A Phase 2/3 Study of GLASSIA for the Treatment of Acute GvHD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase 2/3 Study of GLASSIA for the Treatment of Acute GvHD (NCT02956122), a randomized phase 2, phase 3 study. The primary outcome is Percentage of Participants Achieving Overall Response (OR) At Day 28.

2. Clinical Question

The study was designed to assess GLASSIA as an add-on to conventional steroid treatment for participants with acute graft-versus-host disease (GvHD) involving the lower gastrointestinal tract. Its stated purpose was to evaluate safety and efficacy when GLASSIA was used with standard-of-care steroid treatment as first-line treatment. The primary question concerned the percentage of participants achieving overall response at Day 28.

Overall response was defined as complete response plus partial response. Complete response required complete resolution of acute GvHD signs and symptoms in all organs, without intervening salvage. Partial response required improvement of one stage in one or more involved organs, with no progression in other organs. The endpoint therefore combined improvement with complete resolution, while excluding response accompanied by worsening elsewhere.

The planned study had two parts. All participants in Part 1 were to receive GLASSIA and methylprednisolone or an equivalent steroid. In Part 2, GLASSIA plus steroid treatment was compared with albumin plus steroid treatment. The central comparative question in that part was whether Day 28 overall response differed between those treatment groups. The registry record does not report the number assigned to each part or arm, or any estimate of the comparative response.

The primary endpoint population was described as participants evaluated for overall response at Day 28. To be considered evaluable, participants had to receive at least one dose of study treatment and have a lower GI biopsy consistent with acute GvHD. The record reports total enrollment of 1, but does not report how many participants met this endpoint population definition or had an evaluable Day 28 response.

3. Trial Design

This was a randomized, double-masked, parallel-group phase 2/3 study. It was described as a two-part, multicenter, prospective clinical study of GLASSIA added to conventional steroid treatment. The registry lists graft-versus-host disease as the condition and identifies lower gastrointestinal involvement in the study description. The sponsor is recorded as Baxalta now part of Shire.

Part 1 was a GLASSIA treatment part: all participants were to receive intravenous GLASSIA alongside methylprednisolone or an equivalent steroid, given intravenously or orally at investigator discretion. Part 2 included a GLASSIA group receiving the same combination and an albumin control group receiving intravenous control plus steroid treatment. The record does not report the specific randomization procedure, allocation ratio, or which individuals were masked.

The recorded start date was 2017-04-26, and the primary completion date was 2018-05-03. The study status is terminated. Total enrollment is recorded as 1. The registry record does not explain the termination or report how that participant was distributed across the two parts or treatment groups. The planned comparison in Part 2 should therefore be distinguished from the very limited enrollment recorded; the record does not establish that a comparative analysis could be conducted.

The primary outcome was the percentage achieving overall response at Day 28. Secondary outcomes included GI response at Day 28, overall response at Day 56, GvHD grading at Days 28, 56 and 180, chronic GvHD incidence, duration of overall and GI response, overall survival events, and transplant-related mortality. The registry lists 26 outcome measures, but reports no analysis entries. Details such as the planned statistical model, handling of repeated measurements, and analysis procedures for the secondary outcomes 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: A total of 1 participant was enrolled and completed Part 1 (GLASSIA) of the study, and was analyzed for efficacy and safety with individual PK parameters estimated. Part 2 (GLASSIA versus albumin \[control\]) was not initiated following discontinuation of the study due to operational and business considerations.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Efficacy analysis set included all participants who were evaluated for overall response at Day 28. Participants who received at least 1 dose of study treatment and who had a lower GI biopsy that was consistent with acute GvHD were considered evaluable.

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 Percentage of Participants Achieving Overall Response (OR) At Day 28, measured over Day 28.

The primary endpoint was the percentage of participants achieving overall response at Day 28. It combined complete response and partial response for acute GvHD. Complete response meant that all signs and symptoms of acute GvHD had resolved in all organs, without intervening salvage. Partial response meant improvement of one stage in at least one involved organ without progression in other organs. Thus, the endpoint counted either complete resolution across organs or a defined improvement that was not accompanied by worsening elsewhere.

