A Study of Duvelisib in Participants With Refractory Indolent Non-Hodgkin Lymphoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Duvelisib in Participants With Refractory Indolent Non-Hodgkin Lymphoma (NCT01882803), a na phase 2 study. The primary outcome is Overall Response Rate (ORR).

2. Clinical Question

The study was a Phase 2 evaluation of safety and efficacy in a population described as having disease refractory to prior therapies. The clinical question is therefore framed around observed responses and safety during duvelisib monotherapy in this enrolled group. Without a comparator, the trial does not directly estimate how response or safety would differ under another treatment or under no study treatment.

3. Trial Design

This was a completed Phase 2, single-group study of duvelisib in participants with refractory indolent non-Hodgkin lymphoma. The registry lists enrollment of 129 and describes duvelisib as monotherapy. Eligible disease types named in the record include follicular lymphoma, marginal zone lymphoma, and small lymphocytic lymphoma. The disease was required to be refractory to rituximab and either chemotherapy or radioimmunotherapy.

The design is recorded as single group, with allocation listed as not applicable and masking listed as none. There was no like-for-like control arm in the record. The findings therefore describe participants treated in the study; they do not provide a randomized estimate of a treatment difference. The registry does not report further details about how participants were selected or the specific dosing schedule.

The primary endpoint was overall response rate, assessed locally by investigators. The registry describes response evaluation every 8-16 weeks while participants were on duvelisib, for up to 72 months. The endpoint was based on the best overall response: complete response or partial response counted toward the response measure. The primary analysis population was the Full Analysis Set, defined as all participants who received at least 1 dose of duvelisib.

Other listed outcomes include treatment-emergent adverse events, duration of response, progression-free survival, overall survival, plasma concentration of duvelisib and IPI-656, and time to response. Numerical results for these outcomes are not reported in the registry record. The study start date is 2013-06-17, primary completion is recorded as 2020-11-18, and the overall status is completed. Further information on follow-up completeness, treatment exposure, and the analysis methods for secondary outcomes is 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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (FAS): all participants who received at least 1 dose of duvelisib.

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: Full Analysis Set (FAS): all participants who received at least 1 dose of duvelisib.

6. Primary Endpoint

The registry lists the primary outcome as Overall Response Rate (ORR), measured over Every 8-16 weeks while on treatment with duvelisib for up to 72 months.

Overall response rate measures the percentage of participants whose best overall response was either complete response or partial response. A complete response and a partial response are distinct categories in the response criteria; the endpoint combines them into a single response proportion. The registry says response was assessed locally by the investigator according to the revised IWG Response Criteria for Malignant Lymphoma.

As a binary outcome, each participant contributes to whether the defined response occurred or did not occur within the endpoint definition. The resulting percentage describes the observed frequency of response in the analysis population. The registry specifies a two-sided 95% exact confidence interval to accompany the estimate. The interval communicates statistical uncertainty around the observed proportion; it does not account for every possible source of bias, and it does not turn a single-group result into a comparison with another treatment.

Response rate is used to summarize antitumor activity in a form that is relatively direct to interpret: it records whether a participant met a response category, rather than how long a response lasted. Duration of response is separately listed as a secondary outcome, as are time to response, progression-free survival, and overall survival. These outcomes address different aspects of the course of disease and should not be treated as interchangeable with the response proportion.

The registry specifies assessment every 8-16 weeks while on treatment with duvelisib for up to 72 months. The numerical overall response estimate and its confidence interval are not reported in the registry record, so the observed response frequency and its precision cannot be described from this record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical interpretation starts with the endpoint’s scale and definition. Overall response rate is a binary proportion: participants either meet the definition of response, based on a best overall response of complete response or partial response, or they do not. The registry specifies local investigator assessment according to revised IWG Response Criteria for Malignant Lymphoma. This definition identifies the event being counted, but the numerical count and percentage are not reported in the registry record.

What does a confidence interval describe?

