A Study of FMT in Patients With AML Allo HSCT in Recipients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of FMT in Patients With AML Allo HSCT in Recipients (NCT03678493), a randomized phase 2 study. The primary outcome is Efficacy of FMT in AML Patients and Allo-HCT Recipients - Number of Infections.

2. Clinical Question

3. Trial Design

This was a phase 2, randomized, parallel-group, placebo-controlled trial with a reported enrollment of 100. It involved two distinct cohorts: patients with AML receiving intensive chemotherapy and patients undergoing Allo-HCT. The brief summary describes randomization in a 2:1 ratio within each cohort and plans for up to 3 FMT or placebo treatments after antibacterial antibiotic exposure, continuing until 3 months after randomization. The registry record lists the study as completed, with a start date of 2019-09-09 and primary completion on 2022-02-11.

The record's planned cohort description does not match its stated overall enrollment: the brief summary describes 60 AML patients and 60 Allo-HCT patients, while the enrollment field reports 100. Posted primary endpoint group sizes are 18 and 8 for the AML comparison and 49 and 25 for the Allo-HCT comparison. The record does not explain how the planned cohort totals relate to the enrollment figure or the endpoint group sizes. These details should not be treated as interchangeable measures of recruitment or analysis population.

There is also a discrepancy in how masking is described. The structured field records no masking, while the brief summary calls the study double-blind. The registry record does not resolve that inconsistency. The arm descriptions are blank, and the record does not give further operational details about treatment administration, allocation concealment, or adherence. The published endpoint summaries should therefore be read within the limits of the design information recorded: randomized parallel comparisons, but incomplete and partly inconsistent reporting of masking and cohort accounting.

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

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 Efficacy of FMT in AML Patients and Allo-HCT Recipients - Number of Infections, measured over 4 Months.

The primary endpoint is the number of infections, described as the number occurring until 4 months after the first intervention. Infections were monitored at each visit through safety assessment. The endpoint is reported as a median and a 95% confidence interval for each treatment group in each cohort. A median describes the middle of the observed distribution and can be useful when outcomes are counts and the distribution may be skewed. It does not convey the full range of individual outcomes.

The clinical question is comparative: whether infection counts differ between FMT and placebo. However, the posted primary results are arm-specific summaries rather than an estimated difference between arms. A confidence interval attached to one arm's median describes uncertainty around that arm-level estimate; it is not a confidence interval for the treatment contrast. No primary comparison estimate or p-value is reported in the registry record. Consequently, the summaries provide a description of each group's infection count, but do not by themselves quantify the uncertainty around a between-group effect.

In the AML cohort, the median was 1.12 infections in the FMT group, with a 95% confidence interval from 0.57 to 1.67; the placebo median was 1.5, with an interval from 0.35 to 2.65. In the Allo-HCT cohort, the FMT median was 0.89, with an interval from 0.6 to 1.19; the placebo median was 1.09, with an interval from 0.55 to 1.64. These summaries refer to separate populations and should not be collapsed into a single estimate.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

A median is a measure of the center of a distribution. Unlike a mean, it is less affected by very large observations, which can be useful for outcomes such as counts when values are unevenly distributed. The primary endpoint is reported using medians, with 95% confidence intervals. The registry does not describe the calculation method for those intervals or give a separate measure of spread such as a standard deviation or interquartile range. The confidence interval should therefore be interpreted as the reported uncertainty around the median, not as a range containing most individual participants' infection counts.

What does a confidence interval describe?

In general, a confidence interval expresses uncertainty in an estimate under the assumptions of the method used to calculate it. Its width is influenced by the variability in the data and the amount of information available. A 95% confidence interval is not a statement that 95% of individual observations fall inside the interval. It is also not, by itself, a test of whether two groups differ. In this trial, the intervals attached to the primary endpoint belong to individual arms. A direct treatment comparison would ordinarily be reported as an effect estimate, such as a difference, with its own confidence interval and an appropriate analysis. No such primary comparison is reported in the registry record.

Why are the cohorts analyzed separately?

The registry presents AML patients receiving intensive chemotherapy and Allo-HCT recipients as separate cohorts, with separate intervention and placebo groups. Keeping those comparisons distinct avoids treating the cohorts as if they were one homogeneous population. The conditions and treatment settings differ, and the record does not report a combined primary analysis. The randomized design supports comparisons within the assigned groups, but the registry record does not give enough detail to assess allocation procedures, adherence, exclusions, or the analysis population beyond the posted group sizes.

How should the secondary comparison be read?

The secondary acute GVHD outcome is reported as percentages with confidence intervals, and a risk difference is also listed. The displayed comparison includes an unadjusted p-value of 0.029. This comparison is labelled as recalculated from the posted summary statistics, rather than reported as a sponsor analysis. A recalculated result is a simple, unadjusted comparison and may not match an analysis that accounts for design features or other prespecified factors. The registry reports no analyses for this outcome, so the comparison should not be represented as a reported sponsor analysis.

