CD19 /22 CAR T Cells (AUTO3) for the Treatment of B Cell Acute Lymphoblastic Leukemia (ALL): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for CD19 /22 CAR T Cells (AUTO3) for the Treatment of B Cell Acute Lymphoblastic Leukemia (ALL) (NCT03289455), a na phase 1, phase 2 study. The primary outcome is Number of Patients With Grade 3-5 Toxicities Occurring Within the Dose Limiting Toxicity (DLT) Period of AUTO3 Infusion.

2. Clinical Question

The primary question is about the count of participants experiencing a defined level of toxicity in a specified period. It is not a comparison of AUTO3 with another treatment, and the registry record does not report the primary endpoint result. The record therefore identifies the safety outcome the study set out to assess, but does not establish how many participants met that outcome.

The broader study purpose also included evaluating clinical activity. Listed secondary outcomes include event-free survival, relapse-free survival and overall survival, as well as feasibility of generating AUTO3, CD19- and/or CD22-negative relapse, cell expansion and persistence, and duration of B-cell aplasia. Results for those outcomes are not reported in the registry record. Without reported endpoint estimates or a comparator, the record cannot answer how AUTO3 outcomes differ from those under another treatment or usual care.

3. Trial Design

This was a single-arm, open-label, multicentre Phase I/II study of AUTO3. The registry record lists an enrollment of 23 and describes the study population as paediatric and young adult patients with relapsed or refractory B-cell ALL. The study status is completed. The recorded start date is 2017-06-26 and the primary completion date is 2020-05-18.

All participants were assigned to the same intervention rather than randomized between treatment and control groups. The record lists dose groups of 1x10^6, 3x10^6 and 5x10^6 CD19/CD22 CAR-positive T Cells/kg. These groups describe doses within the study, not independent randomized treatment arms. Differences between them would not, by themselves, establish a causal dose effect: the record does not describe randomized allocation or a comparable control group.

Masking is listed as none, so the study was open-label. The primary endpoint population is patients who received at least one complete or partial dose of AUTO3, termed the infused set in the record. Enrollment and the infused-set population are not interchangeable; the number in the infused set is not reported in the registry record.

The primary endpoint is a count of participants with grade 3-5 toxicities during the dose-limiting toxicity period, within 30 days (+/- 3 days) after the last dose. The registry lists 11 outcome measures and 0 analyses. It names several secondary outcomes, including survival and feasibility measures, but does not report their results in the registry record. Further details of the protocol, statistical analysis plan, sample-size justification, or any interim analyses 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: 23 patients were screened, 20 patients leukapheresed and 15 patients were pre-conditioned with cyclophosphamide and fludarabine and infused with AUTO3 at 3 different dose levels.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Patients who received at least 1 (complete or partial) dose of AUTO3 (infused set)

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: Patients infused with AUTO3

6. Primary Endpoint

The registry lists the primary outcome as Number of Patients With Grade 3-5 Toxicities Occurring Within the Dose Limiting Toxicity (DLT) Period of AUTO3 Infusion, measured over Within 30 days (+/- 3 days) after the last dose of AUTO3..

The primary endpoint counts participants who experienced grade 3-5 toxicities during the dose-limiting toxicity period. The stated period is within 30 days (+/- 3 days) after the last dose of AUTO3. The outcome is defined as a count, rather than as a continuous measurement or a time-to-event estimate. Its population is the infused set: patients who received at least one complete or partial dose.

A toxicity endpoint is used to characterize safety over a defined period after treatment. Restricting the endpoint to grade 3-5 events focuses the count on events meeting the stated severity threshold. The time window makes clear which period the endpoint covers; it does not describe events outside that window. General: a count can communicate how many participants experienced an event, but interpretation also depends on the population in which the count was assessed and the completeness of event ascertainment.

The registry record does not report the number meeting this primary endpoint, its denominator, or a statistical estimate such as a confidence interval. It therefore does not show the observed frequency or the precision of that frequency. The listed serious adverse-event figures elsewhere in the record should not be substituted for this endpoint: the record does not establish that those figures use the same event definition, severity grades, analysis population or time window.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes the study as single-arm, open-label and Phase I/II. These design features shape what can be inferred from results. A single-arm study records outcomes in participants who receive the intervention but does not include a randomized control group. Open-label means masking is listed as none. General: without randomized allocation, observed outcomes cannot separate treatment effects from other influences as reliably as a randomized comparison can. The dose groups listed in this record should not be treated as randomized control arms.

What does a count endpoint convey?

The primary outcome is a count of participants with grade 3-5 toxicities in the specified dose-limiting toxicity period. A count gives the number meeting the endpoint, but it needs the size and definition of the analysis population for interpretation. The endpoint population is the infused set, defined as patients who received at least one complete or partial dose. The number in that set, the endpoint count and a corresponding uncertainty interval are not reported in the registry record.

General: when a count is converted to a proportion, the denominator must refer to the population actually assessed. A count without that denominator cannot establish the event frequency. A confidence interval, when reported, describes the range of values compatible with the data under the method used; it communicates uncertainty and is not a guarantee that the true value lies within that range. No primary-endpoint confidence interval is reported in the registry record.

