Study of Safety, Tolerability and Efficacy of DFV890 in Participants With Familial Cold Auto-inflammatory Syndrome (FCAS): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Safety, Tolerability and Efficacy of DFV890 in Participants With Familial Cold Auto-inflammatory Syndrome (FCAS) (NCT04868968), a na phase 2 study. The primary outcome is Ratio of Fold Change From Pre-challenge to the Highest Post-challenge Value of White Cell Count (WCC) Between Treatment and Screening Period.

2. Clinical Question

In practical terms, the study asked whether the WCC response observed after a cold challenge on Day 4 of treatment differed from the response observed during screening. The challenge lasted 45 minutes, and WCC was assessed from before the challenge through up to 8 hours afterward. The endpoint compares the highest post-challenge value with the pre-challenge value in each period, then compares those two fold changes.

The record also lists safety and disease-activity assessments, including treatment-emergent adverse events, serious adverse events, physician assessments, and a patient global assessment. These are additional questions about tolerability and disease activity, but their numerical results are not reported in the registry record except for the adverse-event counts described in the safety information. Because there was no randomized control group and only 4 participants enrolled, the study is a small within-participant assessment rather than a controlled estimate of a treatment effect.

3. Trial Design

This was an open-label, single-group phase II study of DFV890 in participants with FCAS. The record lists Novartis Pharmaceuticals as the sponsor and gives a study start date of 2021-09-20 and a primary completion date of 2022-12-13. The study status is completed. Four participants were enrolled, compared with 6 planned. The record attributes early closure to an unforeseen screen failure rate and recruitment challenges.

There was no allocation to different treatment groups and no masking. All participants were assigned to the single listed DFV890 group, described as DFV890 100 mg. Since there was no separate randomized comparison group, the screening-period measurement served as the reference period for the primary pharmacodynamic comparison. Such a within-participant comparison differs from a parallel-group trial: it compares measurements from two periods in the study rather than outcomes between treatment and control groups.

The primary endpoint was assessed during screening and on Day 4 of treatment. Participants underwent a 45-minute cold challenge in each period, with WCC measured before the challenge and up to 8 hours afterward. The endpoint was based on the highest post-challenge WCC value relative to the pre-challenge value in each period, and then on the ratio of the treatment-period fold change to the screening-period fold change.

The registry lists four secondary outcome categories: treatment-emergent adverse events and serious adverse events, physician global assessment of autoinflammatory disease activity, physician severity assessment of disease signs and symptoms, and patient global assessment of disease activity. There were 5 outcome measures in total. Numerical results for the secondary assessments are not reported in the registry record. The record does not report a detailed schedule or instrument descriptions for those assessments beyond their listed outcome titles.

The pharmacodynamic analysis population included participants without protocol deviations judged to have relevant impact on the pharmacodynamic data. The number meeting that definition is not reported in the registry record. This distinction matters because enrollment does not necessarily equal the number contributing to a particular analysis.

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: During the screening period the participant's eligibility was assessed at a screening visit and a screening cold challenge was performed.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Pharmacodynamic (PD) analysis set included all participants with no protocol deviations with relevant impact on PD data.

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 Ratio of Fold Change From Pre-challenge to the Highest Post-challenge Value of White Cell Count (WCC) Between Treatment and Screening Period, measured over Screening period and treatment period (Day 4): pre cold challenge and up to 8 hours post cold challenge. The duration of the cold challenge was 45 minutes..

The primary endpoint compares the WCC response to a cold challenge during treatment with the response during screening. In each period, the fold change was defined as the highest WCC value recorded after the challenge divided by the pre-challenge WCC value. The endpoint was the treatment-period fold change divided by the screening-period fold change.

This relative measure is intended to describe whether the WCC increase after challenge was smaller or larger during treatment than during screening. The registry record defines a value below 1 as a lower relative increase during treatment, which it characterizes as favorable. A value of 1 would represent no relative difference between the two fold changes; values above 1 would indicate a higher relative increase during treatment. These are definitions of the measure, not reported findings from the study.

