Study of Relugolix With Estradiol and Norethindrone Acetate in Women With Heavy Menstrual Bleeding Associated With Uterine Fibroids: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Relugolix With Estradiol and Norethindrone Acetate in Women With Heavy Menstrual Bleeding Associated With Uterine Fibroids (NCT03751124), a randomized phase 3 study. The primary outcome is Percentage Of Participants Who Maintained MBL Volume Of <80 Milliliters (mL) At Week 76 During The Randomized Treatment Period.

2. Clinical Question

This is a question about maintaining an earlier response after a treatment transition, not about initiating treatment in an unselected group of people with fibroids. Participants entered the randomized withdrawal study after treatment in a parent study and an open-label extension. The eligibility definition therefore selected people who had already achieved a specified menstrual blood loss response by the extension-study assessment.

The endpoint uses menstrual blood loss measured by the alkaline hematin method. The registry describes 80 mL or more per cycle as the threshold considered diagnostic of heavy menstrual bleeding. The trial asks whether participants remained below that threshold during the randomized period. That threshold-based question is clinically distinct from asking how much menstrual blood loss changed on average, or whether amenorrhea occurred.

The main comparison concerns the randomized treatment assignment among participants in the modified intention-to-treat population, defined as randomized participants who took at least one dose of study treatment. The findings therefore pertain to the enrolled responder population and the treatment conditions described in the registry record; they do not by themselves establish what would occur in other populations or under other treatment schedules.

3. Trial Design

The registry describes NCT03751124 as a Phase 3, randomized, quadruple-masked, parallel-group study with placebo control. It was an international randomized withdrawal study of relugolix with estradiol and norethindrone acetate in women with heavy menstrual bleeding associated with uterine fibroids. The recorded enrollment was 229. The study status is completed; the recorded start date is 2018-10-16 and the primary completion date is 2021-02-19.

The randomized withdrawal design followed an earlier treatment sequence. Participants had completed a total of 52 weeks of treatment, consisting of a 24-week parent-study treatment period and a 28-week open-label extension period, and met the responder definition. In the randomized period, one group received relugolix with estradiol and norethindrone acetate; the other received matching placebo. The intervention descriptions allow treatment for up to 52 weeks, with placebo treatment continuing until heavy menstrual bleeding returned. The registry says open-label retreatment with the combination would be offered if bleeding returned.

The primary outcome timeframe is Week 52/baseline up to Week 76. The registry also lists follow-up outcomes through Week 104, and the study objectives describe evaluation for up to 104 weeks. These statements describe different aspects of the recorded study: the randomized treatment period is described as lasting up to 52 weeks, while the primary comparison is earlier in that period.

The primary analysis population is the modified intention-to-treat population: randomized participants who took at least one dose of study treatment. The posted primary arm estimates use 115 participants in the combination group and 113 in the placebo group. Safety results use different at-risk denominators, which are described with those results. The registry record does not report further detail on sequence generation, allocation concealment, masking procedures, or the original sample-size justification. Those details should not be inferred from the design labels.

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: One participant in the placebo group was randomized in error and did not receive treatment, therefore, not included in the modified intent-to-treat (mITT) population.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: mITT population: all participants randomized to treatment who had taken at least 1 dose of study treatment (relugolix plus E2/NETA or placebo).

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: mITT population: all participants randomized to treatment who had taken at least one dose of study treatment (relugolix plus E2/NETA or placebo).

6. Primary Endpoint

The registry lists the primary outcome as Percentage Of Participants Who Maintained MBL Volume Of <80 Milliliters (mL) At Week 76 During The Randomized Treatment Period, measured over Week 52/Baseline up to Week 76.

The primary endpoint is the percentage of participants who maintained menstrual blood loss below 80 mL at Week 76 during the randomized treatment period. Menstrual blood loss was assessed with the alkaline hematin method. The registry describes processing used feminine products in a sodium hydroxide solution to convert hemoglobin to alkaline hematin, after which blood-loss volume is measured in milliliters. The registry identifies blood loss of 80 mL or more per cycle as the threshold for heavy menstrual bleeding.

The endpoint is a sustained-response measure rather than a mean change in blood-loss volume. The registry says the sustained responder rate was estimated as the cumulative probability of remaining below 80 mL through Week 76 using the Kaplan-Meier method. That method is used for time-related outcomes when participants may have different lengths of observation or an event may occur between scheduled assessments. It uses the observed timing information and accounts for censored observations under its assumptions.

