Study of the Safety and Tolerability of AXA1125 and AXA1957 in Subjects With Non-Alcoholic Fatty Liver Disease (NAFLD): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of the Safety and Tolerability of AXA1125 and AXA1957 in Subjects With Non-Alcoholic Fatty Liver Disease (NAFLD) (NCT04073368), a randomized na study. The primary outcome is Number of Subjects With Incidence of Study Product Emergent Adverse Events (AEs) and Any Serious Adverse Events (SAEs).

2. Clinical Question

The brief summary says the study was intended to assess safety and tolerability and to examine liver biology using blood tests and magnetic resonance imaging. Listed secondary outcomes include MRI-based liver fat, insulin resistance, glucose homeostasis, alanine aminotransferase, and aspartate aminotransferase. These measures could describe biological changes, but the registry record does not post analyses for them. The clinical question that can be addressed from the posted results is therefore limited to the reported event counts and study conduct; treatment effects on the secondary outcomes cannot be assessed from this record.

3. Trial Design

This was a randomized, single-blind, parallel-group study with a placebo comparator. Its official title describes a 16-week food study, and the primary outcome timeframe is baseline to week 16. The record lists the study as completed, with enrollment of 112 and a phase entry of NA. It gives the sponsor as Axcella Health, Inc. and identifies NAFLD as the condition under study.

Four groups appear in the safety results: AXA1957 high dose, AXA1957 low dose, AXA1125, and placebo. The respective listed doses are AXA1957 20.3 g, AXA1957 13.5 g, AXA1125 24 g, and placebo 24 g. The primary outcome description specifies twice-daily administration for the active study products, and the population description says participants received study product or placebo for up to 16 weeks. The record does not give further detail about adherence, dose interruptions, or how exposure duration varied between participants.

Randomization indicates that assignment was intended to occur by chance, but the sequence-generation and allocation-concealment procedures are not reported in the registry record. “Single blind” is recorded, but the identity of the masked party and the practical masking procedures are not reported in the registry record. The record also does not state the planned allocation ratio or explain the difference between total enrollment and the group-specific safety denominators. These details matter when judging whether the groups were comparable and how completely the safety results represent all enrolled participants.

The study was designed around safety and tolerability, with additional biological outcomes listed. Its registry entry does not provide a full protocol or a detailed statistical analysis plan. The design description alone cannot establish whether every stated outcome was analyzed as planned or how missing observations were handled.

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: 112 subjects signed consent and randomized. Ten subjects screen failed prior to day 1 dosing ( after signing consent) and the number of participants who started the study is 102 subjects.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Subjects received AXA1125 at 24 g BID, AXA1957 at 20.3 g BID or 13.5 g BID, or placebo for up to 16 weeks.

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: This is the safety analysis population

6. Primary Endpoint

The registry lists the primary outcome as Number of Subjects With Incidence of Study Product Emergent Adverse Events (AEs) and Any Serious Adverse Events (SAEs), measured over Baseline to week 16.

The primary outcome is the number of participants with study-product-emergent adverse events and any serious adverse events from baseline to week 16. It is a participant-count endpoint: it concerns whether participants experienced an event, rather than the total number of event episodes or the time until an event. The record describes the endpoint as measuring product-related adverse events and any serious adverse events up to 16 weeks. It does not provide a detailed definition of “product-related” or explain how events were adjudicated.

Safety outcomes are used to describe unwanted events observed during exposure to a study product. Counting participants with events can make the results easier to interpret than reporting event episodes alone, because one participant may experience more than one episode. However, a participant count does not convey the number, duration, severity, or timing of all events. Serious and non-serious events also represent distinct categories, so their counts should not be treated as interchangeable.

