AMG 176 First in Human Trial in Participants With Relapsed or Refractory Multiple Myeloma and Participants With Relapsed or Refractory Acute Myeloid Leukemia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for AMG 176 First in Human Trial in Participants With Relapsed or Refractory Multiple Myeloma and Participants With Relapsed or Refractory Acute Myeloid Leukemia (NCT02675452), a non-randomized phase 1 study. The primary outcome is Number of Participants Who Experienced a Dose Limiting Toxicity (DLT).

2. Clinical Question

3. Trial Design

This was a phase 1, first-in-human study of AMG 176 in participants with relapsed or refractory multiple myeloma or acute myeloid leukemia. It was non-randomized, unmasked, and used a parallel model. The registry record lists enrollment of 141 participants and a status of terminated. The study started on 2016-06-13, and primary completion is listed as 2024-05-27.

AMG 176 was administered intravenously in cohorts using different schedules. In the multiple myeloma parts, the listed schedules included infusion on two consecutive days followed by a break, or a once-weekly infusion followed by a break. AML cohorts also used these approaches. Additional AML cohorts received AMG 176 with azacitidine, while one cohort received it with itraconazole. A Japan-only AML cohort is also described. Cohorts therefore represent different treatment settings and schedules rather than randomized alternatives.

The stated main objectives were to evaluate safety and tolerability of AMG 176 monotherapy in participants with multiple myeloma and AML. The primary endpoint was the number of participants with a DLT during the evaluation period. Secondary outcomes included best overall response assessed using disease-specific criteria: IMWG-URC for multiple myeloma participants in Parts 1A and 1B, and 2017 ELN criteria for AML participants in Parts 3A, 3B, 3C, 3D, and 4. Numerical findings for these response outcomes are not reported in the registry record.

Because there was no random allocation or like-for-like control comparison, differences between cohorts should be read descriptively. Disease type, dose, schedule, combination treatment, and cohort membership vary. The record does not report a statistical plan that would support causal comparisons across these groups.

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.

Before assignment: Participants with multiple myeloma (MM) or with acute myeloid leukemia (AML) were enrolled into 29 cohorts. Dosing of AMG 176 was from Dose A (lowest) to Dose I (highest).

Only the first 6 of 29 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: DLT-evaluable analysis set included participants if they experienced a DLT during the DLT evaluation period or if they received at least 75% of the planned doses of AMG 176 (and 100% of the planned doses of azacitidine in Part 4) and completed the DLT evaluation 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.

Population: The full analysis set (FAS) included all participants who received at least 1 dose of AMG 176.

Only the first 6 of 30 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants Who Experienced a Dose Limiting Toxicity (DLT), measured over Days 1 to 28 of cycle 1 (4 weeks).

The primary endpoint counts participants who experienced a dose-limiting toxicity during Days 1 to 28 of cycle 1 (4 weeks). A DLT was assessed using CTCAE v 4.0 and had to be considered by investigators related to AMG 176. The definition included a grade 3 or higher non-hematological adverse event or a grade 4 hematologic adverse event occurring within the DLT observation period.

In general, a count endpoint describes how many people in the defined analysis population met a specified event definition. It is useful in an early phase, dose-finding context because severe toxicity during a defined observation window can help characterize tolerability and inform evaluation of doses and schedules. The endpoint is not a complete account of all safety events: it applies only to events meeting the DLT definition, attribution rule, and time window.

The analysis population was not simply all enrolled participants. The DLT-evaluable set included participants who experienced a DLT during the evaluation period, or who received at least 75% of planned AMG 176 doses and completed the DLT evaluation period. In Part 4, participants also had to receive 100% of planned azacitidine doses to qualify under the exposure route. Numerical primary endpoint counts are not reported in the registry record, so the observed frequency of DLTs cannot be described from this record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a count endpoint describe?

The primary endpoint is a count of participants who met the DLT definition during a fixed observation window. A participant either meets the endpoint definition or does not; the endpoint is not a continuous measurement such as a laboratory value. In general, counts are easiest to interpret alongside the population from which they were counted and the exact event definition. Here, the registry record defines the DLT-evaluable population and specifies the DLT window and severity criteria, but does not report the resulting count.

That distinction matters because the enrolled population and the endpoint analysis population need not be identical. The registry’s DLT-evaluable set required an observed DLT or specified dose exposure and completion of the evaluation period. In Part 4, full planned azacitidine exposure was additionally required for the exposure-based route into the analysis set. A count without its denominator would not provide a complete description of event frequency; the registry record does not give the primary endpoint count or denominator.

Why does the cohort structure matter?

In general, random allocation helps balance measured and unmeasured characteristics between comparison groups. This study was non-randomized and did not have a like-for-like control comparison. Instead, its cohorts varied by cancer type, dose, schedule, and use of azacitidine or itraconazole. Participants in one cohort may therefore differ systematically from participants in another. Observed differences in event counts cannot be attributed to a single treatment feature using these data alone.

