Ibrutinib Before and After Stem Cell Transplant in Treating Patients With Relapsed or Refractory Diffuse Large B-cell Lymphoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Ibrutinib Before and After Stem Cell Transplant in Treating Patients With Relapsed or Refractory Diffuse Large B-cell Lymphoma (NCT02443077), a randomized phase 3 study. The primary outcome is 24-month Progression-free Survival (PFS), Defined as the Proportion of Patients Who Are Alive and Progression-free 2 Years From Randomization.

2. Clinical Question

The comparison concerns the addition of ibrutinib to a transplant-based treatment strategy, rather than ibrutinib in isolation. Both groups receive chemotherapy and autologous transplantation; the assigned study drug differs. The registry’s brief summary describes the study as addressing whether adding ibrutinib around transplantation may help the transplant work better, while noting that this was not known. The posted primary data are proportions for each arm, with confidence intervals. They describe the recorded endpoint in the specified analysis population; they do not by themselves establish a causal difference between assigned groups. The registry record does not report a primary between-arm test or model that would quantify the uncertainty in a direct treatment contrast.

3. Trial Design

This is a randomized phase III study of ibrutinib versus placebo during and following autologous stem cell transplantation for relapsed or refractory diffuse large B-cell lymphoma of the activated B-cell subtype. The record lists randomized allocation, triple masking, and a crossover model. The sponsor is the National Cancer Institute. Enrollment is recorded as 94, and the study status is active, not recruiting.

Both arms include conditioning chemotherapy and autologous transplantation. The registry describes an ibrutinib arm and a placebo arm; it also indicates that investigators may choose between the BEAMi and CBVi conditioning regimens in the ibrutinib arm, with the placebo arm receiving one of the conditioning regimens described for that arm. The brief summary explains the transplant sequence in general terms: stem cells are collected and stored before high-dose chemotherapy, then returned after treatment. The detailed arm descriptions in the registry record are truncated, so complete schedules and crossover procedures are not reported in the registry record.

The record gives a start date of 2016-10-12 and a primary completion date of 2024-06-10. These dates describe the registered study record and do not, on their own, establish the timing or completeness of every outcome analysis. The primary population is defined as participants who completed at least one cycle of treatment, while arm-level endpoint results list 39 participants in the ibrutinib arm and 38 in the placebo arm. The registry enrollment and endpoint analysis counts therefore refer to different record details; the reason for that difference is not reported in the registry record.

Random assignment is intended, in general, to make the groups comparable at baseline in expectation, while masking can reduce the influence of knowledge of assigned treatment on care or outcome assessment. Those design features do not guarantee that every factor is balanced or eliminate uncertainty, especially in a trial with limited observed group sizes. The crossover model is noted in the record, but the timing, eligibility, and analytic handling of crossover are not reported in the registry record. Accordingly, the posted arm-specific proportions should be read as descriptive results rather than as a complete account of the randomized comparison.

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

Analysis population for the primary outcome, as defined in the registry: Participants that completed at least one cycle of treatment

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: Only randomized arms were analyzed. Safety cohort was excluded per protocol: These patients will not be randomized and will be excluded in the statistical analysis to compare the two arms.

6. Primary Endpoint

The registry lists the primary outcome as 24-month Progression-free Survival (PFS), Defined as the Proportion of Patients Who Are Alive and Progression-free 2 Years From Randomization, measured over Time between registration and disease progression or death, whichever comes first, assessed at 24 months.

The primary endpoint is 24-month progression-free survival, described as the proportion of participants who are alive and progression-free 2 years from randomization. PFS combines two kinds of events: disease progression and death. A participant is counted as progression-free only while neither event has occurred by the specified assessment time. The registry says progression is assessed using the Lugano classification.

A fixed-time proportion provides an interpretable description of the share of the defined population meeting the endpoint at a specified point in follow-up. It differs from a median time-to-event result: a proportion at a landmark does not state the typical time to progression or death, and it does not show the full trajectory of events over follow-up. It also does not describe what happened after the landmark.

The registry’s endpoint wording has an important timing distinction. The title describes the endpoint as alive and progression-free 2 years from randomization, while the timeframe says time between registration and progression or death, assessed at 24 months. The record does not clarify whether registration and randomization coincide for this endpoint or explain the wording difference. The listed population is participants who completed at least one treatment cycle. Whether any other analysis populations or additional rules for censoring and missing assessments were used is not reported in the registry record.

