Ruxolitinib and Lenalidomide for Patients With Myelofibrosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Ruxolitinib and Lenalidomide for Patients With Myelofibrosis (NCT01375140), a na phase 2 study. The primary outcome is Participants With Objective Response.

2. Clinical Question

The study was designed to assess the combination of ruxolitinib and lenalidomide in patients with myelofibrosis. Its primary outcome was the number of participants with an objective response, defined using International Working Group for Myelofibrosis Research and Treatment criteria. Those criteria combine complete and partial remission with clinical improvement. They draw on several kinds of clinical findings, including blood counts, bone marrow findings, spleen or liver enlargement, transfusion status, and symptom burden.

This question is clinically descriptive as well as statistical: it asks how often the specified response criteria were met among participants receiving the study regimen. Because the study had a single group and no like-for-like control arm, it cannot estimate a comparative effect against another treatment or against no treatment. The registry record does not report the number of participants meeting the primary endpoint, so the response frequency and its precision cannot be assessed from the posted primary outcome information.

3. Trial Design

This was a phase 2, completed, single-group study sponsored by M.D. Anderson Cancer Center. The registry lists enrollment as 31, allocation as NA, and masking as none. The study began on 2011-09-22, and primary completion was recorded as 2018-09-05. The registry identifies myeloproliferative diseases as the condition category and myelofibrosis as the population in the study description.

The active regimen combined ruxolitinib and lenalidomide. The registry describes ruxolitinib at 15 mg orally twice daily continuously and lenalidomide at 5 mg orally on days 1–21 of a 28-day cycle, followed by 7 days without therapy. Prednisone was to be added for participants who had not responded after 3 cycles. The registry description of the prednisone schedule ends during its account of cycle dosing, so the full schedule is not reported in the registry record.

A single-group design can describe outcomes among participants who received the regimen, but it does not provide a concurrent comparison group. The listed absence of masking means the study was not masked according to the registry. Allocation is recorded as NA, and the registry’s own note cautions that the arms are not a like-for-like control comparison. No separate arm-level comparative interpretation is appropriate on these registry entries.

The primary outcome timeframe was 3 cycles (28 days each) up to 3 months. The registry record does not report a statistical analysis plan, the method used to handle deviations from treatment, or a sample-size justification for the original trial. The planning assumptions associated with sample-size illustrations are not the same as the study’s original design rationale.

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.

Overall Study

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 Participants With Objective Response, measured over 3 cycles (28 days each) up to 3 months.

The primary endpoint is the count of participants with an objective response. The registry defines objective response as clinical improvement, partial remission, or complete remission according to International Working Group for Myelofibrosis Research and Treatment criteria. The assessment timeframe is 3 cycles (28 days each) up to 3 months.

The definition combines different forms of improvement rather than relying on a single measurement. Complete remission incorporates bone marrow and blood-count criteria, immature myeloid cell levels, and whether the spleen and liver are palpable. Partial response is defined in relation to complete remission and specified improvements in blood counts or immature myeloid cells. Clinical improvement can include a rise in hemoglobin, transfusion independence, reduction in spleen or liver enlargement, or a decrease in symptom score.

A composite endpoint can capture clinically relevant changes that may occur in different ways across participants. It can also make interpretation dependent on the exact criteria, assessment timing, and rules for combining response categories. Those details matter because a participant may meet one component without meeting another. The registry record states the response definition but does not report the number meeting the endpoint, the separate counts for complete response, partial response, or clinical improvement, or the statistical analysis used to estimate the response rate. The primary outcome is therefore identifiable, but its observed result is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical interpretation starts with identifying what was measured and which design produced the observations. The primary outcome is a participant count for a binary classification: each participant either meets the objective-response definition or does not. The registry describes a single-group study, so the central descriptive quantity would ordinarily be the observed response count and the proportion of assessed participants meeting the criteria. The registry record does not report the primary count or the analysis denominator, so neither quantity can be evaluated from the posted primary outcome information.

What does a confidence interval add?

In general, a confidence interval expresses uncertainty around an estimate under a specified statistical procedure. For a binary outcome, the interval reflects that the observed proportion is an estimate based on a finite group of participants, rather than a perfectly precise measure of the response frequency in a broader population. A narrower interval indicates greater precision under the method used; a wider interval indicates more uncertainty. The interval does not correct weaknesses in study design, and it cannot establish that the regimen caused the observed outcome. No confidence interval for the primary endpoint is reported in the registry record.

The calculation of uncertainty for a binary response outcome depends on the response count, the denominator, and the chosen interval method. Different methods can behave differently, especially when counts are small or when the observed proportion is near an endpoint of its possible range. The registry record does not report the method used for such an interval or provide the endpoint data needed to evaluate one. It would be inappropriate to infer a method from the study phase or from the endpoint type alone.

Why does the single-group design matter?

In general, an outcome observed in a treated group can reflect the treatment, the course of the disease, participant selection, other care, measurement decisions, or combinations of these factors. A randomized concurrent control group helps support a comparison by balancing, on average, factors that could otherwise differ between groups. This study is identified as single-group, with allocation recorded as NA, so it does not provide that randomized comparison. A response count in this setting would describe what happened in the study group, not the difference attributable to treatment relative to a control.

