This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Long-term Follow-up Study on Safety and Maintenance of Efficacy of ATX-101 (NCT02163902), a non-randomized phase 3 study. The primary outcome is Percentage of Participants Maintaining CR-SMFRS 1-Grade Response During 3 Years of Follow up, i.e. % of Participants Who Were CR-SMFRS 1-Grade Responders at Both Long-term LTFU Baseline and at Subsequent LTFU Visits.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02163902 |
| Phase | Phase 3 |
| Condition | Submental Fat, Healthy |
| Allocation | Non-Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 224 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Kythera Biopharmaceuticals |
| Results posted | Yes |
2. Clinical Question
Population
Participants who completed the predecessor ATX-101-11-22 or ATX-101-11-23 studies. The primary endpoint population was participants with at least a 1-grade reduction on the Clinician-Reported Submental Fat Rating Scale at 12 weeks after last treatment in those studies. For the maintenance analysis, nonresponders at long-term follow-up baseline were excluded.
Intervention
Participants previously treated with ATX-101 in the predecessor studies. The long-term follow-up itself was described as nontreatment; it assessed ongoing response and safety after the earlier study treatment.
Comparator
Participants previously treated with placebo in the predecessor studies. The registry identifies this as a parallel group, but allocation in the follow-up study was nonrandomized. The groups therefore should not be read as a newly randomized, like-for-like treatment comparison.
Primary question
Among participants who had responded after their predecessor-study treatment and were responders at long-term follow-up baseline, what percentage maintained a clinician-rated 1-grade response at subsequent long-term follow-up visits? The planned follow-up extended from 12 weeks after last treatment to up to 36 months after last treatment. Numerical results for this primary outcome are not reported in the registry record.
3. Trial Design
This was a completed phase 3, multicenter, double-blind, nontreatment, long-term follow-up study. It enrolled 224 participants who had completed ATX-101-11-22 or ATX-101-11-23. The registry lists a parallel model and nonrandomized allocation. Its start is recorded as 2013-12, and primary completion as 2016-01.
The follow-up groups reflect treatment received in the predecessor studies: one group had previously received ATX-101 and the other placebo. The study was not a new randomized assignment to ATX-101 or placebo. Because allocation was nonrandomized, group differences during follow-up may reflect differences that existed before follow-up, including the earlier treatment histories or the characteristics of people entering the follow-up. The registry record does not report adjustment methods intended to address such differences.
The study focused on persistence of response and long-term safety after the predecessor treatment, rather than administering treatment during follow-up. The registry describes the masking as double, but who was masked and how masking was maintained are not reported in the registry record. Further details of the follow-up schedule, beyond the stated time frame, are also not reported in the registry record.
Enrollment is not the same as the number included in each endpoint analysis. The primary outcome description excludes participants who were not responders at long-term follow-up baseline, and the analysis denominators and visit-specific results are not reported in the registry record. Thus, the enrollment figure alone cannot establish how many participants contributed data to any particular maintenance estimate.
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: Participants who previously received deoxycholic acid or placebo injections in studies ATX-101-11-22 and ATX-101-11-23 were enrolled in this non-treatment, double-blind, placebo-controlled follow-up study to further evaluate safety and efficacy.
Overall Study
| Stage | ATX-101 (Deoxycholic Acid) Injection | Placebo |
|---|---|---|
| STARTED | 113 | 111 |
| COMPLETED | 102 | 98 |
| NOT COMPLETED | 11 | 13 |
Reasons for not completing this period:
| Reason | ATX-101 (Deoxycholic Acid) Injection | Placebo |
|---|---|---|
| Subject's Request | 3 | 4 |
| Lost to Follow-up | 6 | 7 |
| Administrative Decision | 2 | 2 |
Analysis population for the primary outcome, as defined in the registry: Participants who had at least a 1-grade reduction from baseline on the CRSMFRS (ie, CR-1 responder) at 12-weeks after last treatment in the predecessor studies.