For the ibrutinib arm, the posted proportion is 0.576, with a 95% confidence interval of 0.421 to 0.788. For the placebo arm, it is 0.408, with a 95% confidence interval of 0.265 to 0.628. These estimates describe the recorded endpoint in each arm. The registry record does not provide a reported primary test or confidence interval for the between-arm contrast, so the arm-specific intervals should not be treated as a substitute for that comparison.

7. Results Posted in the Registry

• Arm I (Ibrutinib, Chemotherapy, autoHCT): 0.0% (0/39) versus 0.0% (0/38) on control; risk difference 0.0 percentage points (95% CI 0.0 to 0.0); unadjusted two-proportion z-test p = n/a.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Incidence of Secondary Malignancies

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Arm I (Ibrutinib, Chemotherapy, autoHCT): 2.6% (1/39) versus 0.0% (0/38) on control; risk difference 2.6 percentage points (95% CI -2.4 to 7.5); unadjusted two-proportion z-test p = 0.320.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Arm I (Ibrutinib, Chemotherapy, autoHCT) and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists secondary outcomes covering overall survival, time to hematopoietic engraftment, PFS, response rate using the Lugano classification, treatment-related mortality, hematologic toxicity, and secondary malignancies. Seven secondary results are posted, while the record lists 14 outcome measures overall. These entries address different aspects of outcomes and safety, so they do not all measure the same underlying question as the primary 24-month PFS endpoint.

One posted secondary result concerns secondary malignancies up to 60 months. The record lists 1 participant in the ibrutinib arm and 0 in the placebo arm, and a risk difference of 0.026 with a 95% confidence interval from -0.024 to 0.075. Its unadjusted p-value is 0.32. The uncertainty interval includes values on either side of no difference, and the record lists no reported analyses for that result. This secondary observation is not a substitute for the primary PFS comparison and does not settle the trial’s main question.

More broadly, secondary outcomes can provide context about survival, response, transplant recovery, and adverse outcomes, but their evidential weight depends on how they were prespecified and analyzed. When many outcomes are examined, the chance of finding an apparently unusual result by chance can increase. An unadjusted result does not account for that multiplicity. The registry record does not report a multiplicity adjustment or a hierarchy of confirmatory secondary hypotheses.

Secondary results should therefore be interpreted as additional descriptive or exploratory evidence unless their analysis status and multiplicity handling are clear. They should not be selectively emphasized to override an uncertain primary comparison. The registry record does not provide enough information to establish which secondary outcomes were intended as confirmatory, how missing observations were handled across them, or whether any interim monitoring influenced their interpretation.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Overall Survival (OS)
• Time to Hematopoietic Engraftment
• PFS
• Response Rate Using the Lugano Classification
• Treatment-related Mortality
• Incidence of Hematologic Toxicity of Ibrutinib Therapy
• Incidence of Secondary Malignancies

9. Statistical Interpretation

10. Statistical Methods Explained

Statistical concepts below are general explanations, not claims about analyses used in this trial unless explicitly stated. The registry identifies an endpoint and arm-level estimates, but does not report a primary inferential analysis method.

What does a proportion estimate?

A proportion is the share of an analysis population meeting a defined outcome. Here, the outcome is being alive and progression-free at the specified assessment, and the registry labels the unit as a proportion of participants. The arm values are 0.576 and 0.408. They are estimates for the participants included in each arm’s posted endpoint result, not guaranteed values for all people with the condition or for a different treatment setting.

For a time-to-event outcome, the meaning of a fixed-time proportion depends on how follow-up and incomplete observations are handled. For example, some participants may not have a recorded status at the assessment time. Statistical methods can account for such situations in different ways, but those methods require explicit specification. Whether the posted values use a particular survival estimator, how censoring was handled, and how missing assessments were treated are not reported in the registry record. It would therefore be inappropriate to infer the calculation from the endpoint label alone.

What does randomization contribute?

In general, random allocation aims to reduce systematic differences between study groups at baseline. It supports a comparison based on assigned groups, because treatment assignment is not chosen by participants or clinicians. Randomization does not guarantee that the groups will look identical on every characteristic, and it does not resolve missing data, deviations from assigned treatment, or differences in follow-up. Those issues still depend on the conduct of the study and the analysis.