A response endpoint offers a clinically interpretable way to summarize disease status at a specified assessment time. In general, a percentage expresses how many participants in a defined group meet the endpoint, while its precision depends on the number assessed and the amount of uncertainty around that estimate. A Day 28 assessment also fixes the time point at which response is classified; it does not by itself describe how long a response lasts. The registry separately listed duration of response as a secondary outcome.

The efficacy analysis population was defined as participants evaluated for overall response at Day 28. Participants were considered evaluable if they received at least one dose and had a lower GI biopsy consistent with acute GvHD. The record does not report an observed Day 28 response percentage, an estimate of uncertainty, or the number evaluated in this population. Consequently, the endpoint definition can be described, but the posted record does not permit interpretation of an observed response rate.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the study as randomized, double-masked, and parallel-group. Those labels describe the planned design, not the completeness of the reported analysis. No detailed statistical analysis plan is reported in the registry record, and the record contains zero analysis entries. The distinctions below explain the general statistical ideas relevant to the design and the limits of the posted information.

What does randomization contribute?

In general, randomization assigns participants to treatment groups by chance. It is intended to reduce systematic differences between groups at the time of assignment, so that a later comparison can more credibly be attributed to the assigned treatments. Its value depends on how it was implemented and on adequate enrollment, follow-up, and analysis. The specific randomization method, allocation ratio, and arm-level participant counts are not reported. The randomized designation therefore does not reveal how assignment occurred or whether the planned comparison was carried out.

What do masking and the parallel design mean?

In general, masking means that some people involved in a study are unaware of treatment assignment. It can reduce the influence of expectations on care, assessment, or reporting. The record labels this study double-masked but does not identify which parties were masked or describe how masking was maintained. A parallel-group design means that participants are assigned to a group and followed within that group rather than crossing between treatments. The record describes GLASSIA and albumin groups in Part 2, but does not report their realized sizes or follow-up by group.

How should a response percentage be read?

For a binary outcome such as whether a participant achieved overall response by Day 28, the percentage is based on the number meeting the response definition and the denominator used for analysis. A percentage without its denominator is hard to assess, and a percentage from a very small denominator is generally imprecise. A confidence interval, when reported, gives a range of values compatible with the data under the statistical method used; it communicates uncertainty rather than guaranteeing that the true value lies in that range. No primary response percentage or confidence interval is reported in this record.

A comparison between groups would ordinarily describe the response in each group and a measure of the difference between them, with uncertainty. The registry does not report such estimates. It also does not report a model or test for the primary endpoint, covariate adjustment, or a method for missing Day 28 assessments. None of these procedures should be inferred from the study's design labels.

What do the planning illustrations represent?

The record describes sample-size illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is specified as a fraction of the observed effect. Such calculations are planning scenarios: they describe participant counts under stated assumptions rather than the number observed in the trial. They are not observed power, and they do not show that the trial had a particular probability of detecting an effect.

Those illustrations also depend on having an observed effect estimate that can be used as an input. No primary comparative estimate is reported in the registry record, and the primary outcome was skipped because fewer than two arms were available for it. The planning assumptions should therefore not be read as evidence that a primary treatment effect was estimated or that a confirmatory analysis was performed. No numerical sample-size result is stated in the registry information described for this analysis.

Why do analysis populations matter?

The primary efficacy population was defined by evaluation at Day 28, at least one dose of study treatment, and a lower GI biopsy consistent with acute GvHD. This is an evaluability definition, not a count of participants. In general, different analysis populations can lead to different estimates if participants are excluded or if assessments are missing. The record does not report the number meeting the definition, the number with missing assessments, or how any missing outcomes were handled.

