Onvansertib in Combination With Either Low-dose Cytarabine or Decitabine in Adult Patients With Acute Myeloid Leukemia (AML): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Onvansertib in Combination With Either Low-dose Cytarabine or Decitabine in Adult Patients With Acute Myeloid Leukemia (AML) (NCT03303339), a non-randomized phase 1, phase 2 study. The primary outcome is Number of Participants Who Experienced Dose Limiting Toxicities (DLT).

2. Clinical Question

The study asked an early safety and dose-finding question about onvansertib in combination with established chemotherapy agents. In phase 1b, the onvansertib dose was escalated in cohorts receiving low-dose cytarabine or decitabine. In phase 2, the registry describes onvansertib at the recommended phase 2 dose with decitabine, with further assessment of the combination’s safety profile and preliminary activity.

The primary endpoint focused on dose-limiting toxicities (DLTs) during the first cycle, rather than on a comparison of treatment effectiveness between groups. DLTs were defined as treatment-related events meeting specified hematologic or non-hematologic criteria. The endpoint was therefore intended to inform whether a dose could be taken forward, alongside the study’s broader objectives of assessing tolerability and identifying a maximum tolerated dose and recommended phase 2 dose.

The registry lists additional phase 2 outcomes, including morphologic leukemia-free state, partial response, duration of response, event-free survival, and overall survival, as well as pharmacokinetic measures for onvansertib. Results for these outcomes are not reported in the registry record. The clinical question can therefore be described from the protocol summary and endpoint definitions, but the posted information in this analysis does not establish the observed DLT count or the study’s findings on activity.

3. Trial Design

This completed study is registered as NCT03303339 and is titled “Onvansertib in Combination With Either Low-dose Cytarabine or Decitabine in Adult Patients With Acute Myeloid Leukemia (AML).” It was a phase 1b/2, non-randomized, unmasked, single-group study sponsored by Cardiff Oncology. The registry lists enrollment of 72 participants and the condition as acute myeloid leukemia.

The study began on 2017-11-17, had primary completion on 2021-11-17, and is marked completed. In phase 1b, participants received escalating doses of oral onvansertib with either low-dose cytarabine or decitabine. The low-dose cytarabine regimen was listed as 20 mg/m^2 subcutaneously once daily on Days 1 through 10 of each 28-day cycle. The decitabine regimen was listed as 20 mg/m^2 intravenously over 1 hour on Days 1 through 5 of each 28-day cycle. Onvansertib was administered orally on Days 1 through 5 of each 28-day cycle.

The phase 2 arm combined onvansertib at the recommended phase 2 dose with decitabine. Its registry description also notes that treatment modifications or delays could be based on return of hematopoietic function. The adverse-event groups identify this phase 2 regimen as onvansertib 60 mg/m^2 with decitabine. The record does not describe a phase 2 low-dose cytarabine arm.

Because the study was not randomized and had no masking, comparisons across dose cohorts or regimens should not be interpreted as controlled treatment comparisons. Cohorts could differ for reasons beyond the assigned regimen, and the registry does not identify a concurrent control group. Enrollment is the overall listed study enrollment; the adverse-event group counts are not a substitute for a complete account of enrollment by cohort or of participants included in each endpoint analysis. Those details are not reported in the registry record.

The study’s stated objectives include assessing safety and tolerability, determining the maximum tolerated dose and recommended phase 2 dose, and preliminarily assessing activity in the phase 2 population. The registry record does not report the dose-selection rationale or details of any dose-escalation decision rules.

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 12 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The safety analysis set was comprised of all subjects who received at least one dose of onvansertib.

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 13 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants Who Experienced Dose Limiting Toxicities (DLT), measured over Up to Day 28 of Cycle 1.

The primary endpoint is the number of participants who experienced dose-limiting toxicities, measured up to Day 28 of Cycle 1. The safety analysis set is described as all subjects who received at least one dose of onvansertib. The endpoint is a count of participants, not a measure of how many separate toxic events each participant experienced.

A DLT had to be related to onvansertib and considered an adverse reaction or suspected adverse reaction during the first cycle. The hematologic criterion included persistent pancytopenia resistant to current standards of care, continuing for at least 42 days, and not related to leukemic infiltration or another cause unrelated to study therapy. The non-hematologic criteria included Grade 3 abnormalities that persisted for more than 7 days without decreasing in severity despite standards of care, were clinically significant, or were Grade 4 and symptomatic.