The registry records randomized allocation and triple masking. Masking generally aims to limit the influence of treatment knowledge on participants, clinicians, and outcome assessment, depending on who is masked. The exact masking roles are not reported in the registry record. The record also lists a crossover model, but the crossover procedures and the method used to account for crossover are not reported. It is not possible to determine from the listed design labels alone whether the primary result is based on assignment, treatment received, or another estimand.

How should the confidence intervals be read?

A confidence interval describes the range of values reasonably compatible with the observed data under the statistical procedure used, over repeated samples and under its assumptions. The registry reports 95% confidence intervals for the arm-specific primary proportions: 0.421 to 0.788 for ibrutinib and 0.265 to 0.628 for placebo. These intervals convey uncertainty about each arm’s estimate. They are not a range containing a specified proportion of individual participants, and they do not show the probability that the true value lies within this particular interval.

Separate intervals are not a direct interval for the difference between arms. The uncertainty of a contrast depends on both estimates and their joint sampling variation, as well as the chosen analysis. A formal comparison normally reports the contrast itself and an interval or test for it. The primary record does not report such an inferential result. Comparing the two point estimates by eye, or using overlap of their individual intervals as a significance test, is not a substitute.

Why can endpoint definitions affect inference?

PFS includes progression or death, whichever occurs first, according to the registry timeframe. The event definition, assessment schedule, rules for censoring, and handling of participants without an assessment can affect the estimated result. The record says progression is assessed using the Lugano classification, but does not report the detailed assessment schedule or full statistical rules. It also uses different wording for the time origin: randomization in the endpoint title and registration in the timeframe description. Such details matter because they define when follow-up begins and which observations contribute to the endpoint.

The primary population is described as participants who completed at least one cycle of treatment. In general, excluding participants based on post-randomization treatment completion can make a result differ from an analysis that retains everyone as randomized. The registry record does not explain the rationale for this population definition or report an alternative analysis. This does not establish that the posted estimate is biased; it means that the consequences of the population rule cannot be evaluated from the record alone.

What do the planning assumptions mean?

The registry includes sample-size planning illustrations based on the observed effect, with two-sided significance level 0.05, equal allocation, and a normal approximation. The stated effect to detect is a fraction of the observed effect. These are planning illustrations, not observed power and not the trial’s own design justification. They do not establish what effect the trial was designed to detect, how its enrollment target was chosen, or whether the completed study had adequate precision. The registry record does not report the trial’s sample-size justification.

In general, an illustration that uses an observed effect is conditional on that estimate, which itself is uncertain. Planning for a fraction of an observed effect asks how a hypothetical design would behave for a smaller target effect; it does not change the trial’s observed data. A normal approximation is a simplifying method whose suitability depends on the outcome and assumptions. These illustrations should not be read as evidence of a treatment effect or as a retrospective statement of power.

For time-to-event planning, the record notes that the relevant quantity is a total event count; other planning displays are expressed as participants per arm. These conventions describe how the planning calculations are presented, not the actual number of events observed for the primary endpoint. The registry record does not report the primary event count or a complete analysis plan.

11. Confidence Intervals and Precision

The 95% confidence interval for the ibrutinib-arm PFS proportion is 0.421 to 0.788; for the placebo arm it is 0.265 to 0.628. Both intervals cover a broad range of plausible underlying arm-specific proportions under the method used. The point estimates, 0.576 and 0.408, should therefore be read with substantial uncertainty rather than as exact values for the populations from which the participants were drawn.

Precision concerns the uncertainty around an estimate, not whether the result is important or whether the treatments differ. The registry record does not provide a 95% confidence interval for the primary between-arm contrast. The arm-specific intervals cannot be combined informally to create one. In particular, overlap between separate intervals is not a formal test of whether the arms differ, and their endpoints do not directly indicate the uncertainty of a contrast.

In general, confidence intervals are affected by the amount of information available and the variability of the outcome, along with the analysis method and its assumptions. The endpoint results list 39 participants in the ibrutinib arm and 38 in the placebo arm, while the registry enrollment is 94. The record does not explain the distinction between enrollment and the endpoint analysis population beyond defining the primary population as participants who completed at least one treatment cycle. The amount and pattern of missing follow-up, and the handling of censoring, are not reported in the registry record.

