Rituximab and Pembrolizumab With or Without Lenalidomide in Treating Patients With Relapsed Follicular Lymphoma and Diffuse Large B-Cell Lymphoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Rituximab and Pembrolizumab With or Without Lenalidomide in Treating Patients With Relapsed Follicular Lymphoma and Diffuse Large B-Cell Lymphoma (NCT02446457), a non-randomized phase 2 study. The primary outcome is Overall Response Rate (Complete + Partial Responses).

2. Clinical Question

3. Trial Design

This is a phase II, non-randomized, unmasked, parallel study sponsored by M.D. Anderson Cancer Center. The registry lists enrollment of 53 and status as active, not recruiting. The recorded study start is 2015-07-31, and the primary completion date is 2025-03-31. These dates describe the registry record; they do not establish when all outcome data were collected or analyzed.

The study has two treatment cohorts. Cohort I receives rituximab and pembrolizumab. The registry describes rituximab on days 1, 8, 15, and 22, and pembrolizumab on day 2 every 3 weeks for up to 16 cycles, described as 1 year, in the absence of progression or unacceptable toxicity. Cohort II receives rituximab, pembrolizumab, and lenalidomide. Its rituximab schedule is days 1, 8 and 15 of cycle 1, and day 1 of cycle 2; pembrolizumab is listed on day 2 every 3 weeks for up to 2 years; lenalidomide is listed on days 1-14 every 3 weeks for up to 12 cycles. The Cohort II schedule description ends before its final treatment condition is complete in the registry record.

The study design does not allocate participants at random to the cohorts. Consequently, treatment-group differences could reflect differences in the patients, disease settings, or treatment schedules, as well as the regimens themselves. The record does not report how cohort assignment was determined, whether enrollment criteria differed by cohort, or a protocol-defined adjustment strategy for comparisons. No randomized contrast should be inferred from the parallel-cohort structure.

The primary endpoint is overall response rate. Listed secondary outcomes are toxicity, complete response rate, progression-free survival, and overall survival. Although the registry lists 5 outcome measures, no analysis entries are recorded. The record does not report outcome estimates or an analysis population for these endpoints.

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

Reasons for not completing this period:

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Overall Response Rate (Complete + Partial Responses), measured over Approximately 1 year and 6 months.

Overall response rate (ORR) is the proportion of participants who experience either a complete response or a partial response, as defined by the study's response criteria. The registry expresses the primary endpoint as a count of participants and gives a timeframe of approximately 1 year and 6 months. It does not state the response criteria, the assessment schedule, or the rule for handling participants whose response status is unavailable.

ORR is used in oncology studies as a measure of tumor response during a defined treatment and assessment period. A complete response and a partial response are distinct response categories; combining them gives a broad summary of participants whose disease met either category. This endpoint does not by itself describe how long a response lasts, whether disease later progresses, or how participants fare over longer follow-up. The registry lists progression-free survival and overall survival as secondary outcomes, but does not report their estimates.

The primary-outcome description names separate clinical questions for the two cohorts: relapsed FL treated with rituximab plus pembrolizumab, and relapsed or refractory FL or DLBCL after CAR T cell therapy treated with rituximab, pembrolizumab, and lenalidomide. The registry record does not give a response count or rate for either question. Without those results and their denominators, the endpoint's observed value and its precision cannot be evaluated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Interpreting a clinical trial requires separating the design, endpoint definition, analysis, and uncertainty. The registry record identifies a phase II study and describes an overall response endpoint, but does not report response estimates or analysis entries. Statistical concepts can clarify what such results would mean in general; they cannot replace the missing estimates.

What does an overall response estimate represent?

In general, a binary response endpoint classifies each assessed participant as having a response or not having a response, according to prespecified criteria. ORR combines complete and partial responses. The estimated rate depends on both the number classified as responders and the population used as the denominator. A count without a denominator is not a rate, while a rate without a clearly described analysis population can be difficult to interpret. The registry record does not report the response counts, denominator, response criteria, or analysis population for this endpoint.

