Chemotherapy, Total Body Irradiation, and Post-Transplant Cyclophosphamide in Reducing Rates of Graft Versus Host Disease in Patients With Hematologic Malignancies Undergoing Donor Stem Cell Transplant: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Chemotherapy, Total Body Irradiation, and Post-Transplant Cyclophosphamide in Reducing Rates of Graft Versus Host Disease in Patients With Hematologic Malignancies Undergoing Donor Stem Cell Transplant (NCT03192397), a non-randomized phase 1, phase 2 study. The primary outcome is Extensive Chronic Graft Versus Host Disease (GVHD).

2. Clinical Question

The study’s stated purpose is to examine a transplant approach intended to reduce graft-versus-host disease in patients with hematologic malignancies. Donor stem cells can contribute to blood-cell production, while donor immune cells can also react against a recipient’s normal tissues. The registered regimen combines chemotherapy and total-body irradiation before transplantation with cyclophosphamide after transplantation. These are the clinical rationale and treatment components described in the record, not evidence of an observed treatment effect.

The statistical question is therefore descriptive: how often was the defined chronic GVHD endpoint observed in the stated population and cohorts? A comparison between cohorts would require careful interpretation because the study is non-randomized. The registry record does not report an estimated percentage, uncertainty interval, or formal cohort comparison for the primary endpoint.

3. Trial Design

This is a phase Ib/2, non-randomized, single-group study sponsored by Roswell Park Cancer Institute. The registry lists an enrollment of 35. Its status is active, not recruiting. The study began on 2017-08-09, and its primary completion date is listed as 2023-08-21. These dates describe the registry record; they do not establish when results were analyzed or posted.

The intervention is a transplant strategy built around fludarabine and melphalan conditioning, total-body irradiation, allogeneic hematopoietic stem-cell transplantation, and post-transplant cyclophosphamide. Mycophenolate mofetil and sirolimus also appear among the registered interventions. Cohorts are distinguished by age and regimen labels, including age categories of <=50 years and >50 years and melphalan-dose labels. Some cohort descriptions are truncated or internally unclear in the registry record, so they do not support a complete reconstruction of each regimen.

With no random allocation and no separate control group, the design does not directly estimate a randomized treatment contrast. Outcomes can be described within the treated cohorts, but differences between cohorts may reflect their different characteristics, regimen details, or other factors rather than treatment alone. The record does not report a planned cohort-comparison method or a statistical analysis plan.

The primary endpoint is extensive chronic GVHD, expressed as a percentage of participants, with a timeframe up to 365 days. The primary population is described as all treated and evaluable patients, and the endpoint description says the percentage is considered by HLA matching and cohort. The registry also lists secondary outcomes for clinical response, grade III-IV acute GVHD, relapse, engraftment, overall survival, 1-year progression-free survival, and treatment-related mortality. These outcomes broaden the clinical description, but their detailed results and analysis methods are not reported in the registry record.

The enrollment figure is not itself an outcome estimate or proof that every enrolled participant was included in each endpoint analysis. The record does not report a flow of participants through treatment, evaluation, and outcome follow-up, nor does it state how many patients contributed to the primary endpoint.

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

Analysis population for the primary outcome, as defined in the registry: All treated and evaluable patients

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: All treated and evaluable patients

6. Primary Endpoint

The registry lists the primary outcome as Extensive Chronic Graft Versus Host Disease (GVHD), measured over Up to 365 days.

The primary endpoint is the percentage of participants with extensive chronic graft-versus-host disease, assessed by HLA matching and cohort, up to 365 days. The stated population is all treated and evaluable patients. This is a binary outcome at the participant level: each participant is classified according to whether the endpoint occurred, and a percentage describes the frequency of participants meeting that definition.

In general, a percentage is useful because it expresses how common an event was in the observed population. Its interpretation depends on the endpoint definition, the number of participants assessed, and how follow-up and missing observations were handled. For a time-limited endpoint, consistent follow-up and a clear approach to participants whose status is unknown are particularly important. The registry record names the endpoint and timeframe but does not report its operational definition, cohort-specific estimates, or missing-data procedures.

Chronic GVHD is a distinct outcome from acute GVHD. The record lists grade III-IV acute GVHD as a secondary outcome, while the primary outcome concerns extensive chronic GVHD. These outcomes should not be conflated: they have different labels and are reported as separate objectives. The record does not describe the diagnostic criteria, adjudication process, or measurement schedule used for either endpoint.