The stated endpoint is a percentage of participants with an event, which is a binary-outcome framework. The registry does not report whether the primary analysis used a simple proportion, an adjusted model, or another method. There is also no reported estimate of over-dispersion or other feature that would justify a more complex model. These methodological details remain unknown rather than being recoverable from the trial title or endpoint definition.

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("GLASSIA")
n      <- c(1)
events <- c(1)
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  GLASSIA  1  1  100.0  2.5  100.0  20.7  100.0  15.8  98.7

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 26 outcome measures. Alongside the primary Day 28 overall-response outcome, the secondary outcomes include GI response at Day 28, overall response at Day 56, GvHD grading at Days 28, 56 and 180, chronic GvHD incidence, duration of overall and GI response, overall survival events, and transplant-related mortality. Assessing many outcomes can raise a multiplicity issue: in general, the more statistical claims tested, the greater the chance of at least one apparently positive result arising by chance. Researchers may address this through prespecified testing strategies or adjustments, but the approach used for this study is not reported in the registry record.

The record reports zero analysis entries and does not provide outcome estimates or inferential results for these measures. It therefore does not show which outcomes were analyzed, whether the listed outcomes were all prespecified in their posted form, or how any testing hierarchy was handled. No multiplicity adjustment or interpretation plan is reported.

The primary efficacy population was defined as participants evaluated for overall response at Day 28, who had received at least one dose and had a lower GI biopsy consistent with acute GvHD. The record does not report the number in that population, the number assessed for each secondary outcome, or whether denominators differed across outcomes.

Missing-data methods are not reported. The registry does not state how missing Day 28 response assessments, withdrawals, or incomplete follow-up were handled, nor whether any imputation or sensitivity analyses were planned or conducted. This matters because the proportion achieving response can depend on how unevaluable participants are treated. With enrollment recorded as 1 and no outcome analysis reported, the available record does not allow evaluation of those choices or their effect on results.

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 non-serious adverse event terms

The registry safety summary records one participant at risk for serious adverse events and zero participants affected by a serious adverse event. It lists one participant at risk for other adverse events and one affected. Deaths are recorded as zero. These are counts, not comparative estimates, and the record does not provide event terms in the serious-event category.

The small number at risk limits what these counts can establish. In general, observing no serious event in a small group does not demonstrate that serious events cannot occur; the estimate remains highly uncertain. Likewise, an event count from one participant does not describe how frequently an event might occur in a larger or different population. No percentages are calculated from these counts.

The record does not report a corresponding safety summary for the albumin control group, or clarify whether the adverse-event group shown represents the entire study or a particular treatment group. The GLASSIA label appears in the registry safety summary, but the limited figures should not be treated as a full comparison of harms between the planned groups. The registry also does not provide serious-event terms, timing, severity details beyond the category, or an explanation of the recorded event counts. Zero recorded deaths is a description of the posted data, not a general conclusion about mortality risk.

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

The registry record establishes that a randomized, double-masked, parallel-group phase 2/3 study was planned to evaluate GLASSIA alongside steroid treatment for acute GvHD with lower gastrointestinal involvement. Part 1 assigned all participants to the GLASSIA combination, while Part 2 included a planned albumin control group with steroid treatment. The primary outcome was Day 28 overall response, defined using complete and partial response criteria.

The posted record does not establish whether GLASSIA changed the probability of overall response. It contains no primary response estimate, confidence interval, or p-value, and reports zero analysis entries. The primary outcome was skipped because fewer than two arms were available for it. A trial design and endpoint definition describe the intended question; they do not answer it when a comparative result is absent.

Enrollment is recorded as 1 and the study is marked terminated. These registry entries indicate that the registry reports very limited enrollment relative to the planned two-part design, but the record does not explain the termination or give arm assignments. It is not possible to determine from the posted information whether the participant contributed to the primary endpoint, how follow-up proceeded, or why the study ended.

The safety record reports zero serious adverse events among one participant at risk, one other adverse event among one participant at risk, and zero deaths. Those counts are descriptive and too limited to characterize a broader pattern or compare treatment groups. They do not establish that a treatment is safe or unsafe.