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.
| Characteristic | Unit | ATX-101 (Deoxycholic Acid) Injection | Placebo | Total |
|---|---|---|---|---|
| Participants analysed | 113 | 111 | 224 | |
| Age, Continuous | years | 51.0 (SD 8.22) | 50.1 (SD 9.00) | 50.6 (SD 8.61) |
| Age, Customized: ≤ 30 years | Participants | 3 | 4 | 7 |
| Age, Customized: > 30 to ≤ 50 years | Participants | 52 | 49 | 101 |
| Age, Customized: > 50 years | Participants | 58 | 58 | 116 |
| Age, Customized: < 50 years | Participants | 47 | 47 | 94 |
| Sex: Female, Male: Female | Participants | 96 | 95 | 191 |
| Sex: Female, Male: Male | Participants | 17 | 16 | 33 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 33 | 25 | 58 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 80 | 86 | 166 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 | 1 | 1 |
| Race (NIH/OMB): Asian | Participants | 4 | 1 | 5 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 1 | 0 | 1 |
| Race (NIH/OMB): Black or African American | Participants | 12 | 7 | 19 |
| Race (NIH/OMB): White | Participants | 96 | 101 | 197 |
| Race (NIH/OMB): More than one race | Participants | 0 | 1 | 1 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Body Mass Index (BMI) | kg/m^2 | 29.283 (SD 4.4934) | 28.836 (SD 4.0099) | 29.062 (SD 4.2570) |
| Fitzpatrick Score: I | Participants | 4 | 4 | 8 |
| Fitzpatrick Score: II | Participants | 30 | 39 | 69 |
| Fitzpatrick Score: III | Participants | 33 | 36 | 69 |
| Fitzpatrick Score: IV | Participants | 35 | 20 | 55 |
| Fitzpatrick Score: V | Participants | 9 | 9 | 18 |
| Fitzpatrick Score: VI | Participants | 2 | 3 | 5 |
6. Primary Endpoint
The registry lists the primary outcome as Percentage of Participants Maintaining CR-SMFRS 1-Grade Response During 3 Years of Follow up, i.e. % of Participants Who Were CR-SMFRS 1-Grade Responders at Both Long-term LTFU Baseline and at Subsequent LTFU Visits, measured over From 12 weeks after last treatment (in the predecessor study) to up to 36 months after last treatment.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | percentage of participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint is the percentage of participants maintaining a 1-grade response on the Clinician-Reported Submental Fat Rating Scale during long-term follow-up. The investigator assessed the chin and neck area using a 5-point scale, ranging from 0, absent submental convexity, to 4, extreme submental convexity. A response for this endpoint means that a participant was a CR-1 responder both at long-term follow-up baseline and at subsequent long-term follow-up visits.
This endpoint asks whether an earlier response persisted, rather than whether participants newly responded during follow-up. Its repeated assessments can describe the course of response over time. The registry specifies a time frame from 12 weeks after last treatment in the predecessor study to up to 36 months after last treatment.
The analysis is conditional on response status at long-term follow-up baseline: participants who were not responders then were not included in the analysis in either treatment group, including placebo. This condition matters when interpreting the percentage. It is not a percentage of every participant enrolled, nor does it describe response among people who had already lost response by follow-up baseline. Numerical endpoint estimates and confidence intervals are not reported in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies maintenance of a clinician-rated response as the primary outcome, but it does not provide a numerical estimate for that outcome. It also lists zero analyses. Consequently, the posted record does not establish the observed percentage maintaining response, its precision, or how that percentage varied across follow-up visits. The outcome definitions and assessment period describe what the study set out to measure, not what it found.
The primary population is selected by response history. Participants had to have at least a 1-grade reduction at 12 weeks after their last predecessor-study treatment, and nonresponders at long-term follow-up baseline were excluded from the maintenance analysis. Any maintenance percentage would therefore apply to the eligible baseline responders who were analyzed, not automatically to all enrolled participants or to everyone originally treated in the predecessor studies. The analysis denominators are not reported in the registry record.
The ATX-101 and placebo groups represent prior treatment assignments from the earlier studies. Since allocation in this follow-up study was nonrandomized, an observed difference between those groups would not by itself isolate an effect of prior treatment. Differences might be associated with the earlier study populations, treatment histories, entry into follow-up, or other factors. The registry record does not report a statistical adjustment plan or adjusted results that would help assess these issues.