A fold change expresses a response relative to a participant’s own pre-challenge value, rather than as a simple difference in WCC units. Comparing the treatment and screening fold changes can therefore describe how the challenge response changed across the two periods. This approach may be useful when the scientific question concerns a proportional response, but it does not by itself establish that treatment caused any change. In an open-label single-group study, other period-related differences cannot be separated from treatment effects using a randomized control comparison.

The registry record describes the endpoint as a geometric least-squares mean with a 90% confidence interval and states that log-transformed fold changes were analyzed. However, neither the estimated ratio nor its confidence interval is reported in the registry record. The endpoint’s intended interpretation can be described, but its observed magnitude and precision cannot be evaluated from the reported numerical information.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a ratio built from two within-participant cold-challenge responses. First, for each period, the highest WCC value after the challenge was divided by the pre-challenge WCC value. That is the fold change for the screening period and, separately, for the treatment period. The endpoint then divided the treatment-period fold change by the screening-period fold change. The registry record defines a value below 1 as a lower relative WCC increase during treatment than during screening. This definition explains the direction of the measure; it does not supply the observed result.

Why use a fold change?

In general, a fold change describes a proportional change relative to a starting value. This can be useful when the research question concerns the size of a response relative to a participant’s own baseline, rather than only the absolute difference between two measurements. Here, the endpoint further compares the proportional response in two periods. A ratio of fold changes is not the same as a difference in WCC units, and its interpretation depends on the ratio’s scale and the endpoint definition.

Ratios are often analyzed on a logarithmic scale because logarithms turn multiplicative comparisons into additive ones. In general, taking logs can make a ratio-based analysis more suitable for statistical modeling and can reduce asymmetry in the distribution of positive measurements. The registry record states that the log-transformed fold change from pre-challenge to the highest post-challenge WCC value was analyzed. It does not report diagnostic checks, the distribution of the observed measurements, or the specific rationale for choosing this transformation.

What does a geometric least-squares mean describe?

A geometric mean is based on averaging log-transformed values and converting the result back to the original scale. A least-squares mean is a model-based estimate, rather than necessarily a simple arithmetic average of the raw observations. In general, it can reflect the fitted statistical model and the way that model handles repeated or structured data. The primary endpoint is described as a geometric least-squares mean, but the numerical estimate is not reported in the registry record. The specific terms and adjustments used to derive that estimate are also not reported.

Why a mixed-effects model?

The registry record names a log-linear mixed-effect model and states that the analysis considered data from -2 to 8 hours after challenge. In general, mixed-effects models can account for repeated observations from the same participant, recognizing that measurements from one person may be more alike than measurements from different people. A model can include fixed effects to represent systematic conditions or time patterns and random effects to represent participant-level variation. The record does not specify the exact fixed-effect terms, random-effect structure, covariance assumptions, or handling of measurements at particular time points. Those details should not be inferred from the model label alone.

The challenge took place during screening and on Day 4 of the treatment period. The stated analysis window runs from -2 to 8 hours post-challenge, while the endpoint description defines the fold change using a pre-challenge value and the highest post-challenge value up to 8 hours after challenge. The challenge itself lasted 45 minutes. The registry record does not explain in further detail how the timing of each measurement was incorporated into the fitted model or how the highest post-challenge value was handled within that model.

How should the confidence interval be read?

The endpoint is reported as having a 90% confidence interval. In general, a confidence interval gives a range of values compatible with the observed data and the assumptions of the statistical model, expressing uncertainty in an estimate. A narrower interval indicates greater precision than a wider interval, all else being equal. It is not a range of individual participant responses, and it does not guarantee that the underlying value is within the interval. Because the estimate and interval limits are not reported in the registry record, their width and the values they include cannot be assessed.

Confidence intervals depend on more than the nominal confidence level. Their interpretation also depends on the model, the data included, and the assumptions used for estimation. In a study with 4 enrolled participants, individual observations can have substantial influence on a summary estimate. The registry record defines a pharmacodynamic analysis set based on protocol deviations relevant to PD data, but does not report the number of participants in that set or the individual measurement patterns.

What does over-dispersion mean in general?

Over-dispersion occurs when observed variability is greater than a statistical model expects. It can make standard errors too small and confidence intervals too narrow if the model does not account for the extra variability. This is a general statistical consideration, not a finding reported for this study. The registry record does not report an over-dispersion assessment, and the endpoint’s use of a log-linear mixed-effect model does not establish whether over-dispersion was present or how it was addressed.