Early phase studies commonly explore safety and tolerability across dose and schedule cohorts. Such a design can provide descriptive information about events observed in each cohort, but small or unequal cohorts can make counts unstable and comparisons imprecise. This is a general statistical consideration, not a claim about an unreported trial analysis. The registry record does not report an inferential plan for comparing the cohorts.

What can uncertainty measures add?

In general, a confidence interval communicates uncertainty around an estimate. For a participant-count endpoint, an interval around a proportion or rate would depend on the event count, the analysis denominator, and the method used. A narrow interval would indicate greater precision under the chosen assumptions; a wide interval would indicate less precision. No primary DLT estimate or confidence interval is reported in the registry record, so the precision of the primary result cannot be evaluated.

A p-value, when used, addresses how compatible observed data are with a specified null hypothesis under an analysis model. It does not measure the size or clinical importance of an effect. No analyses or p-values are listed for this trial in the record. It would be inappropriate to infer statistical significance from the cohort event counts alone.

How should the planning illustration be understood?

In general, sample-size planning asks how many participants would be needed under specified assumptions to detect an effect with a chosen probability, or power. The planning illustration associated with the posted results is described as using the observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation; the target effect is a stated fraction of the observed effect. For time-to-event settings, the illustration is described in terms of total events, while other settings use participants per arm.

These are planning assumptions, not a description of how this non-randomized trial was designed or analyzed. The illustration does not establish that the original study had that allocation, significance level, or power target. Sample-size calculations based on an observed effect are hypothetical planning aids, not observed power and not evidence that a result is precise. The registry record does not report the trial’s own sample-size justification or statistical plan.

What does over-dispersion mean in general?

For count data, over-dispersion means that variability in the counts exceeds what a particular statistical model expects. If ignored, a model may understate uncertainty and yield intervals or tests that are too optimistic. This is a general issue in count-data analysis; the registry record does not report a model for the primary endpoint or evidence that an over-dispersion assessment was performed.

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("Part 1A Cohort 1 (MM): AMG 176 Dose A BIW", "Part 1A Cohort 2 (MM): AMG 176 Dose B BIW", "Part 1A Cohort 3 (MM): AMG 176 Dose C BIW", "Part 1A Cohort 4 (MM): AMG 176 Dose D BIW", "Part 1A Cohort 5 (MM): AMG 176 Dose E BIW", "Part 1A Cohort 6 (MM): AMG 176 Dose G BIW")
n      <- c(3, 3, 4, 3, 3, 2)
events <- c(0, 0, 0, 0, 0, 1)
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  Part 1A Cohort 1 (MM): AMG 176 Dose A BIW  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Part 1A Cohort 2 (MM): AMG 176 Dose B BIW  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Part 1A Cohort 3 (MM): AMG 176 Dose C BIW  4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2
ARM  Part 1A Cohort 4 (MM): AMG 176 Dose D BIW  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Part 1A Cohort 5 (MM): AMG 176 Dose E BIW  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Part 1A Cohort 6 (MM): AMG 176 Dose G BIW  2  1  50.0  1.3  98.7  9.5  90.5  9.4  90.6

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 10 outcome measures and identifies secondary response outcomes for multiple myeloma and AML using different disease-specific criteria. It does not report a statistical analysis plan, a hierarchy among outcomes, or a method for handling multiplicity. In general, examining many outcomes or several analyses can increase the chance of at least one apparently notable result arising by chance. Methods such as prespecifying a hierarchy or adjusting significance thresholds can address this concern, but their use in this trial is not reported in the registry record.

Handling of missing endpoint data is also not reported in the registry record. The DLT-evaluable definition requires completion of the DLT evaluation period for participants entering by dose exposure, but this definition alone does not explain how missing assessments, early discontinuation, or incomplete records were handled in the analysis. The record does not report imputation, sensitivity analyses, or reasons for missing primary endpoint data.

The named primary analysis population is the DLT-evaluable set, with different exposure requirements described for Part 4. The registry record does not report an intention-to-treat analysis or other analysis populations for the primary endpoint. It also does not report interim analyses, stopping boundaries, or decision rules. Those features should not be inferred from the phase 1 dose-cohort structure. With no reported primary endpoint analysis, the extent to which multiplicity, missingness, or population selection influenced numerical findings 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: bacteraemia (100.0%), febrile neutropenia (75.0%), sepsis (75.0%), pancreatitis (50.0%), erysipelas (50.0%).

The safety information lists participants at risk, participants with serious adverse events, participants with other adverse events, and deaths by cohort. These are descriptive counts. Cohort sizes and observed counts differ, and the cohorts also differ in disease, dose, schedule, and combination treatment. A difference in counts across cohorts is therefore not a controlled estimate of a treatment-related difference.

The record lists serious-event terms including bacteraemia at 100.0%, febrile neutropenia at 75.0%, sepsis at 75.0%, pancreatitis at 50.0%, erysipelas at 50.0%, and pneumonia at 50.0%. The denominator for these percentages is not reported in the registry record. It also does not state whether one participant could be represented under several terms. The percentages should not be summed or treated as separate, mutually exclusive groups.

