Study of Nivolumab in Patients With Relapsed or Refractory Diffuse Large B-Cell Lymphoma (DLBCL) That Have Either Failed or Are Not Eligible for Autologous Stem Cell Transplant (CheckMate 139): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Nivolumab in Patients With Relapsed or Refractory Diffuse Large B-Cell Lymphoma (DLBCL) That Have Either Failed or Are Not Eligible for Autologous Stem Cell Transplant (CheckMate 139) (NCT02038933), a na phase 2 study. The primary outcome is Objective Response Rate (ORR) Per Independent Radiologic Review Committee (IRRC) Assessment.

2. Clinical Question

3. Trial Design

This was a phase 2, single-arm, open-label study of nivolumab in participants with relapsed or refractory diffuse large B-cell lymphoma. The registry lists 121 participants enrolled. The study began on 2014-03-05, had a primary completion date of 2016-04-08, and is marked completed. These dates describe the recorded study timeline; they do not supply a treatment-effect estimate.

The study included participants after failure of autologous stem cell transplant and participants who were not candidates for transplant after failure of at least two prior multi-agent chemotherapy regimens. Nivolumab was administered intravenously at 3 mg/kg every 2 weeks, until progression or unacceptable toxicity. The primary endpoint population is described as all treated participants. The adverse-event information lists 121 participants at risk, but the primary endpoint denominator and response count are not reported in the registry record.

A single-group design can describe outcomes observed during treatment, but it does not estimate a randomized contrast against a concurrent control. Open-label conduct also means masking was not used. The registry does not report a protocol rationale for the design or a sample-size justification, so those should not be inferred from the recorded enrollment.

The primary endpoint was objective response rate based on independent radiologic review. Secondary outcomes listed include duration of response, complete and partial remission rates and their durations, progression-free survival, and objective response rate by investigator assessment. Numerical results for these outcomes 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: 121 participants entered the treatment period.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All treated participants

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: All treated participants

6. Primary Endpoint

The registry lists the primary outcome as Objective Response Rate (ORR) Per Independent Radiologic Review Committee (IRRC) Assessment, measured over From first dose until date of documented disease progression or subsequent therapy, whichever occurs first (assesed up to April 2016, approximately 25 months).

The primary endpoint was objective response rate by independent radiologic review committee assessment. It is defined as the percentage of participants whose best overall response was complete remission or partial remission, using the 2007 revised International Working Group criteria for malignant lymphoma. Complete remission means disappearance of all evidence of disease, confirmed by PET scan. Partial remission means regression of measurable disease without emergence of new sites.

Response rate is a binary-outcome summary: each participant is classified according to whether the response definition was met, and the endpoint expresses the proportion meeting it. In general, this kind of endpoint can describe how often a prespecified response was observed in the assessed population. Independent review provides an assessment separate from investigator assessment; the registry also lists investigator-assessed objective response rate as a secondary outcome.

The stated timeframe runs from first dose until documented disease progression or subsequent therapy, whichever occurs first, assessed up to April 2016, approximately 25 months. The registry gives the endpoint definition and a 95% confidence interval as its dispersion measure, but the response count, percentage, and interval are not reported in the registry record. Consequently, the endpoint's observed value and precision cannot be interpreted from the posted results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a phase 2, single-group study and specifies objective response rate as the primary endpoint. It states that responses were assessed by an independent radiologic review committee and defined using the 2007 revised International Working Group criteria for malignant lymphoma. The endpoint is a percentage, with a 95% confidence interval named as the measure of dispersion. The registry does not report the response estimate, its interval, or the statistical procedure used to calculate that interval. The details below explain the general statistical ideas relevant to this endpoint; they should not be read as a description of an unreported analysis plan.

What does a response percentage represent?

For a binary endpoint, each participant is classified into one of two categories for the endpoint: meeting the response definition or not meeting it. The response percentage summarizes the share classified as a complete or partial remission among the applicable analysis population. This is a general explanation of how a binary percentage is interpreted. The response count and denominator for this study are not reported in the registry record, so no study-specific percentage can be reconstructed.

Assessment rules matter because they determine how a participant is classified. Here, the record defines complete remission as disappearance of all evidence of disease confirmed by PET scan, and partial remission as regression of measurable disease without new sites. It names independent review as the basis for the primary endpoint and investigator assessment as a separate secondary outcome. The record does not describe how disagreements, unassessable cases, or missing evaluations were handled.

What does a confidence interval add?

In general, a confidence interval gives a range of values compatible with an estimate under a specified statistical method and its assumptions. It expresses sampling uncertainty; it is not a guarantee that the underlying value lies in a particular range, nor does it correct for bias in study design or assessment. A narrower interval generally indicates greater precision than a wider interval, while the interval's interpretation depends on how it was calculated. The registry names a 95% confidence interval but does not post its limits or describe the calculation method. Its statistical precision therefore cannot be evaluated.

For a response percentage, the interval calculation can depend on the method chosen, particularly when the number of observed responses or the denominator is limited. Because neither the observed count nor the denominator is available in the record, a recalculation from summary statistics is not possible. No recalculated result should be substituted for the sponsor's analysis; in any case, the registry does not post a sponsor response estimate to compare with a recalculation.

