Anakinra for the Reduction of CAR-T Toxicity in Patients With Relapsed or Refractory Large B-cell Lymphoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Anakinra for the Reduction of CAR-T Toxicity in Patients With Relapsed or Refractory Large B-cell Lymphoma (NCT04432506), a na phase 2 study. The primary outcome is Incidence of Any Grade Cytokine Release Syndrome (CRS).

2. Clinical Question

The clinical aim described in the registry is to study anakinra’s dose and side effects and whether its use may reduce toxicity associated with axicabtagene ciloleucel in this patient population. That aim is not itself an outcome finding. In particular, a single-group study does not directly estimate how CRS incidence would differ under a separate treatment strategy. Without a randomized comparator, any observed CRS frequency would describe the study group under the recorded regimen, not by itself establish the effect attributable to anakinra.

3. Trial Design

The registry identifies this as a phase 2 study with a single-group model, allocation listed as not applicable, and no masking. It is sponsored by M.D. Anderson Cancer Center. The registered enrollment is 22. The status is completed; the recorded start date is 2020-07-27 and the primary completion date is 2024-05-28. These record details describe the study structure and timeline, but do not give the number of participants included in the primary endpoint analysis.

The treatment sequence includes cyclophosphamide and fludarabine on days -5 to -3, followed by axicabtagene ciloleucel on day 0 and subcutaneous anakinra on days 0-6. The registry describes the chemotherapy as intravenous and specifies infusion durations, as well as intravenous administration for axi-cel. The listed schedule is not a comparison between anakinra and no anakinra: the intervention is a regimen combining these agents.

The primary endpoint is the incidence of any-grade CRS within 30 days after CAR T-cell infusion. The secondary outcomes listed are incidence and duration of different grades of CRS and immune cell-associated neurotoxicity syndrome (ICANS), overall response rate, complete response rate, progression-free survival, and overall survival. The registry record does not report numerical results for these outcomes in the information summarized for this analysis.

The adverse-event record uses Cohort 1 and Cohort 2 labels, but the overall study model remains single group and the record identifies no randomized allocation. Those labels should not be interpreted as randomized treatment arms or as evidence of a controlled comparison. The registry record does not explain how the cohorts relate to one another, or report a statistical comparison between them. Consequently, cohort-level adverse-event counts can be described, but they cannot establish that outcomes differed because of treatment.

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

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 Incidence of Any Grade Cytokine Release Syndrome (CRS), measured over Within 30 days after infusion of CAR T cells.

The primary endpoint counts participants with any grade of cytokine release syndrome (CRS) during the stated window: within 30 days after infusion of CAR T cells. It is a binary outcome at the participant level for that window: a participant either has a recorded event meeting the endpoint definition or does not. The registry specifies that CRS will be tabulated according to NCI CTCAE v5.0 and assessed using Lee 2014 criteria as well as the ASTCT Consensus Grading system.

Counting any-grade CRS provides a measure of how commonly the event occurs under the study regimen during the defined period. It does not, by itself, describe severity, duration, or the timing of onset within the window. Those are distinct clinical features; the registry lists incidence and duration of different grades of CRS and ICANS among the secondary outcomes. A count for the primary endpoint is not reported in the registry record, so the incidence and its precision cannot be described from the record information available for this analysis.

General statistical interpretation: an incidence estimate is meaningful only alongside its denominator and a clear definition of who was assessed. A confidence interval, when reported, describes the uncertainty around an estimate under the method used; it does not remove limitations from the study design or establish a causal treatment effect. The registry record does not report a primary endpoint estimate or confidence interval.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a participant count represent?

For the primary endpoint, the registry specifies a count of participants with any-grade CRS. General statistical explanation: a binary endpoint records whether an event occurred for each participant under a defined outcome rule and observation window. The count is interpreted together with the number of participants assessed. Dividing an event count by its corresponding denominator gives an incidence proportion, but no such calculation can be made from this record because the primary endpoint count and its analysis denominator are not reported.

