Study of Alisertib (MLN8237) in Adults With Aggressive Non-Hodgkin's Lymphoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Alisertib (MLN8237) in Adults With Aggressive Non-Hodgkin's Lymphoma (NCT00807495), a na phase 2 study. The primary outcome is Best Overall Response Rate Based on Investigator's Assessment (Applying the IWG 2007 Response Criteria).

2. Clinical Question

The study's stated purpose was to evaluate alisertib's anti-tumor activity in participants with relapsed or refractory non-Hodgkin's lymphoma. The population included several lymphoma subtypes, so the clinical question concerns a heterogeneous group rather than one narrowly defined disease category. The record describes a planned dose and treatment schedule, but it does not report subtype-specific efficacy results.

Because the design was single-group, the primary question is descriptive: what response was recorded among participants meeting the response-evaluable definition? It cannot, on its own, establish how that response compares with an untreated group or another treatment. The record also lists time to progression, progression-free survival, duration of response, and treatment-emergent adverse events as secondary outcomes. Results for these outcomes are not reported in the registry record.

3. Trial Design

This was a phase 2, completed, single-group study sponsored by Millennium Pharmaceuticals, Inc. The registry lists an enrollment of 48 participants and a start date of 2009-02-10, with primary completion on 2011-01-04. Allocation is recorded as NA, and masking as none. There was no randomized control group; the lymphoma categories listed in the adverse-event results describe participant groups, not treatment comparisons.

Participants received alisertib. The regimen description specifies a 50 mg starting dose, taken orally twice daily for 7 days, followed by a 14-day washout in 21-day cycles. Treatment could continue until disease progression or unacceptable treatment-related toxicity, for up to 24 months or longer with Sponsor approval. The registry record does not report details of dose modifications, adherence, or treatment exposure.

The primary outcome was investigator-assessed best overall response rate under the IWG 2007 response criteria. The outcome description defines this as the percentage of participants with complete response or partial response. Assessments were scheduled at baseline and every 2 cycles up to Month 12, approximately 16 cycles; from Cycle 16, every 4 cycles until disease progression; and after end of treatment every 12 weeks for up to 12 Months from last dose, with the stated follow-up extending up to 4 years.

The safety population was defined as all participants who received any amount of alisertib. The response-evaluable population was a subset of that population, requiring at least baseline imaging and one on-study imaging. The registry does not report the number included in either population, nor does it provide participant flow, reasons for exclusion from efficacy assessment, or a primary response analysis. It lists secondary outcomes including time to progression, progression-free survival, duration of response, and treatment-emergent adverse events, but their numerical results are not reported in the registry record.

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: Participants with a diagnosis of Aggressive Non-Hodgkin's Lymphoma received 50 mg alisertib twice daily for 7 days in 21 day cycles. Results are categorized as disease sub-types: Diffuse Large B-cell lymphoma (DLBL), Mantle Cell lymphoma (MCL), Transformed Follicular lymphoma (TFL), Burkitts lymphoma (BL) and Aggressive T-Cell lymphoma (ATL).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety population was defined as all participants who received any amount of alisertib. Efficacy analysis for the response-evaluable population is a subset of the safety population, with participants having a minimum of baseline imaging and one on-study imaging.

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: Safety population was defined as all participants who received any amount of alisertib.

6. Primary Endpoint

The registry lists the primary outcome as Best Overall Response Rate Based on Investigator's Assessment (Applying the IWG 2007 Response Criteria), measured over Baseline and every 2 cycles up to Month 12 (approximately 16 cycles), from Cycle 16 every 4 cycles until disease progression, after end of treatment every 12 weeks for up to 12 Months from last dose (Up to 4 years).

Best overall response rate is a binary outcome summarized as a percentage of participants. Under the registry's definition, a participant counts as having a response if the investigator assessed a complete response or partial response using the IWG 2007 response criteria. Complete response requires disappearance of all evidence of disease; partial response requires regression of measurable disease with no new sites, as specified in the Cheson 2007 criteria.

Using response as an endpoint gives a compact description of tumor response during the assessment period. It differs from a time-to-event outcome: the response rate does not, by itself, describe how long a response lasts or how long participants remain free of progression. The registry lists duration of response, time to progression, and progression-free survival separately as secondary outcomes, reflecting those distinct questions. Their results are not reported in the registry record.