In dose-finding studies, a prespecified early-cycle toxicity endpoint can help characterize whether a dose is tolerable and inform dose selection. That is a general explanation of the role of a DLT endpoint; it does not establish what dose was selected or what toxicity pattern was observed in this study. The registry record does not report the number of participants who met the primary endpoint, the dose-specific DLT results, or a confidence interval for the count or rate. As a result, the endpoint definition is available, but its observed result cannot be interpreted from the posted information in this analysis.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical approach should be read in light of the study’s phase 1b/2 structure and non-randomized, single-group design. The primary endpoint is a count of participants with a qualifying DLT during the first cycle. The registry defines the safety analysis set as subjects who received at least one dose of onvansertib. The endpoint timeframe is up to Day 28 of Cycle 1. The analysis method used to produce a primary endpoint estimate, and any uncertainty interval, are not reported in the registry record.

What does a participant count describe?

General: A count records how many people meet an endpoint definition. It differs from a count of events, because one person might experience more than one event but still contribute once to a participant-level endpoint. Counts are most interpretable when the analysis population and observation period are clear. Here, the registry gives the endpoint definition and safety-set description, but the observed count and dose-specific primary endpoint results are not reported in the registry record.

For a binary outcome such as whether a participant experienced at least one DLT, analysts may also report a proportion and a confidence interval. General: a confidence interval describes statistical uncertainty around an estimate under the assumptions of the analysis; it does not account for every source of bias, and it is not a guarantee that the true value lies within the interval. No primary endpoint proportion or confidence interval is reported in the registry record, so precision cannot be assessed for that endpoint.

Why does the study structure matter?

General: Random allocation helps balance measured and unmeasured participant characteristics between treatment groups, on average. Without randomization, differences between groups may reflect patient characteristics, treatment selection, calendar time, dose, or other factors as well as the regimen itself. This study is listed as non-randomized and single-group, with no masking. Its different dose cohorts are useful for describing the study’s staged treatment experience, but they are not equivalent to randomized control arms. A difference in observed counts across cohorts would not by itself identify a causal dose or regimen effect.

The phase 1b portion tested escalating onvansertib doses with low-dose cytarabine or decitabine. The phase 2 portion studied the recommended phase 2 dose with decitabine. The registry record does not state the dose-escalation rules, the method used to select the recommended phase 2 dose, or whether any formal dose-response model was used. Those details should not be inferred from the cohort labels.

How should adverse-event counts be read?

The serious adverse-event listings give participants at risk and participants affected for each group, along with deaths. These are descriptive safety counts. General: the denominator matters because the same count can represent different levels of precision in groups of different sizes. Affected counts can also depend on the event definitions, collection period, and analysis population. The registry record does not report all of those details for the listed adverse-event counts, nor does it explain the causal assessment for the deaths. The counts should not be treated as adjusted comparisons across regimens.

The registry lists serious-event terms with percentages. A percentage is only fully interpretable alongside its denominator and calculation rules, including whether participants or events form the basis. Those details are not reported in the registry record. The listed term percentages therefore describe the entries as posted, but should not be used to infer a group-specific incidence or compare risks across treatment cohorts.

What do planning illustrations mean?

General: A sample-size calculation is a planning exercise based on assumptions about the outcome, the effect of interest, variability, allocation, and the chosen statistical test. It is not an observed study result and cannot establish that a completed trial had a particular probability of detecting an effect. In the registry analysis, the planning assumptions are described as a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is expressed as a fraction of the observed effect. For time-to-event outcomes, the stated planning approach concerns total events, while other calculations are expressed as participants per arm.

These are illustrations, not a description of this trial’s design. The study itself was non-randomized and single-group, and its design was not an equal-allocation comparison between two arms. The registry record does not report a trial sample-size justification or the assumptions used when the study was planned. Any planning illustration based on an observed effect is not “observed power,” and it should not be read as evidence that the study was adequately powered for its clinical questions.

