Post Transplant Donor Lymphocyte Infusion: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Post Transplant Donor Lymphocyte Infusion (NCT00167180), a non-randomized phase 2 study. The primary outcome is Number of Patients Alive.

2. Clinical Question

The study's central question is whether a preparative regimen before DLI might enhance its effectiveness in patients with blood cancers after allogeneic transplant. The record describes the hypothesis and treatment groups, but it does not establish a randomized comparison of the preparative regimen against an otherwise equivalent DLI-only strategy. The groups differ in their disease descriptions and treatment histories, and the stated arm descriptions assign different treatment approaches to those groups. As a result, the primary question about survival is not answered simply by comparing the group summaries: such a comparison could reflect differences in who entered each group as well as differences in treatment. The registry lists several additional outcomes, including disease status and complications, but gives no outcome analyses. Thus, the record identifies what investigators intended to examine, rather than reporting evidence that resolves the clinical question.

3. Trial Design

This was a Phase 2, non-randomized, parallel study of donor lymphocyte infusion after allogeneic transplant. The registry lists enrollment of 57 and identifies the study as terminated. The recorded start is 2004-01, and primary completion is dated 2017-11-21. The sponsor is Masonic Cancer Center, University of Minnesota. The registry record does not report why the study was terminated.

Masking is listed as none, so the record does not describe blinded treatment assignment or outcome assessment. Allocation is non-randomized. The arm descriptions identify patients with Chronic Myelogenous Leukemia (CML) who had failed or refused Gleevec(TM) and would receive DLI, and patients with non-CML or CML that relapsed after DLI who would receive induction chemotherapy plus DLI. The adverse-event listing names groups by DLI cell dose: 1.0 x 10^8 CD3+ T-cells/kg and 0.5 x 10^8 CD3+ T-cells/kg. The record does not explain how these dose groups map onto each arm description, or whether other differences in care accompanied the dose distinction.

The non-randomized structure is important when reading any apparent contrast between groups. Random allocation is not reported, and the groups are described in terms of differing disease and treatment histories. A difference in an outcome could therefore reflect baseline differences, treatment selection, follow-up, or other factors, as well as treatment itself. No adjustment strategy or detailed statistical analysis plan is reported in the registry record. The recorded design supports a descriptive account of the study and its outcomes, but the available information does not establish a controlled causal comparison.

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: During the recruitment period of the study, the donor lymphocyte infusion (DLI) was reduced from 1.0 x 10\^8 CD3+ T-cells/kg to 0.5 x 10\^8 CD3+ T-cells/kg due to an excess rate and intensity of graft vs. host disease.

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 Number of Patients Alive, measured over 1 Year.

The primary outcome is listed as the number of patients alive, with a timeframe of 1 Year. Its description defines overall survival as the time from enrollment to death; patients still alive are censored at the date of their last documented contact. In general, overall survival is used because it captures death from any cause and gives a clinically interpretable measure of how long patients remain alive after a specified starting point.

The record uses both a count-oriented outcome label and a time-to-event definition. These represent related but distinct ways to summarize survival. A count of patients alive at a specified time describes status at that time, while a time-to-event analysis uses the timing of deaths and accounts for follow-up that ends before death is observed. Censoring allows follow-up information from patients known to be alive at last contact to be represented without treating them as having died at that point. It does not, by itself, resolve bias if follow-up is incomplete or differs between groups.

The registry record does not report the number alive at the specified time, a survival estimate, a confidence interval, or a comparison between groups. It also does not report the analysis method used to produce a primary outcome result. The endpoint definition therefore clarifies what was intended to be measured, but does not provide an estimate of survival or its precision.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas help explain what can and cannot be learned from the registry record. The record describes the intended endpoint and study structure, but does not report an analysis of the clinical outcomes. The explanations below distinguish general statistical principles from methods actually documented for this study.

What does non-randomized allocation mean for interpretation?

In a randomized study, assignment by chance aims to make treatment groups comparable, on average, with respect to measured and unmeasured characteristics at the start of treatment. In a non-randomized study, treatment assignment can be related to clinical history, disease characteristics, or other factors. Such factors can also relate to outcomes. A group contrast can therefore combine treatment effects with differences in the people receiving each treatment.

That concern is particularly relevant because the registry describes groups using CML status, response or history after DLI, and treatment approach. The arm descriptions are not two versions of an otherwise identical comparison population. To interpret a treatment contrast causally, an analysis would need a clear strategy for addressing relevant differences between groups. The statistical adjustment strategy is not reported in the registry record, so the extent to which any such differences were addressed cannot be assessed.

How should the survival outcome be understood?

The registry defines overall survival as time from enrollment to death, censoring patients who remain alive at their last documented contact. In general, time-to-event methods use both whether an event occurred and when it occurred. Censoring records that a person's event time is not observed after a known point; it does not mean that the person died at that point. These methods rely on assumptions about the relationship between censoring and the event process. If follow-up ends for reasons related to a patient's prognosis, for example, the survival estimate may be biased unless the analysis appropriately addresses that process.

