A Study of HMPL-306 in Advanced Hematological Malignancies With mIDH: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of HMPL-306 in Advanced Hematological Malignancies With mIDH (NCT04764474), a na phase 1 study. The primary outcome is Dose Escalation Part: Number of Patients With Dose-Limiting Toxicities (DLTs).

2. Clinical Question

This question is principally about early safety and dose escalation, rather than a confirmatory comparison of treatment effects. The listed secondary outcomes include best overall response, objective response rate, time to objective response, duration of objective response, clinical benefit rate, progression-free survival, overall survival, and event-free survival. Those outcomes describe additional aspects of response and time until events, but the registry record does not provide their estimates in the information available for this analysis.

Because there is no concurrent control arm, results from this study cannot isolate the effect of HMPL-306 from the natural course of illness, participant selection, or other influences. The central statistical question is therefore what the registry record reports about dose-limiting toxicity and other outcomes, and how much uncertainty can be assessed from those reports. An outcome count without an analysis estimate or a clearly described denominator would not by itself answer that question.

3. Trial Design

This was a phase 1, open-label, multicenter, single-group study of HMPL-306 in advanced hematological malignancies with IDH mutations. The registry record describes oral once-daily dosing and a dose-escalation part with eight named dose-level cohorts. Since masking was none and allocation was not applicable, participants and study personnel were not assigned to masked treatment groups in a randomized comparison.

The registry record reports enrollment of 46. Its status is terminated. The recorded study start is 2021-04-12 and the recorded primary completion date is 2025-01-15. These dates describe the registry record; they do not establish when results were analyzed or posted.

The primary endpoint is the number of participants with DLTs during the specified assessment window in the dose-escalation part. The endpoint description says the DLT-evaluable analysis set was based on decisions by the sponsor’s clinical team and the safety review committee after committee meetings for each cohort. The full eligibility wording for that analysis set is incomplete in the registry record, so the population rule cannot be reconstructed in full.

The registry record lists 17 outcome measures, including the primary endpoint and secondary response and survival outcomes. It reports zero analyses. A formal statistical analysis plan, sample-size justification, interim-analysis plan, and detailed method for analyzing the outcomes are not reported in the registry record.

In a dose-escalation study, cohorts can provide information about safety at different dose levels. However, cohort membership is not described as randomized, and no concurrent control arm is identified. Differences between cohorts should therefore not be interpreted as controlled estimates of dose effects. The listed cohort safety summaries describe observations within those groups; they do not establish a causal comparison between dose levels.

Planning illustrations use assumptions described in the registry information, including equal allocation between two arms and a normal approximation. These are planning illustrations rather than observed power, and a two-arm comparison does not match the single-group structure of the trial. The registry record does not report that these assumptions formed the trial’s own design or sample-size justification.

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: As only 1 patient was enrolled in dose expansion part, all data of patient were pooled in dose escalation part at Dose Level 8 cohort (Cohort 8) for the data analysis. Thus, no separate participant flow period was created for dose expansion part. Study was terminated based on strategic evaluation of clinical development of HMPL-306 with no safety concerns. Dose levels increase from 1 to 8, with Dose Level 1 being the lowest dose and Dose Level 8 the highest dose.

Only the first 6 of 8 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 included all enrolled patients who received at least 1 dose of study drug.

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 safety analysis set included all enrolled patients who received at least 1 dose of study drug.

Only the first 6 of 9 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Dose Escalation Part: Number of Patients With Dose-Limiting Toxicities (DLTs), measured over From the first dose of study drug (Day 1) up to Day 28 of Cycle 1.

The primary endpoint counts participants with dose-limiting toxicities during dose escalation. Its assessment window runs from the first study-drug dose on Day 1 through Day 28 of Cycle 1. DLTs are defined as specified treatment-emergent adverse events unless the investigator considers them clearly unrelated to study drug. The definition covers selected nonhematologic and hematologic toxicities, qualifying dose delays, and confirmed Hy’s law cases. The full clinical definition is detailed in the registry record, although the definition of the DLT-evaluable population is truncated.

A participant count is a direct way to describe how many people in the analysis set met the DLT definition during the assessment window. In dose escalation, such counts are used to characterize early tolerability at studied dose levels and to inform safety review. The count must be interpreted with its denominator: the same count can represent different observed proportions in groups of different sizes. The endpoint description identifies a DLT-evaluable set, but a complete set definition and the primary endpoint result are not reported in the registry record.