Response assessment depends on the stated criteria, the timing and availability of imaging, and the population included in the analysis. The response-evaluable population required baseline imaging and at least one on-study imaging, so it was not necessarily identical to everyone who received treatment. The record does not report how many participants met that definition or the observed percentage with a complete or partial response. Consequently, the endpoint's definition is available, but its numerical result and precision cannot be assessed from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a binary response outcome: each response-evaluable participant either met the complete-response or partial-response definition, or did not. The study-level response rate would express the proportion meeting either definition as a percentage. The registry record describes this endpoint but does not report its numerator, denominator, or calculated percentage. It also reports no analysis results. Consequently, there is no posted efficacy estimate to interpret statistically.

General statistical explanation: A proportion is a summary of how often a defined outcome occurs in a specified population. Its meaning depends on both parts of the calculation: the count classified as having the outcome and the population used as the denominator. Here, the record defines the response-evaluable population as a subset of the safety population: participants had to receive any amount of alisertib to be in the safety population, and also have baseline imaging and at least one on-study imaging to be response-evaluable. The record does not state how many participants met either definition, so the response denominator cannot be reconstructed.

General statistical explanation: A confidence interval describes uncertainty around an estimate under a statistical model and specified assumptions. For a response rate, an interval would help show the range of values compatible with the observed data, rather than treating the point estimate as exact. No interval for the primary response rate is reported in the registry record. Its width cannot be assessed without the response count and denominator, and an interval should not be inferred from enrollment alone.

General statistical explanation: In a single-group study, a response rate is a description of outcomes observed in the treated group. It does not estimate a difference against a concurrent control group. Without random allocation to a comparator, observed response cannot separate the effect of treatment from other explanations such as the natural course of illness, participant selection, or measurement conditions. This is a general limitation of interpreting a single-group outcome, not a finding about what occurred in this study.

The registry lists allocation as NA, masking as none, and the model as single group. Those design features mean the primary response assessment was not a randomized comparison and was not masked, according to the record. The primary outcome was investigator-assessed. The registry does not report whether independent review was also conducted, how disagreements were handled, or whether assessment procedures included safeguards beyond the stated criteria. Such details should not be assumed.

General statistical explanation: Response criteria make outcome definitions more explicit by setting rules for what counts as complete or partial response. Applying a common definition can improve consistency, but the resulting classification still depends on available imaging, assessment timing, and evaluability. The stated schedule included assessments at baseline and every 2 cycles up to Month 12, approximately 16 cycles; from Cycle 16 every 4 cycles until progression; and after treatment every 12 weeks for up to 12 Months from last dose, with follow-up described as up to 4 years. The record does not report whether all scheduled assessments were completed or how missing assessments affected the endpoint classification.

The registry specifies that a complete response requires disappearance of all evidence of disease and a partial response requires regression of measurable disease with no new sites, using the Cheson 2007 response criteria within the IWG 2007 framework. This tells readers what response meant operationally. It does not reveal the number of participants classified in each response category, how many had stable or progressive disease, or whether any participants could not be classified. Those results are not reported in the registry record.

The listed secondary outcomes address different dimensions of outcome. Time to progression and progression-free survival concern time until specified events, while duration of response concerns the period following a response. Treatment-emergent adverse events concern harms recorded after treatment begins. Each requires its own outcome definition, analysis population, and handling of incomplete follow-up. The registry lists these outcomes but does not report their numerical analyses, event counts, censoring rules, or uncertainty estimates.

General statistical explanation: For time-to-event outcomes, censoring occurs when a participant's event time is not observed during the available follow-up. Statistical methods for such outcomes use the observed follow-up while accounting for participants whose event status remains unknown at the end of observation. The interpretation depends on assumptions about why follow-up ended and on the analysis rules. The registry record does not report those rules for the listed time-to-event outcomes, so no specific method should be attributed to the study.

The record's adverse-event summary is presented by lymphoma category and distinguishes serious adverse events from other events. Counts of participants affected are not the same as counts of all event occurrences: one participant may experience more than one event, and the record does not describe whether the reported category summaries count participants, events, or both for every listed term. The top serious-event terms are reported as percentages, but the registry excerpt does not state the denominator or the detailed coding and counting conventions for those percentages. They should therefore be read as registry-reported figures, without reconstructing an underlying denominator.

The registry record lists no analysis results. It does not report the statistical analysis plan, the method used to calculate any efficacy estimate, confidence-interval procedures, or hypothesis-testing procedures. No adjusted or unadjusted recalculation is available to interpret, and no study-specific statistical method beyond the endpoint definitions and populations stated in the record can be described. General concepts such as proportions, confidence intervals, and censoring help explain how such outcomes are commonly interpreted, but they are not claims about methods used in this trial.