The record lists separate serious-event and other-event participant counts by group. It does not post an inferential comparison for the primary endpoint, and the primary outcome is flagged as having several categories or time points rather than being reduced to one comparison. Consequently, the record supports a descriptive reading of the reported safety counts, but it does not establish a comparative statistical effect for this endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Safety results are often presented first as counts of participants with events. Such counts are a direct descriptive summary: they show how many people in each listed group were recorded as affected. The denominators indicate how many participants were included in the corresponding risk set. For this study, the registry lists separate denominators for serious and other adverse events, with the same values for each category within a group. These figures do not reveal whether all participants had identical follow-up or exposure, because participant-level follow-up information is not reported in the registry record.

What can a participant count tell us?

A participant count answers whether an event was recorded for a person, not how many event episodes that person experienced. If a participant had repeated events, a count of affected participants could still count that person once. Counts also do not describe event onset, duration, severity beyond the serious/non-serious classification, or whether events resolved. The registry record does not provide these details for each event. The count is therefore useful as a basic description, but it is not a complete summary of the clinical course or event burden.

Serious adverse events and other adverse events should be kept distinct. Seriousness is a classification, not a synonym for how common an event was or how strongly it was associated with treatment. A small count of serious events may be important to describe, but interpreting it requires context such as the event definition, follow-up time, participant characteristics, exposure, and the method used to assess relatedness. The record provides selected safety counts and terms, but not all of that context.

Why are raw counts not treatment comparisons?

Groups can differ in size, and raw counts do not account for that difference. A comparative analysis might estimate a difference or ratio in event occurrence, or model participant-level outcomes while accounting for the study design. Such analyses require stated methods and suitable data. No inferential analysis is posted in the registry record, so the counts should not be converted into a claim of a difference between products. A general statistical point is that sparse events can yield imprecise estimates: when only a few events are observed, the range of plausible comparative effects can be broad. No confidence intervals are available to quantify that uncertainty.

For continuous outcomes, such as changes in liver fat or laboratory measures, statistical interpretation would normally depend on a defined endpoint, a comparison between groups, an estimate of the effect, and a measure of uncertainty. The registry lists these biological measures as secondary outcomes but does not provide analysis results. The names of outcomes alone do not establish the direction or magnitude of any observed change.

What does randomization contribute?

In general, random allocation is intended to reduce systematic differences between study groups at the start of a trial. Its effectiveness depends on how the sequence is generated and concealed, how participants are assigned, and whether subsequent follow-up and analysis preserve the randomized comparison. The registry identifies the study as randomized but does not report the sequence-generation or concealment methods. The entry also labels the study single blind without identifying which people were masked. These omissions prevent a detailed assessment of safeguards against selection and measurement bias.

The record describes the study as parallel-group. In general, this means each participant is assigned to a group and followed in that group rather than receiving a sequence of study products. The registry does not provide participant-level data or a detailed account of departures from assigned treatment, so it is not possible to assess adherence or crossover from the posted record.

How should planning calculations be read?

Sample-size calculations are planning tools: they estimate how many participants might be needed under stated assumptions about the effect to detect, the statistical test, and the allocation. They are not evidence that an effect was observed, and they are not “observed power.” The record does not report the trial’s own sample-size justification or detailed design assumptions. Any planning illustration based on an observed effect is separate from the original study design and must not be read as a result of the trial. The registry’s primary outcome is not reduced to one comparison, so a single observed-effect planning illustration cannot substitute for a prespecified analysis of all relevant safety categories.

Over-dispersion is a general statistical concern when count data vary more than a model expects. If present, a model that assumes less variability than the data contain can understate uncertainty and make results appear more precise than warranted. The posted information does not include participant-level event counts, model diagnostics, or a count-data analysis, so there is no basis to assess over-dispersion for this study. More broadly, choosing a model requires knowing the outcome definition, observation time, repeated-event structure, and analysis population. Those details are not reported in the registry record.