The regimen also includes a treatment change for participants who had not responded after 3 cycles: prednisone was to be added. Consequently, the treatment received could vary with response status over time. That feature is relevant when interpreting outcomes and safety because later exposure may differ across participants. The registry record does not report detailed exposure, adherence, discontinuation, or analysis rules that would clarify how these changes were incorporated into the primary analysis.

What do the planning assumptions mean?

In general, a sample-size calculation asks how much information a proposed study might need to detect a chosen effect under specified statistical assumptions. The registry’s planning assumptions describe a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They are described as planning illustrations, not as the original trial design. The stated design is single-group, so the hypothetical equal-allocation comparison should not be mistaken for a randomized comparison conducted in this study.

The assumptions also describe the effect to detect as a stated fraction of the observed effect. Such planning scenarios are hypothetical: they explore what a future comparison might require under an assumed effect size. They do not show that a particular sample size was justified for this trial, and they do not measure the probability that the trial’s result is true. In particular, a sample-size calculation is not observed power. The original sample-size justification is not reported in the registry record.

When outcomes are counts or proportions, the model must account for how observations arise. For a binary participant outcome, a binomial framework is a common general approach because each participant contributes an event or non-event classification. That general explanation does not establish which model or test was used in this study. The registry record does not report an analysis method for the primary endpoint. It also does not report whether any adjustment for baseline characteristics or other factors was planned or performed.

Over-dispersion is a general concern when observed variation exceeds what a chosen statistical model expects. For example, participants may differ in underlying risk, or observations may be correlated, making a simple model understate uncertainty. Such issues can affect standard errors and confidence intervals. No over-dispersion assessment is reported in the registry record, and the primary outcome is a participant count rather than a repeated-event rate in the registry record. This general concept should not be read as evidence that over-dispersion occurred in this trial.

Interpretation also depends on the population included in the analysis. An intention-to-treat analysis generally retains participants in their assigned groups, while a per-protocol analysis generally focuses on participants who followed specified treatment requirements. Those general strategies can answer different questions, but the registry record does not report an analysis population for the primary endpoint. Because the study is single-group, the record also does not explain how participants who stopped treatment, missed assessment, or had incomplete response data were handled.

Finally, the primary outcome definition is composite. In general, composite endpoints can increase the range of ways a participant qualifies as having an event, but the overall result can conceal variation among its components. A response count alone would not show how many participants met each component unless component results were also reported. The registry lists the response criteria but does not report component-specific results or the primary response count. Thus, neither the overall response estimate nor the contribution of each response category can 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 + Lenalidomide")
n      <- c(31)
events <- c(7)
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 + Lenalidomide  31  7  22.6  9.6  41.1  11.4  39.8  11.5  40.0

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies one primary outcome measure: participants with objective response. Its definition includes clinical improvement, partial remission, and complete remission. These categories form the components of the composite response outcome, but component-specific counts are not reported in the registry record.

Multiplicity arises when a study makes several statistical comparisons or tests several outcomes, because the chance of at least one apparently notable result can increase as more tests are considered. A prespecified testing hierarchy or adjustment can address some multiplicity concerns. The registry record does not report whether multiplicity adjustments, a testing hierarchy, or interim analyses were planned or used. It would be inappropriate to infer those procedures from the listed endpoint alone.

The record does not report the primary response count, the number assessed for that endpoint, the analysis population, or how missing response assessments were handled. These omissions prevent assessment of whether the endpoint denominator matched enrollment, whether any participants lacked evaluable follow-up, or how missingness could have affected the reported result. Missing outcome data can matter when the chance of an unavailable assessment is related to response or health status; this is a general concern, not a claim that such missingness occurred in this study.

The registry record also does not report a statistical analysis plan or explain whether the primary endpoint was analyzed according to an intention-to-treat, per-protocol, or other population. Without those details and the endpoint result, the response outcome cannot be independently interpreted as an estimate with a defined analysis population and missing-data approach.

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: lung infection (25.8%), skin infection (9.7%), fever (9.7%), thromboembolic event (6.5%), neutropenic fever (3.2%).

The registry reports safety information for the Ruxolitinib + Lenalidomide group. It records 19 participants affected by at least one serious adverse event, with 31 at risk, and 31 participants affected by at least one other adverse event, also with 31 at risk. The registry lists 2 deaths. These are descriptive observations in the study group, not comparisons with another group.

The listed serious adverse event terms include lung infection at 25.8%, skin infection at 9.7%, fever at 9.7%, thromboembolic event at 6.5%, neutropenic fever at 3.2%, and anemia at 3.2%. These percentages refer to the event terms as reported; they should not be added together. A participant may have experienced more than one type of event, so categories need not be mutually exclusive.