A threshold-based endpoint is easy to interpret as the proportion maintaining a prespecified status, but it does not show the full distribution of menstrual blood loss. Participants just below the threshold and those with much lower values are counted alike for this outcome. Nor does this endpoint alone describe symptoms, quality of life, or other clinical outcomes. Those questions require separate outcomes and evidence.

7. Results Posted in the Registry

• Relugolix Plus E2/NETA (Group A): 78.3% (90/115) versus 15.0% (17/113) on control; risk difference 63.2 percentage points (95% CI 53.2 to 73.2), risk ratio 5.20 (3.32–8.15); unadjusted two-proportion z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Time To MBL Volume ≥80 mL During The Randomized Treatment Period

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor (Log Rank); it was not recalculated.
• Result:
  • Relugolix Plus E2/NETA (Group A) vs Placebo (Group B): hazard ratio 0.13 (95% CI 0.08–0.20), reported p < 0.001 (as reported in the registry).
• Interpretation:
  • The 95% confidence interval excludes no difference (a ratio of 1), so the hazard of the event is statistically lower for Relugolix Plus E2/NETA (Group A) vs Placebo (Group B) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage Of Participants Who Maintained MBL Volume Of <80 mL At Week 104 During The Randomized Treatment Period

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Relugolix Plus E2/NETA (Group A): 69.6% (80/115) versus 11.5% (13/113) on control; risk difference 58.1 percentage points (95% CI 47.8 to 68.3), risk ratio 6.05 (3.58–10.23); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Relugolix Plus E2/NETA (Group A) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage Of Participants Achieving Or Maintaining Amenorrhea At Week 76 During The Randomized Treatment Period

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Relugolix Plus E2/NETA (Group A): 57.4% (66/115) versus 13.3% (15/113) on control; risk difference 44.1 percentage points (95% CI 33.1 to 55.1), risk ratio 4.32 (2.63–7.11); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Relugolix Plus E2/NETA (Group A) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The reported secondary outcomes extend the primary question in related but distinct directions. They include maintenance of menstrual blood loss below 80 mL at a later time, time until blood loss reaches at least 80 mL, amenorrhea, and changes in menstrual blood-loss volume. Taken together, the numerical secondary results available in the registry point in a direction consistent with the primary comparison: the reported estimates favor continued combination treatment over placebo for the posted maintenance and amenorrhea outcomes, and the time-to-event result reports a lower hazard of reaching the threshold in the combination group.

These outcomes add context rather than independently restating the primary endpoint. A later maintenance assessment asks whether the threshold-defined response persists for longer. Amenorrhea is a different menstrual outcome, and time to crossing the blood-loss threshold incorporates event timing. Changes in measured volume address the amount of blood loss rather than only whether a threshold is maintained. Their relationships to the primary result make the secondary outcomes relevant, but do not make them interchangeable.

The registry lists 47 secondary outcomes as posted, but numerical details for every listed outcome are not reported in the registry record. The conclusions should therefore be limited to the reported estimates, not generalized to every secondary measure. Secondary analyses also commonly involve multiple comparisons. Unless the analysis plan controls the chance of false-positive findings across those comparisons, an unadjusted p-value carries less evidential weight than a prespecified primary analysis. The registry record does not report a multiplicity adjustment strategy, so the degree of confirmatory status for individual secondary findings is not clear.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Time To MBL Volume ≥80 mL During The Randomized Treatment Period
• Percentage Of Participants Who Maintained MBL Volume Of <80 mL At Week 104 During The Randomized Treatment Period
• Percentage Of Participants Achieving Or Maintaining Amenorrhea At Week 76 During The Randomized Treatment Period
• Change From Week 52/Baseline To Week 76 In MBL Volume During The Randomized Treatment Period
• Change From Week 52/Baseline To Week 104 In MBL Volume During The Randomized Treatment Period
• Percentage Change From Week 52/Baseline To Week 76 In MBL Volume During The Randomized Treatment Period
• Percentage Change From Week 52/Baseline To Week 104 In MBL Volume During The Randomized Treatment Period
• Percentage Of Participants Achieving Or Maintaining Amenorrhea At Week 104 During The Randomized Treatment Period

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary in the sense that each participant’s menstrual blood loss is classified relative to the threshold: below 80 mL or not below 80 mL. The reported quantity is a percentage of participants maintaining the threshold-defined response, rather than an average blood-loss volume. A percentage provides an interpretable estimate of how common the specified status was in each assigned group. It does not preserve information about how far an individual’s result was from the threshold.

