A Study to Evaluate Safety and Efficacy of Elagolix in Participants With Endometriosis With Associated Moderate to Severe Pain: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate Safety and Efficacy of Elagolix in Participants With Endometriosis With Associated Moderate to Severe Pain (NCT03343067), a randomized phase 3 study. The primary outcome is Proportion of Responders at Month 6 Based on DYS Pain Scale.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, quadruple-masked, parallel study sponsored by AbbVie. Its stated purpose was to evaluate the safety and efficacy of elagolix alone and elagolix with E2/NETA in participants with endometriosis-associated moderate to severe pain. The registry describes the study as dose-escalation and planned treatment over 24 months.

All participants initially received open-label elagolix 150 mg QD for three months. The later paths were intended to be double-blind. The record describes efficacy responders at Month 3 continuing elagolix 150 mg QD, and incomplete responders entering randomized treatment paths. One described path used elagolix 200 mg BID with E2/NETA 1/0.5 mg QD. Other arm descriptions include later treatment periods but do not fully specify all later regimens in the available record.

Enrollment was 11. The study status is terminated, with a primary completion date of 2018-10-23 and a start date of 2017-12-27. The registry record does not report why the study was terminated. It also does not report a detailed account of the number assigned to each treatment path or how many completed each period.

This staged design matters for interpretation. The early open-label period was not the same treatment setting as the later double-blind periods. In particular, the later paths depended on efficacy response at Month 3, so participants entering them were not simply an unrestricted group randomized at the start to each possible regimen. The registry record does not report enough detail to reconstruct the participant flow or analyze the consequences of that selection.

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: The study was terminated early for business reasons, not for safety concerns. There are no subject data past Treatment Month 6.

Open Label Period: Day 1 to Month 3

Double Blind Period: Month 3 to Month 6

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analysis was intended to be based on a receiver operating characteristic (ROC) threshold, which could not be calculated, since only 1 participant had a Month 6 assessment for this endpoint at the time of study termination.

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 Proportion of Responders at Month 6 Based on DYS Pain Scale, measured over Month 6.

The primary endpoint was the proportion of participants meeting a responder definition at Month 6 based on the dysmenorrhea (DYS) pain scale. The scale ranges from 0 [none] to 3 [severe] and was recorded using a daily electronic diary. Response also required no increased rescue analgesic use for endometriosis-associated pain. Thus, the endpoint combined a pain-scale criterion with a condition on analgesic use; a pain score alone would not capture the complete definition.

A responder threshold was intended to be determined using a receiver operating characteristic (ROC) approach. General explanation: ROC methods can be used to assess how candidate thresholds distinguish between outcome states, helping define a classification rule. They require adequate observations to estimate that distinction. In this trial, the registry says the threshold could not be calculated because only 1 participant had a Month 6 assessment for this endpoint when the study terminated.

As a result, the record does not report a usable Month 6 responder proportion, a threshold derived from the planned ROC approach, or an estimated treatment contrast for this endpoint. The endpoint’s clinical meaning can be described from its definition, but its comparative result cannot be established from the posted results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The available record identifies a randomized, quadruple-masked, parallel Phase 3 design, but it does not report a complete statistical analysis plan. The primary outcome description gives the intended endpoint and says that its response threshold was to be based on a receiver operating characteristic (ROC) approach. It also states why this threshold was not obtained: only 1 participant had a Month 6 assessment for the endpoint at termination. The explanations below distinguish general statistical principles from the trial-specific information.

What does a responder endpoint represent?

General explanation: a responder endpoint turns a measurement into a classification, such as meeting or not meeting a defined criterion. This can make an outcome easier to interpret as the share of participants reaching a specified response. The classification depends on its threshold and on every part of the responder definition. For this trial, response included a DYS pain-scale criterion and no increased rescue analgesic use for endometriosis-associated pain. Without the threshold, the proportion meeting the full definition cannot be calculated as intended.

Why use an ROC approach?

General explanation: ROC methods compare how well candidate thresholds separate observations into outcome categories. A threshold can be selected according to a prespecified criterion, but its performance depends on the observations available and the definition of the reference outcome. The registry record says the planned threshold could not be calculated because only 1 participant had the Month 6 assessment. It does not report the intended ROC selection criterion, the reference classification, or a substitute method after the planned calculation was unavailable.

What do randomization and masking contribute?

General explanation: randomization is intended to make treatment groups comparable on average at assignment, while masking is intended to reduce the influence of treatment knowledge on care, reporting, or assessment. The registry identifies the trial as randomized and quadruple-masked, and describes later treatment periods as double-blind after an open-label initial period. It does not identify the specific roles covered by quadruple masking or give details of implementation. The initial open-label period should not be described as masked.

