Umbilical Cord Blood NK Cells, Rituximab, High-Dose Chemotherapy, and Stem Cell Transplant in Treating Patients With Recurrent or Refractory B-Cell Non-Hodgkin's Lymphoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Umbilical Cord Blood NK Cells, Rituximab, High-Dose Chemotherapy, and Stem Cell Transplant in Treating Patients With Recurrent or Refractory B-Cell Non-Hodgkin's Lymphoma (NCT03019640), a na phase 2 study. The primary outcome is Treatment-related Mortality Within 30 Days (TRM30).

2. Clinical Question

3. Trial Design

This completed phase II study was sponsored by M.D. Anderson Cancer Center and enrolled 22 participants. The registry describes one treatment group combining cord blood-derived expanded allogeneic natural killer cells with rituximab, high-dose chemotherapy, and stem cell transplantation. The arm description names chemotherapy, NK infusion, and stem cell transplant as the treatment. The trial does not have a separate control arm.

The registry records allocation as NA, masking as none, and the model as single group. In practical terms, the study was not a randomized comparison and did not mask treatment assignment. These design features matter when interpreting outcomes: observations describe participants receiving the treatment combination, but they cannot by themselves establish how outcomes would compare under another treatment or under no treatment.

The listed conditions include mantle cell lymphoma, recurrent or refractory diffuse large B-cell lymphoma, recurrent or refractory follicular lymphoma, and recurrent or refractory indolent adult non-Hodgkin lymphoma. The registry summary describes the study as examining side effects as well as the treatment combination in this population. The primary outcome is treatment-related mortality within 30 days; a secondary outcome is the number of participants who survived.

The registry lists 22 enrolled participants and notes that four did not receive treatment because of screen failures. It does not report a separate analysis population for each outcome, or explain how those participants were handled in outcome analyses. The record gives study start and primary completion dates and marks the study completed, but those dates do not resolve the missing outcome results or analysis details.

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: Four participants did not received treatment due to screen failures.

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 Treatment-related Mortality Within 30 Days (TRM30), measured over Up to 30 days.

The primary outcome is the count of participants with treatment-related mortality within 30 days. It is a binary participant-level outcome: for each participant included in the relevant assessment, the event is whether treatment-related mortality occurred within the specified timeframe. The registry describes the unit as participants and the timeframe as up to 30 days.

A short-term mortality outcome is used to describe a serious safety event associated with treatment. A count can convey how many participants experienced the event, while a denominator clarifies the population in which that count was assessed. Estimates of uncertainty, when reported, help show how imprecisely the underlying event frequency is known. These are general statistical considerations, not details of an analysis documented for this trial.

The registry record does not report the primary outcome count, its denominator for analysis, an event rate, or a confidence interval. It states that four participants did not receive treatment because of screen failures, but does not specify the analysis population used for this endpoint. Consequently, the posted record does not permit a numerical estimate of treatment-related mortality within 30 days.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a single-group phase II study, not a randomized comparison. This design determines what kinds of statistical interpretation are available. A single-group result can describe the outcomes recorded among participants in that group. Without a separate comparison group, it cannot estimate a treatment contrast within the trial. General statistical explanations below distinguish those principles from analyses actually documented in the registry; the record does not report a statistical analysis plan.

What does a participant count tell us?

The primary outcome is a count of participants with treatment-related mortality within 30 days. For a binary outcome, each participant is classified according to whether the specified event occurred. A count describes the number meeting that definition. To interpret it as a frequency in a study population, the denominator—the number assessed for that outcome—also needs to be clear. Here, the record gives an enrollment of 22 and notes four screen failures, but does not state the primary endpoint analysis denominator or its result. It would therefore be inappropriate to infer an endpoint rate from the enrollment information.

Counts for adverse events are also descriptive. An affected-participant count does not necessarily describe the number of separate event episodes: a participant could experience more than one event term, and the registry's summary does not give episode-level detail. The reported adverse-event categories and their percentages should be interpreted according to their recorded labels, rather than assumed to be mutually exclusive.

Why do confidence intervals matter?

In general, a confidence interval expresses the statistical uncertainty around an estimate. A narrow interval indicates greater precision under the assumptions of the method; a wide interval indicates less precision. For a binary outcome, the estimate and interval depend on the observed event count and the population denominator, as well as the interval method. The registry record does not report a confidence interval or an interval method for the primary outcome. No interval should be inferred from the other safety counts.

For a treatment comparison, an interval around a difference or ratio can show the range of effects compatible with the data under the analysis assumptions. This trial has no separate comparator group, so an in-trial randomized treatment contrast is not available. General external comparisons would involve additional assumptions and are not reported in the registry record.

