A Study Evaluating the Safety and Efficacy of Atezolizumab in Combination With Obinutuzumab Plus Lenalidomide in Patients With Relapsed or Refractory Follicular Lymphoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Evaluating the Safety and Efficacy of Atezolizumab in Combination With Obinutuzumab Plus Lenalidomide in Patients With Relapsed or Refractory Follicular Lymphoma (NCT02631577), a non-randomized phase 1, phase 2 study. The primary outcome is Percentage of Participants Achieving Complete Response (CR) at End of Induction (EOI), as Determined by the Independent Review Committee (IRC) Using Modified Lugano 2014 Criteria.

2. Clinical Question

3. Trial Design

This completed study was a Phase Ib/II investigation of atezolizumab in combination with obinutuzumab and lenalidomide in patients with relapsed or refractory follicular lymphoma. Hoffmann-La Roche is listed as the sponsor. The registry reports a start date of 2015-12-31, a primary completion date of 2018-10-23, and an overall status of completed. Enrollment was 38.

The design was non-randomized, unmasked, and parallel. The registry describes an arm receiving obinutuzumab, atezolizumab, and 15 mg lenalidomide, and another receiving the same named agents with 20 mg lenalidomide. Because allocation was not randomized, these groups should not be treated as exchangeable comparison groups. Differences in participant characteristics, treatment decisions, or other factors could contribute to differences in outcomes; a causal dose comparison is not established by the design.

The study summary describes induction treatment followed by maintenance treatment for participants with complete response, partial response, or stable disease at the end of induction. The registry states that no dose-limiting toxicities were observed and that 20 mg lenalidomide was confirmed as the recommended Phase II dose. The efficacy-evaluable population was restricted to participants who received induction at that 20 mg dose. Thus, the primary efficacy endpoint does not describe a pooled estimate across both dose arms according to the population description in the record.

The primary endpoint was complete response at end of induction, determined by an independent review committee using modified Lugano 2014 criteria and PET-CT scans. The timeframe is listed as 6 months, up to the clinical cut-off date of 23 October 2018. Secondary outcomes listed in the record include complete response and objective response assessments using different criteria, reviewers, or imaging bases, best response, adverse events, dose-limiting toxicities, and obinutuzumab serum concentration. The registry record does not report full treatment schedules, detailed eligibility criteria, or a statistical analysis plan.

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: All participants received daily low-dose aspirin (81-100 mg) during lenalidomide treatment and until 28 days after the last dose of lenalidomide. Participants who are unable to tolerate aspirin, who have a history of thromboembolism (TE), and who are at high risk of TE, received warfarin or low-molecular-weight heparin (LMWH).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: As no DLTs were observed, the dose of 20 mg lenalidomide was confirmed as the recommended Phase II dose (RP2D) for lenalidomide. Only participants who received lenalidomide induction at the 20 mg RP2D were included in the Efficacy Evaluable population, hence participants in the Atezo-G-L 15 mg arm were not included in the efficacy analysis.

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 Percentage of Participants Achieving Complete Response (CR) at End of Induction (EOI), as Determined by the Independent Review Committee (IRC) Using Modified Lugano 2014 Criteria, measured over 6 months (up to clinical cut-off date (CCOD) of 23 October 2018).

The primary endpoint is the percentage of participants achieving complete response at end of induction, assessed by an independent review committee using modified Lugano 2014 criteria. The description says response was evaluated with PET-CT scans. The timeframe is 6 months, up to the clinical cut-off date of 23 October 2018.

A complete response is a categorical response outcome: each participant included in the endpoint assessment is classified according to whether the assessment meets the stated criteria. Reporting a percentage gives the proportion of the assessed population classified as having complete response. The value is meaningful only alongside the population definition, the number assessed, the response count, and the approach to participants without an evaluable assessment. Those details are not reported in the registry record.

Using an independent review committee provides an assessment separate from the treating investigator, which can help reduce reliance on a single unblinded investigator's judgment. It does not, by itself, remove limitations arising from the study's non-randomized design or establish a comparative treatment effect. The registry lists a 90% confidence interval for this endpoint, but the response percentage and interval limits are not reported in the registry record. Consequently, the endpoint definition can be explained, but the observed response level and its precision cannot be evaluated from the registry entry.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a percentage endpoint represent?