Why does the observation window matter?

The primary toxicity window is within 30 days (+/- 3 days) after the last dose. That definition determines which events are included in the primary endpoint. It does not address events outside the specified window. General: time windows make outcomes more comparable by setting a common period for counting events, but the result still depends on how events were collected and classified. The registry record does not provide further details about event ascertainment or handling of incomplete follow-up.

How should dose-group entries be read?

The record lists three dose groups and, for each, counts at risk and affected for serious and other events, as well as deaths. These are descriptive entries. The study does not list randomized allocation, so the groups do not provide a randomized dose comparison. General: differences in counts across groups can arise from group size, participant characteristics, follow-up, or chance, in addition to any treatment-related difference. A comparative causal interpretation would require design and analysis details not reported in the registry record.

The serious-event terms are accompanied by percentages, but the record does not state their denominators or explain whether a participant could contribute more than one term. General: adverse-event categories may overlap, so percentages for separate terms should not automatically be added. Their interpretation also depends on the period of observation, coding approach and analysis population; those details are not reported in the registry record.

What can the analysis record establish?

The registry lists 11 outcome measures and 0 analyses. It identifies the primary endpoint and secondary outcomes, but does not provide a detailed statistical analysis plan or endpoint estimates. The statistical plan, including any prespecified models, covariate adjustments, missing-data methods or interim analyses, is not reported in the registry record. Those methods should not be inferred from the endpoint names or from the study phase.

The record describes a sample-size illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms and a normal approximation. It says time-to-event illustrations use a total number of events and other illustrations use participants per arm. This is a planning illustration, not the study's own design justification, and no resulting planning values are reported in the registry record. General: planning calculations estimate the sample size needed under stated assumptions; they are not observed power and do not replace uncertainty estimates for results actually observed.

General: over-dispersion occurs when variability in count data exceeds what a simple model assumes. If relevant, it can make standard errors too small and confidence intervals too narrow unless the analysis accounts for the extra variability. The registry record does not report a count model or an assessment of over-dispersion for this endpoint, so it is not possible to say whether either was used.

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("1x10^6 CD19/CD22 CAR-positive T Cells/kg", "3x10^6 CD19/CD22 CAR-positive T Cells/kg", "5x10^6 CD19/CD22 CAR-positive T Cells/kg")
n      <- c(4, 5, 6)
events <- c(4, 4, 6)
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  1x10^6 CD19/CD22 CAR-positive T Cells/kg  4  4  100.0  39.8  100.0  51.0  100.0  47.8  99.5
ARM  3x10^6 CD19/CD22 CAR-positive T Cells/kg  5  4  80.0  28.4  99.5  37.6  96.4  35.9  95.7
ARM  5x10^6 CD19/CD22 CAR-positive T Cells/kg  6  6  100.0  54.1  100.0  61.0  100.0  59.0  99.6

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 11 outcome measures, including the primary toxicity endpoint and secondary measures of survival, relapse, manufacturing feasibility, cell expansion and persistence, and duration of B-cell aplasia. It lists 0 analyses and does not report a statistical analysis plan. The record therefore does not state how multiple outcomes were prioritized or whether any adjustment for multiple testing was planned or applied. This matters because examining many outcomes can increase the chance of at least one apparently notable finding by chance. No multiplicity-adjusted results are reported in the registry record.

Missing-data methods are not reported in the registry record. The record does not describe how participants without complete follow-up, unavailable endpoint assessments, or incomplete adverse-event records were handled. The infused set is defined for the primary endpoint, but its size and the number assessed for the endpoint are not reported. The enrollment of 23 should not be assumed to be the denominator for an endpoint defined in the infused set.

The registry record does not report an interim-analysis plan or statistical stopping rules. It also does not provide enough information to determine whether outcome reporting covers every planned assessment. Those details are not reported in the registry record, so it is not possible to assess their impact on interpretation.

General: missing observations can affect both the estimated result and its uncertainty, particularly if the chance of missingness is related to outcome status. Transparent reporting of analysis populations, missingness and sensitivity analyses helps readers judge the robustness of a result. No such estimates or sensitivity analyses are reported in the registry record.

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: anaemia (33.3%), neutropenia (33.3%), thrombocytopenia (33.3%), encephalopathy (25.0%), febrile neutropenia (20.0%).

The registry lists safety figures for three dose groups. For the 1x10^6 CD19/CD22 CAR-positive T Cells/kg group, it records 4 at risk and 1 affected for serious events, 4 at risk and 4 affected for other events, and 3 deaths. For the 3x10^6 CD19/CD22 CAR-positive T Cells/kg group, it records 5 at risk and 2 affected for serious events, 5 at risk and 5 affected for other events, and 3 deaths. For the 5x10^6 CD19/CD22 CAR-positive T Cells/kg group, it records 6 at risk and 3 affected for serious events, 6 at risk and 6 affected for other events, and 3 deaths.

These entries should be read as descriptive counts, not as a comparison of dose safety. Allocation was not randomized, and the record does not establish that the groups are comparable. It also does not say whether the recorded deaths were related to treatment, or give their causes or timing. The serious-event figures are not the primary endpoint result: the registry does not report whether they use the same toxicity grades, population or time window as the primary endpoint.