What does the Kaplan-Meier estimate represent?

The registry describes the primary sustained responder rate as a Kaplan-Meier estimate of the cumulative probability of menstrual blood loss remaining below 80 mL through Week 76. Kaplan-Meier methods are generally used to estimate event-time distributions when observation ends at different times for different participants. A participant whose event time is not observed during follow-up may be censored; the method retains information up to the last observed time rather than treating that person as if the event had occurred at that time. Valid interpretation depends on assumptions about censoring and the way the event is defined and recorded. Detailed censoring rules are not reported in the registry record.

For the primary outcome, the event of concern is no longer maintaining the specified low blood-loss status. The reported group estimates are cumulative probabilities of maintaining the response through the stated time, rather than raw proportions calculated only from participants with a measurement at that exact visit. That distinction matters: a time-to-event estimate can incorporate follow-up time, whereas a simple visit-specific percentage uses a different denominator and may answer a different question.

How should the treatment contrast be read?

The registry reports the sponsor’s treatment difference using a log-log transformation, with an estimate of 63.36 and a 95% confidence interval from 52.85 to 73.86. The record labels the quantity a treatment difference and reports p < 0.001. The treatment difference compares the estimated response rates between randomized groups. The transformed method is the sponsor’s stated approach for estimation and inference; it is not the same calculation as an unadjusted difference reconstructed directly from the posted group summary statistics.

The unadjusted recalculation is a simple risk-difference comparison from those posted summaries. It yields 0.632, with a 95% confidence interval from 0.532 to 0.732 and p <0.001. This is an analytical check, not a replacement for the trial analysis. Because the registry does not identify a discrepancy, the sponsor’s own analysis remains the primary reported inference. The two p-values are reported at different levels of precision and come from differently specified analyses; their displayed forms should not be treated as evidence of a substantive disagreement.

What does a p-value tell the reader?

In general, a p-value measures how incompatible the observed data are with a specified null model, assuming that model and the analysis assumptions hold. A small p-value is not the probability that the treatment has no effect, and it does not measure the size or clinical importance of an effect. The estimate and its confidence interval are needed to understand direction and precision. The p-value also does not account automatically for every analysis choice or for the number of outcomes tested.

Why can over-dispersion matter?

In general, over-dispersion means that observed variability is greater than a particular statistical model expects. It can make standard errors too small and confidence intervals too narrow if a model assumes less variation than the data contain. The registry record does not report an over-dispersion model for this primary binary outcome, so it would be inappropriate to claim that one was used or needed. For this endpoint, the stated sponsor method is a log-log transformation, while the primary sustained response is estimated with Kaplan-Meier methods.

What do the time-to-event secondary methods mean?

For time to menstrual blood loss of at least 80 mL, the sponsor reports a log-rank analysis and a hazard ratio of 0.13, with a 95% confidence interval from 0.08 to 0.20 and p < 0.001. A hazard ratio compares event rates over time under the model’s interpretation; it is not a ratio of probabilities at a single visit. The registry also reports the placebo median time as 5.9 weeks, with a 95% confidence interval from 5.4 to 6.3. A median for the combination group is not reported in the registry record. These results should therefore be interpreted using the reported hazard ratio rather than by inventing a group median or a time difference.

The posted Week 104 maintenance outcome and Week 76 amenorrhea outcome use binary percentage estimates, with sponsor analyses reported using a log-log transformation and Cochran-Mantel-Haenszel methods, respectively. Each method has assumptions and targets a particular contrast. The registry record does not provide enough detail to reconstruct all implementation choices, including covariate or stratification details, so those should not be assumed.

11. Confidence Intervals and Precision

A confidence interval gives a range of values compatible with the data and the method used, at the stated confidence level. It is not a guarantee that the true value lies in the interval, and it does not describe the range of outcomes for individual participants. Its width reflects statistical uncertainty under the analysis assumptions; it does not include every possible source of bias, measurement error, or limited generalizability.