The outcome definition matters. CRS is to be tabulated under NCI CTCAE v5.0 and assessed using Lee 2014 criteria and the ASTCT Consensus Grading system. Different grading frameworks can classify or grade clinical events differently. Reporting the framework helps readers understand the intended measurement, but it does not reveal how many participants met the endpoint or whether the classifications agreed. Agreement results are not reported in the registry record.

What would an interval estimate tell a reader?

General explanation: a confidence interval expresses statistical uncertainty around an estimate under its calculation method and assumptions. A wider interval indicates less precision than a narrower interval, all else being equal. It does not mean that the true value has a particular probability of lying in the observed interval under the usual frequentist interpretation, and it cannot correct for bias, incomplete follow-up, or the absence of a control group. No primary endpoint interval is reported in the registry record.

A p-value, if reported, would address how compatible observed data are with a specified null hypothesis under the statistical model. It would not measure clinical importance, the probability that a treatment works, or the likelihood that a result is true. The registry record lists 0 analyses. This does not establish that no analyses were conducted; it means that a set of analysis results is not listed in the record information. No primary p-value is reported.

Why is the study design important to interpretation?

General explanation: random allocation can help balance measured and unmeasured prognostic factors between treatment groups, on average. A single-group design has no concurrent randomized control group, so it cannot separate treatment-associated changes from other explanations such as underlying disease, participant characteristics, clinical care, or the effects of other components of the regimen. This is especially important when the treatment sequence includes cyclophosphamide, fludarabine, axi-cel, and anakinra. An observed event rate would be descriptive of the studied regimen and population, not an unconfounded estimate of anakinra’s effect.

The registry lists two cohorts in adverse-event reporting, but gives no randomized allocation and no like-for-like control comparison. General explanation: labels such as cohort do not by themselves imply random assignment, comparable follow-up, or comparable baseline risk. A difference between descriptive counts could reflect cohort composition, follow-up, or other factors. The registry record does not report the cohort definitions or a statistical model comparing them, so the figures should not be treated as a causal contrast.

How should the planning illustrations be understood?

The sample-size planning assumptions specify a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They describe an effect to detect as a stated fraction of the observed effect. For time-to-event planning, the stated method uses the total number of events under Schoenfeld; other planning results are expressed as participants per arm. These are planning illustrations, not the trial’s own design justification and not observed power.

General explanation: sample-size planning asks what study size might be needed under selected assumptions about an effect, variability, allocation, and statistical error rates. Changing those assumptions changes the planning result. A planning scenario does not demonstrate that an observed result has adequate power after the fact, and it is not evidence that the assumed effect is real or clinically important. The registry record does not report the original sample-size justification or a statistical analysis plan. The planning assumptions should therefore be kept distinct from the registered enrollment and from any actual outcome analysis.

Why do event definitions and observation windows matter?

General explanation: an event count depends on what qualifies as an event, who is included in the denominator, how the event is assessed, and how long participants are observed. The primary endpoint names both a 30-day window after CAR T-cell infusion and grading approaches. Those details support interpretation of the intended endpoint, but they do not establish complete follow-up or consistent assessment. The completeness of primary endpoint ascertainment and the handling of missing assessments are not reported in the registry record.

Different outcomes also require different statistical summaries. Binary outcomes such as whether CRS occurred are naturally described by event counts and incidence estimates. Time-to-event outcomes such as progression-free survival and overall survival require information about event times and censoring, meaning the last known follow-up when an event has not been observed. General explanation: censoring methods rely on assumptions about follow-up and the reasons observation ends. The registry lists these outcomes but does not report their estimates, event counts, censoring details, or analysis methods in the information summarized for this 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("Cohort 1", "Cohort 2")
n      <- c(10, 10)
events <- c(10, 9)
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  Cohort 1  10  10  100.0  69.2  100.0  72.2  100.0  71.5  99.8
ARM  Cohort 2  10  9  90.0  55.5  99.7  59.6  98.2  58.7  97.7

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and several secondary outcomes: incidence and duration of different grades of CRS and ICANS, overall response rate, complete response rate, progression-free survival, and overall survival. When a study evaluates multiple outcomes or makes multiple statistical comparisons, the chance of at least one apparently positive result can increase. General statistical practice may address this through a prespecified testing hierarchy or multiplicity adjustment, depending on the aims. The registry record does not report a multiplicity procedure or a statistical testing hierarchy.