For a binary outcome such as complete response, each participant with an evaluable assessment contributes to one of two categories: the response criterion was met, or it was not met. The response percentage expresses the number meeting the criterion relative to the population used for that endpoint. It is essential to know both the response count and the denominator. If participants are missing an assessment or are excluded from the efficacy population, the choice of denominator can affect the reported percentage. The registry record does not report the primary endpoint count or denominator.

The population definition is especially consequential in this study. The record says that only participants who received lenalidomide induction at the 20 mg recommended Phase II dose were included in the efficacy-evaluable population. It does not provide a complete account of how many participants were assessed, how many had missing response evaluations, or how exclusions were handled. The percentage cannot be reconstructed or independently checked from the available registry entry.

What does a confidence interval communicate?

A confidence interval describes uncertainty around an estimated quantity under the assumptions of the method used to calculate it. For a response percentage, a narrower interval indicates greater statistical precision, while a wider interval indicates more uncertainty. A confidence interval is not a guarantee that the true value lies within the particular reported limits, nor does it describe the range of individual participant outcomes.

The endpoint lists a 90% confidence interval. The interval limits and the method used to calculate them are not reported in the registry record. Without the estimate and its interval, neither the observed response level nor the precision of that estimate can be assessed. The confidence level alone does not supply the missing result.

Why does the design affect interpretation?

Random allocation is one way to make treatment groups comparable in expectation. This study was non-randomized, and its arm descriptions distinguish lenalidomide dose. As a result, treatment assignment may be associated with participant characteristics or clinical decisions. Even a numerical difference between arms would not automatically represent a causal effect of dose or regimen. The registry itself warns that the arms are not a like-for-like control comparison.

The trial is described as parallel, meaning the arm descriptions refer to separate groups rather than a crossover sequence. It is also unmasked. Masking can reduce the influence of knowledge about treatment assignment on some assessments or decisions, but the record says masking was none. Independent review of the primary endpoint provides an assessment separate from investigators, but it does not turn the study into a randomized or masked trial.

How should several response definitions be understood?

The primary outcome uses an independent review committee, modified Lugano 2014 criteria, and PET-CT at end of induction. Secondary outcomes include complete response or objective response assessed by investigators and/or the independent review committee, criteria described as Lugano 2014, PET-CT, CT alone, and best response during the study. These outcomes can differ because they use different definitions, reviewers, imaging bases, and assessment periods. An outcome using CT alone should not be presumed to yield the same classification as one using PET-CT.

When a study evaluates many endpoints, there is a general statistical concern that the chance of at least one apparently notable result can increase if many separate hypotheses are tested. This is the general issue of multiplicity. It does not mean that every list of endpoints represents a formal family of hypothesis tests. The registry does not report a multiplicity strategy or analysis results for these endpoints.

What does over-dispersion mean for this endpoint?

Over-dispersion generally concerns data whose variability is greater than expected under a specified statistical model. It is often relevant to count outcomes or clustered data, where observations may be more variable than a simple model assumes. The primary outcome is a participant-level response percentage, and the registry record does not describe a model involving over-dispersion. It would not be appropriate to assume a particular model or correction for this study.

What can planning calculations say?

General sample-size planning uses assumptions about the effect to detect, the significance level, allocation, and the statistical approximation. The registry's planning illustration assumptions specify a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation, with effects defined as fractions of the observed effect. For time-to-event outcomes, the stated illustration uses a total number of events; other outcomes are expressed as participants per arm.

These are planning illustrations based on an observed effect, not the trial's own design justification. The registry record does not report the actual illustration values, a sample-size justification for the trial, or observed power. Planning calculations do not establish that a study was adequately powered, and they do not substitute for the estimate and confidence interval from the endpoint analysis.