For a binary outcome, a risk difference expresses the difference in the proportion with the outcome between groups. In general, this gives an absolute contrast on the probability scale. A confidence interval for that contrast indicates the range of effect sizes compatible with the estimate and method. The registry's comparison also lists relative measures, but these are not a substitute for understanding the underlying event counts, the uncertainty, or the analysis approach. The record does not report adjustment for confounding factors, nor does it describe a prespecified method for analyzing this secondary outcome.

What do the planning illustrations mean?

The sample-size planning illustration is based on the observed secondary effect, not on the trial's own design justification. It assumes a two-sided significance level of 0.05, equal allocation, and a normal approximation. It gives participants per arm for 80% and 90% power at specified fractions of the observed effect. For planning at 100% of the observed effect, the illustration gives 47 participants per arm for 80% power and 62 for 90% power. At 75% of the observed effect, it gives 74 and 99 participants per arm, respectively; at 50%, it gives 144 and 192. These are planning illustrations, not observed power and not evidence that the trial was designed around those assumptions. The registry record does not report the original sample-size justification.

In general, planning calculations depend on the chosen outcome, target effect, variability or event rate, significance level, allocation, and analysis method. When an effect estimated in a small or uncertain study is used as the planning target, the resulting illustration can be sensitive to that estimate. The fractions shown describe smaller effects relative to the observed effect; they are scenarios for future planning, not alternative estimates of what happened in this trial. These calculations concern the secondary comparison and should not be mistaken for a sample-size analysis of the primary infection endpoint.

The registry record does not report the primary statistical model, handling of repeated measurements, missing-data methods, analysis populations, or any interim-analysis plan. General statistical considerations such as distributional assumptions and over-dispersion may matter for infection counts: over-dispersion occurs when observed variability exceeds what a simple count model assumes, and can affect standard errors and interval estimates. But no particular count model, adjustment, or over-dispersion assessment is reported in the registry record. These possibilities cannot be attributed to the trial's analysis without documentation.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 8 outcome measures and 7 posted secondary outcomes, including FMT engraftment twice. Several outcomes address different aspects of infection, while another concerns acute GVHD. Considering multiple outcomes raises a multiplicity issue: in general, the more statistical comparisons made, the greater the chance of at least one apparently notable result arising by chance. A prespecified strategy might define a primary outcome, a testing order, or an adjustment method. The registry record does not report a multiplicity plan, a hierarchy for secondary outcomes, or an adjustment for multiple testing.

The record's analyses list is empty, and it reports 0 analyses. This means there is little documentation of the inferential methods used for the posted outcomes. In particular, the registry record does not report the statistical analysis plan, the primary comparison method, covariate adjustment, handling of repeated infection observations, or whether analyses followed an intention-to-treat or another defined population. It also does not report interim analyses or stopping rules. Their absence from the record should not be taken to show that they did or did not occur; these details are not reported in the registry record.

Missing-data handling is not described. The posted group sizes for the primary endpoint are available, but the record does not explain whether they represent all randomized participants, participants with an observed endpoint, or another analysis set. It also does not state how withdrawals, incomplete follow-up, or unobserved infection assessments were handled. Without those details, it is not possible to judge whether missing outcomes could have influenced the arm summaries or comparisons.

The enrollment field reports 100, while the brief summary describes two planned cohorts of 60 each. The registry does not reconcile those figures. Nor does it explain the relationship between enrollment and the group sizes attached to the primary outcome. This reporting gap limits assessment of recruitment, retention, and the populations contributing data to each result. The registry record also gives conflicting masking descriptions: the structured field says none, whereas the brief summary says double-blind. These uncertainties matter when considering risk of bias and interpretation, but the record does not provide enough detail to resolve them.

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

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: bloodstream infection (55.6%), acute gvhd (18.4%), fever (12.5%), urinary tract infection (12.5%), pneumonia (10.2%).

The safety record reports serious adverse events and other adverse events by group. Serious events affected 16 of 18 AML patients receiving intensive chemotherapy with FMT and 4 of 8 in the AML placebo group. Among Allo-HCT recipients, serious events affected 22 of 49 in the FMT group and 4 of 25 in the placebo group. The same affected-participant counts are listed for other adverse events. These figures describe participants recorded as affected; they do not provide event rates per person-time or distinguish how many events each person experienced.

The listed serious-event terms include bloodstream infection (55.6%), acute GVHD (18.4%), fever (12.5%), urinary tract infection (12.5%), and pneumonia (10.2%). The registry record does not clarify the denominator or whether participants could contribute to more than one term, so these percentages should not be summed or treated as mutually exclusive categories. Deaths are reported as 0 in each AML group, 6 among Allo-HCT recipients assigned FMT, and 3 among Allo-HCT placebo recipients. The record does not attribute deaths to treatment or provide a causal assessment. These safety counts are descriptive and should be interpreted with the group sizes and incomplete reporting context in mind.