The record lists 0 analyses, and the numerical results for the primary and secondary outcomes are not reported in the registry information. This absence does not establish that analyses were never performed. It does mean that the analyses, their prespecification, and any adjustment for multiple testing cannot be assessed from this record.

The registry record does not report the analysis populations for the primary or secondary endpoints, the number assessed for each outcome, or how participants with missing observations were handled. General explanation: excluding participants with incomplete outcome data can affect an estimate if missingness is related to prognosis or outcome; the direction and size of any resulting bias cannot be known without information about the missingness and analysis. The record also does not state whether analyses used an intention-to-treat, per-protocol, or other population definition.

Consequently, the registered enrollment should not be assumed to equal the denominator for every outcome. The completeness of follow-up, treatment discontinuations, and handling of unavailable outcome assessments are not reported in the registry record. These omissions limit assessment of both the reliability of event estimates and the comparability of outcomes across the listed cohorts.

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 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: encephalopathy (20.0%), cytokine release syndrome (20.0%), headache (10.0%), fever (10.0%), paresthesia (10.0%).

The adverse-event record reports Cohort 1 and Cohort 2 separately. For serious adverse events, 6 of 10 participants at risk were affected in Cohort 1 and 4 of 10 were affected in Cohort 2. For other adverse events, 10 of 10 at risk were affected in each cohort. These figures describe recorded event categories; the registry does not report a randomized comparison between the cohorts.

The record lists deaths as 5 in Cohort 1 and 2 in Cohort 2. Cause, timing, and relationship to treatment are not reported in the registry record, so these counts cannot be attributed to the regimen from the information available. The serious-event terms listed include encephalopathy and cytokine release syndrome, each at 20.0%, and headache, fever, paresthesia, and dysarthria, each at 10.0%. The denominator and detailed method for these percentages are not reported. They should not be treated as event counts or as evidence of a difference between cohorts.

General explanation: adverse-event summaries can count participants affected, events, or event terms, and those quantities are not interchangeable. A participant may experience more than one event, and an event term is not necessarily equivalent to a distinct participant. The registry figures should therefore be interpreted using the labels and denominators explicitly reported, without inferring totals or treatment causality.

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

The registry describes a phase 2 study of anakinra given with axi-cel and other treatment components in patients with relapsed or refractory large B-cell lymphoma. Its primary endpoint is any-grade CRS within 30 days after CAR T-cell infusion. The endpoint definition and assessment frameworks are stated, but the primary outcome count and incidence estimate are not reported in the registry record. This means the primary result cannot be quantified or evaluated for precision from the record information.

The adverse-event data provide descriptive counts for Cohort 1 and Cohort 2, along with listed serious-event terms and death counts. They do not establish a treatment contrast. The study is identified as single group, has no randomized allocation, and does not present a like-for-like control comparison. The cohort labels do not change that design. Differences in descriptive figures cannot be assigned to anakinra or another treatment component based on these data alone.

The record’s stated aim includes studying dose, side effects, and potential reduction of CAR T toxicity. An aim is a research question, not a result. Likewise, the presence of secondary outcome names does not provide findings for response, CRS or ICANS duration, progression-free survival, or overall survival. Those numerical outcomes are not reported in the registry record information.

General statistical interpretation: an estimate describes the data collected under a particular outcome definition and analysis population. Its uncertainty, susceptibility to bias, and relevance to other patients depend on factors such as sample size, follow-up, missing data, and design. Here, the registered enrollment is 22, but the primary endpoint analysis denominator and missing-data handling are not reported. The completed status does not fill those gaps. The appropriate conclusion is therefore limited to the study’s stated design, endpoint, and recorded descriptive safety figures, without a claim that anakinra did or did not reduce toxicity.