General statistical explanation: a confidence interval expresses uncertainty around an estimate under the assumptions of the statistical method. For a response proportion, the interval reflects uncertainty associated with observing responses in a finite group of participants. A narrower interval generally indicates greater statistical precision, while a wider interval indicates more uncertainty. The interval does not correct for systematic differences, assessment bias, or limitations created by a single-group design.

The registry specifies a two-sided 95% exact confidence interval. An exact interval is a method for estimating uncertainty around a binomial proportion without relying on a normal approximation to the distribution of that proportion. The record does not give the endpoint estimate or interval limits, so their width and the precision of the observed response rate cannot be assessed. The confidence level describes the method’s long-run coverage properties under its assumptions; it is not the probability that a particular interval contains the true value after it has been calculated.

Why does the analysis population matter?

The Full Analysis Set is defined as participants who received at least 1 dose of duvelisib. This definition anchors the stated primary endpoint population to treatment receipt. The enrollment count is 129, but the number who received a dose and the number included in the endpoint analysis are not reported in the registry record. Nor is it reported how participants with missing or unevaluable response assessments were handled. Those details are needed to understand the denominator used for the observed percentage and whether exclusions could affect interpretation.

General statistical explanation: an analysis population should match the question being asked. A population defined by treatment receipt can describe outcomes among participants who started the study intervention, while other population definitions may preserve different features of the study design. The registry does not report additional analysis-population definitions or a sensitivity analysis for the primary endpoint.

How should a single-group result be read?

A single-group response proportion is descriptive of the study group. Because there is no randomized comparator, it does not directly estimate a difference between duvelisib and another treatment. Randomization is a general method for making comparison groups similar, on average, with respect to measured and unmeasured characteristics. That method is not part of the design recorded. Cross-study comparisons would also be difficult to interpret because eligibility, assessment, follow-up, and participant characteristics may differ; the registry does not provide a controlled comparison to resolve those differences.

The lack of masking is relevant to assessment and reporting. General statistical explanation: when treatment and assessment are unmasked, expectations may influence subjective decisions, even when formal response criteria are specified. Use of defined criteria can support consistency, but it does not by itself show that assessments were independent or blinded. Independent review and other assessment procedures are not reported in the registry record.

What do the safety counts represent?

The serious adverse-event summary reports affected participants relative to participants at risk. The record gives 83 affected and 129 at risk for the duvelisib group. These are participant counts, not event totals: a participant may experience more than one event, but the registry information does not describe event-level counting or recurrence. The listed serious-event percentages describe specific terms, but the record does not provide their event counts or clarify whether a participant could contribute to more than one term. The summary also records 68 deaths without reported timing or attribution details.

General statistical explanation: adverse-event proportions can be influenced by the length of exposure and follow-up, as well as by how events are defined, collected, and classified. A participant-level proportion does not capture how often an event recurred or how long it lasted. Comparing event percentages across studies is especially difficult when populations, exposure, surveillance, and definitions differ. The registry does not report enough detail to make those comparisons reliably.

What can planning calculations establish?

General statistical explanation: sample-size planning calculations describe hypothetical requirements under stated assumptions about the target effect, significance level, allocation, and statistical method. They are not the same as observed results, and they are not observed power. A planning illustration based on an observed effect is not evidence that the trial was designed prospectively around that effect. The registry does not report the trial’s own sample-size justification, nor does it report a full statistical analysis plan or interim-analysis procedures.

The record’s primary outcome description specifies an exact interval for the response proportion. Details about testing hypotheses, adjustment for multiple outcomes, missing-data methods, and prespecified subgroup analyses are not reported in the registry record. Without those details and the endpoint results themselves, a reader cannot assess whether the reported analyses followed a prespecified plan or how uncertainty was handled beyond the stated confidence interval.

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("Duvelisib")
n      <- c(129)
events <- c(77)
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  Duvelisib  129  77  59.7  50.7  68.2  51.1  67.8  51.0  67.8

11. Multiplicity, Missing Data and Analysis Populations

The registry lists overall response rate as the primary outcome and names several secondary outcomes: treatment-emergent adverse events, duration of response, progression-free survival, overall survival, plasma concentration of duvelisib and IPI-656, and time to response. It does not report a multiplicity strategy for interpreting several outcomes. When many outcomes are examined, general statistical explanation is that the chance of at least one apparently unusual finding can increase if each is assessed separately. Adjustment methods or a hierarchy of testing can address that issue, but no such plan is reported in the registry record. The listed outcomes should therefore be read as separate measures, not as a documented sequence of confirmatory tests.