Statistical methods can also address multiple outcomes, missing observations, and analysis populations. General: testing many outcomes can increase the chance of at least one apparently notable result unless the analysis plan accounts for multiplicity. Missing outcomes can affect estimates if the reasons for missingness are related to participants’ outcomes. Those concepts are relevant to interpreting clinical-trial results, but the corresponding procedures for this study 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("Phase 1b: Onvansertib 12 mg/m^2 + Low-dose Cytarab", "Phase 1b: Onvansertib 18 mg/m^2 + Low-dose Cytarab", "Phase 1b: Onvansertib 27 mg/m^2 + Low-dose Cytarab", "Phase 1b: Onvansertib 40 mg/m^2 + Low-dose Cytarab", "Phase 1b: Onvansertib 60 mg/m^2 + Low-dose Cytarab", "Phase 1b: Onvansertib 12 mg/m^2 + Decitabine")
n      <- c(3, 3, 3, 3, 5, 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  Phase 1b: Onvansertib 12 mg/m^2 + Low-dose Cytarab  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Phase 1b: Onvansertib 18 mg/m^2 + Low-dose Cytarab  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Phase 1b: Onvansertib 27 mg/m^2 + Low-dose Cytarab  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Phase 1b: Onvansertib 40 mg/m^2 + Low-dose Cytarab  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Phase 1b: Onvansertib 60 mg/m^2 + Low-dose Cytarab  5  0  0.0  0.0  52.2  0.0  43.4  0.4  45.9
ARM  Phase 1b: Onvansertib 12 mg/m^2 + Decitabine  4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary DLT endpoint and multiple secondary outcomes. The secondary outcomes include morphologic leukemia-free state, partial response, duration of response, event-free survival, overall survival, and three pharmacokinetic parameters for onvansertib. The primary endpoint result and the results for these secondary outcomes are not reported in the registry record.

Multiplicity refers, in general, to the interpretive issue that arises when several hypotheses or outcomes are assessed: more opportunities to identify an apparently notable result can increase the chance of a false-positive finding. A prespecified hierarchy, adjustment, or other strategy can address that concern, but the registry record does not report whether a multiplicity strategy was planned or applied. The order in which the outcomes are listed does not establish a testing hierarchy.

Missing-data methods are also not reported in the registry record. The primary safety population is described as all subjects who received at least one dose of onvansertib, but this alone does not explain how participants with incomplete follow-up, discontinued treatment, or unavailable endpoint information were handled. The record does not report missingness counts, reasons for missingness, imputation methods, or sensitivity analyses.

The analysis population for each secondary outcome is not reported in the registry record. This matters because the number of participants assessed may differ across outcomes, particularly when outcomes require follow-up or a specific measurement. The listed enrollment of 72 should not be assumed to be the denominator for every endpoint. General: transparent reporting of analysis populations, missing observations, and multiplicity procedures helps readers understand what each estimate represents and how much confidence to place in it. Without those details or posted outcome results, the registry information supports a description of planned outcomes but not a complete assessment of the statistical analysis.

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 (83.3%), arthralgia (66.7%), lung infection (33.3%), pneumonia (33.3%), septic shock (33.3%).

The registry lists serious adverse-event group counts for phase 1b dose cohorts and the phase 2 combination of onvansertib 60 mg/m^2 with decitabine. In the phase 2 group, 27 of 32 participants at risk are listed as affected by serious adverse events, and 32 of 32 are listed as affected by other adverse events. The group listing also records 24 deaths. In phase 1b, the counts vary across the dose cohorts and combinations; those cohorts are not randomized comparison groups.

The registry also lists febrile neutropenia at 83.3%, arthralgia at 66.7%, and lung infection, pneumonia, septic shock, and asthenia at 33.3% each among the top serious-event terms. The denominator and calculation basis for these percentages are not reported in the registry record. The terms and percentages should therefore be interpreted as registry listings, not as comparable rates across treatment groups.

Counts of participants affected by an adverse event are not counts of separate episodes. A participant can experience more than one type of event, and the event listings do not establish that treatment caused an event. The record does not report a causal explanation for the deaths listed in the groups. Because cohort sizes and treatment regimens differ, the counts do not provide a like-for-like estimate of comparative safety.

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

The record establishes that a completed phase 1b/2 study evaluated onvansertib in combination with low-dose cytarabine or decitabine in adults with acute myeloid leukemia. It describes escalating onvansertib doses in phase 1b, a decitabine combination in phase 2, and a primary endpoint based on participants experiencing DLTs during the first cycle. It also contains group-level adverse-event counts and lists serious-event terms.

The record does not report the primary endpoint result or results for the secondary outcomes listed. It therefore does not show how many participants experienced a DLT, what the response outcomes were, or what the survival and pharmacokinetic findings were. Completion status and enrollment do not answer those questions. Nor do the adverse-event counts establish that a treatment caused an event or death.

