ATI-450 vs Placebo in Participants With Moderate to Severe Hidradenitis Suppurativa (HS): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for ATI-450 vs Placebo in Participants With Moderate to Severe Hidradenitis Suppurativa (HS) (NCT05216224), a randomized phase 2 study. The primary outcome is Change From Baseline in Abscess and Inflammatory Nodule (AN) Count at Week 12.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group Phase 2a study comparing ATI-450 with placebo in participants with moderate to severe hidradenitis suppurativa. The registry lists quadruple masking and a total enrollment of 95. It records the study as completed. The recorded study start was 2021-12-29, and the primary completion date was 2022-12-23.

The ATI-450 arm received a 50 milligrams (mg) oral tablet twice daily; the placebo arm received a placebo oral tablet twice daily. The primary endpoint was assessed at Week 12. The registry record describes the study’s broader objectives as investigating efficacy, safety, tolerability, pharmacokinetics, and pharmacodynamics.

The primary endpoint results are reported for 48 participants in the ATI-450 arm and 47 in the placebo arm. These are the analysis counts reported for that endpoint, not a substitute for the registry’s overall enrollment figure. The primary analysis population is defined as participants who were randomized and administered at least 1 dose of study drug. The registry does not report further detail about randomization procedures, how masking was maintained, allocation concealment, or the study’s sample-size justification.

Other posted outcomes include clinical response, disease severity, a participant-reported measure identified as NRS30, physician assessment, quality of life, treatment-emergent adverse events, and ATI-450 and metabolite concentrations. Outcome-specific analysis counts vary in the posted results. The registry record does not explain all of those differences.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: FAS population: All participants who were randomized and administered at least 1 dose of study drug.

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: Full analysis set (FAS) population included all participants who have been randomized and administered at least 1 dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Abscess and Inflammatory Nodule (AN) Count at Week 12, measured over Baseline, Week 12.

The primary endpoint was change from baseline in abscess and inflammatory nodule (AN) count at Week 12. The count combines the number of abscesses and inflammatory nodules across anatomical regions. A change from baseline describes how the count at the follow-up assessment differs from the baseline count; a negative change indicates a lower count than at baseline.

The endpoint focuses on two lesion types and provides a numeric measure of change over the study period. It is distinct from a responder endpoint, which classifies participants according to whether they meet a specified response definition. The registry also lists response-based outcomes, but the primary measure is the model-estimated change in AN count.

The posted description says the least-squares mean change was estimated using a mixed model for repeated measures (MMRM). This approach is intended to estimate group-level changes while using repeated outcome measurements in the model. The registry gives the timeframe as Baseline, Week 12 and reports a 90% confidence interval for the arm estimates.

The record does not report a clinically important threshold for AN-count change or explain how the primary endpoint was selected. Accordingly, the numeric result describes the estimated change and its precision; it does not by itself establish how meaningful a particular change would be to participants.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary comparison uses a mixed model for repeated measures (MMRM), as stated in the registry record. In general, a repeated-measures model analyzes outcome measurements collected over time and accounts for the fact that measurements from the same participant are related. It estimates patterns of change using the measurements included in the model rather than treating each measurement as if it came from an unrelated person. The registry describes the primary result as a least-squares mean change from baseline at Week 12.

What does a least-squares mean represent?

In general, a least-squares mean is a model-based estimate of a group’s mean outcome. It can reflect the model’s structure and included information, rather than being identical to a simple arithmetic average of observed values. The registry reports the primary arm estimates as -2.93 for ATI-450 and -5.94 for placebo. It does not list the model’s covariates, assumptions, repeated-measure covariance structure, or estimation details, so those aspects cannot be assessed from the registry record.

How should the confidence intervals be read?

The primary arm estimates and their contrast are accompanied by 90% confidence intervals. In general, a confidence interval describes the statistical uncertainty around an estimate under the model and assumptions used. A narrower interval indicates greater precision than a wider interval, while the interval’s position shows which values are compatible with the estimate at the stated confidence level. A confidence interval is not a prediction interval for individual outcomes and does not guarantee that the true value lies within the particular reported interval.