General statistical context: a count is descriptive, and by itself does not express the precision of an estimated risk. Precision depends in part on the number of participants assessed and on the analysis method. A confidence interval, when reported, would show a range of values compatible with the data and method; no confidence interval for this primary endpoint is reported in the registry record. Nor is a treatment-versus-control effect estimated by a single-group DLT count.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The trial’s design determines what statistical conclusions its data can support. This is a single-group, open-label, phase 1 study, not a randomized comparison. It is therefore suited to describing observations during dose escalation and collecting early safety, pharmacokinetic, pharmacodynamic, and preliminary efficacy information. Without a control arm, a difference from an external expectation cannot be attributed to treatment using the design alone.

What does a participant count tell us?

The primary endpoint is a count of participants with DLTs in a defined assessment window. A count has meaning in relation to the analysis population and its denominator. The registry record describes a DLT-evaluable set selected based on review committee decisions after cohort meetings, but the complete eligibility rule is not reported. The endpoint result and its denominator are also not reported. Consequently, a rate, interval estimate, or comparison cannot be reconstructed from the available information.

General statistical context: a confidence interval describes the uncertainty around an estimate under a specified statistical method. Wider intervals indicate less precision, while narrower intervals indicate more precision. The interval does not remove possible bias from selection, outcome definitions, missing observations, or study design. No primary endpoint confidence interval is reported in the registry record.

Why does the assessment window matter?

The DLT window is defined from the first dose on Day 1 through Day 28 of Cycle 1. A fixed window clarifies which events count toward the endpoint, but it also means the count is specific to that period and definition. Events outside the window may not count as DLTs under this endpoint, even if they are recorded elsewhere as adverse events. The listed serious-event summaries should not be assumed to use the same endpoint definition or time window; further details are not reported in the registry record.

How should dose cohorts be understood?

The record identifies eight dose-level cohorts. Cohorts are groups used in the dose-escalation part, but the record does not describe random assignment to these groups. Their safety summaries are descriptive. Comparing counts across cohorts without accounting for the group sizes, eligibility, exposure, and follow-up would not constitute a controlled treatment-effect analysis. A trend across dose levels cannot be asserted from counts alone, and a causal dose-response relation is not established by the information reported.

General statistical context: unequal group sizes affect how much information each group contributes. A raw count is especially difficult to compare across groups when their sizes differ. Even rates may be difficult to interpret if follow-up or the likelihood of being included in an analysis differs. Methods for adjustment, time-at-risk, or formal cohort comparison are not reported in the registry record.

What do the planning illustrations represent?

The planning illustrations use a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. For time-to-event outcomes, the stated planning quantity is the total number of events; for other outcomes, it is participants per arm. These assumptions describe planning illustrations, not the trial’s own reported design. The single-group study has no two-arm randomized comparison, so such illustrations should not be read as evidence that the trial was designed or powered for a controlled comparison. Sample-size illustrations are not observed power.

General statistical context: power is a property of a planned design under specified assumptions, not a measure of whether a completed study found an effect. Observed power is not a substitute for reporting an estimate and its uncertainty. The registry record does not report a sample-size justification for this trial, and the numerical sample-size illustration results are not repeated in this explanation.

What can be inferred from the analysis status?

The registry record lists 17 outcome measures and zero analyses. That information indicates that analysis results are not represented in the record as described, but it does not prove that no analyses were conducted outside the registry. Statistical methods for the primary endpoint and secondary outcomes, including models, confidence-interval procedures, and handling of repeated or time-to-event observations, are not reported in the registry record. No inferential result should be inferred from a stated endpoint or from the existence of a safety summary alone.

