty. The registry record also does not report the endpoint analysis population, missing-data rules, or inferential method. Consequently, it does not support a numerical conclusion about the observed response rate or its precision.

Even if a response estimate were available, the design would remain important to interpretation. Because the study was single-group and non-randomized, the outcomes would be descriptive of the analyzed participants, not a randomized comparison against another treatment. Differences between dose cohorts would not automatically identify a causal dose effect.

The safety counts likewise describe reported events rather than treatment-attributable effects. Without a control group and further details on event attribution, exposure, and follow-up, the counts cannot show how the event burden compares with an alternative or what caused individual events. Overall, the registry record provides context for the study and selected safety information, but it does not provide the primary efficacy result needed to quantify response in the target population.

14. Why This Trial Matters Statistically

This study is statistically notable for the transition from dose escalation to an expansion segment and a stated primary focus on response in PIF/ER AML. That structure highlights how early-phase studies can have evolving objectives: initial cohorts characterize dosing and safety, while later enrollment may address an activity question. The registry describes these stages, but the statistical procedures governing the transition and the later efficacy assessment are not reported.

The study also illustrates why endpoint definitions and analysis populations matter. The primary endpoint combines CR categories and CRh, and its timeframe is specified as up to 14 months. Yet the registry record does not post the number of participants meeting the definition, the denominator used, or how missing assessments were handled. Without those elements, readers cannot evaluate the estimate or its precision, even when the endpoint itself is clearly described.

Finally, the design emphasizes a distinction between descriptive evidence and comparative evidence. A single-group study can document responses and safety events among participants receiving study treatment, but it does not provide a randomized estimate of how outcomes differ from a control. Dose cohorts do not substitute for randomized arms. The posted record makes the design and some safety information visible, while leaving the primary efficacy outcome and its analysis unavailable for statistical interpretation.

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 limitations follow from the registry information. The study is listed as terminated, but the reason for termination is not reported in the registry record. The record lists an enrollment of 244, but does not identify how many participants were included in the primary endpoint analysis or how many belonged to the PIF/ER AML population targeted by the later primary objective. The official title also includes relapsed or refractory AML and intermediate-2/high-risk MDS, so the precise relationship between the enrolled population and the endpoint population is not fully specified.

The primary response count and proportion are absent, as are confidence intervals, p-values, and a reported analysis method. The analysis population, handling of missing response assessments, response-review procedures, and multiplicity approach are not reported in the registry record. The detailed cohort allocation and dose-selection rules are likewise not described.

Safety counts are provided by cohort, but the record does not report exposure-adjusted rates, event attribution, or the denominator behind the listed serious-event percentages. Cohorts differ in regimen and size, so unadjusted cohort counts should not be treated as direct comparisons. The listed deaths are not identified as treatment-related in these data. Finally, the study’s open-label, non-randomized, single-group design limits causal interpretation even if descriptive outcomes are available.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Flotetuzumab in Primary Induction Failure (PIF) or Early Relapse (ER) Acute Myeloid Leukemia (AML) — NCT02152956.
• Registry record last updated 2024-01-30; status last verified 2024-01.
• Results first posted 2024-01-30.
• 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

NCT02152956 is an open-label, non-randomized, single-group Phase 1/2 study of flotetuzumab in AML and related populations. The study included single-patient and multi-patient dose escalation, followed by an MTDS expansion segment. Enrollment is listed as 244; the study started on 2014-06-09, the primary completion date is 2022-07-05, and the status is terminated.

The primary outcome is the number of participants whose best response was CR or CRh, assessed up to 14 months. The registry record does not report the primary response result, its denominator, or an analysis method. It lists 23 outcome measures and no analyses. Cohort-level adverse event counts, death counts, and selected serious adverse event terms are available, but do not provide a controlled safety comparison or establish causation.