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("Atezolizumab-G-lena 20mg")
n      <- c(32)
events <- c(23)
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  Atezolizumab-G-lena 20mg  32  23  71.9  53.3  86.3  54.6  84.4  54.5  84.4

11. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple secondary outcomes, including complete response and objective response using different reviewers, response criteria, imaging bases, and assessment periods. It also lists adverse events, serious adverse events, dose-limiting toxicities during Cycle 2, and obinutuzumab serum concentration. The presence of multiple outcomes raises a general question about multiplicity if formal statistical tests are conducted across several endpoints. A prespecified testing hierarchy or adjustment can help control the interpretation of repeated testing, but the registry record does not report a multiplicity strategy or a statistical analysis plan.

The registry record reports no analysis entries. It does not provide a complete set of outcome estimates, confidence intervals, hypothesis tests, or analysis methods for the primary and secondary outcomes. In particular, the primary complete-response percentage and its 90% confidence interval are not reported. There is therefore no basis in this record to determine whether statistical testing was planned, how endpoint families were defined, or whether any adjustment was applied.

Missing-data handling is also not reported in the registry record. The efficacy population is described as including only participants who received induction lenalidomide at the 20 mg recommended Phase II dose, but the record does not give the denominator assessed, the number with unevaluable scans, reasons for missing assessments, or rules for classifying missing response data. These details matter because the analysis population and missing-outcome rules can affect a percentage estimate.

The registry does not report whether the endpoint analysis followed an intention-to-treat, per-protocol, or other named approach. Nor does it give a participant flow accounting for the efficacy analysis population. Conclusions should therefore remain limited to the endpoint definition and population description actually recorded, without assuming unreported analysis choices.

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: abdominal pain (25.0%), epididymitis (25.0%), pneumonia parainfluenzae viral (25.0%), dehydration (25.0%), orthostatic hypotension (25.0%).

The registry reports serious adverse events and other adverse events by the two lenalidomide-dose arms. In the 15 mg arm, 4 participants were listed as at risk for serious adverse events and 2 as affected; 4 were listed as at risk for other adverse events and 4 as affected. In the 20 mg arm, 34 were listed as at risk for serious adverse events and 16 as affected; 34 were listed as at risk for other adverse events and 34 as affected. These are descriptive registry counts, not randomized comparisons.

Deaths are listed as 1 in the 15 mg arm and 7 in the 20 mg arm. The registry record does not report the causes of death or their relationship to treatment. The counts should not be interpreted as evidence that treatment caused the deaths, or as a comparative risk estimate.

The serious-event terms listed include abdominal pain, epididymitis, parainfluenza viral pneumonia, dehydration, orthostatic hypotension, and lung disorder. The associated percentages are 25.0% for each of the first five listed terms and 5.9% for lung disorder. The registry record does not give further context for interpreting these term-specific percentages, including a full event-level account. The counts and percentages describe what the record reports; they do not establish comparative safety or causation.

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

The study addressed a clinically specific question: complete response at end of induction following a combination regimen in relapsed or refractory follicular lymphoma. The registry defines the primary assessment as independent review committee evaluation using modified Lugano 2014 criteria and PET-CT. It also reports that 20 mg lenalidomide was confirmed as the recommended Phase II dose after no dose-limiting toxicities were observed, and that the efficacy-evaluable population included only participants receiving that dose for induction.

However, the registry record does not report the primary response percentage, its denominator, or its 90% confidence interval. Without these results, the observed frequency of complete response and the statistical precision of that frequency cannot be described. An endpoint definition is not an endpoint result, and the stated confidence level does not replace the missing estimate.

The non-randomized design also limits what can be concluded from comparisons between dose arms. The registry explicitly cautions that the arms are not a like-for-like control comparison. The study does not provide a randomized estimate of the relative effect of the 15 mg and 20 mg regimens. Any unadjusted contrast, if encountered elsewhere, would need cautious interpretation and would not by itself establish causation.

The listed safety counts and event terms provide descriptive information but do not establish that an event was caused by a study drug. The record does not provide the relationship of deaths to treatment, nor enough detail to treat the arm counts as comparative risk estimates. The endpoint and safety information should therefore be read as a partial registry account rather than a complete assessment of efficacy or safety.