The endpoint can describe what occurred among treated and evaluable patients, but by itself it cannot establish that the regimen caused a particular outcome frequency. The single-group, non-randomized design has no randomized control arm for a causal treatment comparison. The posted record does not report the primary endpoint result, so the observed percentage and its precision cannot be interpreted from these details.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registered primary outcome is a percentage of participants with an event. For this type of binary endpoint, the basic descriptive quantity is the number of participants meeting the endpoint divided by the number assessed, expressed as a percentage. That general description does not supply an estimate for this study: the registry record does not report either the primary-event count or the endpoint denominator.

What does a confidence interval add?

In general, a confidence interval communicates the uncertainty around an estimated quantity. It helps distinguish a relatively precise estimate from one that is compatible with a broad range of underlying event frequencies. The interval’s meaning depends on the statistical method and assumptions. For a participant-level percentage, the calculation also depends on the number of participants and the number with the event. Neither a primary-endpoint estimate nor its confidence interval is reported in the registry record, so precision cannot be assessed.

Why does the design affect interpretation?

Random allocation is a design method intended to make comparison groups similar, on average, with respect to known and unknown characteristics. This study is registered as non-randomized and single-group. The age- and regimen-defined cohorts therefore do not form a randomized treatment comparison. In general, observed differences between such cohorts may be associated with the intervention, but may also reflect differences in participants, clinical circumstances, follow-up, or other factors. Statistical adjustment can address measured factors under appropriate assumptions, but cannot guarantee removal of unmeasured confounding. The registry record does not report an adjustment strategy or formal between-cohort analysis.

How should the timeframe be understood?

The primary endpoint timeframe is up to 365 days. This identifies the period referenced by the outcome, but does not by itself state how follow-up was scheduled, how incomplete follow-up was handled, or whether all participants had an opportunity for the same observation period. In general, an event percentage over a fixed horizon can be misleading if participants have different follow-up or if outcome status is unknown for some participants. Time-to-event methods can account for differing follow-up under specified assumptions, but the registry record does not say that such methods were used for the primary endpoint.

What do the registered analyses tell us?

The registry record lists no statistical analyses. It does not report a statistical analysis plan, model specification, hypothesis test, confidence-interval method, or adjustment for multiple outcomes. The record also does not report whether the primary endpoint was analyzed by HLA match, by cohort, or by both, beyond the endpoint description’s statement that the percentage is assessed by HLA matching and cohort. Detailed results and methods are therefore not available for assessment.

In general, a statistical analysis plan explains how outcomes, analysis populations, missing data, and comparisons will be handled. Prespecification can reduce the risk that analytic choices are made in response to the observed results. The registry record does not report enough information to determine whether such a plan existed or how closely any analysis followed it.

Why are planning illustrations not study results?

Sample-size calculations are planning tools that describe how many participants might be needed under stated assumptions to detect a specified effect with a chosen error rate and power. They do not describe the observed outcome, and they are not a substitute for a confidence interval or treatment comparison. The record’s sample-size assumptions describe an illustration based on an observed effect, two-sided significance level 0.05, equal allocation between two arms, and a normal approximation. The assumptions also state that the target effect is a stated fraction of the observed effect. These are planning assumptions, not the trial’s own design justification or observed power. The registry record does not report the trial’s sample-size justification, and its single-group, non-randomized design should not be mistaken for the equal-allocation two-arm planning setup.

The listed outcomes also include event counts, survival measures, and percentages. In general, different outcome types call for different methods. Time-to-event outcomes account for when an event occurs and for censoring; binary outcomes summarize whether an event occurred within a defined period. No specific statistical method for the secondary outcomes is reported in the registry record. Consequently, readers cannot assess model assumptions, handling of competing events, or the precision of those outcomes from the registered descriptions alone.

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("<=50 Years of Age Cohort A", "<=50 Years of Age Cohort B", ">50 Years of Age Cohort A -25 mg/m2 Melphalan", ">50 Years of Age Cohort B 50 mg/m2 Melphalan", ">50 Years of Age Cohort C .75 mg/m2 Melphalan")
n      <- c(3, 1, 3, 6, 22)
events <- c(2, 0, 0, 2, 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  <=50 Years of Age Cohort A  3  2  66.7  9.4  99.2  20.8  93.9  19.4  93.2
ARM  <=50 Years of Age Cohort B  1  0  0.0  0.0  97.5  0.0  79.3  1.3  84.2
ARM  >50 Years of Age Cohort A -25 mg/m2 Melphalan  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  >50 Years of Age Cohort B 50 mg/m2 Melphalan  6  2  33.3  4.3  77.7  9.7  70.0  9.9  71.0
ARM  >50 Years of Age Cohort C .75 mg/m2 Melphalan  22  9  40.9  20.7  63.6  23.3  61.3  23.2  61.5

