Methods to Enhance the Safety and Effectiveness of Stem Cell Transplants: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Methods to Enhance the Safety and Effectiveness of Stem Cell Transplants (NCT00378534), a non-randomized phase 2 study. The primary outcome is Survival and Non-relapse Mortality at Day +200 Using the Miltenyi Reagent System.

2. Clinical Question

3. Trial Design

This was a Phase 2, non-randomized, unmasked, parallel study. The registry lists enrollment of 116 and describes both transplant participants and co-enrolled stem cell donors. The study began in 2006-09, had primary completion in 2014-04, and is marked completed. It was sponsored by the National Heart, Lung, and Blood Institute (NHLBI).

The transplant approach combined myeloablative conditioning with a stem cell product prepared using the Miltenyi CliniMacs system for CD34 selection. The registry describes donor T cells being given back later as a donor lymphocyte infusion at day 90. The stated rationale was to remove most T cells from the initial transplant product and delay their return. This is a description of the study strategy, not evidence about its comparative effect.

Because allocation was non-randomized, the registry does not describe an assigned control treatment. The donor group is not a substitute for a control group: donors and transplant participants differ in their roles in the study and are not a like-for-like comparison. The record also describes stem cell collection for donors as not investigational.

The listed secondary outcomes concern relapse and forms of graft-versus-host disease (GVHD). The record names relapse, acute GVHD grades I through IV, limited chronic GVHD, and extensive GVHD. Although it lists five outcome measures, it reports zero statistical analyses. A detailed statistical analysis plan, including the planned comparisons, is 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.

Overall Study

Reasons for not completing this period:

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 Survival and Non-relapse Mortality at Day +200 Using the Miltenyi Reagent System, measured over Day 200.

The primary outcome is titled “Survival and Non-relapse Mortality at Day +200 Using the Miltenyi Reagent System.” The registry says the study aimed to determine overall survival and non-relapse mortality at Day +200 among people receiving allogeneic stem cell transplantation with a product prepared using the Miltenyi CliniMacs system. The outcome is expressed as a number of participants.

As a general statistical concept, overall survival describes whether participants are alive at a specified assessment time. Non-relapse mortality concerns death without a preceding relapse, but the precise operational definition, event handling, and calculation used in this study are not reported in the registry record. The record also does not clarify whether the two outcomes were presented separately or combined under the primary outcome title.

These measures address different aspects of outcome after transplantation: survival captures death from any cause within the defined assessment framework, while non-relapse mortality distinguishes deaths not attributed to relapse. Such distinction can be important when relapse is also an outcome of interest. However, interpretation depends on the analysis definitions and time window, which are not fully.

No numerical result for the Day +200 primary outcome is reported in the registry record. The endpoint can therefore be described, but its observed value and precision cannot be assessed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods are not fully described in the registry record. It identifies a non-randomized parallel design, a participant-count primary outcome, and several event-count secondary outcomes, but lists zero statistical analyses. The explanations below are general statistical context, not a description of methods confirmed to have been used in this study.

What does a non-randomized design mean?

In a randomized study, chance allocation is used to make comparison groups more alike on average, including on factors that may be difficult to measure. In a non-randomized study, group membership is not assigned through that process. Observed differences may therefore reflect differences in who entered each group, in addition to any effect of an intervention. Statistical adjustment can address measured differences under appropriate assumptions, but it cannot guarantee removal of bias from unmeasured or poorly measured factors. Whether adjustment was planned or conducted is not reported in the registry record.

The listed donor group is not a control group for the transplant participants. Donors and recipients have different study roles, and a comparison between them would not isolate the effect of the transplant strategy. The registry itself cautions that the arms are not a like-for-like control comparison.

How should participant counts be understood?

For a binary outcome, each participant is classified according to whether an event occurred within the defined observation period. The event count and the number assessed form the basis for an event proportion. A comparison of proportions may use a difference or ratio, but the choice of measure and any adjustment should be specified in the analysis plan. The registry record does not report a primary event count, a primary denominator, or an effect estimate for Day +200 survival and non-relapse mortality.

Uncertainty is commonly expressed with a confidence interval. In general, a confidence interval describes a range of estimates compatible with the data under the method used; it does not remove design bias or turn an association into a causal effect. No confidence interval for the primary endpoint is reported in the registry record. Without an estimate and interval, readers cannot judge the precision of the primary finding.

Why do endpoint definitions matter?