The primary endpoint is specified with a two-sided 95% exact confidence interval. The registry does not report further details about hypothesis testing, prespecified thresholds, subgroup analyses, or interim analyses. It is not possible to infer whether any such procedures were part of the study’s statistical plan.

The primary analysis population is the Full Analysis Set, defined as all participants who received at least 1 dose of duvelisib. Enrollment is listed as 129, but the analyzed population size is not reported in the registry record. Handling of participants with missing response evaluations, incomplete follow-up, withdrawal, or unevaluable disease assessments is also not reported. These omissions leave uncertainty about the endpoint denominator and how incomplete observations may have affected the response estimate.

The record names time-to-event outcomes, including duration of response, progression-free survival, and overall survival, but does not report their numerical results or methods for handling censoring. General statistical explanation: time-to-event analyses need information about when follow-up ends for participants without an observed event, and how those participants are treated in the analysis. That information is not reported in the registry record. The absence of these details limits assessment of both the completeness and precision of the listed outcomes.

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: diarrhoea (7.8%), febrile neutropenia (7.0%), disease progression (7.0%), pneumonia (5.4%), colitis (4.7%).

The safety summary lists 83 participants with serious adverse events among 129 at risk in the duvelisib group. It gives serious-event terms with percentages: diarrhoea, 7.8%; febrile neutropenia, 7.0%; disease progression, 7.0%; pneumonia, 5.4%; colitis, 4.7%; and pancytopenia, 3.1%. These percentages are reported for named serious-event terms. The record does not state whether the terms count participants or events, or whether one participant could appear under more than one term.

A separate adverse-event count is 122 among 129 at risk, but its exact category definition is not reported in the registry record. The summary also records 68 deaths. It does not report when these deaths occurred, their causes, or whether they were considered related to treatment. The serious-event counts and term percentages therefore should not be interpreted as proof of causation or as a comparison with another treatment.

General statistical explanation: adverse-event summaries depend on definitions, surveillance, exposure time, and follow-up. Counts of affected participants do not necessarily convey how many total episodes occurred or how severe and persistent events were. The registry record does not report enough detail to assess those features or to compare safety rates across studies.

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

The registry record describes a Phase 2, single-group evaluation of duvelisib monotherapy in participants with refractory indolent non-Hodgkin lymphoma. It defines overall response rate as the percentage of participants whose best overall response was complete response or partial response, assessed locally by investigators using revised IWG Response Criteria for Malignant Lymphoma. It also specifies a two-sided 95% exact confidence interval. These details clarify the intended endpoint and method of expressing its uncertainty.

The record does not report the numerical overall response rate or its confidence interval. It therefore does not allow the observed response frequency or its precision to be assessed. Although secondary outcomes are listed, their numerical results are not reported in the registry record. The information on safety includes serious adverse-event counts, named serious-event percentages, another adverse-event count, and a death count; it does not establish treatment causation or comparative safety.

The single-group design also shapes what can be concluded. Outcomes in the treated study group are not a randomized comparison against another intervention. Differences that might arise between this group and participants in another study could reflect differences in eligibility, disease features, assessment, follow-up, or other factors. The record does not provide a concurrent control group to separate these possibilities.

The study population was described as having disease refractory to rituximab and either chemotherapy or radioimmunotherapy, with follicular lymphoma, marginal zone lymphoma, and small lymphocytic lymphoma named. Results from this selected group, if reported, would describe that study population and would not automatically apply to people with other disease characteristics. The primary analysis population includes participants who received at least 1 dose, but its size and handling of missing response assessments are not reported in the registry record.

In short, the record establishes the trial question, design, endpoint definition, and selected safety counts. It does not report the primary efficacy result, the other outcome estimates, or enough detail to judge missing-data handling and the analysis plan. The available information supports a careful account of what was measured, but not a numerical conclusion about the primary endpoint or a controlled estimate of treatment effect.