Finally, a complete interpretation would usually distinguish the planned analysis from exploratory analyses, state how missing data were addressed, and report estimates with uncertainty. The registry record does not provide those elements. The appropriate statistical reading is therefore descriptive and limited to the available participant counts, study structure, and outcome listings.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary safety outcome and several secondary biological outcomes, but the analysis count is 0. It does not post statistical comparisons or a detailed analysis plan. The primary outcome is identified as involving several categories or time points and is not reduced to one comparison. This means the posted entry does not define a single inferential result that can be interpreted as the primary treatment contrast.

Multiplicity arises when several outcomes, groups, time points, or comparisons are tested. In general, evaluating many tests without accounting for their number can increase the chance of finding an apparently notable result by chance. A prespecified hierarchy, adjustment procedure, or clear designation of exploratory analyses can help readers interpret a family of tests. The registry record does not report whether any multiplicity strategy was planned or used. Since inferential results are not posted, there is no reported multiplicity adjustment to assess.

Missing-data methods and analysis populations are also not described. The record gives total enrollment of 112 and group-specific safety denominators of 32, 26, 29, and 15, but does not explain the relationship between these figures or identify the population used for each analysis. It does not state how participants with incomplete follow-up, treatment discontinuation, or missing outcome measurements were handled. These omissions matter because the set of participants included can affect both the estimated event frequencies and biological outcome comparisons.

For the secondary outcomes, the record does not provide the number analyzed, the amount of missing data, or any method for handling missing observations. It is therefore not possible to determine whether those analyses used all randomized participants, participants who received treatment, or another analysis population. The registry record also does not report imputation methods or sensitivity analyses. Without these details, the robustness of any outcome analysis cannot be assessed.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: nephrolithiasis (3.4%), laryngeal squamous cell carcinoma (3.1%).

The serious adverse event counts are 1 of 32 in the AXA1957 high-dose group, 0 of 26 in the AXA1957 low-dose group, 1 of 29 in the AXA1125 group, and 0 of 15 in the placebo group. The record lists no deaths. These are the numbers of participants affected, not necessarily the total number of serious event episodes. Zero recorded participants in a group means none were listed in this registry result; it does not establish that the underlying risk is zero.

The registry’s top serious adverse event terms are nephrolithiasis at 3.4% and laryngeal squamous cell carcinoma at 3.1%. The record does not attach these terms to a specific study group or explain the basis for those percentages in the safety summary. They cannot be used to infer a group-specific rate or a causal relationship to a study product.

Safety counts are descriptive observations over the study period. A comparison requires attention to denominators, follow-up, event definitions, and uncertainty. No inferential comparison or confidence interval is posted for these serious-event figures. The fact that the trial was randomized does not by itself turn the observed counts into a definitive estimate of comparative risk, particularly when the analysis details are not reported.

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

The registry record documents a randomized, single-blind, placebo-controlled study in participants with NAFLD, with three active-product groups and a placebo group. It records safety outcomes through week 16 and gives counts of participants with serious events, other events, and deaths. The reported counts provide a descriptive account of events in the listed safety groups; they are not a complete evaluation of comparative safety because no inferential analysis or uncertainty estimates are posted.

The primary outcome is safety-focused. It does not establish whether AXA1125 or AXA1957 changed liver fat, insulin resistance, glucose homeostasis, ALT, or AST. Those measures are named as secondary outcomes, but their results are not posted in the registry record. Neither the study title nor the list of intended measurements can stand in for outcome estimates.

The record also does not establish that a product caused a particular event. Event counts observed during a trial can reflect many factors, and causal interpretation depends on the design, conduct, event assessment, participant follow-up, and analysis. Randomization is a useful design feature in general, but the registry does not give enough methodological detail to assess all these factors. The group counts are not accompanied by a reported treatment-effect estimate or a quantified range of uncertainty.

These results should not be treated as a claim that a product is safe or unsafe, or that it improves or fails to improve a clinical or biological outcome. They are a limited registry summary of observed safety counts. A fuller interpretation would require a complete report with participant flow, event definitions and details, outcome estimates, statistical methods, missing-data handling, and analysis populations. The record’s completed status does not supply those missing elements.