In general, clinical trial findings need to be interpreted in light of the population studied, the quality and completeness of outcome assessment, the size and precision of estimates, and the analysis methods. For this record, the outcome methods and analysis results are not reported in enough detail to support an efficacy conclusion. The appropriate reading is that the registry documents a planned clinical question and limited posted enrollment and safety information, rather than a demonstrated treatment effect.

14. Why This Trial Matters Statistically

This record is statistically instructive because it separates a trial's planned design from the evidence available for analysis. Randomization, masking, a comparator group, and a defined response endpoint are important design features, but they do not by themselves create an interpretable treatment-effect estimate. The registry reports a randomized, double-masked, parallel study and a planned GLASSIA-versus-albumin comparison in Part 2, yet it also records enrollment of 1, termination, and zero analysis entries.

The primary endpoint illustrates the importance of a clear outcome definition. Complete response and partial response were specified in relation to organ-level GvHD status, including a requirement that improvement not coincide with progression elsewhere. A reader can understand the intended measure, but cannot judge its observed frequency or compare groups because the registry does not report a response result or analyzed denominator.

The record also highlights why sample-size planning scenarios must be kept distinct from trial outcomes. The stated illustrations rely on assumptions including a two-sided significance level of 0.05, equal allocation, and an effect expressed as a fraction of the observed effect. They are planning illustrations, not evidence of achieved power, enrollment, or treatment benefit. With no primary comparative estimate reported, the planned design and the posted results remain different kinds of information.

Finally, sparse safety counts illustrate the limits of descriptive data from very small enrollment. The posted figures can be reported accurately, but they cannot support broad estimates of risk or comparisons between groups. For clinical researchers, the record is a reminder to examine actual analysis populations, denominators, missing-data handling, and uncertainty—not merely the design labels or endpoint definitions.

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 is that the registry reports total enrollment of 1 while describing a two-part study with a planned control group in Part 2. It does not report which part or treatment group included the participant, or provide arm-specific enrollment. The primary outcome was skipped because fewer than two arms were available for it, and the record lists zero analyses. No comparative efficacy estimate can therefore be assessed from the posted information.

The record does not state why the study was terminated. It also does not provide a detailed account of recruitment, follow-up, protocol changes, or whether any planned assessments were completed. The primary completion date is recorded as 2018-05-03, but that date alone does not describe the amount or completeness of outcome data.

The primary endpoint population is defined, but its size is not reported. No Day 28 response count, response percentage, confidence interval, or p-value is given. The record also does not state how missing or unevaluable assessments were handled. Analysis methods, covariate adjustment, multiplicity procedures, and sensitivity analyses are not reported in the registry record.

The safety information is limited to counts for one participant at risk in each displayed adverse-event category and zero deaths. Serious-event terms are absent, and a control-group safety summary is not reported. These limitations prevent assessment of the frequency, pattern, or comparative incidence of adverse events. The registry's design and endpoint descriptions clarify what was intended, but they do not resolve these gaps in the posted results.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Phase 2/3 Study of GLASSIA for the Treatment of Acute GvHD — NCT02956122.
• Registry record last updated 2021-01-13; status last verified 2020-12.
• Results first posted 2019-04-24.
• 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 describes a randomized, double-masked, parallel-group phase 2/3 study of GLASSIA added to steroid treatment for acute GvHD with lower gastrointestinal involvement. Part 1 planned GLASSIA plus steroid treatment; Part 2 planned GLASSIA plus steroid treatment versus albumin plus steroid treatment. The primary outcome was overall response at Day 28.

The recorded start date was 2017-04-26, the primary completion date was 2018-05-03, and the status is terminated. Enrollment is recorded as 1. The primary outcome was skipped because fewer than two arms were available for it, and the registry lists zero analyses. No primary efficacy estimate is reported. The safety summary records zero serious adverse events among one participant at risk, one other adverse event among one participant at risk, and zero deaths.