Survival and non-relapse mortality are time-dependent outcomes. A complete analysis description would ordinarily clarify the time origin, event definition, follow-up window, handling of participants whose status is unknown, and how relapse relates to non-relapse death. For non-relapse mortality, relapse may be treated as a competing event: a participant who relapses may no longer be eligible to experience a death classified as non-relapse mortality. General methods for competing risks differ from methods that simply censor participants at relapse. The registry record does not report which approach, if any, was used.

For GVHD and relapse outcomes, the timing and definition of assessment also matter. The registry lists acute GVHD grades I through IV, limited chronic GVHD, extensive GVHD, and relapse, but does not provide their numerical results or detailed analysis definitions. It is not reported whether outcomes were analyzed as counts, proportions, time-to-event measures, or another form.

What do planning illustrations represent?

Sample-size illustrations are planning scenarios, not observed power. The stated illustration assumptions use a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. These are planning assumptions, not evidence that the trial was designed using them or that it had a particular probability of detecting an effect. The record does not report the trial’s own sample-size justification. Any illustration based on an observed effect is conditional on that effect estimate and should not be read as a retrospective measure of the study’s success.

In general, power is a design property calculated under assumptions about an effect, variability, analysis, and sample size. It is not a probability that a particular result is true. Retrospective observed power does not add useful information beyond the observed estimate and its uncertainty. Planning scenarios can instead show how requirements change under different assumed effects, provided the assumptions are transparent.

What can be concluded from the posted analysis information?

The registry lists zero statistical analyses. It does not report a test statistic, p-value, effect estimate, confidence interval, model, adjustment strategy, or statistical software. No inferential analysis can be reconstructed from the safety counts alone. Recalculating a comparison from summary counts would not reproduce a sponsor analysis, and it would not address the non-randomized design. In this record, the sponsor’s own analysis is not reported, so there is no sponsor analysis to compare with a simple recalculation.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists five outcome measures, including the primary outcome and secondary measures concerning relapse and GVHD. It does not report how the primary outcome was prioritized in the analysis, whether secondary outcomes were confirmatory or exploratory, or whether any multiplicity adjustment was planned. When several outcomes are tested, the chance of at least one apparently notable result can increase. General approaches include prespecifying a primary comparison, adjusting for multiple tests, or interpreting secondary findings as exploratory. Which approach was used is not reported in the registry record.

The record lists zero statistical analyses and gives no numerical primary or secondary outcome results. It does not report whether missing outcome data occurred, how missingness was defined, or how missing values were handled. It also does not describe censoring, follow-up completeness, or sensitivity analyses. These details matter because an analysis restricted to participants with recorded outcomes can be biased if the availability of data is related to prognosis or treatment experience.

The analysis populations are not reported in the registry record. In general, a study may define populations based on enrollment, receipt of treatment, eligibility, or availability of outcome data. Those choices affect the interpretation of denominators and estimates. No population definition or participant flow for the primary outcome is available in the record.

Without a prespecified analysis plan and outcome results, it is not possible to assess whether outcome selection, missingness, or multiple testing influenced the findings. These are reporting gaps, not evidence that a particular analysis choice was made.

11. 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

Serious events with the highest arm-level rate: infection (12.1%), gvhd (6.9%), graft failure (5.2%), posterior reversible encephalopathy syndrome (1.7%), hip fracture (1.7%).

The adverse-event record distinguishes serious adverse events from other adverse events. For T Cell Depletion Transplant Participants, it lists 58 at risk for serious events, 14 affected by serious events, 58 at risk for other events, and 0 affected by other events; the listed death count is 14. For Stem Cell Donors, the corresponding figures are 58 at risk, 0 affected by serious events, 58 at risk, 0 affected by other events, and 0 deaths. These are registry counts, not comparative estimates of treatment effect.

The listed serious-event terms include Infection at 12.1%, GVHD at 6.9%, Graft failure at 5.2%, Posterior reversible encephalopathy syndrome at 1.7%, Hip fracture at 1.7%, and Deep vein thrombosis at 1.7%. The record does not give additional details about the observation period, event-level counting, or the denominators underlying each listed percentage.

These figures should not be interpreted as proving that the transplant strategy caused an event, or that a zero count means an event is impossible. The donor group is not a like-for-like comparator for transplant participants. The registry also does not specify whether the deaths occurred within the primary endpoint’s Day +200 window, so the death count should not be treated as the Day +200 mortality result.