In general, evidence from an early-phase, non-randomized study can help describe treatment experience and inform further study questions. That general role does not imply a particular clinical benefit. For this record, the absence of the primary result and analysis details prevents a quantitative conclusion about complete response or its uncertainty.

14. Why This Trial Matters Statistically

This study is statistically notable for the way its dose-finding and efficacy descriptions intersect. The record states that no dose-limiting toxicities were observed and that 20 mg lenalidomide was confirmed as the recommended Phase II dose. It then defines the efficacy-evaluable population as participants who received induction at that dose. The choice of analysis population is therefore central to understanding which participants the primary endpoint is intended to describe.

It is also a useful example of why study arms should not automatically be interpreted as controlled comparisons. The trial was non-randomized and unmasked, and the arm descriptions differ by lenalidomide dose. The registry cautions that the arms are not a like-for-like control comparison. Study design, rather than the presence of multiple arm labels, determines whether a contrast supports a causal interpretation.

The endpoint definition illustrates how response results depend on more than the label “complete response.” The primary assessment specifies an independent review committee, modified Lugano 2014 criteria, PET-CT, and end of induction. Other listed outcomes involve other criteria, reviewers, or imaging bases. A result from one definition cannot simply be substituted for another.

Finally, the registry record shows why transparent reporting of denominators, missing assessments, estimates, and confidence intervals matters. It names the primary endpoint and its 90% confidence interval, but does not report the estimate or interval limits. Without those values, readers can understand the intended statistical question but cannot assess the observed response level or its precision.

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 primary limitation for interpreting efficacy is that the registry record does not report the primary complete-response percentage, its denominator, or its 90% confidence interval. It therefore does not permit assessment of the observed response frequency or the precision of that estimate. The registry also does not report results for the listed secondary response outcomes.

The study was non-randomized and unmasked. Its lenalidomide-dose arms are explicitly not described as a like-for-like control comparison. Differences between the arms cannot be attributed to dose alone on the basis of this design. The efficacy-evaluable population description is restricted to participants receiving induction at the 20 mg recommended Phase II dose, but a complete participant flow and denominator are not reported.

Additional endpoint details that could affect interpretation are absent. The registry record does not report how missing or unevaluable response assessments were handled, whether response assessments were conducted consistently across participants, or the statistical method used to calculate the confidence interval. It does not report a multiplicity strategy, a full analysis plan, or a sample-size justification.

Safety interpretation is limited by the descriptive nature of the arm counts and by missing context. The record does not report the causes of deaths or their relationship to treatment. It lists serious-event terms and percentages without a detailed event-level account. These limitations preclude causal conclusions about adverse events or comparative safety from the reported counts alone.

The registry describes a completed Phase Ib/II study, but completion status does not fill gaps in the outcome reporting. Conclusions should be restricted to the design, endpoint definition, population description, and safety information recorded.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study Evaluating the Safety and Efficacy of Atezolizumab in Combination With Obinutuzumab Plus Lenalidomide in Patients With Relapsed or Refractory Follicular Lymphoma — NCT02631577.
• Registry record last updated 2022-04-13; status last verified 2022-04.
• Results first posted 2019-12-30.
• 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

The registry record describes a completed, non-randomized, unmasked, parallel Phase Ib/II study of atezolizumab, obinutuzumab, and lenalidomide in relapsed or refractory follicular lymphoma. Enrollment was 38. The arm descriptions used lenalidomide at 15 mg or 20 mg. The primary endpoint was complete response at end of induction, assessed by an independent review committee using modified Lugano 2014 criteria and PET-CT, with a 90% confidence interval specified.

The record states that no dose-limiting toxicities were observed and that 20 mg lenalidomide was confirmed as the recommended Phase II dose. It says the efficacy-evaluable population included only participants receiving induction at that dose. The primary response estimate, denominator, and confidence interval are not reported in the registry record. The arm comparison is not a like-for-like control comparison. Safety counts are reported descriptively, without reported causes or treatment relationships for deaths.