The record lists anaemia, neutropenia and thrombocytopenia at 33.3% each; encephalopathy at 25.0%; febrile neutropenia at 20.0%; and pyrexia at 16.7% among its top serious-event terms. The denominators and possible overlap between terms are not reported in the registry record. These percentages therefore should not be added or assumed to represent separate groups of participants.

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

The record establishes that AUTO3 was studied in a single-arm, open-label Phase I/II study for paediatric and young adult patients with relapsed or refractory B-cell ALL. It identifies the primary safety question and its observation window, and it lists several secondary outcomes and dose-group safety entries. Those details describe the study and what it set out to assess.

The record does not report the primary endpoint result. It therefore does not show how many participants in the infused set had grade 3-5 toxicities during the defined period, nor the precision of such an estimate. The listed serious-event counts and percentages provide some descriptive safety information, but the record does not establish their relation to the primary endpoint definition or indicate whether deaths were related to treatment.

The study has no randomized comparator. As a result, the registry record cannot establish whether outcomes differ from those under another treatment, or attribute differences across dose groups to dose. The dose-group entries should not be read as a controlled comparison. The secondary outcomes named in the record, including survival and relapse measures, have no reported results in the registry record.

General: a study's stated purpose and its outcome definitions are not the same as evidence that an outcome was observed at a particular rate. Interpreting a result requires the estimate, its denominator, uncertainty and the analysis methods. Here, the registry record gives important design information but does not report the primary result or detailed statistical plan. It consequently supports a limited description of the study, not a conclusion about efficacy, comparative benefit, or the size and precision of the primary safety outcome.

14. Why This Trial Matters Statistically

This study raises a statistical question that is common in early clinical research: how to describe outcomes from a treatment study when there is no randomized control group. The single-arm design can document events among treated participants, but it cannot provide a like-for-like comparison with an untreated or differently treated group. That distinction is important when interpreting both safety listings and any future clinical-activity results.

The primary endpoint has a defined severity threshold, analysis population and observation window. Those features make the question specific, but the registry record does not report the endpoint count or its denominator. Without those elements, readers cannot assess the frequency or precision of the primary safety result. The record's serious adverse-event entries add descriptive context, while their relationship to the primary endpoint remains unspecified.

The listed secondary outcomes span several kinds of data: binary feasibility and relapse outcomes, survival measures, and measures of cell expansion, persistence and B-cell aplasia. Each type requires careful definition and analysis. A statistical analysis plan would clarify how these outcomes were handled, how missing assessments were treated, and how interpretation across multiple outcomes was controlled. Those details are not reported in the registry record.

More broadly, the record illustrates why reporting the result is as important as registering the research question. Study design and endpoint definitions explain what the investigators intended to evaluate; outcome estimates and measures of uncertainty are needed to understand what the data show. In this record, the stated purpose, endpoint definitions and selected safety listings are available, but the primary outcome result and detailed analyses are not.

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 principal limitation for interpretation is that the primary endpoint result is not reported in the registry record. The endpoint is defined as a count in the infused set, but the record does not state the number who met it, the number assessed, or an uncertainty estimate. The enrollment of 23 does not identify the size of the infused set.

The study is single-arm, with no randomized control group. The three dose groups are not a like-for-like control comparison, so their event counts cannot establish a causal dose relationship. The registry record also does not report enough information about the timing, causes or treatment relationship of the listed deaths to support further interpretation.

Although the record lists serious-event terms and percentages, it does not report their denominators or whether participants may appear under more than one term. It also does not clarify whether these safety entries share the primary endpoint's population, severity threshold or observation window. They should therefore be treated as separate descriptive information rather than as a substitute for the primary endpoint result.

The registry lists 11 outcome measures and 0 analyses, but does not report the statistical analysis plan, missing-data methods, multiplicity approach or interim analyses. Results for the listed secondary outcomes are not reported in the registry record. Further protocol details and the sample-size justification are also not reported. These gaps limit assessment of analysis choices, completeness of outcome reporting and the precision of any treatment-effect conclusions.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: CD19 /22 CAR T Cells (AUTO3) for the Treatment of B Cell Acute Lymphoblastic Leukemia (ALL) — NCT03289455.
• Registry record last updated 2021-02-01; status last verified 2021-01.
• Results first posted 2021-02-01.
• 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 NCT03289455, a completed, single-arm, open-label, multicentre Phase I/II study of AUTO3, a CAR T cell treatment targeting CD19 and CD22. The study concerned paediatric and young adult patients with relapsed or refractory B-cell ALL, with enrollment listed as 23. The recorded start date is 2017-06-26 and primary completion date is 2020-05-18.

The primary endpoint is the count of participants in the infused set with grade 3-5 toxicities within 30 days (+/- 3 days) after the last dose. The registry record does not report that endpoint result. It lists 11 outcome measures, 0 analyses, selected serious adverse-event figures and secondary outcomes, but does not report results for the listed secondary measures. A detailed statistical plan, missing-data approach and multiplicity strategy are not reported in the registry record.