The interval interpretation also depends on the endpoint’s time origin and assessment rules. The primary endpoint title refers to 2 years from randomization, whereas the timeframe refers to time from registration and assessment at 24 months. The record does not clarify that wording difference. These limitations do not invalidate the reported intervals, but they constrain how precisely the result can be interpreted and compared.

12. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary endpoint and multiple secondary outcomes, but does not report a statistical analysis plan describing multiplicity control. When researchers test or estimate many outcomes, the probability of at least one apparently notable result can rise even if there is no underlying effect. General approaches include defining a hierarchy of outcomes, adjusting significance thresholds, or treating secondary analyses as exploratory. The record does not say which, if any, approach was used. The posted secondary p-value is explicitly unadjusted, so it should not be interpreted as though it accounts for the other outcomes.

Missing data and analysis populations also affect interpretation. The primary population is described as participants who completed at least one cycle of treatment. The registry record does not state how participants who did not meet that criterion were handled, whether an intention-to-treat analysis was performed, or whether other populations were analyzed. The endpoint-specific participant counts are 39 and 38, while total enrollment is recorded as 94; reasons for the difference are not reported in the registry record.

The primary endpoint is time-to-event in concept, but the posted result is a fixed-time proportion. The registry does not report the method used to estimate that proportion, censoring rules, the amount of missing follow-up, or any sensitivity analysis for missing outcomes. It also does not describe how a participant with incomplete endpoint assessment was classified. These omissions prevent evaluation of whether the assumptions about follow-up and missingness were reasonable.

The trial is labeled crossover, but crossover timing and its treatment in the primary analysis are not reported in the registry record. The record also does not report interim analyses, stopping boundaries, or the detailed statistical plan. These items should not be inferred from the phase, design labels, or observed estimates. In general, an analysis plan is needed to distinguish prespecified confirmatory tests from analyses selected after observing results.

13. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: abdominal pain (50.0%), creatinine increased (50.0%), metabolism, nutrition disord - oth spec (50.0%), acute kidney injury (50.0%), sepsis (33.3%).

The registry reports serious and other adverse-event counts for several safety groups. In the ibrutinib arm, 22 of 45 participants at risk are listed as affected by serious adverse events, and 36 of 45 by other adverse events. In the placebo arm, the corresponding counts are 23 of 43 and 38 of 43. Separate entries are shown for a safety cohort and a crossover arm: 4 of 6 and 6 of 6 in the safety cohort, and 1 of 2 and 1 of 2 in the crossover arm, for serious and other adverse events respectively. These are reported counts within the listed groups, not a direct measure of comparative risk adjusted for follow-up or exposure.

The registry lists deaths of 14 in the ibrutinib arm, 10 in the placebo arm, 3 in the safety cohort, and 2 in the crossover arm. The record does not provide a cause-specific interpretation for these counts in this summary. The listed serious-event terms include abdominal pain, creatinine increased, metabolism and nutrition disorder, acute kidney injury, sepsis, and hypotension; their displayed percentages are term-level summaries, and their denominators are not reported in the registry record.

Safety counts require context about exposure time, event definitions, overlap among participants, and attribution. Those details are not reported in the registry record. The figures should not be added across groups, treated as mutually exclusive, or interpreted as a treatment comparison without an appropriate analysis.

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

The posted primary data describe 24-month PFS proportions of 0.576 for the ibrutinib arm and 0.408 for the placebo arm, with broad 95% confidence intervals for each estimate. They show what was recorded for the defined endpoint population in each arm. They do not, by themselves, establish that assignment to one strategy caused a difference in PFS.

The registry record does not report an inferential primary comparison, a treatment-effect confidence interval, or a primary analysis model. It is therefore not possible to make a statistical conclusion about a between-arm difference from the arm-specific values alone. Nor do these data establish equivalence: a failure to report a direct test is not evidence that the strategies have the same effect. The result should not be turned into a recommendation about treatment.

The endpoint is limited to the stated assessment time and definition. It does not summarize every aspect of survival, response, quality of life, toxicity, or transplant recovery. Some of those outcomes appear among the secondary measures, but their presence in the registry does not mean their results are complete or confirmatory. The secondary malignancy result has an unadjusted analysis and a confidence interval that spans values on either side of no difference, so it is a limited finding rather than a definitive comparative conclusion.