The registry record does not report whether each event was considered related to treatment, when events occurred, how long they lasted, or whether they led to treatment changes. It also does not give the circumstances or attribution of the deaths. Serious-event frequencies can describe the burden recorded during study follow-up, but without a control group they do not establish how much risk was caused by the regimen or how the risk compares with another approach.

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

The record documents a phase 2 study of ruxolitinib plus lenalidomide in patients with myelofibrosis, with prednisone added for participants who had not responded after 3 cycles. It identifies objective response as the primary outcome and gives the criteria and assessment timeframe. It also reports serious and other adverse-event summaries for the study group. These details show what the study set out to assess and describe some of the safety information registered.

The record does not report the primary response count, an estimated response proportion, or an analysis of the primary endpoint. That absence should not be interpreted as evidence that no participant responded. It means the posted record does not provide the result needed to judge the frequency or precision of response. The registry also does not report component-specific response counts, so the separate contributions of clinical improvement, partial remission, and complete remission are not available.

The design places important limits on causal interpretation. There was no separate control group, and the study is identified as single-group rather than a randomized comparison. Any outcome observed among participants in the study could reflect influences beyond the regimen. A comparison with another treatment, or an estimate of the treatment’s effect relative to a control, cannot be made from these registry entries.

The safety data should likewise be interpreted as descriptive. The listed event terms and deaths are recorded for the study group, but the registry does not report treatment attribution, timing, or comparative safety results. It is not possible to determine from these summaries whether the regimen caused a particular event or how event frequencies compare with another care approach.

Overall, the registry supports a description of the trial’s design, intended response definition, and reported safety events. It does not provide the primary efficacy result or a controlled estimate of treatment effect. These limits are about what can be concluded from the registry record, not a judgment about the treatment or the full set of study findings that may exist elsewhere.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of the distinction between a registered research question and a reportable result. The registry specifies an objective-response endpoint and gives a detailed clinical definition, but the primary response count and analysis are not reported in the registry record. A clearly defined endpoint is necessary for interpretation, yet it is not enough to evaluate an effect without the observed result and its uncertainty.

The design also illustrates why a single-group result answers a narrower question than a controlled comparison. A response outcome among participants receiving a regimen can describe the experience of that study group. It cannot, on its own, separate treatment effects from other influences or show how outcomes would have differed under an alternative. This is especially important when a registry also presents hypothetical sample-size planning scenarios based on equal allocation between two arms: such assumptions do not turn a single-group study into a comparative trial.

Composite endpoints raise another useful statistical issue. Combining complete remission, partial remission, and clinical improvement can capture several clinically meaningful ways to meet a response definition. But an overall count can obscure which components account for the result. Component counts and a transparent analysis plan would help readers understand what the composite outcome represents; those results and details are not reported in the registry record.

The safety summaries similarly demonstrate the distinction between describing events and attributing them. Event frequencies can identify what was recorded during follow-up, but without a control group and information on attribution, timing, and analysis, they cannot establish comparative risk or causation. The record’s missing primary result and limited analysis details therefore constrain both efficacy and safety interpretation.

15. Limitations

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

The primary limitation for interpreting efficacy is that the registry record does not report the primary objective-response count or a statistical analysis. The assessment timeframe and response definition are stated, but the analysis denominator, uncertainty estimate, and results for the response components are not. Consequently, the primary question cannot be answered from the posted primary outcome information.

The single-group design has no like-for-like control arm. The registry lists allocation as NA and explicitly cautions that arms are not a like-for-like control comparison. The study can describe outcomes in its treated group, but it cannot provide a randomized between-group estimate of treatment effect. The registry record also does not report a statistical analysis plan, the original sample-size justification, interim analyses, or the analysis population.

Prednisone was to be added for participants who had not responded after 3 cycles, meaning later treatment could vary according to response. The registry description of the prednisone schedule is incomplete, and full subsequent dosing is not reported in the registry record. Detailed treatment exposure, adherence, discontinuation, and handling of incomplete response assessments are also not reported.

Safety interpretation is limited by the lack of a control group and the absence of information on event attribution, timing, duration, and circumstances surrounding the deaths. The registry reports adverse-event frequencies but does not describe whether events overlapped within participants or how safety follow-up was handled. These limitations do not negate the observations; they define what those observations can support.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Ruxolitinib and Lenalidomide for Patients With Myelofibrosis — NCT01375140.
• Registry record last updated 2025-06-08; status last verified 2025-05.
• Results first posted 2019-07-31.
• 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

NCT01375140 is a completed phase 2, single-group study sponsored by M.D. Anderson Cancer Center. It enrolled 31 participants and evaluated ruxolitinib with lenalidomide for myelofibrosis. Prednisone was to be added for participants who had not responded after 3 cycles. The primary outcome was the count of participants with objective response over 3 cycles (28 days each) up to 3 months.

The registry record does not report the primary response count or its statistical analysis. Safety information lists 19 participants with at least one serious adverse event, 31 with at least one other adverse event, and 2 deaths. The record does not report treatment attribution for events or provide a controlled comparison. Therefore, the posted information describes the study question and some safety observations, but does not establish a comparative treatment effect or quantify the primary response result.