What can a single-group design estimate?

In general, a single-group study can describe outcomes among the participants treated and assessed under its study conditions. It does not create a concurrent randomized comparison. As a result, a response proportion from such a study is not a randomized estimate of the difference between nivolumab and another intervention. Natural disease course, participant selection, prior treatment, and assessment context can affect the observed proportion. The record lists allocation as not applicable and the model as a single group; it does not report a comparator arm.

The open-label design means masking was not used. Independent radiologic review is specified for the primary endpoint, while investigator assessment is listed for a secondary response endpoint. In general, independent review can provide a separate assessment process, but it does not make the overall study randomized or remove all possible sources of bias. The record does not describe review procedures, adjudication, or whether the independent assessors were masked to other information.

What analysis details are unavailable?

The registry does not report the statistical analysis plan, the confidence-interval method, a hypothesis test, a threshold for interpreting the response rate, or a sample-size justification. It also does not report interim analyses. These details cannot be inferred from the phase designation, the endpoint, or the enrollment count. The record lists zero analyses, and no numerical primary endpoint result is posted. Accordingly, there is no basis for describing a reported p-value or for claiming that a statistical test demonstrated a particular conclusion.

The listed secondary outcomes include duration of response, complete remission rate, duration of complete remission, partial remission rate, duration of partial remission, progression-free survival, and investigator-assessed objective response rate. The record does not provide numerical results for these outcomes or explain how their analyses were prioritized. In general, assessing multiple outcomes can raise questions about multiplicity: when many hypotheses are evaluated, the chance of at least one apparently notable result may increase unless the analysis plan addresses that issue. That general concern does not establish which methods, if any, were used in this study.

Likewise, the record does not specify censoring rules for time-to-event outcomes, missing-data methods, sensitivity analyses, or a per-protocol population. These details can affect how secondary results are estimated and interpreted. Since the relevant results and methods are not reported in the registry record, the appropriate conclusion is limited to the design and endpoint definitions that are stated.

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("ASCT-failed", "ASCT-ineligible")
n      <- c(87, 34)
events <- c(9, 1)
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  ASCT-failed  87  9  10.3  4.8  18.7  5.5  18.5  5.6  18.5
ARM  ASCT-ineligible  34  1  2.9  0.1  15.3  0.5  14.9  0.7  14.9

11. Multiplicity, Missing Data and Analysis Populations

The registry lists objective response rate by independent radiologic review as the primary endpoint and several secondary outcomes, including response duration, complete and partial remission measures, progression-free survival, and investigator-assessed objective response rate. It lists eight outcome measures and zero analyses. The numerical results and a statistical analysis plan are not reported in the registry record. It is therefore not possible to determine whether outcomes were tested in a specified order, whether any adjustment for multiple testing was planned, or how conclusions were intended to account for multiple endpoints.

As a general statistical issue, evaluating several outcomes can increase the chance of finding at least one result that appears notable by chance. Prespecified prioritization or adjustment methods can address aspects of that concern, but the record does not say whether such methods were used. No inference about the study's actual multiplicity procedure should be made.

The primary endpoint population is described as all treated participants, but the registry does not provide the endpoint denominator or explain the handling of missing, incomplete, or unassessable response evaluations. It also does not report missing-data methods, sensitivity analyses, or exclusions from the efficacy assessment. Those details are important because the endpoint is a percentage whose interpretation depends on both classification and denominator. The enrollment and adverse-event at-risk counts do not establish how many participants contributed to the primary response estimate.

The registry also does not specify the analysis populations for the secondary outcomes or the handling of censoring for time-to-event measures. Consequently, the available information supports neither a detailed account of missing-data treatment nor a conclusion about the robustness of any efficacy analysis.

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: malignant neoplasm progression (24.8%), diffuse large b-cell lymphoma (6.6%), pneumonia (5.8%), hypercalcaemia (5.0%), abdominal pain (4.1%).

The adverse-event record lists 121 participants at risk for serious events and 83 affected in the serious-event category. It separately lists 121 at risk and 109 affected in the other-event category, along with 98 deaths. These are registry counts, not comparative estimates. The record does not provide a control group, so the counts cannot establish how event occurrence differed from another treatment or from what would have happened without nivolumab.

The listed serious-event terms include malignant neoplasm progression at 24.8%, diffuse large B-cell lymphoma at 6.6%, pneumonia at 5.8%, hypercalcaemia at 5.0%, abdominal pain at 4.1%, and acute kidney injury at 4.1%. These percentages describe terms in the recorded serious-event information. They should not be read as proof that the study drug caused the events. The registry information does not provide event-level timing, duration, severity detail beyond the serious-event classification, or attribution to treatment.

The death count is not accompanied by a cause or a treatment-attribution breakdown in the record. It therefore should not be interpreted as a count of deaths caused by nivolumab. Safety interpretation requires the study context and event definitions, and a single-group record cannot supply a randomized safety comparison.