Finally, any planning illustration based on a fraction of an observed effect is a hypothetical design calculation, not an observed result or observed power. The registry record describes assumptions of a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation for those illustrations. These assumptions do not describe the actual single-group design, and they should not be interpreted as the trial’s own sample-size justification. A sample-size justification is not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The record lists 5 outcome measures: the primary WCC response measure and secondary measures concerning adverse events, physician-assessed disease activity, physician-assessed severity of signs and symptoms, and patient-assessed disease activity. When a study examines multiple outcomes, the chance of at least one apparently unusual result can increase if each outcome is tested separately without an adjustment or a prespecified hierarchy. This is a general statistical issue; it does not establish that a false-positive finding occurred in this study.

A multiplicity strategy, such as a testing hierarchy or an adjustment across outcomes, is not reported in the registry record. The record also lists no analyses. Numerical findings for the physician and patient assessments are not reported in the registry record, and no numerical primary endpoint estimate is reported. Consequently, there is no set of reported efficacy estimates with which to assess how multiple comparisons were handled.

The record defines the pharmacodynamic analysis population as participants without protocol deviations with relevant impact on PD data. The number included in that population and the specific deviations considered relevant are not reported in the registry record. This means the reader cannot determine whether all 4 enrolled participants contributed to the primary endpoint analysis.

Methods for handling missing measurements, incomplete cold-challenge assessments, or other missing data are not reported in the registry record. It is therefore not possible to tell whether missing values were excluded, imputed, or handled through the model’s likelihood-based estimation. These possibilities have different assumptions and should not be presumed. The effect of missing data on the estimated endpoint cannot be evaluated from the available record.

In general, analysis populations and missing-data rules matter because they define which observations inform an estimate and what assumptions support it. With a small enrollment, even a limited amount of unavailable data could be consequential, but the extent of missingness is not reported. No conclusion about the direction or magnitude of any resulting bias can be drawn from the registry record.

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 non-serious adverse event terms

The safety information lists 4 participants at risk for serious adverse events and 0 participants with a serious adverse event. For other adverse events, it lists 4 participants at risk and 4 participants affected. These are participant counts for the DFV890 100 mg treatment group; they are not event totals, and the record does not give the specific terms for the non-serious events.

A participant count with at least one event does not show how many events that participant experienced, when events occurred, how long they lasted, or how severe non-serious events were. Those details are not reported in the registry record. The listed counts also do not provide a comparison with a control group, because the study had a single treatment group.

Zero participants with a reported serious adverse event in a group of 4 is a description of the observed record, not evidence that serious events cannot occur. A small study has limited ability to characterize uncommon harms or to compare safety between treatments. Similarly, the count of 4 participants with other adverse events should not be treated as a rate that applies to people outside this study.

The registry record lists treatment-emergent adverse events and serious adverse events as a secondary outcome, but the detailed event terms and timing are not reported. These counts describe what was recorded for this small group; they do not establish a comparative safety profile.

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

The study provides a defined pharmacodynamic question: whether the WCC fold-change response to cold challenge during treatment differed from the response during screening. The record describes how that endpoint was calculated, the analysis population definition, and the model used. It also reports that 4 participants enrolled and gives counts for participants with serious and other adverse events.

The record does not report the numerical primary endpoint estimate or its 90% confidence interval. It therefore does not show the observed size, direction, or precision of the WCC comparison. The definition that a ratio below 1 would be favorable explains how an estimate would be interpreted, but it is not evidence that the study’s estimate was below 1. Numerical findings for the listed physician and patient assessments are also not reported in the registry record.

Even if a numerical within-participant difference were available, this design would not provide a randomized comparison against a separate control group. The screening period supplies a reference measurement, but it cannot by itself rule out period-related changes or other explanations for a difference between periods. The open-label design also means treatment was not masked.

The enrollment of 4 participants, below the 6 planned, and early closure due to screen failure and recruitment challenges further constrain interpretation. A study this small can describe the observations made in its participants, but its estimates may not precisely characterize variation in a wider population. The size of that uncertainty cannot be quantified from the primary result because the estimate and interval are not reported.

