Ruxolitinib and Pracinostat Combination Therapy for Patients With Myelofibrosis (MF): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Ruxolitinib and Pracinostat Combination Therapy for Patients With Myelofibrosis (MF) (NCT02267278), a na phase 2 study. The primary outcome is Objective Response Rate (ORR).

2. Clinical Question

The study was intended to examine whether the combination could help control myelofibrosis and to study its safety. Those aims are not interchangeable: response criteria describe disease-related changes, while safety outcomes describe harms and tolerability. A response endpoint in a single-group study can describe what happened among participants receiving the combination, but without a comparison group it cannot establish how outcomes would have differed under another treatment or no study treatment.

The registry identifies pracinostat as investigational and says it was not FDA-approved or commercially available for this use. It describes ruxolitinib as FDA-approved and commercially available to treat myelofibrosis. These statements explain the study context, not the study's response findings. No primary response result is reported in the registry record.

3. Trial Design

This is a Phase 2, single-group study of ruxolitinib and pracinostat in participants with myelofibrosis. The record lists allocation as NA and masking as none. There is no randomized assignment and no concurrent control arm, so the study is not a like-for-like comparison of treatment groups. The sponsor is M.D. Anderson Cancer Center, and the brief summary says all participants would take part at MD Anderson.

The registry enrollment is 25. The brief summary describes enrollment of up to 25 participants. The study is marked completed, with a start date of 2015-01-12 and a primary completion date of 2018-06-01. Completion status indicates the trial's status in the registry; it does not itself mean that outcome results are posted.

The regimen description gives a ruxolitinib starting dose of 15 mg orally twice/day in a 28-day cycle, with dose assignment based on platelet count. For a low platelet count, the description says gradual up-titration from a starting dose of 5 mg. Ruxolitinib was given alone for the first 3 months, after which pracinostat was added at a starting dose of 60 mg orally once/day. The recorded schedule then ends with the incomplete phrase “for 3 alternating days eve”; the full schedule is not reported in the registry record. A questionnaire is also listed as an intervention, but details about its use are not reported in the registry record.

The primary outcome is objective response rate, defined using clinical improvement, partial remission, and complete remission under International Working Group criteria, with a 3-month timeframe. Toxicity of the combination is listed as a secondary outcome. Although the study is marked completed, the registry record reports no primary response estimate and no statistical analyses.

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: Of the twenty-five participants who started the study, five never began pracinostat and are not considered evaluable for response.

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.

6. Primary Endpoint

The registry lists the primary outcome as Objective Response Rate (ORR), measured over 3 months.

The primary endpoint is objective response rate (ORR), recorded as a count of participants and assessed at 3 months. A participant qualifies under the stated definition by meeting criteria for clinical improvement (CI), partial remission (PR), or complete remission (CR) according to International Working Group criteria. These categories combine changes in blood measures and disease-related findings.

The criteria describe CR using bone marrow blasts <5%, hemoglobin >= 10, absolute neutrophil count (ANC) >= 1000, platelets >= 100, <2% immature myeloid cells, and spleen and liver not palpable. PR is defined as CR plus one or more listed features: ANC >= 1000, decreased platelets by 50%, hemoglobin >= 8.5 but <10, or <2% immature myeloid cells. CI includes a hemoglobin increase of 2g/dl, transfusion independence, reduction in splenomegaly and/or hepatomegaly >= 50%, or >=50% reduction in MPN-SAF TSS.

Response criteria make an endpoint more explicit than an undefined judgment of improvement: they state which clinical findings count toward the outcome. However, the registry's definition alone does not tell how many participants met any category. The observed response count and rate, and any uncertainty interval for them, are not reported in the registry record. The endpoint is therefore defined, but its result cannot be assessed from the posted results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The main statistical feature of this study is its single-group design. The record lists no allocation and no masking, and there is no separate comparator arm. The primary endpoint is a count of participants meeting a binary response definition at 3 months: a participant either meets at least one of the specified response categories or does not. General statistical explanation: a response proportion is commonly estimated by dividing the number who meet the criterion by the number assessed, but neither the response count nor the resulting rate is reported in the registry record. No response rate should be inferred from enrollment alone.

What does a confidence interval add?

General statistical explanation: an estimate based on a finite group of participants has sampling uncertainty. A confidence interval gives a range of values compatible with the observed data under the chosen method and its assumptions. A narrow interval indicates greater precision than a wide interval, but it does not correct for bias, missing outcome information, or the absence of a control group. The registry record does not report a confidence interval for ORR, so the precision of the response estimate cannot be evaluated.

What can a single-group result establish?