In general, a confidence interval describes uncertainty around an estimate under the statistical method used. Its width depends on the amount of information and the observed data, among other features of the analysis. A confidence interval does not guarantee that the true value lies within its limits, and it does not resolve design limitations such as confounding. No confidence interval for ORR is reported in the registry record.

Why does allocation matter?

Random allocation is a method for making treatment groups comparable, on average, with respect to both known and unknown factors. This study is listed as non-randomized, so that protection is not part of its design. Cohort assignment may be related to disease type, prior treatment, or other patient characteristics, but the assignment process and cohort characteristics are not reported. These are possible sources of group differences in general, not claims about what happened in this trial.

The two cohorts also receive different regimens and have distinct descriptions in the primary-outcome text. Accordingly, a comparison across cohorts would mix treatment differences with potential differences in populations and care pathways. A parallel design means the groups are studied during the same overall trial structure; it does not make them exchangeable or create a randomized comparison. The registry record does not state a comparative estimand or an adjusted analysis plan.

What can counts and percentages say about safety?

In general, an adverse-event count describes how many participants experienced an event under the registry's reporting definitions. A denominator gives the number considered at risk or assessed. Such figures do not, by themselves, indicate when an event occurred, how long participants were observed, whether an event was related to treatment, or whether the event led to a treatment change. The record reports serious and other adverse-event counts by cohort, but does not provide these additional details in the information available.

A percentage attached to a serious event term communicates the registry's reported frequency for that term. It should be read alongside its definition and denominator. The registry lists serious-event terms with percentages, but does not give additional event-level detail or explain whether an individual could contribute to more than one term. The percentages therefore should not be added together or treated as mutually exclusive categories.

What is the role of a planning illustration?

In general, a sample-size calculation asks what amount of information might be needed to detect a specified effect under stated assumptions. The registry's planning illustration is described as using the observed effect, a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. For time-to-event outcomes, the stated calculation concerns the total number of events; for other outcomes, it concerns participants per arm. The illustration is not the trial's own design justification, and its assumptions do not show that the study enrolled equal numbers in its cohorts.

The illustration also considers a stated fraction of the observed effect. Such scenarios ask how the planning requirement changes when the effect targeted for detection is smaller than the observed effect; they do not show that the smaller effect was observed. Because the registry record does not report the observed ORR effect or its estimate, the numerical illustration cannot be independently interpreted as a clinically meaningful target. A calculated sample size is a planning quantity, not a measure of study success, and not observed power.

What does the absence of analysis entries imply?

The registry record lists 5 outcome measures but no analysis entries. An endpoint listing identifies what the study intended to assess; an analysis result would additionally report the analyzed population, observed data, statistical estimate, and potentially uncertainty or a hypothesis test. Without those results, it is not possible to determine whether the planned outcome was analyzed, what method was used, or how missing assessments were handled. These details are not reported in the registry record.

In general, results are most informative when they include the estimate, its uncertainty, the denominator, and a clear description of the participants included. For a non-randomized study, they also benefit from transparent description of cohort characteristics and any adjustment methods. These features support a careful account of precision and possible bias, but they cannot turn an observational cohort comparison into a randomized one.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists several outcomes: ORR as the primary endpoint, and toxicity, complete response rate, progression-free survival, and overall survival as secondary outcomes. It does not report a multiplicity strategy, such as a hierarchy of tests or an adjustment for testing multiple outcomes. It is also not reported in the registry record whether the separate cohort-specific ORR questions were treated as separate confirmatory tests. Without that information and without analysis results, no claim about statistical significance across endpoints can be assessed.

The record does not report how missing response assessments were handled, whether participants without an evaluable response were counted as nonresponders, or whether any alternative missing-data analysis was planned. It does not describe the response-analysis population, exclusions after enrollment, or the number of participants contributing data to each endpoint. These omissions matter because the denominator and missingness rules can affect an estimated response rate. No missing-data method should be inferred from the endpoint description alone.