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("Alisertib 50 mg BID Starting Dose (DLBL)", "Alisertib 50 mg BID Starting Dose (MCL)", "Alisertib 50 mg BID Starting Dose (TFL)", "Alisertib 50 mg BID Starting Dose (BL)", "Alisertib 50 mg BID Starting Dose (ATL)")
n      <- c(16, 13, 5, 1, 6)
events <- c(4, 3, 2, 1, 3)
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  Alisertib 50 mg BID Starting Dose (DLBL)  16  4  25.0  7.3  52.4  10.2  49.5  10.3  49.9
ARM  Alisertib 50 mg BID Starting Dose (MCL)  13  3  23.1  5.0  53.8  8.2  50.3  8.4  50.8
ARM  Alisertib 50 mg BID Starting Dose (TFL)  5  2  40.0  5.3  85.3  11.8  76.9  11.8  77.7
ARM  Alisertib 50 mg BID Starting Dose (BL)  1  1  100.0  2.5  100.0  20.7  100.0  15.8  98.7
ARM  Alisertib 50 mg BID Starting Dose (ATL)  6  3  50.0  11.8  88.2  18.8  81.2  18.4  81.6

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary response endpoint and secondary outcomes including time to progression, progression-free survival, duration of response, and treatment-emergent adverse events. It does not report a multiplicity strategy: whether and how the analysis accounted for assessing more than one outcome is not reported in the registry record. The record also does not report a statistical analysis plan, the number or order of any formal tests, or whether any outcome was treated as exploratory.

Missing-data handling is similarly not described. The response-evaluable population required baseline imaging and at least one on-study imaging, but the record does not state how many enrolled participants met that requirement, why any participant might have lacked evaluable imaging, or how missing response assessments were classified. The number included in the safety population is not reported either, although its definition is stated as all participants who received any amount of alisertib.

These omissions affect interpretation because analysis populations and missing-data rules determine which observations contribute to an estimate. A response rate calculated in the response-evaluable group may answer a different descriptive question from one calculated among all treated participants. Without the reported denominator and the missing-data approach, the primary response result cannot be reconstructed or compared across possible analysis populations.

The record does not report interim analyses, stopping rules, or whether the primary endpoint analysis was conducted at a prespecified time. It also does not report how censoring was handled for the listed time-to-event outcomes. These details are not reported in the registry record and should not be inferred from the phase, enrollment, or outcome schedule.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

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: febrile neutropenia (27.3%), neutropenia (20.0%), pneumonia (20.0%), cellulitis (20.0%), osteomyelitis (20.0%).

The serious adverse-event summaries are organized by lymphoma category. The diffuse large B-cell lymphoma group lists 16 affected participants among 22 at risk; the mantle cell lymphoma group lists 4 among 13; the transformed follicular lymphoma group lists 5 among 5; the Burkitt's lymphoma group lists 0 among 1; and the T-cell lymphoma group lists 4 among 7. These are group-specific figures, not randomized comparisons. The groups differ by lymphoma type and do not represent alternative treatment arms.

The registry lists febrile neutropenia at 27.3% among the top serious adverse-event terms. Neutropenia, pneumonia, cellulitis, osteomyelitis, and oropharyngeal candidiasis are each listed at 20.0%. These percentages are registry-reported, and the denominator and detailed counting conventions for the term-level percentages are not stated in the record. They should not be reverse-engineered or treated as estimates for a broader population.

Serious adverse events are distinct from the broader category of other adverse events. The record lists affected counts for both categories by lymphoma group, but these counts do not indicate how many total event episodes occurred. The other-event counts are 22 of 22, 13 of 13, 5 of 5, 1 of 1, and 7 of 7 across the listed groups. Without more detail, these summaries do not establish event severity beyond the serious-versus-other classification or show how long participants were observed.

These safety results describe events recorded in the study; the absence of a comparator means they do not establish comparative safety. The registry record does not report deaths in the adverse-event summary.

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

The record establishes that this completed phase 2 study evaluated alisertib in adults with relapsed or refractory aggressive non-Hodgkin's lymphoma and specified investigator-assessed best overall response rate as its primary outcome. It also supplies the response criteria, assessment schedule, analysis-population definitions, treatment regimen, and safety summaries by lymphoma category. These details make the study question and some of its measurement procedures understandable.

What the record does not establish is the numerical primary response result. No response count, denominator, percentage, or uncertainty interval is reported. The registry also does not report analysis results for the listed secondary outcomes. Therefore, the response magnitude, its precision, and the time-to-event outcomes cannot be judged from this record. A defined endpoint is not evidence of a particular result.

The single-group design also limits what an observed response rate, if reported, could say. A response proportion in participants receiving alisertib would describe outcomes in that group, but without a randomized comparator it would not quantify a treatment difference. The lymphoma categories in the safety summaries do not supply a control group, and their figures should not be interpreted as comparative efficacy or safety results.