The endpoint is based on a clinician rating scale. Such a scale records judgments using ordered categories, and the endpoint turns a change in that rating into a responder classification. That makes the result interpretable as a proportion meeting a defined threshold, but it does not communicate every degree of change or the participant’s own assessment. The registry record does not report additional information about agreement among investigators or the handling of borderline ratings.
The record states that a 95% confidence interval was the specified measure of dispersion. In general, a confidence interval communicates the precision of an estimate under the analysis assumptions; it does not remove bias from nonrandomized group assignment, missing observations, or outcome selection. No primary estimate or interval is reported in the registry record, so precision cannot be evaluated from the posted results.
The long-term safety information is also descriptive. The record lists participants at risk and counts with serious adverse events, other adverse events, and deaths for each prior-treatment group. These counts can be read as recorded safety summaries, but they do not establish that any event was caused by a prior treatment. With no randomized follow-up assignment and no reported inferential safety analysis, the figures do not support a causal comparison.
9. Statistical Methods Explained
The registry defines the primary outcome as a percentage of participants, making it a binary-response summary: each eligible participant either meets the stated maintenance definition or does not. In general, a percentage is formed by relating the number meeting a criterion to the number evaluated under that criterion. The registry record does not report the primary outcome counts, analysis denominators, or resulting percentages, so none can be interpreted from the posted results.
What does the response definition capture?
The Clinician-Reported Submental Fat Rating Scale is described as a 5-point scale, with 0 indicating absent submental convexity and 4 indicating extreme submental convexity. The primary outcome uses a 1-grade response threshold and requires response at long-term follow-up baseline and at subsequent visits. This converts an ordinal assessment into a responder status. In general, a threshold can make a result easier to describe, but it also discards detail: people with different underlying rating patterns can fall into the same responder category. The registry does not report the distribution of scale ratings or other endpoints that would show the full range of change.
How should a confidence interval be understood?
The primary outcome lists a 95% confidence interval as its dispersion measure. In general, a confidence interval describes uncertainty around an estimated quantity under the chosen statistical method and assumptions. A narrower interval indicates greater precision than a wider interval, but precision is not the same as freedom from bias. An interval cannot correct for nonrandom group assignment, selective availability of follow-up data, or limits in how the outcome was defined. Since the interval itself is not reported in the registry record, its width and the precision of the primary estimate cannot be assessed.
Why does the group assignment matter?
The study is labeled nonrandomized. Its groups are based on whether participants received ATX-101 or placebo in predecessor studies, rather than a new randomized allocation in this long-term study. In general, random assignment can support a fair comparison by balancing measured and unmeasured characteristics on average. Without that assignment, differences between groups may be related to other characteristics as well as their earlier treatment. The registry record does not report a covariate-adjustment method, a model for group comparisons, or adjusted estimates. Accordingly, any difference between group summaries would need cautious descriptive interpretation.
The parallel-group label indicates that the record organizes participants into groups, but it does not make those groups exchangeable. The study’s double masking is a separate design feature. The registry does not report who was masked, whether assessors remained masked throughout follow-up, or how masking was protected. Those details are relevant to interpreting a clinician-rated outcome, but cannot be inferred from the masking label alone.
What is the role of repeated follow-up assessments?
The maintenance outcome refers to baseline and subsequent long-term follow-up visits. In general, observations from the same participant over time are related; they should not automatically be treated as independent observations. Statistical methods for repeated measurements may account for within-participant dependence, but the method used for this analysis is not reported in the registry record. The record also does not provide visit-specific denominators, the handling of missed visits, or a rule for determining maintenance when a participant has some but not all follow-up assessments.
Participants who were not responders at long-term follow-up baseline were excluded from the maintenance analysis. This is an explicitly stated analysis-population restriction, not a general rule that can be assumed for all other outcomes. The registry record does not report how other missing outcome data were handled, whether participants were removed after withdrawal, or whether any imputation or sensitivity analysis was conducted.
What can planning calculations tell readers?
Any sample-size figures associated with planning are illustrations based on assumptions, not observed trial results or observed power. The stated illustration assumptions include a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation, with the effect to detect described as a fraction of the observed effect. These are planning assumptions and should not be mistaken for the study’s actual design rationale. The sample-size justification and statistical plan are not reported in the registry record. A planning calculation cannot substitute for the observed outcome estimate, its uncertainty, or a transparent description of the analysis that produced it.