Likewise, the registry record does not report whether analyses followed an intention-to-treat principle, a treated-population definition, or another analysis set. It does not give an interim-analysis plan, monitoring rules, or a statistical analysis plan. These items are not reported in the registry record; their absence from the record does not establish whether they existed elsewhere.

For general interpretation, multiplicity concerns arise when multiple statistical claims are considered together: the chance of at least one false-positive finding can increase if each is judged independently. Missing-data methods address a different issue: how incomplete outcome information is incorporated into an estimate. Neither issue can be evaluated for this study without the relevant analysis details. The available registry information supports description of the design and endpoint, not reconstruction of the unreported analysis plan.

11. 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: febrile neutropenia (neutropenia fever) (21.4%), fever (14.3%), hyperglycemia (7.1%), respiratory failire (7.1%), pneumonitis (7.1%).

The safety record separates serious adverse events from other adverse events and reports counts by cohort. In Cohort I, 7 participants were affected by serious adverse events among 30 at risk; the other-adverse-event count is 30 among 30 at risk, and deaths are listed as 0. In Cohort II, 6 participants were affected by serious adverse events among 14 at risk; the other-adverse-event count is 14 among 14 at risk, and deaths are listed as 2. These are descriptive registry counts, not estimates of a causal difference between regimens.

The registry also lists febrile neutropenia (neutropenia fever) at 21.4%, fever at 14.3%, hyperglycemia at 7.1%, respiratory failire at 7.1%, pneumonitis at 7.1%, and pulmonary embolism at 7.1% among the top serious-event terms. The record does not give event-level timing, duration, severity grades, or treatment attribution for these terms. It also does not establish whether a participant could have experienced more than one event term.

Because allocation was non-randomized and the cohorts have different treatment regimens, their counts should not be read as a controlled safety comparison. The registry does not state that the listed deaths were caused by treatment. Counts and percentages describe what was recorded; they do not, by themselves, establish cause, comparative risk, or the overall balance of benefit and harm.

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

The registry record establishes that investigators studied rituximab plus pembrolizumab, with or without lenalidomide, in a phase II setting for relapsed lymphoma. It specifies a primary endpoint based on complete and partial responses and lists additional outcomes related to response, toxicity, and survival. It also records safety counts for the two cohorts. Those design and reporting details identify the clinical questions being addressed, but they are not the same as evidence of an observed treatment effect.

The record does not report an ORR result, response counts, survival estimates, or inferential analyses. It therefore does not establish the frequency of response, its precision, how long responses lasted, or how outcomes compared between the cohorts. An outcome being named in a registry is not equivalent to a result being reported. The lack of a result in this registry record should not be interpreted as a result of zero or as proof that the endpoint was not assessed.

The non-randomized design also limits what can be inferred from a comparison between cohorts. Cohort I receives a two-drug regimen, while Cohort II receives a three-drug regimen, and the registry describes different clinical settings in the primary-outcome text. If outcome results were available, differences could reflect patient selection or other differences as well as treatment. A direct causal claim about the addition of lenalidomide would require stronger support than an unadjusted difference between these cohorts.

The safety counts show that serious adverse events and deaths were recorded, but they do not identify cause or provide a controlled comparison. The record does not report the event timing, duration of exposure, or treatment attribution. These counts are useful descriptive information, but they are not enough to quantify comparative safety or determine the source of an event.

In general, a statistical result should be read with its endpoint definition, analysis denominator, estimate, uncertainty interval, and design context. Here, the estimates and analysis details are not reported in the registry record. The planning illustration is not an observed outcome and is explicitly distinct from the trial's design justification. Taken together, the registry supports a description of the study structure and recorded safety information, but not a conclusion about efficacy or a causal difference between regimens.