What can the design establish?

Random allocation is a method for creating comparable treatment groups, on average, by chance. Since the record identifies a single group and allocation as NA, that mechanism is absent. Masking is recorded as none. Lack of masking does not change the arithmetic of a participant count, but, in general, knowledge of treatment assignment can affect some assessments or reporting. The registry does not describe whether outcome assessment was standardized or independently adjudicated.

The treatment includes several components: cord blood-derived expanded allogeneic natural killer cells, rituximab, high-dose chemotherapy, and stem cell transplantation. A single-group study of the combination can describe outcomes following that regimen, but cannot separate the effects of its components. The registry record does not describe a component-specific comparison.

What do the planning assumptions mean?

The record includes assumptions for a sample-size planning illustration, not for the trial's original design. It specifies a two-sided significance level of 0.05, equal allocation between two hypothetical arms, and a normal approximation. It describes the effect to detect as a stated fraction of the observed effect. The illustration is therefore a hypothetical planning calculation based on an observed effect; it is not evidence that the trial itself was designed with those assumptions, and it is not observed power.

In general, a significance level sets a threshold used in a specified testing procedure, while a two-sided test considers departures in either direction. Equal allocation is an assumption about how participants would be assigned across hypothetical arms. A normal approximation uses a normal distribution to approximate the sampling behavior of an estimate or test statistic. These assumptions describe the planning illustration only; the record does not state that they were used to analyze the trial.

Planning for a fraction of an observed effect asks how a hypothetical future comparison might be sized for a smaller or larger target effect than the one used as a reference. It does not verify that the effect is reproducible. The registry record does not report the calculated sample-size outputs or the observed primary outcome needed to evaluate the result of such a calculation from the information available.

Which analyses were reported?

The record lists two outcome measures and zero analyses. It does not report a statistical analysis plan, a primary outcome estimate, a confidence interval, a hypothesis test, or a method for handling missing outcome data. The interpretation should therefore stay with the stated design and descriptive results, without attributing an unreported analytical method to the investigators.

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("Treatment (Chemotherapy, NK Infusion, Stem Cell Tr")
n      <- c(18)
events <- c(0)
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  Treatment (Chemotherapy, NK Infusion, Stem Cell Tr  18  0  0.0  0.0  18.5  -0.0  17.6  0.1  17.6

11. Multiplicity, Missing Data and Analysis Populations

The registry lists two outcome measures: treatment-related mortality within 30 days as the primary outcome and the number of participants who survived as a secondary outcome. It reports zero analyses. The registry record does not describe a multiplicity plan, such as whether or how interpretation would account for assessing more than one outcome. It also does not report hypothesis tests or p-values for these outcomes.

Multiplicity is a general statistical concern when several outcomes or comparisons are evaluated. If each is tested separately, the chance of at least one apparently notable result can increase; analysis plans can address this in different ways. That general issue does not establish that any adjustment was required or performed in this study. The relevant plan is not reported in the registry record.

The record notes four screen failures who did not receive treatment, but does not describe the analysis populations for the primary or secondary outcome. It does not specify whether participants were analyzed according to enrollment, treatment received, or another rule. Missing-data definitions, follow-up completeness, and methods for handling unavailable outcome information are also not reported in the registry record.

Without those details, readers cannot determine how the analysis population was defined or whether missing outcome information could affect the reported results. The available safety counts remain descriptive registry entries; they do not establish the analysis approach for the unreported mortality or survival 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: white blood cell decreased (68.2%), neutrophil count decreased (54.5%), platelet count decreased (54.5%), lymphocyte count decreased (40.9%), alanine aminotransferase increased (9.1%).

The registry's treatment group lists 22 participants at risk for serious adverse events, with 15 affected. For other adverse events, it lists 22 at risk and 22 affected. It also records two deaths. These counts refer to the registry's safety categories; they should not be interpreted as the primary treatment-related mortality result, which is not reported.

The listed serious-event terms include white blood cell decreased (68.2%), neutrophil count decreased (54.5%), platelet count decreased (54.5%), lymphocyte count decreased (40.9%), alanine aminotransferase increased (9.1%), and anemia (4.5%). These percentages describe the registry's entries for those terms. The record does not provide confidence intervals or comparative estimates for them.

Adverse-event terms may overlap within participants, so the percentages should not be added together or treated as distinct groups. The registry record does not supply enough detail to determine event duration, severity grades for each term, or whether events were attributed to a particular treatment component. Its summary should be read as descriptive safety reporting, not as a causal comparison.