The staged design also affects the question a comparison answers. Participants began with the same open-label elagolix regimen. Subsequent paths were related to response at Month 3 and included randomized treatment for incomplete responders. General explanation: comparisons among participants selected by an earlier response can answer a different question from a comparison among all participants randomized at the outset. The registry record does not report a full participant-flow account or the analysis populations needed to clarify which comparison was intended for each outcome.

How should estimates and uncertainty be read?

General explanation: an estimate is a summary of the observed data, such as a response proportion or a difference between groups. A confidence interval expresses uncertainty around an estimate under the chosen model and assumptions. A p-value, when used, assesses how compatible the data are with a specified null hypothesis under a statistical model; it is not a measure of clinical importance. None of these measures can be interpreted for the primary comparison when the endpoint estimate itself is unavailable.

The registry record does not report a primary effect estimate, confidence interval, or p-value for the endpoint described. It reports 0 analyses, so the listed outcomes should not be mistaken for completed statistical comparisons. The record also does not report the statistical model, covariate adjustment, handling of repeated measurements, or approach to non-normal or over-dispersed outcomes. General explanation: over-dispersion occurs when variation is greater than a model expects; if present, it can make standard errors and inferential results unreliable unless the model accounts for it. There is no trial-specific basis in the record to say whether over-dispersion was assessed or relevant.

What can planning calculations tell the reader?

General explanation: sample-size calculations are planning exercises based on assumptions about the effect to detect, variability, significance level, allocation, and analysis method. They describe what a future or planned design might require under those assumptions. They do not describe the strength of evidence in a completed trial and are not observed power. The registry record’s planning assumptions specify a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of an observed effect. The sample-size justification for the trial’s original design is not reported in the registry record.

Because no primary effect estimate is reported, planning illustrations should not be treated as a substitute for a trial result. Nor can they resolve the absence of the Month 6 endpoint assessment. The record does not provide enough detail to determine which outcomes, if any, were included in a completed analysis or whether any analysis differed from the intended approach.

What statistical details remain unavailable?

The registry record does not report interim analyses, stopping rules, multiplicity control, missing-data methods, analysis populations, or the handling of participants who discontinued treatment. It does not report a full statistical analysis plan or a reason for termination. These are not details that can be reconstructed from the trial design label or outcome list. Their absence limits evaluation of how the intended analyses were meant to control bias and uncertainty, but does not establish that any particular method was or was not used outside the information in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 40 outcome measures, including several Month 6 pain and analgesic outcomes and responder measures over time. It reports 0 analyses. The primary endpoint was not calculated because only 1 participant had a Month 6 assessment at termination. Consequently, the record does not provide a completed primary result alongside which secondary findings could be interpreted.

Multiplicity arises when several outcomes, time points, or comparisons are tested. General explanation: as the number of tests increases, the chance of obtaining at least one apparently notable result by chance can increase. Statistical plans may address this through prespecified outcome hierarchies or adjusted testing procedures. The registry record does not report whether multiplicity was addressed, what testing order was intended, or whether any adjustment method was used. No claim about control of false-positive findings can therefore be made.

The record also does not report how missing outcome data were handled, which participants were included in each analysis population, or whether results were analyzed according to randomized assignment. The endpoint description documents the limited Month 6 assessment availability, but does not describe reasons for missing assessments or a method for addressing them. General explanation: missing-data assumptions matter because analyses can differ depending on whether missingness is related to observed or unobserved outcomes.

The registry record does not report interim analyses or stopping rules, and it does not state why the study terminated. These omissions limit assessment of the decision process and the intended inferential framework. They do not establish that a specific procedure was omitted from all study documents; they mean it is not reported in 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 registry reports serious adverse events for three safety groups. Group A had 7 participants at risk and 0 affected by serious adverse events; Group B had 3 at risk and 0 affected; Group C had 1 at risk and 0 affected. No deaths are listed for any of these groups. The registry lists no serious adverse-event terms.

For other adverse events, Group A had 7 participants at risk and 3 affected; Group B had 3 at risk and 3 affected; Group C had 1 at risk and 1 affected. These are participant counts, not estimates of treatment effect. The adverse-event group labels do not provide enough information to map each group confidently to a specific regimen or treatment path, so they should not be read as a direct comparison of the listed interventions.

General explanation: an at-risk count gives context for the number of participants affected, but small groups leave substantial uncertainty about how representative the observed counts are. Counts also do not describe event severity, timing, duration, or recurrence unless those details are reported. The registry record does not provide those details in the safety information. The absence of listed serious events or deaths in these groups should not be interpreted as proof that serious events cannot occur or that the regimens have equivalent safety.

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

The posted results establish that this randomized Phase 3 study was terminated after enrollment of 11 participants. The primary Month 6 responder endpoint was not calculable because only 1 participant had an assessment for it at the time of termination. Therefore, the registry record does not provide the intended response threshold, a primary responder proportion, or a primary comparison between treatment paths.