The registry lists six outcome measures, including the primary maintenance outcome and secondary measures of maintenance at different response thresholds and composite definitions. In general, examining multiple outcomes or definitions can raise multiplicity concerns: the chance of a seemingly notable finding can increase when many results are assessed. The multiplicity strategy, if any, is not reported in the registry record. With no reported numerical analyses, there are no outcome estimates available to judge either individually or as a family.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists six outcome measures. The primary measure concerns maintenance of a CR-SMFRS 1-grade response; the secondary outcomes address maintenance using other response thresholds and composite response definitions. The registry record does not report whether a hierarchy was prespecified, whether statistical testing was planned, or whether any adjustment for multiple outcomes or repeated assessments was used. The effect of multiplicity on interpretation therefore cannot be assessed from the posted results.
Missing-data methods are not reported in the registry record. In particular, the record does not explain how missed long-term follow-up visits, withdrawal, or incomplete outcome assessments were handled. It does not state whether analyses used observed cases, imputation, or another approach, and no sensitivity analyses are reported. These details matter because the availability of follow-up data can affect the population represented by a maintenance estimate.
The primary endpoint description does specify an analysis-population condition: nonresponders at long-term follow-up baseline were not included in the analysis in either group, including placebo. The eligible population is also described as participants who had at least a 1-grade reduction at 12 weeks after last treatment in the predecessor studies. The record does not provide visit-specific denominators or the number excluded under these criteria. The registry reports zero analyses, so the actual analytic populations and missing-data handling cannot be verified from outcome results.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| ATX-101 (Deoxycholic Acid) Injection | 5 | 113 | 4.4% |
| Placebo | 1 | 111 | 0.9% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| ATX-101 (Deoxycholic Acid) Injection | 0 of 113 |
| Placebo | 0 of 111 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| ATX-101 (Deoxycholic Acid) Injection | 0 of 113 |
| Placebo | 0 of 111 |
Most frequent serious adverse event terms
| Event term | Organ system | ATX-101 (Deoxycholic Acid) Injection | Placebo |
|---|---|---|---|
| Diverticulitis | Gastrointestinal disorders | 2 of 113 | 0 of 111 |
| Small cell lung cancer stage unspecified | Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 1 of 113 | 0 of 111 |
| Breast reconstruction | Surgical and medical procedures | 1 of 96 | 0 of 95 |
| Ovarian cancer recurrent | Reproductive system and breast disorders | 1 of 96 | 0 of 95 |
| Breast cancer in situ | Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 0 of 96 | 1 of 95 |
Serious events with the highest arm-level rate: diverticulitis (1.8%), breast cancer in situ (1.1%), breast reconstruction (1.0%), ovarian cancer recurrent (1.0%), small cell lung cancer stage unspecified (0.9%).
The registry lists 113 participants at risk in the ATX-101 group and 111 in the placebo group for serious adverse events. It records 5 and 1 participants with serious adverse events, respectively. For other adverse events, the recorded counts are 0 in each group; deaths are also recorded as 0 in each group.
These are counts, not comparative effect estimates. The groups reflect prior treatment in predecessor studies and were not randomized in this follow-up study. The record does not report a causal analysis, an exposure-adjusted rate, or the timing and duration of individual events. The counts therefore should not be read as evidence that prior treatment caused or prevented an event, or as a direct comparison of risk.
The registry also lists serious-event terms with percentages, including diverticulitis at 1.8% and breast cancer in situ at 1.1%. The record does not make the denominator or group attribution for those listed percentages clear in the information available for these safety summaries. They should be interpreted as registry-listed terms rather than as arm-specific rates.
12. What the Result Does, and Does Not, Mean
The study’s central statistical question is whether a clinician-rated response present at long-term follow-up baseline was maintained at later visits after earlier ATX-101 or placebo treatment. That question is distinct from whether participants newly responded during follow-up. It is also distinct from whether the earlier treatment caused a lasting difference, because the follow-up study did not randomly assign participants to the recorded groups.