The endpoint also names a count of patients alive at 1 Year. A fixed-time count and a time-to-event estimate are not interchangeable summaries: the former focuses on status at a particular time, while the latter can incorporate event timing and censored follow-up. The registry record does not state which estimator or test was used, nor does it report a survival curve or fixed-time estimate. The definition alone does not tell readers how the analysis handled incomplete records or follow-up.

What do confidence intervals and tests usually add?

In general, a confidence interval describes the range of values reasonably compatible with an estimate under the assumptions of the analysis. Its width reflects statistical precision: a wider interval indicates less precise information than a narrower interval. A hypothesis test addresses compatibility between data and a specified null model, but a p-value is not a measure of clinical importance and does not show the probability that a treatment works. Both estimates and uncertainty intervals are typically more informative than a significance label alone.

No confidence interval, p-value, or primary outcome estimate is reported in the registry record. There is thus no reported estimate to interpret for magnitude, and no reported interval from which to assess precision. No statistical test can be inferred from the endpoint definition or the study's Phase 2 designation.

What is the role of sample-size planning?

In general, a sample-size calculation describes how much information a planned study would need under specified assumptions about the effect to detect, significance level, allocation, and statistical method. It is a planning exercise, not a result of the completed study and not evidence that a study had a particular probability of detecting an effect after the fact. The registry record characterizes its planning illustrations as based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. It says that the effect to detect is a stated fraction of the observed effect. For time-to-event outcomes, the stated convention is to express the planning requirement as a total number of events; for other outcomes, it is expressed as participants per arm.

Those assumptions describe the planning illustration, not necessarily the trial's original design. The record does not report that these assumptions were used to plan enrollment, and does not report a sample-size justification. It also does not report the outcome estimate needed to understand a planning illustration based on an observed effect. Such planning values, where available, would remain hypothetical requirements under their assumptions rather than observed power or proof of adequate power in this study.

What analysis details remain unknown?

The registry record reports no outcome analyses. It does not specify an estimator for survival, a model for comparisons, covariates, treatment of competing events, or methods for dealing with missing outcome data. It does not describe an interim analysis or a prespecified statistical analysis plan. These details matter because analytic choices can affect both the estimate and its uncertainty. Without reported results and methods, readers cannot reproduce or independently evaluate a treatment-effect estimate. General statistical principles can frame the questions to ask, but they cannot fill in methods or findings that the registry record does not report.

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("DLI Cell Dose of 1.0 x 10^8 CD3+ T-cells/kg", "DLI Cell Dose of 0.5 x 10^8 CD3+ T-cells/kg")
n      <- c(15, 42)
events <- c(3, 20)
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  DLI Cell Dose of 1.0 x 10^8 CD3+ T-cells/kg  15  3  20.0  4.3  48.1  7.0  45.2  7.3  45.6
ARM  DLI Cell Dose of 0.5 x 10^8 CD3+ T-cells/kg  42  20  47.6  32.0  63.6  33.4  62.3  33.3  62.3

11. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures in total: the primary outcome of patients alive and secondary outcomes concerning patients alive without disease, complete remission, acute graft-versus-host disease, and bone marrow aplasia. When a study assesses multiple outcomes, the chance of at least one apparently unusual result can increase if each is tested separately. In general, multiplicity methods can control this issue, or investigators can distinguish a primary outcome from exploratory outcomes and interpret the latter with appropriate caution.

The record identifies one primary outcome and four secondary outcomes, but it reports no analyses and no multiplicity procedure. Whether any adjustment or hierarchical testing approach was planned is not reported in the registry record. No claim about statistical significance across the outcomes can therefore be evaluated.

The primary outcome definition specifies censoring for patients still alive at last documented contact. The record does not describe how missing outcome values, incomplete follow-up, or other missing data were handled. Nor does it identify the analysis population for the primary or secondary outcomes, or state whether analyses included all enrolled patients or a defined subset. These omissions limit assessment of selection and attrition. The registry record also does not report an interim-analysis plan or statistical analysis plan. None of these details can be inferred from the outcome names or study design.

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: acute graft vs. host disease (46.7%), death, nos (6.7%), infection, nos (6.7%), acute sepsis (6.7%), pulmonary failure (2.4%).

The adverse-event listing separates two DLI cell-dose groups. For the group titled “DLI Cell Dose of 1.0 x 10^8 CD3+ T-cells/kg,” the record lists 15 at risk for serious events and 8 affected; it lists 15 at risk for other events and 12 affected, with 12 deaths. For the group titled “DLI Cell Dose of 0.5 x 10^8 CD3+ T-cells/kg,” it lists 42 at risk for serious events and 3 affected; it lists 42 at risk for other events and 11 affected, with 22 deaths. These are descriptive counts, not adjusted comparisons.