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

The record establishes that this was a completed, single-group phase II study of a combination involving cord blood-derived natural killer cells, rituximab, high-dose chemotherapy, and stem cell transplantation in recurrent or refractory B-cell non-Hodgkin lymphoma. It reports an enrollment of 22 and provides selected adverse-event counts and percentages. Those details characterize the study and some recorded safety observations.

The record does not report the primary outcome result: the count of participants with treatment-related mortality within 30 days. It also does not report the result of the secondary survival outcome. Without those results, the registry entry cannot quantify either outcome. The note about screen failures does not supply the missing outcome count or the analysis denominator.

Even if outcome counts were available, a single-group study would describe what happened among participants in that group, not provide a randomized estimate of how outcomes differ from an alternative treatment. General statistical reasoning also cautions against attributing observed outcomes to any one component of a multi-component treatment when the components are not separately compared.

The adverse-event percentages are descriptive and do not by themselves establish comparative risk, treatment causation, or the precision of an estimate. No confidence intervals, comparative analyses, or detailed analysis plan are reported in the registry record. The available information therefore supports a limited account of design and recorded safety data, not a complete assessment of the primary endpoint or a causal treatment comparison.

These distinctions concern what the registry reports and what its design can statistically establish. They do not determine how the treatment should be used. The record's stated question, outcomes, and adverse-event summaries should be interpreted within the limits of the missing results and single-group design.

14. Why This Trial Matters Statistically

This trial is statistically notable because it combines a clearly defined short-term safety outcome with a treatment strategy studied in a single group. The primary endpoint—treatment-related mortality within 30 days—is framed as a participant count and a specific timeframe. That makes the intended outcome understandable, but the registry record does not report the count, denominator, or uncertainty estimate needed to interpret it quantitatively.

The design also illustrates a central distinction between describing outcomes and estimating comparative effects. A single-group study can document events among participants receiving the regimen. It cannot use randomization to balance measured and unmeasured characteristics between a treatment group and a control group. Therefore, results from this design do not yield the same kind of causal comparison as a randomized trial.

The combination has multiple treatment components, which creates a further interpretive boundary. Without component-specific groups, an observed outcome cannot be assigned statistically to the NK cells, chemotherapy, rituximab, transplantation, or a particular combination of them. This matters whenever a complex regimen is discussed as one intervention.

The record's adverse-event counts and term-specific percentages demonstrate the value of keeping descriptive safety summaries distinct from the primary endpoint. The two kinds of information answer different questions, and the listed deaths cannot be assumed to represent the primary outcome without a reported relationship between them. The missing outcome results, analysis populations, and statistical plan also show why registry design information alone is not a substitute for a complete report of results.

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 is that the registry record does not report the primary endpoint result, despite identifying treatment-related mortality within 30 days as the primary outcome. It does not report the secondary survival result either. No confidence intervals, p-values, or analysis estimates are reported, and the record lists zero analyses.

The record notes that four participants did not receive treatment because of screen failures, but does not state the denominator or analysis population used for either outcome. It also does not describe missing-data handling, follow-up completeness, or whether all outcome assessments were available. These omissions limit interpretation of both the intended endpoint and the participant counts.

The study is recorded as single group, with allocation NA and no masking. There is no concurrent control group for a like-for-like comparison, and outcomes cannot be attributed to a specific component of the multi-part regimen. The registry does not report an analysis plan, multiplicity handling, or details of outcome adjudication.

For safety, the record gives affected-participant counts and selected percentages, but the summary does not provide event-level detail or comparative estimates. The two deaths are not identified as the primary endpoint events, and the registry does not report their timing or relationship to treatment in this summary. These limits are specific to the information recorded; they do not establish that such details were absent from other study documentation.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Umbilical Cord Blood NK Cells, Rituximab, High-Dose Chemotherapy, and Stem Cell Transplant in Treating Patients With Recurrent or Refractory B-Cell Non-Hodgkin's Lymphoma — NCT03019640.
• Registry record last updated 2023-02-16; status last verified 2022-12.
• Results first posted 2023-02-16.
• 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

NCT03019640 is a completed phase II, single-group study sponsored by M.D. Anderson Cancer Center. It enrolled 22 participants with recurrent or refractory B-cell non-Hodgkin lymphoma and evaluated a treatment combination involving cord blood-derived expanded allogeneic natural killer cells, rituximab, high-dose chemotherapy, and stem cell transplantation. The registry records no allocation and no masking.

The primary outcome is treatment-related mortality within 30 days; the secondary outcome is the number of participants who survived. The registry record does not report either outcome result and lists zero analyses. It notes four screen failures who did not receive treatment. Safety summaries list 15 participants affected by serious adverse events among 22 at risk, 22 affected by other adverse events among 22 at risk, and two deaths. These summaries do not replace the unreported primary endpoint result or provide a randomized comparison.