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 1: HMPL-306 Dose Level 1", "Cohort 2: HMPL-306 Dose Level 2", "Cohort 3: HMPL-306 Dose Level 3", "Cohort 4: HMPL-306 Dose Level 4", "Cohort 5: HMPL-306 Dose Level 5", "Cohort 6: HMPL-306 Dose Level 6")
n      <- c(3, 3, 9, 5, 3, 3)
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 1: HMPL-306 Dose Level 1  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Cohort 2: HMPL-306 Dose Level 2  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Cohort 3: HMPL-306 Dose Level 3  9  0  0.0  0.0  33.6  -0.0  29.9  0.3  30.8
ARM  Cohort 4: HMPL-306 Dose Level 4  5  0  0.0  0.0  52.2  0.0  43.4  0.4  45.9
ARM  Cohort 5: HMPL-306 Dose Level 5  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Cohort 6: HMPL-306 Dose Level 6  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 17 outcome measures, including the primary DLT count and secondary response and survival outcomes. It reports zero analyses. A multiplicity-control strategy—how error rates would be handled across several outcomes or comparisons—is not reported in the registry record. Without that information and without analysis results, the secondary outcomes cannot be treated as a coordinated set of confirmatory tests.

Missing-data rules are not reported in the registry record. The record does not explain how participants who discontinue, do not complete the DLT window, or lack outcome assessments were handled in analyses. It also does not report whether imputation, censoring rules, sensitivity analyses, or other methods were planned for incomplete observations. These details matter because the estimate can depend on which participants contribute data and how incomplete follow-up is treated.

The DLT-evaluable population is described as determined with the safety review committee after review of each cohort, and the description includes a criterion beginning with receipt of at least 75% of the assigned study drug. The remainder of that criterion is truncated, so the full population definition is not reported in the registry record. The number of participants included in that set is not reported in the registry information.

General statistical context: an analysis population should be defined in advance and applied transparently, because different inclusion rules can yield different estimates. For a time-to-event outcome, censoring rules and follow-up completeness also affect interpretation. The registry record does not report these analysis details for the listed secondary outcomes. The absence of these details limits assessment of both the robustness and the precision of any outcome 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: pneumonia (50.0%), sepsis (50.0%), septic shock (33.3%), cellulitis (33.3%), escherichia bacteraemia (33.3%).

The safety summaries are organized by dose-level cohort. For each cohort, the record lists numbers at risk and numbers affected for serious adverse events and for other adverse events, alongside a death count. For instance, Cohort 5 lists 6 participants at risk and 5 affected for serious adverse events, while Cohort 8 lists 8 at risk and 5 affected. These figures describe the recorded cohorts and should not be added together or treated as a single overall estimate without a reported analysis population and method.

The top serious-event terms are reported as percentages: Pneumonia and Sepsis are each listed at 50.0%; Septic shock, Cellulitis, Escherichia bacteraemia, and Perirectal abscess are each listed at 33.3%. The corresponding term counts, the denominator used for these percentages, and details about whether participants could contribute more than one event are not reported in the registry record.

General statistical context: an adverse-event count is not the same as the number of events, since one participant may experience more than one event. “At risk” also needs to be understood according to the registry’s event-specific definition and observation period. The available information does not establish that the serious-adverse-event summaries use the primary endpoint’s DLT window or DLT-evaluable population.

Deaths are listed by cohort, but the record does not report causes or attribute deaths to treatment. These counts alone cannot establish a treatment-related risk or support comparison with an untreated group, which is absent from the study design.

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

The registry record establishes that the study was designed as an open-label, single-group phase 1 dose-escalation study of HMPL-306 in advanced hematological malignancies with IDH mutations. It defines a primary safety endpoint based on DLTs in the first cycle and lists additional response and survival outcomes. The cohort-level adverse-event summaries provide descriptive safety information for the named dose levels.

It does not report an estimate for the primary DLT endpoint, its denominator, or a confidence interval. It also does not report analyses of the secondary outcomes. The record therefore does not provide a quantitative basis for judging the magnitude or precision of DLT frequency, objective response, survival outcomes, or treatment effects relative to a control. A study’s endpoint definition is not itself an outcome result.

The absence of a comparator is important. This study cannot provide a randomized estimate of how outcomes differ between HMPL-306 and another treatment or no treatment. Even if outcomes vary between dose cohorts, those differences are not controlled comparisons and may have other explanations. The record does not report methods to adjust for cohort differences.

General statistical context: an estimate is most informative when accompanied by its population, denominator, uncertainty, and analysis method. A count, percentage, or observed event can be useful descriptively, but it does not by itself establish causality. Similarly, an absence of a reported estimate should not be taken to mean that an outcome did not occur.