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

The study is described as an investigation of nivolumab in participants with relapsed or refractory diffuse large B-cell lymphoma after autologous stem cell transplant failure, or in those who were not transplant candidates after failure of at least two prior multi-agent chemotherapy regimens. Its primary question focused on objective response assessed by an independent radiologic review committee using the stated lymphoma criteria. This establishes what the study set out to measure, but the posted record does not give the response result.

Because the primary response count, percentage, and 95% confidence interval are not reported in the registry record, the observed response frequency and its precision cannot be described. The listed secondary outcomes likewise have no numerical results in the record. A reader therefore cannot use this registry entry alone to quantify response, response duration, or progression-free survival.

The single-group design also limits what any observed response rate could establish. In general, an outcome observed among treated participants does not provide a randomized estimate of how outcomes would differ under another treatment or a control condition. The registry lists no concurrent comparator, and no comparative efficacy conclusion is supported by the study design information.

The record does provide selected safety counts and serious-event terms. These describe reported events in the study but do not establish causation or comparative risk. The listed death count is not accompanied by a cause or attribution breakdown, so it cannot be treated as a treatment-related death count.

In practical statistical terms, the record documents the study question, treatment schedule, endpoint definition, timeframe, and selected safety information. It does not provide the primary efficacy estimate, its confidence interval, or detailed analysis methods. The result therefore does not establish either a positive or negative efficacy finding from the information recorded. It identifies what was measured and the limits of what can be concluded from the posted results.

14. Why This Trial Matters Statistically

This study illustrates why an endpoint definition and a numerical result serve different purposes. The registry clearly defines objective response rate by independent radiologic review, including the remission criteria and the observation timeframe. Yet without the response count, percentage, and confidence interval, the endpoint cannot be used to assess the observed response level or its precision. A well-described question does not replace the estimate needed to answer it.

The design also matters. A single-group study can characterize outcomes among treated participants, but it cannot provide a randomized comparison with another treatment. This distinction applies even when the endpoint is assessed by an independent review committee. Assessment independence and treatment allocation answer different methodological questions: the former concerns how outcomes are evaluated, while the latter determines whether a concurrent randomized contrast is available.

The record also demonstrates the value of examining the information accompanying an endpoint. Analysis populations, missing assessments, interval methods, and treatment of multiple outcomes affect interpretation, yet they are not reported. The registry names secondary outcomes and a 95% confidence interval but does not provide their numerical results or the associated analysis details.

Finally, safety counts require careful framing. The record reports serious-event and other-event counts, selected serious-event terms, and deaths, but these are not comparative causal estimates. This distinction helps prevent descriptive event reporting from being mistaken for proof of treatment attribution. The statistical significance of this study record is therefore methodological: it shows how design, measurement, uncertainty, missing reporting, and causal interpretation constrain what can be concluded.

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 main limitation for statistical interpretation is that the registry does not report the primary endpoint estimate or its 95% confidence interval. The response count and denominator are also absent. As a result, neither the magnitude nor the precision of the primary response result can be assessed from the record. Numerical results for the listed secondary outcomes are not reported either.

The study was single group, with no concurrent comparator. This prevents a randomized treatment comparison and limits causal interpretation of outcomes observed during treatment. The open-label design did not use masking, although the primary endpoint was assessed by an independent radiologic review committee. The record does not describe the committee's procedures, adjudication, or masking arrangements.

The registry describes the primary endpoint population as all treated participants, but does not state the response denominator or how missing and unassessable evaluations were handled. It also does not provide a statistical analysis plan, sample-size justification, confidence-interval method, multiplicity approach, or interim-analysis details. Those omissions leave important aspects of the planned and completed analysis unclear.

The adverse-event information is descriptive and lacks a concurrent comparison. The record does not provide event attribution, event timing, or a cause and attribution breakdown for the listed deaths. Accordingly, those counts cannot establish treatment causation or comparative safety. These limitations concern what can be inferred from the registry record and do not, by themselves, determine the results of analyses that are not posted.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study of Nivolumab in Patients With Relapsed or Refractory Diffuse Large B-Cell Lymphoma (DLBCL) That Have Either Failed or Are Not Eligible for Autologous Stem Cell Transplant (CheckMate 139) — NCT02038933.
• Registry record last updated 2021-10-14; status last verified 2021-10.
• Results first posted 2017-05-30.
• 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.

19. Record Summary

This completed phase 2 study evaluated nivolumab in a single group of participants with relapsed or refractory diffuse large B-cell lymphoma. The registry lists 121 enrolled participants and an open-label design without a comparator. Nivolumab was administered intravenously at 3 mg/kg every 2 weeks until progression or unacceptable toxicity.

The primary endpoint was objective response rate by independent radiologic review, defined by complete or partial remission under the 2007 revised International Working Group criteria. The endpoint timeframe ran from first dose until progression or subsequent therapy, whichever occurred first, with assessment up to April 2016, approximately 25 months. The primary response estimate and 95% confidence interval are not reported in the registry record.

Several secondary outcomes are listed, but their numerical results and analysis methods are not reported. The adverse-event information includes serious-event counts, selected serious-event terms, and a death count; these are descriptive and do not establish causation or comparative risk.