Interpretation is also constrained by the population definition, the difference between total enrollment and endpoint counts, and the lack of reported details on missing assessments and censoring. The record’s different references to randomization and registration as the time origin, as well as its crossover label without analytic details, further limit reconstruction of the estimand. These are uncertainties in the registry information; they are not proof that the trial was improperly conducted.

In short, the registry reports arm-specific endpoint estimates and safety counts, but the comparative statistical evidence needed for a firm conclusion is not reported in the registry record. The result is best treated as a description of the posted data, with uncertainty, rather than as a definitive answer about the effect of adding ibrutinib.

15. Why This Trial Matters Statistically

This trial poses a statistically demanding question because it compares two complex treatment strategies around autologous transplantation, with the study drug differing between arms. The main endpoint is a time-to-event concept expressed as a fixed-time PFS proportion. That combination makes clarity about the time origin, progression assessment, censoring, and analysis population especially important. The registry identifies some of these elements, including use of the Lugano classification, but does not report all the details needed to reconstruct the analysis.

Randomization and masking are important design features, but their value in supporting a comparison depends on how outcomes are analyzed and reported. The registry offers arm-level estimates and confidence intervals, yet does not report a primary contrast or model. This illustrates a central statistical principle: separate descriptions of groups are not the same as an estimate of the difference between them, and uncertainty for the contrast needs to be quantified directly.

The endpoint population is defined by treatment completion, while the record also gives a larger enrollment figure. Understanding who contributes to each analysis is essential for interpreting any randomized comparison. The registry record does not explain the difference or describe missing-data and crossover methods. These details matter because they can affect both the meaning and precision of an estimate.

Finally, multiple secondary outcomes are listed, including outcomes related to survival, engraftment, response, toxicity, and secondary malignancies. Their variety may make the evidence more informative, but also raises questions about prespecification and multiplicity. Without a reported analysis plan, secondary findings should be weighed cautiously and kept distinct from the primary question.

16. 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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

Several limitations are specific to the registry record for this trial. The primary endpoint title defines the assessment as 2 years from randomization, while the timeframe description refers to time between registration and progression or death, assessed at 24 months. The record does not explain whether registration and randomization coincide or reconcile the two descriptions.

The primary population is described as participants who completed at least one cycle of treatment. Arm-level endpoint counts are 39 and 38, whereas enrollment is 94. The record does not explain the difference, report the handling of participants outside the stated endpoint population, or identify another primary analysis population. Missing follow-up, censoring, and sensitivity analyses are not described.

Although the trial is listed as randomized, triple-masked, and crossover, the masking roles and crossover procedures are not reported. The registry does not report how crossover was incorporated into the primary analysis, or whether the reported estimates correspond to assignment or treatment received. The arm descriptions are truncated, limiting detail about regimen implementation.

The registry gives primary arm-specific proportions and confidence intervals but no reported primary between-arm analysis. It also does not report the detailed statistical analysis plan, multiplicity strategy, interim analyses, or sample-size justification. The planning illustrations are based on the observed effect and should not be confused with the trial’s design rationale. For secondary outcomes, the registry lists seven posted results and no reported analyses for the secondary malignancy result. Taken together, these omissions limit independent assessment of comparative inference, missing-data robustness, and the status of secondary findings.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Ibrutinib Before and After Stem Cell Transplant in Treating Patients With Relapsed or Refractory Diffuse Large B-cell Lymphoma — NCT02443077.
• Registry record last updated 2026-07-31; status last verified 2026-06.
• Results first posted 2025-09-30.
• 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.

20. Record Summary

The registry record describes NCT02443077, a randomized phase III study of ibrutinib versus placebo with chemotherapy and autologous stem cell transplantation in relapsed or refractory activated B-cell subtype diffuse large B-cell lymphoma. The record lists triple masking, a crossover model, enrollment of 94, and active, not recruiting status. The sponsor is the National Cancer Institute.

The primary endpoint is 24-month PFS. Posted arm-specific proportions are 0.576 (95% confidence interval 0.421 to 0.788) for ibrutinib and 0.408 (95% confidence interval 0.265 to 0.628) for placebo. The registry record does not report a primary between-arm inferential analysis. Secondary outcomes and adverse-event counts are also recorded, but multiplicity, missing-data handling, crossover analysis, and detailed statistical methods are not reported in the registry record.