The primary contrast is 3.01, with a 90% confidence interval from 1.50 to 4.52. The registry also reports p > 0.999. Because the registry does not explain how that p-value relates to the reported contrast and interval, the numerical inconsistency should not be resolved by substituting a new analysis. The sponsor’s reported analysis takes precedence when the record does not flag a discrepancy.

What do the secondary comparisons add?

For the two binary secondary outcomes with posted comparisons, recalculated risk differences compare the observed response proportions between ATI-450 and placebo. These are simple, unadjusted recalculations from the posted summary statistics. The HiSCR50 risk difference is -0.198, with a 95% confidence interval from -0.392 to -0.004 and an unadjusted p-value of 0.05. The NRS30 risk difference is -0.146, with a 95% confidence interval from -0.370 to 0.077 and an unadjusted p-value of 0.204. The 90% arm confidence intervals displayed for these endpoints are separate from the 95% confidence intervals for the recalculated risk differences.

These recalculations do not replace sponsor analyses. The registry lists no reported comparative analyses for those secondary outcomes. An unadjusted p-value does not account for the fact that several outcomes were assessed. In general, testing multiple outcomes increases the opportunity for chance findings, so secondary results without a reported multiplicity strategy are less conclusive than a prespecified primary comparison.

What are the planning figures?

The planning illustration is based on the observed effect and is not the trial’s own design justification or observed power. The recorded assumptions are a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation; the effect to detect is expressed as a stated fraction of the observed effect. For planning for 100% of the observed HiSCR50 effect, the illustration gives 96 participants per arm for 80% power and 128 participants per arm for 90% power. For planning for 75% of the observed effect, it gives 174 participants per arm for 80% power and 232 participants per arm for 90% power. For planning for 50% of the observed effect, it gives 397 participants per arm for 80% power and 530 participants per arm for 90% power. These are planning illustrations, not evidence that the completed study had a particular power.

For NRS30, planning for 100% of the observed effect gives 182 participants per arm for 80% power and 243 participants per arm for 90% power. Planning for 75% of the observed effect gives 324 participants per arm for 80% power and 433 participants per arm for 90% power. Planning for 50% of the observed effect gives 725 participants per arm for 80% power and 970 participants per arm for 90% power. These figures depend on the stated planning assumptions and the observed effect used as their starting point; they do not establish what effect would occur in another study.

Why can over-dispersion matter?

In general, over-dispersion occurs when outcome variability is greater than a statistical model assumes. If present but ignored, uncertainty can be understated and inferential results can become too optimistic. The registry does not report an over-dispersion assessment for these analyses, and the posted endpoint description does not identify a count-distribution model. It is therefore not possible to determine from the registry record whether over-dispersion was assessed or relevant to the reported model.

The record also does not report interim analyses, a sample-size justification, the full statistical analysis plan, or detailed missing-data procedures. These omissions constrain evaluation of model specification, outcome selection, and the handling of incomplete observations.

10. Analyses Reported by the Sponsor

The sponsor-reported primary analysis uses MMRM and gives an LS mean difference of 3.01, with a 90% confidence interval from 1.50 to 4.52 and a p-value of >0.999. The registry record does not flag a discrepancy between that analysis and recalculated results. Under the stated reporting convention, the sponsor’s own analysis takes precedence; no recalculated primary result should be substituted for it.

The reported p-value is difficult to reconcile with the positive contrast and confidence interval, but the registry record does not explain the relationship. This should be treated as an unresolved reporting issue rather than as grounds to infer a different p-value or a different analysis. The sponsor’s analysis and the posted statistics should be read as recorded, with that limitation made explicit.

For the binary secondary outcomes, no sponsor-reported comparative analyses are listed. The risk differences and unadjusted p-values are recalculations from posted summary statistics, not sponsor analyses. They do not include a reported adjustment for multiple outcomes and should carry less inferential weight than a prespecified, fully described analysis.