13. Why This Trial Matters Statistically

This study is statistically notable for the questions its design raises about interpreting multiple cohorts without random allocation. A parallel structure can organize different regimens, but it does not by itself make the groups comparable. In this record, the cohorts receive different combinations and the primary-outcome description associates them with different disease and treatment contexts. That makes clear definition of each cohort's target population essential when interpreting any eventual response estimates.

The endpoint also illustrates why a registry description and an outcome report serve different purposes. ORR is defined as complete plus partial responses, but the record does not post response counts or estimates. Without a denominator and uncertainty estimate, a reader cannot judge the observed response frequency or its precision. A named endpoint signals what was to be measured, not what the results were.

The safety information provides counts by cohort and a list of serious-event terms. Those data can support descriptive reading, while also showing why denominators, event definitions, follow-up, and attribution matter. Non-randomized safety counts do not establish that one regimen caused more or fewer events than another. The different group sizes and treatment settings further limit a simple comparison, without changing the fact that the counts are part of the registry record.

Finally, the sample-size illustration is explicitly framed as a planning exercise based on the observed effect, rather than the trial's original design justification. This distinction matters: a planning calculation is conditional on assumptions and is not an observed treatment result or observed power. With no ORR estimate reported, the illustration cannot substitute for outcome reporting. The record therefore offers a useful case in separating study design, descriptive safety data, planned endpoints, and the inferential evidence needed to answer a clinical question.

14. Limitations

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

The registry title refers to rituximab and pembrolizumab with or without lenalidomide in relapsed FL and DLBCL, while the official title specifies a phase II study of rituximab plus pembrolizumab in relapsed FL. The primary-outcome description then identifies a second cohort involving relapsed or refractory FL and DLBCL after CAR T cell therapy. These descriptions indicate the scope recorded in different parts of the registry, but the record does not explain their relationship in greater detail.

The Cohort II treatment schedule is incomplete in the registry record: its final condition is cut off after the statement about absence of disease progression. No completion of that wording should be inferred. The record also does not report how cohort assignment occurred, the characteristics of each cohort, the number assessed for the primary endpoint, or the criteria and schedule used to determine response.

The enrollment figure is 53, while the safety groups list 30 and 14 participants at risk. The registry does not explain how those safety denominators relate to overall enrollment or to outcome-specific analysis populations. No outcome results are reported among the analysis entries, and the record does not give response counts, confidence intervals, survival estimates, or analysis methods.

Important statistical details are not reported in the registry record, including a multiplicity strategy, missing-data rules, analysis populations, interim analyses, and a statistical analysis plan. The record also does not give follow-up detail or event attribution for safety outcomes. These limitations restrict assessment of precision, bias, and causal interpretation; they do not establish that the information was never collected or documented elsewhere.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Rituximab and Pembrolizumab With or Without Lenalidomide in Treating Patients With Relapsed Follicular Lymphoma and Diffuse Large B-Cell Lymphoma — NCT02446457.
• Registry record last updated 2026-10-22; status last verified 2026-10.
• Results first posted 2026-02-27.
• 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.

18. Record Summary

NCT02446457 is listed as a phase II, non-randomized, unmasked, parallel study sponsored by M.D. Anderson Cancer Center. The registry records enrollment of 53, a start date of 2015-07-31, a primary completion date of 2025-03-31, and active, not recruiting status.

The study has Cohort I, treated with rituximab and pembrolizumab, and Cohort II, treated with rituximab, pembrolizumab, and lenalidomide. The primary endpoint is overall response rate, defined as complete plus partial responses, with a timeframe of approximately 1 year and 6 months. Listed secondary outcomes include toxicity, complete response rate, progression-free survival, and overall survival.

The record lists 5 outcome measures but no analysis entries. It reports serious adverse event counts and deaths by cohort, but does not report an ORR estimate, response counts, or inferential results. Because allocation is non-randomized, the cohorts are not a like-for-like control comparison.