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

The record describes a transplant strategy and identifies survival and non-relapse mortality at Day +200 as the primary outcome. It also lists safety counts and secondary outcomes related to relapse and GVHD. These details explain what the study set out to assess, but the numerical primary result is not reported in the registry record. The record therefore does not establish the observed value or precision of Day +200 survival or non-relapse mortality.

The reported safety counts indicate that serious adverse events and deaths were recorded among transplant participants, while the donor group has zero listed serious events, other events, and deaths. These are descriptive observations. They cannot be interpreted as a treatment comparison because donors and transplant participants have different roles, and the study was not randomized. The record does not establish that the listed death counts share the primary endpoint’s time window.

The non-randomized design is important for interpreting any observed outcome. General statistical reasoning is that participant selection, disease characteristics, and other prognostic factors can contribute to outcome differences when allocation is not random. The registry does not report whether such factors were measured or adjusted for. Even a precise estimate would not, on its own, overcome limitations in design or analysis.

Nothing in the registry record supports a conclusion that the approach did or did not improve outcomes. The information supports describing the intervention, design, outcome definitions as stated, and selected adverse-event counts. It does not support medical advice, an estimate of comparative effectiveness, or a claim that the donor group is an appropriate control. A fuller interpretation would require the primary outcome results, their uncertainty, endpoint definitions, analysis methods, and follow-up information, none of which is reported in the registry record.

13. Why This Trial Matters Statistically

This study raises several statistical issues that are relevant to clinical research. First, the design is non-randomized, so outcome differences cannot automatically be attributed to the transplant strategy. Comparisons require attention to group roles and baseline differences; the donor group is not a suitable substitute for a randomized treatment control.

Second, the primary outcome combines concepts that need clear definitions. Overall survival and non-relapse mortality do not necessarily use the same event classification, and relapse can affect how non-relapse death is analyzed. The registry gives the Day +200 endpoint title but does not report the detailed definitions or analysis method. That gap limits interpretation even before considering the absence of a numerical result.

Third, safety counts are not self-interpreting. Event counts need a defined observation period, clear denominators, and context about how events were collected. The posted figures describe events recorded in each study group, but they do not establish causation or a comparative treatment effect.

Finally, the record lists zero statistical analyses. Without effect estimates, uncertainty intervals, analysis populations, and a multiplicity plan, readers cannot evaluate precision, robustness, or the role of secondary outcomes. The study’s statistical importance lies not in a result that can be quantified from this record, but in the need for transparent endpoint definitions and analysis reporting for complex, non-randomized transplant studies.

14. 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.

A central limitation is that the registry record does not report the numerical primary outcome. Although the primary endpoint is identified as survival and non-relapse mortality at Day +200, there is no estimate or uncertainty measure to interpret. The record also lists zero statistical analyses, so no reported model or inferential method can be assessed.

The design is non-randomized and has no like-for-like control group. The stem cell donors are a separate study group, not a treatment comparator. This precludes a straightforward comparative estimate of the transplant strategy from the listed arms. The registry does not report whether analyses adjusted for participant characteristics or other potential sources of confounding.

Endpoint details are limited. The precise definitions of survival and non-relapse mortality, the treatment of relapse as a competing event, and the follow-up or censoring rules are not reported in the registry record. The safety observation period and the denominators underlying listed serious-event percentages are also not reported. The record does not establish whether the death count corresponds to the Day +200 endpoint.

The record lists five outcome measures but does not report a multiplicity strategy, missing-data methods, analysis populations, or outcome-specific results. It also does not report the trial’s own sample-size justification. These reporting gaps prevent evaluation of statistical precision and robustness. They do not establish that a particular bias or analysis problem occurred; they define what cannot be assessed from the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Methods to Enhance the Safety and Effectiveness of Stem Cell Transplants — NCT00378534.
• Registry record last updated 2021-06-16; status last verified 2014-12.
• Results first posted 2015-01-06.
• 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.

18. Record Summary

The registry record describes a completed, non-randomized, unmasked, parallel Phase 2 study of a stem cell transplant approach using CD34 selection and delayed T-cell add-back. It lists enrollment of 116, with transplant participants and stem cell donors co-enrolled. The primary outcome is survival and non-relapse mortality at Day +200.

The record lists five outcome measures and zero statistical analyses. It does not report a numerical primary outcome or its precision. Safety counts include 14 serious adverse events and 14 deaths among transplant participants, and zero serious adverse events and zero deaths among donors. Because the design is non-randomized and donors are not a like-for-like control group, these figures do not establish a comparative treatment effect.