The registry record does not provide enough detail to determine whether additional sponsor analyses were conducted but not posted. Their existence, methods, and results are not reported in the registry record.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and several secondary outcomes, but it does not report a multiplicity adjustment or a hierarchy for testing outcomes. Multiplicity matters because, in general, conducting several statistical tests creates more opportunities for a chance finding. The unadjusted secondary p-values therefore provide limited evidence on their own and should not be read as if they were protected by a reported family-wise testing strategy.

The registry defines the primary analysis population as the FAS population: all participants who were randomized and administered at least 1 dose of study drug. This definition does not by itself explain how participants with missing Week 12 measurements were handled. The registry record does not report the amount or pattern of missing primary outcome data, the assumptions used for missing observations, or any imputation or sensitivity analyses.

Analysis counts differ across the posted outcomes. For example, the primary outcome reports 48 participants in the ATI-450 arm and 47 in the placebo arm, while NRS30 reports 37 and 38. These outcome-specific counts are stated in the record, but the reason for the differences is not reported in the registry record. They should not be treated as evidence of a particular missing-data mechanism.

The record describes MMRM for the primary analysis but does not report the model’s missing-data assumptions or how robustness to those assumptions was evaluated. In general, repeated-measures models can use available observations under specified assumptions; they do not automatically remove bias if missingness is related to unobserved outcomes. Without the model details and missing-data sensitivity analyses, the robustness of the estimate cannot be fully assessed.

The registry also does not report interim analyses, a full statistical analysis plan, or a sample-size justification. Their methods and existence are not reported in the registry record. The planning illustrations based on observed secondary effects should not be mistaken for the trial’s original design calculations or for observed power.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hidradenitis (2.1%).

The posted safety counts include serious adverse events, other treatment-emergent adverse events, and deaths. In the ATI-450 group, the record lists 0 serious adverse events among 48 at risk, 23 with other treatment-emergent adverse events among 48 at risk, and 0 deaths. In the placebo group, it lists 1 serious adverse event among 47 at risk, 13 with other treatment-emergent adverse events among 47 at risk, and 0 deaths. The record also identifies Hidradenitis at 2.1% among the top serious terms.

These are descriptive counts. They do not establish that an event was caused by study treatment, and they do not by themselves determine the comparative safety profile. The number of serious events is small, so the posted counts provide limited precision for comparing uncommon harms. The registry does not provide event-level detail, timing, duration, or an assessment of relatedness in the information presented.

Other treatment-emergent adverse events are reported as counts of participants affected, not as a complete account of event severity or clinical course. Zero deaths in the posted groups means no deaths are listed for those groups in this record; it does not establish that the treatment is free of risk. Safety findings should be interpreted alongside the study’s size, duration, and the limitations of the posted detail.

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

The primary endpoint describes modeled changes in abscess and inflammatory nodule count from baseline to Week 12. The reported arm estimates are negative, and the reported contrast is positive for ATI-450 relative to placebo. Numerically, the estimated reduction was smaller in the ATI-450 arm than in the placebo arm. This is a description of the posted endpoint estimates; it is not a recommendation about treatment.

The primary result does not establish a universal effect for all people with hidradenitis suppurativa. It comes from a Phase 2a study, a defined analysis population, a particular endpoint, and the follow-up period recorded in the registry. The record does not report a clinically important threshold for the endpoint, so the clinical meaning of the magnitude cannot be determined from the estimate alone.

The reported primary p-value of >0.999 does not appear to align with the contrast and its confidence interval, and the registry record does not explain why. That unresolved issue limits a straightforward reading of the primary inference. The sponsor-reported analysis remains the analysis to quote, while the unexplained reporting relationship should remain visible to readers.

Secondary outcomes and safety counts provide additional descriptive information but do not settle the primary question. The secondary comparisons are unadjusted, and multiplicity handling is not reported. Some listed outcomes do not have usable posted two-arm comparisons. Safety counts are not proof of causation and are too limited to characterize uncommon events precisely.