14. Why This Trial Matters Statistically

This study raises a clinically relevant statistical question: how to characterize CRS when anakinra is administered alongside CAR T-cell therapy for relapsed or refractory large B-cell lymphoma. A defined primary endpoint and a specified observation window are important because toxicity measures depend on consistent event classification and follow-up. The registry identifies any-grade CRS within 30 days after CAR T-cell infusion and names the grading approaches planned for assessment.

The design also illustrates why the estimand—the specific treatment effect a study aims to estimate—must be distinguished from the treatment regimen being observed. In this single-group study, an event frequency would describe outcomes under a regimen including cyclophosphamide, fludarabine, axi-cel, and anakinra. Without a randomized comparator, the study cannot isolate anakinra’s contribution from other treatment components or from differences in patient characteristics and care.

The record also highlights the difference between a registered outcome and a reported result. The primary endpoint and secondary outcomes are named, but numerical primary and secondary outcome estimates are not reported in the registry information. Statistical interpretation depends on those estimates, their denominators, and measures of precision; endpoint definitions alone cannot establish the observed frequency or its uncertainty.

Finally, descriptive safety data require careful reading. Cohort-specific counts may be useful as records of events, but without cohort definitions and a valid comparison they do not show that one regimen or group is safer. In statistical terms, clear reporting of populations, event definitions, follow-up, missing data, and analysis methods is essential for deciding what conclusions the data can support. These are not reported in full in the registry record.

15. Limitations

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

The primary limitation for interpreting efficacy or toxicity is the absence of a reported primary endpoint result. Although the record defines any-grade CRS within 30 days after CAR T-cell infusion, it does not report the number meeting that endpoint, the analysis denominator, or a precision estimate. The registered enrollment of 22 cannot substitute for those details.

The single-group model and lack of randomized allocation limit causal interpretation. The regimen combines chemotherapy, axi-cel, and anakinra, so any observed outcome would not isolate anakinra’s effect. Although adverse-event data are separated into Cohort 1 and Cohort 2, the registry record does not explain the cohort definitions or provide a randomized, like-for-like comparison. The cohort counts should remain descriptive.

Other limitations are incomplete reporting of analysis methods and populations. The statistical analysis plan, sample-size justification, endpoint analysis populations, missing-data handling, and multiplicity approach are not reported in the registry record. The secondary outcomes are listed without numerical estimates or corresponding analysis details. The adverse-event record does not report the cause or timing of deaths, or their relationship to treatment. The denominator and detailed calculation method for the listed serious-event percentages are also not reported.

The registry identifies the study as completed, but completion status does not establish complete outcome ascertainment or make unreported results available. These gaps prevent assessment of the precision and robustness of the primary result and limit what can be inferred from the safety summaries. They also preclude conclusions about treatment effects from the information recorded.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Anakinra for the Reduction of CAR-T Toxicity in Patients With Relapsed or Refractory Large B-cell Lymphoma — NCT04432506.
• Registry record last updated 2025-01-28; status last verified 2024-12.
• Results first posted 2025-01-28.
• 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 is a completed phase 2, single-group study sponsored by M.D. Anderson Cancer Center, with registered enrollment of 22. The record gives a start date of 2020-07-27 and a primary completion date of 2024-05-28. It describes cyclophosphamide and fludarabine before axi-cel, with anakinra administered on days 0-6.

The primary endpoint is any-grade CRS within 30 days after CAR T-cell infusion, assessed using the stated grading frameworks. The registry record does not report a primary endpoint result or numerical results for the listed secondary outcomes. Adverse-event counts are reported separately for Cohort 1 and Cohort 2, but the record does not describe a randomized comparison or establish causal differences between cohorts. Analysis populations, missing-data procedures, multiplicity methods, and the statistical analysis plan are not reported in the registry record.