13. Why This Trial Matters Statistically

This study illustrates an important distinction in clinical-trial statistics: a stated endpoint and a set of observed counts are not the same as a comparative result. The registry identifies safety and tolerability as the main purpose and reports participant counts for serious and other adverse events. Without an analysis that quantifies group differences and uncertainty, the counts remain descriptive. This distinction is particularly important in safety studies, where sparse events can make apparent group contrasts unstable.

The study also lists biological outcomes alongside safety outcomes. MRI-PDFF, HOMA-IR, glucose homeostasis, ALT, and AST address different aspects of liver biology and metabolism. Each would require a clearly defined analysis, including the measurement time, comparison, analysis population, and handling of missing observations. Their inclusion in the registry indicates that they were planned outcomes, but does not reveal their results.

The unequal group-specific safety denominators and the difference from total enrollment highlight why clear participant accounting matters. Readers need to know who was randomized, who received study product, who completed follow-up, and who contributed to each outcome. Those details help distinguish differences in observation from differences in events. They are not explained in the registry record.

Finally, this record shows why interpretation depends on transparent methods as well as trial design. Randomization and masking are meaningful design features, yet their implementation and the analysis plan are not detailed in the registry. Reporting estimates with uncertainty, documenting missing data, and explaining how multiple outcomes were handled are essential for evaluating the strength of evidence. In their absence, a careful statistical account should stay within the limits of the posted descriptive safety information.

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.

The registry record does not provide a full statistical analysis plan, and no analyses are posted. It does not report comparative estimates or confidence intervals for the primary safety outcome, nor does it provide results for the listed secondary outcomes. As a result, the record cannot support an assessment of statistical precision or comparative treatment effects.

The primary outcome description refers to product-related adverse events and any serious adverse events, while the posted safety results separate serious events from other events. Detailed definitions, event attribution procedures, adjudication methods, and the handling of participants with multiple events are not reported in the registry record. The specific primary endpoint categories or time points are not reduced to a single comparison.

Participant accounting is incomplete in the posted information. Enrollment is 112, while the safety denominators are reported separately as 32, 26, 29, and 15. The registry does not explain the difference, provide a participant-flow account, or specify the analysis population for each outcome. Follow-up completeness, adherence, discontinuation, and missing-data handling are also not reported.

Important design details are absent. The record does not describe random-sequence generation, allocation concealment, who was masked, or how masking was maintained. It also does not report the study’s own sample-size justification, interim analyses, or multiplicity plan. The phase entry is NA, and the record does not provide further clarification. These omissions limit independent assessment of bias, planned precision, and the relationship between intended and reported analyses.

The listed serious-event terms and percentages are not linked to study groups in the safety summary. They cannot be interpreted as arm-specific risks. Overall, the available information is a registry-level description rather than a complete clinical study report.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Study of the Safety and Tolerability of AXA1125 and AXA1957 in Subjects With Non-Alcoholic Fatty Liver Disease (NAFLD) — NCT04073368.
• Registry record last updated 2021-07-06; status last verified 2021-06.
• Results first posted 2021-07-02.
• 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

NCT04073368 is recorded as a completed, randomized, single-blind, parallel-group, placebo-controlled study of AXA1125 and AXA1957 in participants with NAFLD. The official title describes a 16-week food study; total enrollment is 112. The four groups were AXA1957 high dose, AXA1957 low dose, AXA1125, and placebo, with doses listed as 20.3 g, 13.5 g, 24 g, and 24 g, respectively.

The primary outcome concerns participants with study-product-emergent adverse events and any serious adverse events from baseline to week 16. The safety results list serious-event denominators of 32, 26, 29, and 15 and other-event counts by group. No deaths are recorded. The registry also lists liver fat, insulin resistance, glucose homeostasis, ALT, and AST as secondary outcomes, but reports no analyses. Comparative estimates, uncertainty measures, analysis populations, and missing-data methods are not reported in the registry record.