The serious adverse-event counts and term-level percentages describe reported events in the study groups. They do not establish whether alisertib caused each event, how risks compare with another treatment, or what would occur in a different population. The record does not report key context such as exposure duration and detailed event-counting conventions, which further limits interpretation of the safety summaries.

In short, the registry record supports a description of the study design, the intended response measure, and the recorded adverse-event summaries. It does not support a conclusion about the size or precision of the primary efficacy result, the secondary efficacy outcomes, or comparative treatment effects. The unreported analysis details and results should remain unknown rather than be inferred from the study's completion status or its enrollment.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of how a clearly named endpoint and an outcome result are different things. The registry defines best overall response rate, specifies the response criteria, and describes when assessments were planned. Yet the numerical response analysis is not reported in the registry record. Readers can identify the intended estimand in broad terms, but cannot assess its observed value or uncertainty.

The study also illustrates why the analysis population matters. The safety population included participants who received any amount of alisertib, while the response-evaluable population additionally required baseline imaging and at least one on-study imaging. Those definitions are not interchangeable. Since the record does not report the numbers in either population or the response denominator, the effect of the eligibility rule on the result cannot be assessed.

The design is relevant to interpretation as well. A single-group study can describe outcomes among treated participants, but it does not provide a randomized control comparison. This makes it especially important to distinguish a descriptive response rate from an estimate of a treatment effect. The multiple lymphoma categories in the record also underline that an overall summary would concern a mixed population; the registry does not report subtype-specific efficacy results.

Finally, the safety entries show why counts need context. The serious adverse-event summaries are separated by lymphoma category, and the listed serious-event terms are expressed as percentages. Without a comparator and without the denominator and detailed counting conventions for those percentages, these figures are descriptive rather than comparative. The record's strengths as a source are the stated design, endpoint definition, and safety counts; its absent analysis results leave the principal efficacy question unanswered in this registry record.

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 central limitation for statistical interpretation is that the registry record does not report the primary response analysis. The response definition and assessment schedule are described, but the number of response-evaluable participants, number with complete or partial response, response percentage, and precision estimate are absent. This prevents readers from evaluating the primary endpoint result.

The study included several lymphoma categories, but no subtype-specific efficacy results are reported. The registry does not provide enough information to assess the distribution of response-evaluable participants across those categories or whether estimates varied by disease type. The 48 enrolled participants should not be treated as the primary endpoint denominator because the response-evaluable population was defined separately and its size is not reported.

The single-group design and lack of masking limit comparative interpretation. There is no randomized control group, and the record does not report a masked response assessment. The registry does not report whether an independent review was used, or the procedures for resolving assessments or handling incomplete imaging. These details cannot be assumed.

The safety summaries have limits of their own. They are arranged by lymphoma category rather than by randomized treatment assignment. The registry does not report exposure-adjusted event rates, detailed event-counting rules, or the denominator for the percentages attached to top serious-event terms. The record's adverse-event summary also does not report deaths.

More broadly, the statistical analysis plan, multiplicity handling, interim analysis procedures, missing-data rules, and analysis methods for the listed secondary outcomes are not reported in the registry record. The record lists time to progression, progression-free survival, and duration of response, but supplies no corresponding numerical results or censoring details. These limitations leave the primary efficacy result and several aspects of outcome interpretation unresolved.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study of Alisertib (MLN8237) in Adults With Aggressive Non-Hodgkin's Lymphoma — NCT00807495.
• Registry record last updated 2018-03-27; status last verified 2018-02.
• Results first posted 2018-03-27.
• 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

The registry record describes a completed phase 2, single-group study of alisertib in adults with relapsed or refractory aggressive non-Hodgkin's lymphoma. Enrollment was 48. The listed study start date is 2009-02-10, and primary completion is recorded as 2011-01-04. The regimen was alisertib 50 mg orally twice daily for 7 days, followed by a 14-day washout in 21-day cycles.

The primary endpoint was investigator-assessed best overall response rate under the IWG 2007 response criteria, defined as the percentage with complete or partial response. The response-evaluable population required baseline imaging and at least one on-study imaging. The registry record does not report the primary response result, its denominator, or an uncertainty estimate.

Secondary outcomes listed include time to progression, progression-free survival, duration of response, and treatment-emergent adverse events. Numerical analyses for these outcomes are not reported in the registry record. Serious adverse-event counts are reported by lymphoma category, and selected serious-event terms are listed with percentages. The design has no randomized comparator, so the recorded outcomes are descriptive rather than comparative.