The listed death counts do not by themselves establish cause or treatment relationship. The record does not report cause-specific or exposure-adjusted mortality analyses. Serious adverse events are not identical to DLTs: a DLT had to meet the stated severity, timing, and investigator-relatedness criteria. The serious-event counts therefore cannot substitute for the missing primary DLT result.

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

The record describes a first-in-human phase 1 study that evaluated AMG 176 across multiple myeloma and AML cohorts, with varying schedules and some combination-treatment cohorts. It establishes that safety and tolerability were central objectives and defines the primary DLT endpoint, observation window, and DLT-evaluable population. It also reports cohort-level adverse-event counts and a list of serious-event terms.

However, the registry record does not report numerical results for the primary DLT endpoint, and it lists no analyses. It therefore does not establish how many DLT-evaluable participants experienced a DLT or how precisely that outcome was estimated. The secondary response outcomes are named, but their numerical results are not reported in the record. The posted adverse-event counts offer descriptive information by cohort, not randomized evidence about the relative safety of different doses, schedules, or combinations.

Non-randomized cohort differences can reflect many factors, including differences in participant groups and treatment context. Without a comparable control group and a reported analysis plan, the counts cannot establish that one schedule or combination caused more or fewer events than another. Likewise, the registry’s status of terminated does not explain why the study ended, and the reason is not reported in the registry record.

In general, a result should be interpreted in light of its definition, denominator, uncertainty, and study design. Here, the primary endpoint definition and analysis population are described, but the numerical endpoint result and uncertainty measures are absent. The record supports a description of study structure and reported safety counts; it does not support a quantitative conclusion about the primary DLT outcome or a causal comparison between cohorts.

14. Why This Trial Matters Statistically

This trial is statistically notable for the way its early phase structure combines several populations and treatment settings. Multiple myeloma and AML are represented, and the AML program includes different infusion schedules as well as cohorts with azacitidine or itraconazole. These differences are relevant to interpretation: cohort-level counts describe what was recorded in each setting, but do not create exchangeable comparison groups.

The primary outcome also illustrates why analysis-population definitions matter. A DLT was assessed over a specified window, but the DLT-evaluable set was based on event occurrence or meeting exposure and follow-up requirements. In general, selection into an evaluable population can affect which participants contribute to a reported endpoint. The record defines the population, yet does not report the DLT count or explain how incomplete assessments were handled.

The registry record also distinguishes the DLT endpoint from broader adverse-event reporting. Serious-event counts, deaths, and the listed serious-event terms provide different descriptive views of safety, but they are not interchangeable outcomes. DLTs have specific severity, timing, and attribution rules. Without numerical primary results and an inferential analysis, the record cannot quantify the primary outcome’s uncertainty or support comparative claims.

Finally, the record names secondary response criteria but supplies no numerical response findings. This underlines a general lesson in registry interpretation: an endpoint’s presence in a study description is not the same as a reported result. Clear separation between planned questions, observed counts, and statistical comparisons is essential, especially in non-randomized, multi-cohort studies.

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 most important limitation for interpreting efficacy or primary safety findings is that the registry record does not report numerical results for the primary DLT endpoint or the listed secondary response outcomes. It lists 0 analyses. The numerical adverse-event information is descriptive and organized by cohort, without a randomized control group or a reported comparative analysis.

The study includes multiple myeloma and AML participants, varied dose and infusion schedules, and combination cohorts. Cohorts are not like-for-like groups, so their event counts should not be used to rank schedules or combinations. The record does not report baseline characteristics by cohort, follow-up duration for the safety counts, or adjusted analyses. It also does not report the denominator underlying the percentages attached to serious-event terms.

The primary endpoint’s analysis population depends partly on treatment exposure and completion of the DLT evaluation period. The record does not report how participants with incomplete evaluations or missing data were handled, nor does it list alternative analysis populations or sensitivity analyses. The trial’s statistical analysis plan, sample-size justification, and interim-analysis procedures are not reported in the registry record.

The registry status is terminated, but the reason for termination is not reported in the registry record. That status alone does not explain the study’s conduct or outcomes. These limitations mean the available record supports an account of design and some cohort-level safety reporting, but not a complete statistical assessment of the stated primary or secondary endpoints.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: AMG 176 First in Human Trial in Participants With Relapsed or Refractory Multiple Myeloma and Participants With Relapsed or Refractory Acute Myeloid Leukemia — NCT02675452.
• Registry record last updated 2025-07-09; status last verified 2025-06.
• Results first posted 2025-07-09.
• 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

NCT02675452 is a phase 1, first-in-human, non-randomized study of AMG 176 in participants with relapsed or refractory multiple myeloma or AML. Enrollment is listed as 141, and the record lists the study as terminated. The primary endpoint was the number of participants experiencing a DLT during Days 1 to 28 of cycle 1 (4 weeks), using a defined DLT-evaluable population. AMG 176 was given in varied schedules, with some AML cohorts receiving azacitidine or itraconazole. Secondary outcomes included disease-specific response criteria. The registry record does not report numerical primary endpoint results, secondary response results, or analyses. Cohort-level adverse-event counts are descriptive and do not provide randomized comparisons.