14. Why This Trial Matters Statistically

This trial illustrates the difference between defining an outcome and reporting its result. The registry provides a clear operational description of overall response rate: complete response or partial response as best overall response, determined by local investigator assessment under specified response criteria. It also identifies the intended analysis population and confidence-interval approach. Yet the numerical estimate and interval are not reported in the registry record. Without them, the central efficacy question cannot be answered from the record, regardless of how clearly the endpoint itself is defined.

The design is also important statistically. A single-group study can describe the outcomes observed among participants who received the intervention, but it does not provide a randomized treatment contrast. That distinction helps prevent a descriptive response proportion from being interpreted as an estimate of benefit relative to a control. The open-label design and local assessment add further context for how the endpoint was measured.

The safety information shows why denominators and definitions matter. The record gives 83 participants with serious adverse events among 129 at risk and lists percentages for selected serious-event terms. These are useful descriptive elements, but they do not show event timing, recurrence, attribution, or comparative rates. The recorded death count likewise needs context that the registry does not report.

Finally, the record names multiple outcomes but does not report their results or describe how multiplicity, missing observations, and time-to-event censoring were handled. These are not technical footnotes: they affect how much confidence a reader can place in estimates and how outcomes should be interpreted together. This study is therefore a useful example of reading a trial by separating endpoint definitions, design features, observed data, and unreported analysis details rather than treating them as interchangeable.

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 most direct limitation for interpretation is that the numerical result for the primary endpoint is not reported in the registry record. The overall response rate and its specified two-sided 95% exact confidence interval are therefore unavailable for assessment. Numerical findings for the listed secondary outcomes are also not reported in the registry record.

The design has no control group and is not masked. As a result, the study record does not provide a randomized estimate of a difference between duvelisib and an alternative. The primary endpoint is assessed locally by investigators; whether an independent review was performed is not reported in the registry record. These features limit the conclusions that can be drawn from a response proportion alone.

The registry reports enrollment of 129, but the number included in the Full Analysis Set is not reported in the registry record. Nor does it describe the number of participants with evaluable response assessments, missing assessments, treatment discontinuation, or incomplete follow-up. The handling of missing outcome data is not reported in the registry record, so its possible influence on the estimate cannot be evaluated.

Several details relevant to interpreting safety are also absent. The record does not report event-level counts, recurrence, timing, duration, or attribution for the listed serious events. The category represented by the separate adverse-event count is not defined in the registry record. The timing, causes, and treatment relationship of the 68 deaths are not reported in the registry record.

Finally, the statistical analysis plan, trial-specific sample-size justification, multiplicity procedures, interim analyses, and methods for time-to-event outcomes are not reported in the registry record. These omissions restrict assessment of prespecification, censoring, and the relationship among multiple outcomes.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of Duvelisib in Participants With Refractory Indolent Non-Hodgkin Lymphoma — NCT01882803.
• Registry record last updated 2023-09-07; status last verified 2023-09.
• Results first posted 2018-11-20.
• 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 duvelisib monotherapy in 129 participants with refractory indolent non-Hodgkin lymphoma. The record names follicular lymphoma, marginal zone lymphoma, and small lymphocytic lymphoma and describes the disease as refractory to rituximab and either chemotherapy or radioimmunotherapy. The design was single group, with no masking and no control arm.

Overall response rate was the primary endpoint, defined as the percentage with best overall response of complete response or partial response by local investigator assessment. The Full Analysis Set included participants who received at least 1 dose. The registry specifies a two-sided 95% exact confidence interval, but the primary result and its interval are not reported in the registry record. Numerical results for the listed secondary outcomes are also not reported.

The safety summary lists 83 participants with serious adverse events among 129 at risk, selected serious-event percentages, another adverse-event count of 122 among 129, and 68 deaths. Further details on event attribution and timing are not reported in the registry record. The study started on 2013-06-17, reached primary completion on 2020-11-18, and is recorded as completed.