The status is terminated, but the reason is not reported in the registry record. Neither termination nor the listed deaths identify whether the study stopped because of safety, efficacy, feasibility, or another cause. The safety counts are not causal attributions, and the available information does not establish whether the listed adverse events were considered related to study drug.

Accordingly, the defensible reading is limited: the registry describes the study design, intended outcomes, DLT definition, and descriptive cohort safety summaries. It does not report the inferential results needed to quantify the primary endpoint’s precision or make comparative claims about efficacy or safety.

14. Why This Trial Matters Statistically

This trial illustrates the distinction between a study’s stated question and the results needed to answer it. The primary endpoint is clearly framed as a count of participants with DLTs during a defined early-treatment window. Yet the registry information does not include an analyzed count, denominator, or uncertainty interval for that endpoint. A precise endpoint definition is valuable, but it cannot replace reporting the observed data and analysis.

The design also sets an important boundary on inference. A single-group dose-escalation study can describe events observed in its cohorts, but it does not create a randomized control comparison. This matters when interpreting differences between dose levels or considering whether an outcome is attributable to treatment. Descriptive patterns can motivate further questions without answering them causally.

The cohort safety summaries show why denominators and populations matter. Cohort sizes differ, while adverse-event and death counts are reported separately. The percentages for serious-event terms are not accompanied by their denominators in the available information. Without those details, readers cannot fully evaluate the precision or comparability of the figures.

There is also an important registry-reporting lesson: planned outcomes, analysis methods, missing-data rules, multiplicity handling, and sample-size rationale are distinct elements. The record lists many outcomes but reports zero analyses, and several analysis details are not reported. That makes it difficult to assess whether the reported summaries, if any, correspond to prespecified methods.

General statistical context: transparent reporting helps readers distinguish descriptive evidence from confirmatory evidence, and uncertainty from absence of information. For this study, the record supports a careful description of design and selected safety observations, while leaving the central primary endpoint result and its precision unreported.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The primary endpoint result is not reported in the registry record. Neither the number of DLTs nor the DLT-evaluable denominator is available in the information analyzed. The DLT-evaluable population description is truncated after the criterion that a participant had received at least 75% of study drug, so its full inclusion rule cannot be assessed.

The study has no separate control group and is not randomized. This limits causal interpretation and prevents a like-for-like comparison with another treatment. Dose cohorts should not be treated as randomized arms. The registry does not report an adjusted comparison between cohorts or a method for controlling differences among them.

The record lists 17 outcome measures but reports zero analyses. Estimates, uncertainty intervals, and analysis methods for the primary and secondary outcomes are not reported in the registry record. The absence of a reported analysis does not establish whether analyses took place outside the registry.

Details needed to assess completeness are also missing: the missing-data approach, analysis populations for the outcomes, follow-up and censoring rules for time-to-event outcomes, and multiplicity strategy are not reported. The serious-event term percentages lack their associated counts and denominators in the available information. Death counts are listed without causes or attribution.

The reason for termination is not reported in the registry record. A sample-size justification and interim-analysis plan are also not reported. Planning illustrations based on two-arm assumptions do not establish the design basis or power of this single-group trial. These limitations restrict both the interpretation of the descriptive safety figures and the ability to make precise statistical conclusions about the intended outcomes.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of HMPL-306 in Advanced Hematological Malignancies With mIDH — NCT04764474.
• Registry record last updated 2026-01-29; status last verified 2026-01.
• Results first posted 2026-01-29.
• 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

NCT04764474 is a phase 1, open-label, multicenter, single-group study of HMPL-306 in advanced hematological malignancies with IDH mutations. The registry record reports enrollment of 46, oral QD dosing, and eight dose-level cohorts. Its status is terminated; the recorded start date is 2021-04-12 and the recorded primary completion date is 2025-01-15.

The primary endpoint is the number of participants with DLTs from the first dose on Day 1 through Day 28 of Cycle 1. Secondary outcomes include response, progression-free survival, overall survival, and event-free survival measures. The record lists 17 outcome measures and zero analyses. It includes cohort-level adverse-event and death counts, but the primary endpoint estimate, analysis methods, and reason for termination are not reported in the registry record.

Because the design has no control group, the cohort safety summaries are descriptive rather than controlled treatment comparisons. The available record supports description of the study’s design and stated outcomes, but not a precise estimate of the primary endpoint or a comparative conclusion about treatment effects.