11. Multiplicity, Missing Data and Analysis Populations

The registry lists eight outcome measures, including the primary extensive chronic GVHD endpoint and secondary measures covering acute GVHD, relapse, response, engraftment, survival, progression-free survival, and treatment-related mortality. Considering multiple outcomes can raise a multiplicity issue: in general, the more hypotheses tested, the greater the chance of at least one apparently notable result arising by chance if tests are interpreted without adjustment. Whether multiplicity adjustment is appropriate depends on the study objectives and analysis plan. The registry record does not report a hierarchy of outcomes, a multiplicity procedure, or a statistical analysis plan.

The primary endpoint population is described as all treated and evaluable patients. The record does not define “evaluable,” report the primary endpoint denominator, or state how participants who did not complete follow-up were handled. Missing outcome status can affect a percentage if participants with unavailable assessments differ from those assessed. In general, analyses may use explicit assumptions or sensitivity analyses to examine the effect of missing data, but the registry record does not report such methods.

The record does not report whether an intention-to-treat, per-protocol, or other analysis population was used for the primary or secondary outcomes. Nor does it report participant flow by cohort or reasons for exclusion from an endpoint analysis. Enrollment and the stated primary population are not sufficient to reconstruct those details.

No interim analyses, stopping rules, or outcome-specific analysis methods are reported in the registry record. This absence does not establish that none were used; it means their use and details cannot be determined from the registry record. The study’s non-randomized design and lack of listed analyses further limit interpretation of any unreported cohort differences.

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: infections and infestations - other, specify (33.3%), sepsis (33.3%), febrile neutropenia (16.7%), neutrophil count decreased (16.7%), platelet count decreased (16.7%).

The safety record reports serious and other adverse-event counts separately for five age- and regimen-defined cohorts. For each cohort it gives the number at risk and the number affected for each category, as well as a death count. These are descriptive cohort figures. The registry record does not establish that events were caused by the transplant regimen, and the figures should not be interpreted as comparative treatment effects.

Among the serious-event terms listed, infections and infestations—other, specify, and sepsis are each reported as 33.3%. Febrile neutropenia, decreased neutrophil count, decreased platelet count, and neoplasms benign, malignant and unspecified (including cysts and polyps)—other, specify, are each reported as 16.7%. These are the recorded percentages for the listed serious-event terms; the registry record does not clarify in this summary whether the percentages represent participants or events, or how repeat events were handled.

The cohort-level figures should be read with their own denominators rather than compared as if the cohorts were randomized and equally followed. The registry record does not report detailed exposure time, event timing, severity grading beyond the serious-event classification, or the causal assessment for each event. Death counts are reported, but the record does not provide enough detail to determine their timing or relationship to treatment. The adverse-event information therefore describes what is recorded, not a causal safety comparison.

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

The registry record identifies an important clinical question: the frequency of extensive chronic GVHD after a donor stem-cell transplant strategy using chemotherapy, total-body irradiation, and post-transplant cyclophosphamide. It also identifies secondary outcomes relevant to transplantation, including acute GVHD, relapse, engraftment, survival, and treatment-related mortality. These registered outcomes establish what the study set out to assess, but they are not themselves outcome findings.

The record does not report the primary endpoint percentage or its uncertainty. It therefore does not show how many treated and evaluable patients experienced extensive chronic GVHD, how that frequency varied by cohort or HLA matching, or how precisely any estimate was measured. It also does not report statistical analyses for the listed outcomes. No treatment effect or absence of effect can be inferred from the endpoint definition or study rationale.

The non-randomized, single-group design can provide descriptive information about participants treated in the study. It does not offer a randomized control comparison. If cohort outcomes are eventually reported, differences between cohorts should be interpreted cautiously because age and regimen define the cohorts, and other characteristics may differ as well. An observed contrast would not, by itself, identify the cause of a difference.

The safety figures describe serious and other adverse-event records by cohort and include death counts. They can help characterize the recorded events, but they do not establish causality or permit a controlled comparison. The cohort-specific denominators and incomplete information about follow-up, exposure, and event attribution matter when interpreting those figures.