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

The posted results provide arm-specific medians and confidence intervals for the number of infections in two separate cohorts. They describe the observed endpoint summaries and show that the registry reports uncertainty around each median. They do not provide the primary treatment contrast, its confidence interval, or a primary p-value. Without a direct comparative estimate, the record does not quantify how precisely the groups differed in infection counts.

A difference between two displayed medians is not, by itself, a complete statistical analysis. General statistical inference for a treatment contrast depends on the data distribution, sample size, and method used to estimate uncertainty. The primary confidence intervals are attached to separate arm medians, not to a difference. Their overlap or separation cannot substitute for a comparison analysis, and the medians should not be interpreted as causal effects without the missing inferential context.

The results also do not establish that the same pattern applies across AML chemotherapy and Allo-HCT. These are separate cohorts, with different group sizes and distinct treatment settings. The record does not report a combined estimate or a formal comparison between cohorts. The secondary acute GVHD comparison is a separate outcome and is listed as an unadjusted recalculation; it does not answer the primary infection question or resolve the absent primary analysis.

The sample-size illustration is for planning, based on a secondary observed effect and stated assumptions. It is not observed power and does not establish whether the actual trial had adequate power for its primary endpoint. No original design justification is reported. Similarly, the adverse-event counts describe the safety record but do not establish that treatment caused or prevented any event. Overall, the registry supports a descriptive account of the posted outcomes, while important comparative, analysis, and reporting details remain unavailable.

13. Why This Trial Matters Statistically

This trial raises a central statistical issue common to randomized studies: a treatment question requires a comparison estimate, not only a summary for each group. The registry reports medians and confidence intervals for infection counts but does not report a primary between-arm effect or its uncertainty. Readers can see the endpoint summaries, yet cannot determine from the registry record how precisely the randomized groups differed according to the primary analysis.

The two-cohort structure makes clear reporting especially important. AML patients receiving intensive chemotherapy and Allo-HCT recipients are presented separately, with separate group sizes and a secondary outcome specific to Allo-HCT. A single overall statement could obscure meaningful differences between these populations, while separate descriptions without formal comparisons leave uncertainty about the evidence in each cohort. The record does not report a pooled analysis or a test comparing cohort effects.

The record also illustrates why trial metadata and results need to align. The stated enrollment differs from the total implied by the planned cohort description, and the masking field conflicts with the brief summary. The relationship between enrollment and the endpoint group sizes is not explained. Such inconsistencies do not determine the results, but they make it harder to assess who contributed data and how the study was conducted.

Finally, the secondary outcomes and sample-size illustration show why inferential context matters. A secondary p-value without a documented multiplicity plan has a different interpretation from a prespecified primary test. A planning calculation based on an observed secondary effect is not a retrospective measure of trial power. Clear reporting of the analysis plan, populations, missing-data methods, and primary contrast would be needed for a fuller statistical interpretation.

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.

Several limitations are specific to the registry record for this trial. First, the primary endpoint is reported only by arm. There is no primary difference estimate, confidence interval for the difference, or p-value, and the record does not describe the method used to produce the median confidence intervals. This prevents assessment of the precision of the main randomized comparison.

Second, the cohort and enrollment accounting is unclear. The enrollment field states 100, while the brief summary describes 60 AML patients and 60 Allo-HCT patients. The primary endpoint group sizes are 18 and 8 in the AML comparison and 49 and 25 in the Allo-HCT comparison. The record does not explain these differences or establish whether the endpoint groups include all randomized participants. Missing-data and analysis-population definitions are not reported.

Third, masking information is inconsistent: the structured field records none, but the brief summary describes a double-blind trial. Arm descriptions are blank, and details of allocation concealment, treatment adherence, and protocol deviations are not reported. The record also does not provide the statistical analysis plan, interim-analysis information, multiplicity handling, or a sample-size justification.

Finally, the secondary comparison is an unadjusted recalculation, and no reported analyses are listed. The seven secondary outcomes cover varied questions, but no overall strategy for interpreting them is reported. These reporting limitations do not demonstrate that specific methods were absent from the conduct of the trial; they mean those methods cannot be assessed from the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of FMT in Patients With AML Allo HSCT in Recipients — NCT03678493.
• Registry record last updated 2024-05-16; status last verified 2024-04.
• Results first posted 2024-05-16.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

The registry record describes a completed phase 2 randomized, parallel, placebo-controlled trial of FMT in AML patients receiving intensive chemotherapy and in Allo-HCT recipients. It reports enrollment of 100 and primary completion on 2022-02-11. The primary outcome is the number of infections through 4 months after the first intervention, reported as median values with 95% confidence intervals for each group in each cohort.

No primary between-arm analysis is reported. Secondary outcomes include FMT engraftment, acute GVHD, and infectious outcomes. The registry record contains conflicting masking descriptions, does not reconcile planned cohort numbers with enrollment, and does not report the analysis plan, missing-data methods, or multiplicity strategy. These gaps limit the interpretation of comparative effects and analysis populations.