The serious-event terms with reported percentages include Acute Graft vs. Host Disease at 46.7%, Death, NOS at 6.7%, Infection, NOS at 6.7%, Acute Sepsis at 6.7%, Pulmonary Failure at 2.4%, and Graft vs. Host Disease, NOS at 2.4%. The registry record does not explain the full event-counting rules, timing, or whether an individual could contribute to multiple terms. Because treatment groups are non-randomized and differ in their descriptions, these figures should not be interpreted as a causal comparison of safety. They also do not establish how an adverse event affected survival or other outcomes.

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

The registry record documents a study of DLI after allogeneic transplant, its intended primary survival outcome, its treatment-group descriptions, and selected adverse-event information. It does not report the primary outcome results or analyses of the listed secondary outcomes. Accordingly, the record cannot quantify survival at 1 Year, the size of a treatment contrast, or the precision of any such contrast. It does not support a conclusion that the stated preparative regimen did or did not improve DLI effectiveness.

The study's non-randomized design also limits what a simple group comparison could establish. Participants are described in different clinical categories and with different treatment histories. Those differences may influence outcomes independent of treatment. An analysis that adjusts for relevant factors could address some measured differences, but the registry record does not report whether an adjustment was performed, which variables were considered, or how any model was specified. Unmeasured differences would remain a concern even if measured covariates were adjusted.

The adverse-event counts and percentages provide safety-related descriptions, but they do not establish comparative risks between otherwise equivalent groups. The record does not provide a full account of event definitions, timing, or the analysis population. In general, counts can help describe what was recorded, but their interpretation depends on ascertainment and follow-up as well as on the underlying group composition.

These limits do not mean that the study produced no findings; they mean the registry record does not report the analyses required to evaluate them. The design and endpoint definitions identify the question, while the missing outcome estimates prevent assessment of the study's answer. Any conclusion about benefit, harm, or comparative effectiveness would require results and methods beyond those reported in the registry record.

14. Why This Trial Matters Statistically

This trial matters statistically because its clinical question involves treatment selection in a setting where patient groups are not exchangeable by design. The registry describes a non-randomized study and groups defined by disease status and treatment history. That structure makes the distinction between a descriptive association and a causal treatment effect central. Readers need to know not only which outcomes were recorded, but also how differences between groups were addressed before interpreting any contrast.

The primary endpoint also illustrates the need for clarity about outcome form. “Number of Patients Alive” is a count-oriented label, while the description defines overall survival through time from enrollment to death with censoring at last documented contact. Counts at a fixed time and time-to-event analyses answer related but different questions. The selected estimand and the handling of follow-up determine what an estimate means. The registry record defines the endpoint but does not report the analysis or estimate.

Finally, this record highlights the difference between planning calculations and observed evidence. Planning assumptions can illustrate the information a future study might require under a specified model; they do not replace outcome estimates or establish observed power. Similarly, serious adverse-event counts provide descriptive safety context, but cannot resolve a treatment comparison in non-randomized groups. The combination of an explicitly stated hypothesis, a defined survival endpoint, and absent outcome analyses makes transparent reporting of estimates, uncertainty, analysis populations, and missing-data methods especially important for interpretation.

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 registry record does not report primary or secondary outcome analyses, so survival, remission, disease status, and complication outcomes cannot be evaluated from the reported results. It does not give effect estimates, confidence intervals, or statistical tests. The record also does not report a statistical analysis plan, analysis populations, covariate adjustment, or methods for handling missing outcome data. These gaps prevent assessment of how the intended outcomes were analyzed.

Non-randomized allocation creates potential for confounding, and the group descriptions indicate differences in disease status or prior treatment. No method for addressing those differences is reported. The record lists DLI dose groups in the safety data, but does not explain their relationship to the treatment-arm descriptions. It does not report the reason for termination, and that reason should not be inferred from the status label.

The primary endpoint description includes censoring at last documented contact for patients still alive, but the record does not state follow-up completeness or how censoring was assessed. It does not report event-counting rules for adverse events, their timing, or whether participants could appear in multiple event categories. These details matter when comparing safety summaries. The registry record therefore offers design and event-description information, but not enough information to assess efficacy, precision, or the robustness of comparative safety conclusions.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Post Transplant Donor Lymphocyte Infusion — NCT00167180.
• Registry record last updated 2019-07-30; status last verified 2019-07.
• Results first posted 2017-07-11.
• 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

NCT00167180 is a Phase 2, non-randomized, parallel study titled “Post Transplant Donor Lymphocyte Infusion,” sponsored by Masonic Cancer Center, University of Minnesota. It is listed as terminated, with enrollment of 57, a recorded start of 2004-01, and primary completion on 2017-11-21. Masking is listed as none.

The study's stated hypothesis concerns cyclophosphamide and fludarabine before DLI. The primary outcome is the number of patients alive at 1 Year, with overall survival defined from enrollment to death and censoring at last documented contact for patients still alive. The registry lists four secondary outcomes and adverse-event summaries for two DLI cell-dose groups. It reports no outcome analyses, so the clinical effectiveness question and the precision of any treatment contrast cannot be assessed from the registry record.