Overall, the registry record supports a careful account of the reported estimates, intervals, and counts. It does not, on its own, provide enough statistical-plan detail to judge all model assumptions, missing-data handling, or the full confirmatory status of secondary findings. Results should be interpreted within those boundaries.

14. Why This Trial Matters Statistically

This trial is statistically informative because it reports both model-based changes in a count endpoint and responder outcomes. Those measures answer different questions: the primary endpoint estimates average change in lesion count, while a responder measure classifies participants using a response definition. Looking across measures can reveal whether the broad direction is consistent, but it does not make them interchangeable.

The record also illustrates why estimates should be read together with their uncertainty and analysis context. The primary contrast has a reported 90% confidence interval, but its accompanying p-value is difficult to reconcile with that interval and the contrast. A statistical result is not fully interpretable from a single number; transparent reporting of estimates, intervals, model details, and analysis populations matters.

Secondary findings require particular care when several outcomes are recorded. Without a reported multiplicity strategy, unadjusted comparisons are better treated as exploratory than as independent confirmation. Similarly, planning illustrations based on observed effects can show how study-size requirements vary with the effect assumed, but they are not evidence of observed power and do not explain the original design.

Finally, the outcome-specific analysis counts and incomplete reporting of missing-data methods show why population definitions and analysis procedures matter. The registry identifies the primary analysis population but does not detail the reasons for differing counts or the assumptions behind incomplete observations. These are important considerations for assessing precision and potential bias, beyond the numerical point estimates themselves.

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.

Several limitations are specific to the posted record. First, the primary p-value is reported as >0.999 alongside an LS mean difference of 3.01 and a 90% confidence interval from 1.50 to 4.52. The registry record does not explain this relationship. Although no discrepancy is flagged, the reported statistics are difficult to reconcile, limiting a clear interpretation of the primary inference.

Second, the record does not report the full MMRM specification, including covariates, covariance structure, or diagnostic assessments. It does not provide the detailed missing-data pattern, missing-data assumptions, or sensitivity analyses. The primary population is defined, but that definition alone does not establish how incomplete outcome data affected the estimate.

Third, comparative sponsor analyses are not listed for the secondary outcomes. The posted recalculations for the two binary outcomes are unadjusted, and the registry does not report a multiplicity procedure. Several other secondary measures are listed without usable two-arm comparisons, so the overall secondary evidence is incomplete.

Fourth, the registry does not report a sample-size justification, a full statistical analysis plan, or interim-analysis information. The planning illustrations use observed effects and should not be treated as the original design rationale. Safety reporting gives group counts but not detailed event-level information or a full account of event relatedness and timing.

These reporting limits do not establish that analyses or procedures were absent; they mean those details are not reported in the registry record. They constrain how independently the statistical choices and robustness of the results can be evaluated.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: ATI-450 vs Placebo in Participants With Moderate to Severe Hidradenitis Suppurativa (HS) — NCT05216224.
• Registry record last updated 2026-01-07; status last verified 2025-12.
• Results first posted 2026-01-07.
• 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 NCT05216224 as a completed, randomized, parallel-group, quadruple-masked Phase 2a study comparing ATI-450 with placebo in participants with moderate to severe hidradenitis suppurativa. Enrollment was 95. ATI-450 was administered as a 50 milligrams (mg) oral tablet twice daily, and placebo was administered as an oral tablet twice daily.

The primary endpoint was change from baseline in abscess and inflammatory nodule count at Week 12, estimated using MMRM in the FAS population. The reported LS mean changes were -2.93 for ATI-450 and -5.94 for placebo; the reported LS mean difference was 3.01, with a 90% confidence interval from 1.50 to 4.52 and p > 0.999. The registry does not explain the apparent tension among these statistics.

Secondary outcomes cover clinical response, severity, participant-reported and physician assessments, quality of life, adverse events, and drug concentrations. Safety counts include 0 serious adverse events and 0 deaths in the ATI-450 group, and 1 serious adverse event and 0 deaths in the placebo group. Important analysis-plan, multiplicity, and missing-data details are not reported in the registry record.