The posted registry record does not report numerical results for the primary maintenance endpoint, its confidence interval, or the secondary outcomes. It therefore cannot show the observed proportion maintaining response, how estimates changed over visits, or how precise the estimates were. The design and endpoint description establish the intended measurement, not the outcome.
Even if numerical maintenance results were available, their interpretation would be limited to the population included in the analysis. The primary endpoint excludes people who were not responders at long-term follow-up baseline and is framed around participants who had responded in the predecessor studies. It does not automatically describe all enrolled participants, all participants from the predecessor trials, or people outside those study populations.
The safety counts provide a descriptive record of events among participants at risk in each prior-treatment group. They are not proof of causation and do not establish comparative safety. The nonrandomized group structure, absent analysis details, and missing numerical efficacy results should remain central to any reading of the record. The study’s stated follow-up purpose is relevant, but it does not fill in results that the registry record does not report.
13. Why This Trial Matters Statistically
Long-term follow-up addresses a different statistical question from an assessment made soon after treatment: whether a response is still present over time. Repeatedly assessing a response can describe its maintenance trajectory, provided that the analysis and follow-up completeness are reported clearly. This study’s endpoint makes that persistence question explicit by requiring a response at long-term follow-up baseline and at subsequent visits.
The record also illustrates why the analysis population matters. A maintenance estimate restricted to baseline responders answers a conditional question about people who had maintained response until that baseline. It is not interchangeable with an estimate among everyone enrolled or everyone treated earlier. Clear denominators and handling of missing visits would be needed to understand precisely who contributes to each estimate; those details are not reported in the registry record.
The prior-treatment groups offer a way to describe follow-up outcomes by earlier treatment history, but nonrandomized allocation limits causal interpretation of group differences. Statistical summaries can describe what was observed in each group; they cannot, by themselves, show that the earlier treatment produced any difference. That distinction is important when interpreting long-term follow-up studies.
Finally, the registry lists an outcome framework and safety summaries but no numerical efficacy analyses. This makes the record useful for understanding the intended question and design while leaving the answer to that question unavailable in the posted results. A clear separation between outcome definitions, observed estimates, and causal claims is essential when reading such records.
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 main limitation for interpreting efficacy is that numerical primary and secondary outcome results are not reported in the registry record, which lists zero analyses. Without estimates, denominators, and confidence intervals, the magnitude and precision of maintenance cannot be assessed.
The study was nonrandomized, and the groups were defined by treatment received in predecessor studies. As a result, group comparisons may reflect differences beyond prior treatment. The registry record does not report adjustment methods or other analyses that would address this concern. Double masking is recorded, but details about who was masked and how masking was maintained are not reported in the registry record.
The primary maintenance analysis excludes nonresponders at long-term follow-up baseline, and the specified population also depends on response in the predecessor study. This limits the scope of any estimate to the eligible analyzed population. The registry does not report the analysis denominators or the number of participants with data at each visit, so the extent of selection or loss to follow-up cannot be evaluated.
Further limitations include unreported methods for handling missing observations, unreported statistical methods for repeated visits, and unreported multiplicity procedures for the listed outcomes. The safety summaries are counts without a reported causal analysis, and the context for the listed serious-event percentages is unclear. These omissions leave important aspects of precision, bias, and safety interpretation unresolved.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Long-term Follow-up Study on Safety and Maintenance of Efficacy of ATX-101 — NCT02163902.
- Registry record last updated 2020-02-17; status last verified 2020-01.
- Results first posted 2020-02-17.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT02163902 is a completed phase 3, multicenter, double-blind, nontreatment follow-up study of participants who completed ATX-101-11-22 or ATX-101-11-23. The registry records enrollment of 224, nonrandomized allocation, a parallel model, a start of 2013-12, and primary completion in 2016-01.
The primary outcome concerns maintenance of a clinician-rated 1-grade response from long-term follow-up baseline to subsequent visits, over a period extending from 12 weeks after last predecessor-study treatment to up to 36 months. Nonresponders at long-term follow-up baseline were excluded from the analysis. Numerical efficacy analyses and their methods are not reported in the registry record.
Safety counts are recorded for the prior ATX-101 and placebo groups. Because the groups were not randomized in this follow-up study, and because the registry does not report a causal safety analysis, these figures are descriptive.