A single-group study can describe outcomes among the participants who received the study regimen. It cannot directly estimate a between-group treatment contrast because there is no concurrent control group. General statistical explanation: causal comparisons require a credible counterfactual, meaning an estimate of what would have happened to similar participants under an alternative condition. Random assignment is one way to support such a comparison, but this study is not randomized. A response observed in this group could reflect the regimen, participant characteristics, the natural course of disease, other care, or combinations of these influences. The registry record does not report an adjustment strategy that would address these issues.

Why do analysis details matter?

The registry reports no statistical analyses. The statistical plan, including the method used to analyze ORR, is not reported in the registry record. Without an analysis description, readers cannot determine whether any result was an unadjusted estimate, an adjusted estimate, or based on a particular model. General statistical explanation: for a binary outcome, a binomial model is one common way to describe the number of participants meeting a criterion, but its use in this trial is not reported. Model assumptions and the definition of who was included in the analysis affect interpretation.

Over-dispersion is a general issue in count and proportion modeling. It means that variability in the data is greater than a model assumes; when present, conventional uncertainty estimates can be too small. Dependence among repeated observations or clustering can contribute to this problem. The registry record does not report an over-dispersion assessment or a model for repeated observations. The primary endpoint is listed as a participant count, and no further analytic details are available.

How should planning calculations be read?

The stated planning assumptions describe an illustration based on a fraction of an observed effect, a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The assumptions specify participants per arm for non-time-to-event outcomes. They are planning assumptions, not a report of the study's own design justification and not observed power. They also describe a two-arm calculation, whereas the registered study is single-group. Thus such an illustration should not be interpreted as a calculation that validates this study's enrollment or as evidence about its outcome.

The protocol's sample-size justification is not reported in the registry record. No calculated planning estimates or observed response effect are reported in the registry record. General statistical explanation: sample-size planning depends on assumptions about the effect worth detecting, variability, significance level, power, allocation, and analysis method. If those assumptions differ from what occurs, the achieved precision can differ from the planning target. Planning calculations cannot replace an observed estimate and its uncertainty.

Finally, a significance test, if one were reported, would answer a question defined by its null hypothesis and assumptions; it would not by itself indicate clinical importance or establish causation. No p-value or primary-outcome test is reported. With the result and its uncertainty absent, statistical significance and precision for ORR cannot be assessed from 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("Ruxolitinib + Pracinostat")
n      <- c(20)
events <- c(16)
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  Ruxolitinib + Pracinostat  20  16  80.0  56.3  94.3  58.4  91.9  58.1  91.8

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies ORR as the primary outcome and toxicity of the combination as a secondary outcome. It does not report the complete set of statistical hypotheses, an adjustment for multiple testing, or a prespecified testing sequence. The multiplicity strategy is not reported in the registry record. General statistical explanation: when several outcomes or analyses are tested, the chance of at least one apparently notable result can increase; prespecified methods can address that issue. No such method can be attributed to this study on the information available.

Missing-data rules and analysis populations are also not reported in the registry record. The record does not state how participants with missing response assessments, incomplete follow-up, or treatment discontinuation were handled. It does not identify whether the primary analysis included all enrolled participants, only participants with an evaluable assessment, or another population. These details affect the denominator and interpretation of an ORR estimate.

The posted enrollment is 25, but that number alone does not establish how many participants had an assessment at the primary timepoint. The response count, denominator for an analysis, handling of missing observations, and sensitivity analyses are not reported. The registry reports no statistical analyses, so it is not possible to assess whether missingness or multiple outcomes influenced the reported conclusions. In this case, the primary response result itself is not reported in the registry record.

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: infection (8.0%), fever (8.0%), benign mass kidney (4.0%), bronchial infection (4.0%), chest pain - cardiac (4.0%).

The registry lists 8 of 25 participants with serious adverse events and 21 of 25 with other adverse events; deaths are listed as 0. These are participant counts in the combination group. They describe events recorded in the registry and should not be read as a comparison between treatments, because there is no control group.

The listed serious-event terms include Infection and Fever at 8.0% each, and Benign Mass Kidney, Bronchial Infection, Chest Pain - Cardiac, and Conduction disorder at 4.0% each. The percentages are reported as term-level figures. The registry does not report how events were distributed over time or provide enough detail to infer whether listed terms occurred in the same participants. The figures therefore should not be summed or treated as mutually exclusive categories.

Serious adverse events and other adverse events are different reporting categories; the record does not explain whether participants could appear in both. A serious event is not automatically caused by a study drug. The registry record does not provide an attribution analysis, event-level descriptions, or a comparative safety estimate. These figures are a descriptive account of reported safety events, not evidence by themselves of the combination's causal safety profile.