This means the record cannot answer the trial’s main comparative question. It does not show that elagolix alone and elagolix with hormonal add-back produced the same response, different responses, or no response. A missing estimate is not a neutral or null result. It is an unavailable result, with too little endpoint information to make the planned calculation.

The outcome list includes changes in several pain measures and rescue analgesic use, as well as responder measures over time. But the record reports 0 analyses, and no treatment-effect estimates for these outcomes are reported in the information summarized. The presence of an outcome in a registry is not evidence that its result was analyzed or that it supports a conclusion.

Safety counts offer limited descriptive information: no serious adverse events and no deaths are listed in the three reported groups, while other adverse events affected participants in each group. The group sizes are small, and their relationship to specific treatment paths is not fully reported. These counts cannot establish comparative safety or rule out uncommon harms.

General explanation: randomized design, masking, and a prespecified endpoint can support reliable comparisons when adequate outcome information is collected and analyzed. They do not compensate for an endpoint that could not be calculated or for unreported analysis details. In this record, design intent and observed evidence must be kept separate. The study was designed to evaluate efficacy and safety, but the posted results do not provide a completed primary efficacy comparison.

13. Why This Trial Matters Statistically

This trial illustrates why a registry entry has to be read as more than a list of planned outcomes. The study was randomized and quadruple-masked, with later double-blind treatment paths after an open-label initial period. Those design features describe how the study was intended to proceed. They do not establish that enough outcome data were available for the planned analysis.

The primary endpoint depended on a Month 6 assessment and a responder threshold based on an ROC approach. The registry states that the threshold could not be calculated because only 1 participant had the relevant Month 6 assessment at termination. That single detail changes the statistical interpretation: there is no endpoint estimate to compare across treatment paths, regardless of the intended design.

The trial also used a staged treatment structure in which later paths depended on Month 3 response. This makes participant flow and analysis-population definitions especially important for understanding any comparison. The registry record does not report enough detail to reconstruct those paths fully or to identify the populations used in completed analyses.

Finally, the listed outcome measures and safety counts illustrate the distinction between data collection plans and analyzed evidence. Forty outcome measures are listed, but the record reports 0 analyses. The safety counts provide descriptive observations, but the small reported groups and missing treatment mapping constrain comparisons. Statistically, the main lesson is not a treatment conclusion; it is that design labels, outcome listings, and adverse-event counts cannot replace a calculable primary endpoint and clearly reported analysis.

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 most direct limitation is the lack of a calculable primary endpoint. Only 1 participant had a Month 6 assessment for the endpoint when the study terminated, preventing the intended ROC threshold from being calculated. No primary response estimate or treatment contrast is reported.

Enrollment was 11, and the registry does not report the number assigned to each treatment path, the number completing each study period, or the number contributing to each secondary outcome. It also does not report why the study was terminated. These gaps prevent a detailed assessment of participant flow and follow-up.

The treatment structure is complex: all participants began with open-label elagolix 150 mg QD, and later double-blind paths depended on Month 3 response. Some arm descriptions do not fully specify later treatment periods. The registry record does not provide sufficient detail to reconstruct every regimen or make a clear comparison across all paths.

The record lists 40 outcome measures but reports 0 analyses. Statistical analysis details, including missing-data handling, analysis populations, multiplicity procedures, interim analyses, and stopping rules, are not reported in the registry record. A sample-size justification for the original design is also not reported. The reason for termination is not reported in the registry record.

Safety information is limited to counts in three registry groups, with no detailed mapping to treatment regimens or descriptions of event timing and severity in this account. The small group sizes make these counts descriptive rather than a robust basis for comparing adverse-event risk. Taken together, these limitations mean the registry record does not support a completed efficacy comparison or a precise comparative safety assessment.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study to Evaluate Safety and Efficacy of Elagolix in Participants With Endometriosis With Associated Moderate to Severe Pain — NCT03343067.
• Registry record last updated 2019-11-27; status last verified 2019-07.
• Results first posted 2019-11-27.
• 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

NCT03343067 was a Phase 3, randomized, quadruple-masked, parallel study sponsored by AbbVie, evaluating elagolix alone and with E2/NETA for endometriosis-associated moderate to severe pain. The study planned treatment over 24 months, beginning with open-label elagolix 150 mg QD for three months and then moving to intended double-blind treatment paths.

The registry lists enrollment of 11 and status as terminated. The primary endpoint was the Month 6 proportion of responders on the DYS pain scale with no increased rescue analgesic use. Its planned ROC threshold could not be calculated because only 1 participant had a Month 6 assessment for the endpoint at termination. The record reports 40 outcome measures and 0 analyses. It does not report a primary treatment-effect estimate, the reason for termination, or a detailed statistical analysis plan.