The study’s non-randomized, single-group design also limits comparative interpretation. Cohort-to-cohort differences cannot be treated as evidence that one dose or combination is better or safer than another, because there is no randomized control comparison and the cohorts may not be comparable. The posted data do not support a causal conclusion about the effects of onvansertib or either combination.

General: an early-phase safety and dose-finding study can contribute descriptive information about tolerability and help inform later research. Its interpretation depends on the endpoint results, analysis population, follow-up, and uncertainty around estimates. Here, the DLT definition and safety population are described, but the primary result and its precision are not reported in the registry record. The appropriate statistical conclusion is therefore limited to what the record describes: the design, intended questions, endpoint definition, and posted adverse-event listings. It does not establish treatment efficacy, comparative benefit, or a complete safety profile.

14. Why This Trial Matters Statistically

This study is statistically notable for combining dose escalation with a later phase 2 assessment while using a non-randomized, single-group design. That structure makes the distinction between descriptive safety information and comparative evidence especially important. The registry’s phase 1b cohorts involve different onvansertib doses and chemotherapy combinations, while the phase 2 portion focuses on onvansertib with decitabine. Those groups can describe treatment experience, but they are not interchangeable control arms.

The DLT endpoint illustrates how an early-cycle, participant-level safety measure can be used in dose-finding research. Its definition sets a clinical threshold for qualifying toxicity and its timeframe focuses on the first cycle. At the same time, a detailed endpoint definition does not substitute for the observed result. Since the primary DLT count is not reported in the registry record, readers cannot evaluate the endpoint estimate or its precision from the information.

The posted safety counts also demonstrate why denominators, cohort composition, and event definitions matter. Group sizes differ, and the record does not report the calculation basis for the listed serious-event percentages. Comparing percentages without that context could overstate what the data can show. General: small cohort counts are often uncertain, and descriptive differences can arise by chance or because participants and treatment circumstances differ.

Finally, the study lists several secondary clinical and pharmacokinetic outcomes, but their results are not reported in the registry record. That distinction separates the questions a study set out to address from the evidence available for interpretation. For clinical researchers, the record is a useful example of why trial design, analysis populations, endpoint results, and uncertainty should be considered together rather than treating a completed status or a list of outcomes as evidence of a finding.

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 direct limitation for interpretation is that the primary endpoint result is not reported in the registry record. Results for the listed secondary response, survival, and pharmacokinetic outcomes are also not reported. Consequently, the endpoint definition and study aims cannot be connected to observed estimates or their precision from the information.

The study is non-randomized, unmasked, and described as single-group. It does not provide a randomized control comparison, so differences across dose cohorts or combinations cannot be interpreted as causal treatment contrasts. The phase 1b and phase 2 cohorts also address different stages and regimens; the design does not make them like-for-like groups.

The record lists enrollment of 72, but does not report the analysis population or denominator for each outcome in the information. The safety set is defined as participants receiving at least one dose of onvansertib, but further details on missing observations, follow-up, and endpoint-specific inclusion are not reported in the registry record. The denominator and calculation basis for serious-event term percentages are also not reported.

The registry record does not report the statistical analysis plan, dose-escalation decision rules, missing-data methods, multiplicity procedures, or a sample-size justification. It also does not give a causal explanation for the deaths listed in the adverse-event groups. These absences limit interpretation of the posted counts and make it inappropriate to infer treatment effects, comparative safety, or the rationale behind design choices.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Onvansertib in Combination With Either Low-dose Cytarabine or Decitabine in Adult Patients With Acute Myeloid Leukemia (AML) — NCT03303339.
• Registry record last updated 2023-02-27; status last verified 2022-12.
• Results first posted 2023-02-27.
• 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

NCT03303339 is a completed phase 1b/2, non-randomized, unmasked, single-group study sponsored by Cardiff Oncology. It enrolled 72 participants and evaluated onvansertib with low-dose cytarabine or decitabine in adults with acute myeloid leukemia. Phase 1b tested escalating doses with both combinations; phase 2 studied the recommended phase 2 dose with decitabine.

The primary endpoint was the number of participants with a DLT up to Day 28 of Cycle 1. The safety analysis set was defined as participants who received at least one dose of onvansertib. The record lists serious adverse-event counts by group and several serious-event terms with percentages. The primary endpoint result, secondary outcome results, and several analysis details are not reported in the registry record.