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

The posted information establishes the design and the outcomes the study intended to assess. It records a completed, single-group Phase 2 study of ruxolitinib with pracinostat, with ORR at 3 months as the primary endpoint and toxicity as a secondary outcome. It also provides participant counts for serious and other adverse events and a list of serious-event terms.

The main limitation for interpreting the intended response question is that the primary response result is absent. The registry record does not report how many participants met the ORR definition, the response rate, or the uncertainty around that result. Consequently, the posted information cannot show the observed response frequency or its precision. A study's completed status does not fill this gap.

Even if a response estimate were available, the single-group design would limit what could be concluded. The study has no randomized comparison group and no masking. Outcomes in the treated group could not be contrasted with a concurrent control group to separate treatment-related change from other explanations. The design therefore supports description of this group, not a definitive estimate of comparative treatment effect.

The safety counts indicate that serious and other adverse events were recorded among participants, and the listed terms identify some reported serious events. They do not establish that the study drugs caused those events or provide a comparison of risk against another treatment. The registry does not report event timing or a detailed attribution analysis.

The record also lacks information about missing response assessments, the analysis population, multiplicity, and the statistical analysis plan. These omissions make it difficult to evaluate how the intended outcomes would have been analyzed. Overall, the registry record documents the trial's question, design, and selected safety counts, but it does not provide the primary efficacy result or sufficient analysis detail for a precise assessment of response.

14. Why This Trial Matters Statistically

This trial illustrates the difference between an outcome definition and an outcome finding. ORR is carefully described through clinical improvement, partial remission, and complete remission criteria, and the assessment timeframe is stated as 3 months. Yet the registry does not report the number of participants who met those criteria. A well-specified endpoint helps readers understand what was intended to be measured, but without the observed result it cannot answer how often the endpoint occurred.

The single-group structure is also important statistically. A response count in one group can describe that group's experience, but it does not create a counterfactual comparison. The registry marks allocation as NA and masking as none. These features should shape interpretation even if further response results become available: a result in this design would not carry the same comparative meaning as a result from a randomized, controlled comparison.

The safety record offers a second lesson in reading trial data. Counts and event terms provide useful descriptive information, but they are not automatically causal evidence. Without a control group, event timing, and attribution details, the listed adverse-event figures cannot quantify excess risk due to the combination. The distinction between a participant count and an event-level account also matters; the registry reports affected participants, not a complete narrative of event occurrence.

Finally, missing statistical details are themselves relevant. The analysis plan, missing-data handling, and analysis population are not reported in the registry record. Statistical interpretation depends on these choices, not only on endpoint labels. This study therefore demonstrates why readers need the design, analysis approach, and actual outcome estimates together to assess what evidence a trial contributes.

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 registry record does not report the primary ORR result, any confidence interval, or a statistical test. This is the central limitation for interpreting the study's stated response question. It also does not report the primary endpoint analysis method, the analysis population, missing-data handling, or sensitivity analyses. Without these details, the response endpoint cannot be evaluated for either magnitude or precision.

The design has no concurrent comparator and is not randomized. This limits the ability to distinguish outcomes associated with the combination from changes that might have occurred for other reasons. The record lists no masking, so it does not describe a blinded assessment process. Whether any other safeguards were used is not reported in the registry record.

The regimen description is incomplete: the pracinostat schedule ends mid-phrase after “for 3 alternating days eve”. The full dosing schedule is therefore not available in the registry record. The record lists a questionnaire but does not report its content or role. Further protocol details, including eligibility criteria and follow-up procedures, are not reported in the registry record.

The registry provides safety counts and selected serious-event terms, but not event timing, detailed participant-level information, or attribution to study treatment. It also does not report the protocol's sample-size justification or statistical plan. These limitations mean the posted record supports a basic description of study design and selected safety data, but not a complete assessment of response, comparative effects, or the analysis process.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Ruxolitinib and Pracinostat Combination Therapy for Patients With Myelofibrosis (MF) — NCT02267278.
• Registry record last updated 2019-06-12; status last verified 2019-06.
• Results first posted 2019-06-05.
• 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

NCT02267278 is a completed Phase 2, single-group study titled “Ruxolitinib and Pracinostat Combination Therapy for Patients With Myelofibrosis (MF).” M.D. Anderson Cancer Center is the sponsor. The registry lists 25 participants, no masking, and allocation as NA. The study start date is 2015-01-12, and the primary completion date is 2018-06-01.

The primary outcome is ORR at 3 months under International Working Group criteria. Toxicity of the combination is listed as a secondary outcome. The registry reports 8 participants with serious adverse events, 21 with other adverse events, and 0 deaths. The primary response result and statistical analyses are not reported in the registry record.