Alisertib With and Without Rituximab in Treating Patients With Relapsed or Refractory B-Cell Non-Hodgkin Lymphoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Alisertib With and Without Rituximab in Treating Patients With Relapsed or Refractory B-Cell Non-Hodgkin Lymphoma (NCT01812005), a non-randomized phase 2 study. The primary outcome is Best Overall Response Rate (ORR) to Alisertib Alone.

2. Clinical Question

3. Trial Design

This was a phase II, parallel-cohort study with non-randomized allocation and no masking. The registry lists an enrollment of 14 and identifies Ohio State University Comprehensive Cancer Center as the sponsor. The study began on 2013-05-21, had a primary completion date of 2015-10-14, and is marked terminated. The reason for termination is not reported in the registry record.

The listed cohorts both received alisertib. In Cohort A, patients received alisertib by mouth twice daily on days 1–7. Patients unable to achieve complete response after course 4 also received rituximab intravenously on day 1 of courses 5–12. Courses were to repeat every 21 days in the absence of disease progression or unacceptable toxicity.

Cohort B also began with alisertib. Patients achieving stable disease or asymptomatic progressive disease after 2 courses were to receive rituximab on day 1 of courses 3–10. Patients unable to achieve complete response by course 4 were to receive rituximab as in Cohort A. The record does not give a complete description of the remaining course conditions for Cohort B.

Because allocation was non-randomized, observed differences between cohorts cannot be interpreted as if the groups had been assigned to comparable treatment strategies at random. The record does not report how participants were assigned to cohorts, whether assignment depended on clinical characteristics, or whether the cohorts were intended for formal comparative testing. It also does not report cohort-specific enrollment targets or a rationale for the overall enrollment.

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 Best Overall Response Rate (ORR) to Alisertib Alone, measured over Up to 18 weeks (6 courses).

The primary endpoint was best overall response rate (ORR) to alisertib alone, expressed as the percentage of patients. The assessment window was up to 18 weeks (6 courses). Response was defined using RECIST v1.0 for target lesions assessed by MRI. A complete response meant disappearance of all target lesions. A partial response meant at least a 30% decrease in the sum of the longest diameters of target lesions. Overall response combined complete and partial responses.

Response rate is a binary outcome at the participant level: each assessed person is classified as having an overall response or not having one under the stated criteria. The percentage summarizes the share classified as responders, while a confidence interval conveys uncertainty around that estimate. It does not describe how long a response lasted, whether disease later progressed, or overall survival.

The registry record names this endpoint as a response to alisertib alone, while its cohort descriptions include rules for adding rituximab during later courses. The record does not explain how the endpoint was attributed when rituximab had been added within the assessment window. Most importantly, it does not report the number of responders, the estimated response percentage, or the stated 95% confidence interval. The endpoint definition is therefore available, but its observed result cannot be assessed from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record describes a binary response endpoint and specifies a confidence interval, but does not report the estimator, interval method, analysis population, or outcome values. The statistical concepts below explain how such measures are generally interpreted; they do not identify procedures that the investigators used unless the registry record says so.

What does a response percentage represent?

For a binary endpoint, each participant with an evaluable outcome is classified according to whether the defined event occurred. In this study, the event was an overall response, defined as complete response or partial response under RECIST v1.0. The response percentage describes the proportion of the analysis group classified as responders. Its interpretation depends on the group included in the denominator and on the rules for incomplete or unevaluable assessments. Those analysis details are not reported in the registry record.

A response percentage is a summary of observed outcomes, not a guarantee that every participant has the same probability of response. It also does not capture the duration of response. Someone classified as having responded within the assessment window may later have had disease progression; response rate by itself does not answer that later-outcome question. The record lists progression-free survival and overall survival among secondary outcomes, but does not report their results.

What does a confidence interval add?

In general, a confidence interval represents uncertainty around an estimated quantity under a specified statistical procedure. For a response percentage, it gives a range of values compatible with the observed data and the assumptions of that procedure. A narrower interval indicates greater precision under those assumptions; a wider interval indicates more uncertainty. The interval does not fix limitations in study design, correct bias from non-randomized allocation, or show that a treatment caused an outcome.

The registry record specifies a 95% confidence interval for the primary endpoint but does not post the interval limits or name the method used to calculate it. It is therefore not possible to assess the precision of the response estimate or to determine whether a particular interval method was suitable for the data. With no reported response estimate, the confidence interval cannot be interpreted numerically.

Why does the allocation method matter?

Random allocation is a general design approach for reducing systematic differences between treatment groups. In a non-randomized study, differences in patient characteristics, disease features, or clinical decisions may influence which group a participant enters. Those factors can also relate to outcomes. As a result, a comparison between non-randomized cohorts may combine treatment differences with pre-existing group differences.