Overall, the registry record gives a study design, treatment description, outcome definitions, and safety counts, but does not provide the primary clinical result or inferential analysis needed to quantify the endpoint or evaluate its precision. The appropriate interpretation is limited to the information recorded, without translating the study’s aim into a claim about treatment effect.

14. Why This Trial Matters Statistically

Statistically, this trial illustrates the distinction between registering an outcome and reporting an outcome result. The primary endpoint is clearly named as a percentage of participants with extensive chronic GVHD, assessed by HLA matching and cohort up to 365 days. Yet the registry record does not report the event count, denominator, estimate, confidence interval, or analysis. Without those elements, the central outcome cannot be quantitatively interpreted.

The design also makes the comparison question important. The study is non-randomized and single-group, with cohorts defined by age and regimen. Such a design can describe experience in the treated population, but does not provide the protection against confounding that random allocation aims to provide. Any future cohort contrast would need to distinguish description from causal inference and make clear which factors were considered.

Outcome definitions, analysis populations, and follow-up rules are essential to interpreting a percentage. The record identifies an all-treated-and-evaluable population but does not define evaluability, provide endpoint denominators, or describe missing-data handling. These omissions prevent assessment of whether the observed population represented all enrolled participants and how incomplete follow-up might affect the estimate.

The multiple registered outcomes raise another general statistical concern: interpretation across several endpoints depends on the prespecified outcome hierarchy and approach to multiplicity. Neither is reported. No statistical analyses are listed, so methods and assumptions cannot be evaluated. The sample-size assumptions are planning illustrations and do not establish the trial’s own justification or its ability to detect a particular effect.

Together, these features show why registry descriptions, results, and analysis plans serve different purposes. A thorough statistical reading needs the observed estimates, denominators, uncertainty, population definitions, and methods—not only the clinical rationale and intended outcomes.

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 main limitation for interpretation is that the registry record does not report primary-endpoint results or statistical analyses. The endpoint is defined as a percentage up to 365 days, but the event count, denominator, cohort-specific percentages, uncertainty intervals, and analysis method are not reported. The record therefore cannot support a quantitative assessment of the primary question.

The design is non-randomized and single-group. There is no randomized comparator, and the age- and regimen-defined cohorts are not like-for-like control groups. This limits causal interpretation of differences between cohorts. The record does not report an adjustment strategy that might address measured differences, nor does it establish how unmeasured differences would be handled.

The registry lists an enrollment of 35, but does not report how many participants were treated, evaluable for the primary endpoint, or followed through the stated timeframe. The definition of evaluability, participant flow, and missing-data methods are not reported in the registry record. These details are needed to understand the population underlying any endpoint percentage.

Some cohort regimen descriptions are incomplete or unclear, including truncated text. This limits the ability to reconstruct the exact interventions across cohorts from the registry record. The safety summaries also lack details about event timing, exposure, repeat events, and attribution, so they cannot establish comparative or causal safety effects.

Finally, the record does not report a statistical analysis plan, multiplicity approach, interim analyses, or the trial’s sample-size justification. The planning assumptions are illustrations, not confirmation of the study’s design rationale or observed power. These limitations arise from what is recorded and do not establish what may have been specified elsewhere.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Chemotherapy, Total Body Irradiation, and Post-Transplant Cyclophosphamide in Reducing Rates of Graft Versus Host Disease in Patients With Hematologic Malignancies Undergoing Donor Stem Cell Transplant — NCT03192397.
• Registry record last updated 2026-04-14; status last verified 2026-04.
• Results first posted 2026-04-14.
• 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

NCT03192397 is a phase Ib/2, non-randomized, single-group study of a transplant strategy involving fludarabine, melphalan, total-body irradiation, and post-transplant cyclophosphamide in patients with hematologic malignancies. The registry lists an enrollment of 35 and a primary endpoint of extensive chronic GVHD among all treated and evaluable patients, assessed by HLA matching and cohort up to 365 days.

The record lists eight outcome measures, including acute GVHD, relapse, response, engraftment, survival, and treatment-related mortality. It also reports adverse-event counts by cohort and percentages for several serious-event terms. However, primary-endpoint estimates and statistical analyses are not reported in the registry record. The non-randomized, single-group design does not provide a randomized treatment comparison, and the cohort descriptions do not establish equivalent control groups.

Accordingly, the registry supports a description of the study’s aims, design, registered outcomes, and posted safety information, but not a quantitative conclusion about the primary endpoint or a causal treatment effect.