The safety counts show that no participant was recorded with a serious adverse event and that all 4 were recorded with other adverse events. These observations should be read as counts from a small study, not as a forecast of event frequency in other settings. Overall, the registry record supports a description of the design and endpoint but does not provide the numerical efficacy-related results needed to judge the magnitude or precision of the primary comparison.

13. Why This Trial Matters Statistically

This trial illustrates the distinction between a clearly defined endpoint and a reportable estimate. The registry record explains the cold-challenge procedure, the construction of the WCC fold-change ratio, the direction designated as favorable, and the statistical model. Yet without the numerical estimate and its 90% confidence interval, readers cannot evaluate the observed result or its precision. Detailed methods do not substitute for the outcome estimate itself.

The within-participant comparison also highlights a design distinction. Comparing treatment-period measurements with screening-period measurements can address whether responses differed across those periods, but it is not the same as a randomized comparison with a control group. The study was open-label and single-group, so the record does not support a direct controlled treatment contrast.

Enrollment and analysis populations are important to statistical interpretation. Only 4 participants enrolled, fewer than the 6 planned, and early closure was attributed to screen failure and recruitment challenges. The record defines a PD analysis set but does not state its size. These details make transparent reporting of participant flow and the number contributing to each endpoint especially relevant.

The safety information likewise shows why counts need context. The record reports 0 participants with serious adverse events and 4 with other adverse events, but does not report the non-serious event terms. Small-group counts can describe the study experience without establishing how often events may occur more broadly.

Finally, the record lists multiple secondary assessments but does not report their numerical results or a multiplicity strategy. This limits the ability to understand whether different outcome domains offered consistent evidence. The study is statistically informative as an example of a small, open-label pharmacodynamic investigation and of the importance of complete outcome reporting, but the reported information does not permit a numerical appraisal of its primary endpoint.

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 this study. First, only 4 participants enrolled, although 6 were planned. The registry record attributes early closure to an unforeseen screen failure rate and recruitment challenges. This enrollment limits the amount of participant-level variation that can be represented and makes the reported safety counts particularly difficult to generalize.

Second, the trial was open-label and single-group, with no randomized control group. The screening-period response provides a within-participant reference, but there is no separate control group to help distinguish treatment-period changes from other differences between periods. The record does not report additional design features that would address those alternative explanations.

Third, the primary endpoint estimate and 90% confidence interval are not reported in the registry record. Without them, readers cannot assess the size, direction, or precision of the WCC comparison. The listed secondary physician and patient assessments also lack numerical results in the registry record, limiting assessment across outcomes.

Fourth, the number of participants in the pharmacodynamic analysis set is not reported. The analysis-set definition excludes participants with protocol deviations considered relevant to PD data, but the record does not describe the number or nature of such deviations. Missing-data methods are also not reported, so the treatment of incomplete measurements cannot be assessed.

Finally, the registry record names a log-linear mixed-effect model but does not report its full specification, diagnostic checks, or the exact handling of measurements within the analysis window. A multiplicity strategy is not reported. These reporting gaps do not demonstrate that any particular analysis was inappropriate; they mean that the analysis cannot be independently evaluated in detail from the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Study of Safety, Tolerability and Efficacy of DFV890 in Participants With Familial Cold Auto-inflammatory Syndrome (FCAS) — NCT04868968.
• Registry record last updated 2024-10-09; status last verified 2024-10.
• Results first posted 2024-05-03.
• 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, open-label, single-group phase II study of DFV890 in participants with FCAS. The sponsor was Novartis Pharmaceuticals. The study started on 2021-09-20 and reached primary completion on 2022-12-13. Four participants enrolled, compared with 6 planned, and the record states that the study closed early because of an unforeseen screen failure rate and recruitment challenges.

The primary endpoint compared WCC fold changes after cold challenge during screening and on Day 4 of treatment. The registry record defines a ratio below 1 as a lower relative WCC increase during treatment. It describes a log-linear mixed-effect model and a 90% confidence interval, but the numerical primary estimate and interval are not reported in the registry record.

The safety record lists 0 of 4 participants with serious adverse events and 4 of 4 with other adverse events. Specific non-serious event terms, detailed secondary outcome results, the size of the PD analysis set, and missing-data methods are not reported in the registry record.