Here the registry identifies non-randomized allocation and parallel cohorts, but does not describe the assignment process or provide baseline characteristics. The cohorts also had different rules for when rituximab was added. Accordingly, cohort outcomes, if available, would need cautious descriptive interpretation; the design information does not support treating one cohort as an unbiased control for the other.

What can be inferred from the analysis listing?

The record lists 0 analyses. This means the registry entry does not present analysis results in the information described. It does not establish whether analyses were conducted outside the registry, whether data were collected for every endpoint, or why results are absent. The primary response estimate and results for the listed secondary outcomes are not reported in the registry record.

The record also does not report a statistical analysis plan, the handling of missing observations, methods for adjusting for baseline characteristics, or any interim analysis procedures. Without such information, it is not possible to assess whether analyses followed prespecified rules or how potentially incomplete data were addressed. No specific method should be attributed to the trial beyond the endpoint and confidence-interval description explicitly recorded.

How should planning calculations be distinguished from observed results?

In general, a sample-size planning calculation asks how much information a future study might need under stated assumptions. A calculation based on an observed effect is an illustration, not evidence that the study had a particular probability of detecting an effect. The registry describes planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation; it also says that the effect to detect is a stated fraction of the observed effect. These are planning assumptions, not trial outcome results or proof of observed power.

The registry does not report values for the planning illustrations in the registry record for this study. No participant requirement can therefore be quoted from them. More broadly, planning calculations depend on assumptions about the effect and statistical procedure, while actual study interpretation depends on the observed data, the analysis population, and the design. They answer different questions and should not be conflated.

Overall, the endpoint definition allows readers to understand what the study intended to count as response. The absence of analysis results and key analysis details means that the response proportion, its precision, and the robustness of any estimate cannot be evaluated 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("Cohort A (Alisertib, Rituximab)", "Cohort B (Alisertib, Rituximab)")
n      <- c(1, 13)
events <- c(0, 2)
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 A (Alisertib, Rituximab)  1  0  0.0  0.0  97.5  0.0  79.3  1.3  84.2
ARM  Cohort B (Alisertib, Rituximab)  13  2  15.4  1.9  45.4  4.3  42.2  4.7  42.8

11. Multiplicity, Missing Data and Analysis Populations

The registry lists several secondary outcomes, including overall response rate, complete response rate, overall survival, and progression-free survival. Some outcome labels appear more than once in the list, but the record does not explain whether these repetitions correspond to different timepoints, populations, or analyses. It lists 9 outcome measures and 0 analyses; results for the secondary outcomes are not reported in the registry record.

When a study evaluates several outcomes or makes multiple comparisons, the chance of at least one apparently notable result can increase. Statistical approaches can address this issue, but the registry record does not report a multiplicity strategy, a hierarchy of outcomes, or whether any secondary outcomes were intended for confirmatory testing. It is therefore not possible to determine how multiplicity was handled or what inferential status the secondary outcomes were intended to have.

The record specifies a 95% confidence interval for the primary endpoint but does not report its limits or the method used. It does not report an analysis population, the number assessed for response, or rules for missing or unevaluable measurements. It also does not say whether missing outcomes were excluded, classified in a particular way, or addressed through another method. These details are not reported in the registry record.

The enrollment figure does not establish the denominator used for any outcome. Without a reported participant flow, analysis population, or missing-data approach, readers cannot judge whether the primary response summary, if one were posted, represented all enrolled participants or only a subset. The record does not report a statistical analysis plan or explain whether it changed over the course of the study. These gaps limit assessment of both the reliability and the intended interpretation of any outcome analyses.

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: dehydration (15.4%), hypoxia (7.7%), pneumonitis (7.7%), duodenal obstruction (7.7%), anemia (7.7%).

The registry lists serious adverse-event figures by cohort. For Cohort A, the serious adverse-event group lists 1 participant at risk and 0 affected. For Cohort B, it lists 13 at risk and 5 affected. The separate other-adverse-event figures list 1 at risk and 1 affected for Cohort A, and 13 at risk and 10 affected for Cohort B. The record lists deaths as 0 for Cohort A and 10 for Cohort B.

These figures should be read as registry counts, not as proof that treatment caused an event. The record does not provide event-level attribution, timing, severity details beyond the serious-event classification, or an explanation of the death counts. The cohorts were non-randomized, and the figures do not establish a comparative safety effect.

The listed serious-event terms include dehydration at 15.4%, and hypoxia, pneumonitis, duodenal obstruction, anemia, and fatigue at 7.7% each. The registry does not specify in this listing how terms were grouped by cohort, whether participants could contribute more than one event term, or the time period used for the percentages. These percentages should therefore not be treated as rates attributable to a treatment without additional information. The registry record does not report a causal assessment for these events.

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

The record establishes that this was a phase II study of alisertib, with rituximab introduced according to cohort-specific rules, in patients with relapsed or refractory B-cell non-Hodgkin lymphoma. It defines a primary response endpoint using RECIST v1.0 and identifies the assessment window as up to 18 weeks (6 courses). These details clarify the intended clinical outcome and treatment context.

It does not establish how many participants responded. The primary response percentage and its 95% confidence interval are not reported in the registry record. Consequently, the posted information does not show the magnitude or precision of the response outcome. It also does not provide results for the listed secondary outcomes, including overall survival and progression-free survival.

The study's non-randomized allocation means that the cohorts should not be read as randomized treatment groups. Their different rituximab schedules, along with the unreported assignment process and baseline characteristics, prevent a straightforward causal comparison. The record does not show that any difference between cohorts, if one were calculated from other sources, would be due to the treatment schedule rather than other differences.

The safety counts provide descriptive information, but do not establish that the recorded events were caused by alisertib or rituximab. The record does not provide an event attribution analysis or enough detail to interpret the listed deaths as treatment-related. Likewise, the terminated status is a registry status, not an explanation of why the study ended or what effect that had on data collection.

Thus, the registry record documents a study question, endpoint definition, cohort treatment rules, and selected safety counts. It does not provide enough outcome or analysis information to estimate response, determine precision, or support a comparative treatment conclusion. Missing results should not be interpreted as evidence of either benefit or lack of benefit.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of why a clearly defined endpoint and a reported result are both necessary. The registry states what counted as overall response, the imaging assessment framework, and the observation window. Yet it does not post the response estimate or its confidence interval. Without those results, the endpoint definition alone cannot answer how often a response was observed or how uncertain an estimate might be.

The non-randomized parallel-cohort design also makes the treatment question more complex than a simple comparison. Both cohorts received alisertib, but rituximab was introduced under different response- and course-based rules. Cohort differences could reflect those treatment rules, participant characteristics, clinical decisions, or other factors. The registry does not describe assignment or baseline characteristics, so the design cannot be treated as a randomized contrast.

The study also illustrates the distinction between an efficacy endpoint and other outcomes. Overall response assesses tumor response under stated criteria; it does not substitute for response duration, progression-free survival, overall survival, or safety. The record lists several secondary outcomes but does not report their analyses. Separate outcomes address different questions and need their own results and interpretation.

Finally, a terminated status and posted adverse-event counts are not self-explanatory. A reason for termination, analysis details, missing-data handling, and event attribution would help readers assess what was observed and how confidently it can be interpreted. Their absence leaves important uncertainty about both the evidence available and its limitations. The statistical value of the registry entry is therefore chiefly in documenting design and endpoint definitions, while also making clear which outcome conclusions remain unavailable.

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.

Several limitations specific to this study are apparent from the registry record. The allocation was non-randomized, and the process for assigning participants to Cohort A or Cohort B is not reported. The cohorts also followed different rules for adding rituximab, so they do not represent a randomized comparison of otherwise identical treatment strategies. Baseline characteristics that might help assess cohort comparability are not reported in the registry record.

The primary endpoint is labeled as best overall response rate to alisertib alone, but the cohort descriptions permit rituximab during the stated assessment period under specified conditions. The record does not explain how response was attributed if rituximab had been added. It also does not report the number assessed, the primary response estimate, the confidence interval limits, or the analysis population.

The registry lists the study as terminated but does not report a reason. It does not explain whether termination affected follow-up, planned enrollment, or outcome collection. The overall enrollment is reported, but cohort-specific enrollment targets and a sample-size justification are not reported in the registry record.

Details that would aid evaluation of endpoint reliability are also absent, including the handling of missing or unevaluable scans, the assessment process, and any independent review procedures. The record does not report a statistical analysis plan, multiplicity approach, interim analyses, or analysis of secondary outcomes. The adverse-event listing lacks information on event attribution and the context for deaths. These gaps do not demonstrate that procedures or analyses were absent; they mean that the relevant details are not reported in the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Alisertib With and Without Rituximab in Treating Patients With Relapsed or Refractory B-Cell Non-Hodgkin Lymphoma — NCT01812005.
• Registry record last updated 2018-06-06; status last verified 2018-05.
• Results first posted 2018-06-06.
• 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

This phase II study evaluated alisertib alone and in cohorts with rules for adding rituximab in patients with relapsed or refractory B-cell non-Hodgkin lymphoma. It used non-randomized allocation, a parallel-cohort model, and no masking. The registry lists an enrollment of 14 and marks the study terminated.

The primary endpoint was best overall response rate to alisertib alone, assessed under RECIST v1.0 by MRI through 18 weeks (6 courses), with a 95% confidence interval specified. The registry record does not report the response estimate, interval limits, or results for the listed secondary outcomes. It lists serious and other adverse-event counts by cohort, but does not establish event causation. The reason for termination, analysis populations, missing-data methods, and statistical analysis plan are not reported in the registry record.