For the primary group estimates, the 95% confidence interval is 69.33% to 85.12% for relugolix plus estradiol and norethindrone acetate, and 8.91% to 22.76% for placebo. The sponsor’s treatment-difference interval is 52.85 to 73.86. These intervals communicate uncertainty in the estimated response rates and their contrast. The reported intervals are separated for the two group rates, and the interval for the treatment difference does not include zero. These features are consistent with the reported small p-value, but the confidence interval is more informative about the range and scale of the estimated contrast.

The sponsor interval and the recalculated unadjusted interval arise from different methods. The recalculated risk-difference interval is 0.532 to 0.732, while the sponsor reports a treatment difference on its analysis scale with interval 52.85 to 73.86. Differences in method and scale mean that apparent differences in the endpoints of the intervals should not be interpreted as a conflict without accounting for those definitions. The registry record does not flag a discrepancy.

Precision is also population-specific. These estimates come from participants who met the responder criteria before randomization and took at least one dose during the randomized phase. The intervals do not resolve uncertainty about outcomes among people who were not eligible for withdrawal randomization, nor do they address unreported outcomes or the study’s unreported multiplicity and missing-data procedures.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning illustrations use the observed effect as a reference and show how hypothetical sample-size requirements change when planning for 100%, 75%, or 50% of that observed effect. For the binary comparisons, the figures are participants per arm. For the time-to-event comparison, the figures are total events, not participants per arm. These are planning illustrations computed with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the target effect is the stated fraction of the observed effect.

The 80% power column means the illustrative design has an 80% chance of detecting the specified effect under the planning assumptions if that effect is the true one. The 90% power column uses a higher probability of detection for the same stated effect, and therefore generally requires more information. Smaller effects are harder to detect and require more participants or events than larger effects under these assumptions. For time-to-event outcomes, the event count is the relevant displayed planning quantity; the actual number of participants needed would depend on how many experience the event and on follow-up.

These calculations are not observed power. They do not describe the original trial’s planning, and the original design assumptions and sample-size justification are not reported in the registry record. The observed effect can be an unstable basis for future planning, especially when estimated in a single study. The illustrations are best read as conditional comparisons of planning requirements, not as claims about the adequacy of the completed trial or a recommendation for a future study.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.150; treatment proportion p₁ = 0.783; average p̄ = 0.467.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7055 = 1.3828.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5458 = 0.4594.
4. n = (1.3828 + 0.4594)² / (0.6322)² = 8.5, rounded up to 9 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.150442477876106                          # control proportion
p1 <- 0.782608695652174                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(3751124)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 9 (80% power), 11 (90% power)
 75% of observed effect: n per arm = 16 (80% power), 21 (90% power)
 50% of observed effect: n per arm = 33 (80% power), 43 (90% power)
power.prop.test: 8.5 
Simulated power at n = 9 per arm: 0.821 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary analysis as a treatment difference estimated with a log-log transformation: 63.36, with a 95% confidence interval from 52.85 to 73.86 and p < 0.001. The simple unadjusted recalculation from the posted group summaries gives a risk difference of 0.632, with a 95% confidence interval from 0.532 to 0.732 and p <0.001. The registry record does not flag a discrepancy between the recalculation and the sponsor result.

The sponsor analysis takes precedence as the trial’s own reported analysis. It uses the stated transformed method for the treatment comparison, while the recalculation is a check based on posted summary values and does not reproduce all details of the sponsor’s analysis. The different displayed scales and methods explain why the estimates and confidence intervals should not be compared as if they were generated by an identical procedure. The p-values also have different reported precision.

For the Week 104 maintenance outcome, the sponsor reports a log-log transformation analysis and a treatment difference of 58.04, with a 95% confidence interval from 46.97 to 69.11 and p < 0.001. For Week 76 amenorrhea, the sponsor reports a Cochran-Mantel-Haenszel analysis, a treatment difference of 44.12, a 95% confidence interval from 33.13 to 55.11, and p < 0.001. Those sponsor analyses should likewise be distinguished from unadjusted recalculations from arm summaries.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates several outcomes, time points, or contrasts. Testing many hypotheses can increase the chance that at least one result appears statistically significant by chance, even if each individual test uses a conventional threshold. A prespecified hierarchy, adjustment procedure, or clearly defined distinction between confirmatory and exploratory outcomes can help readers interpret the collection of results. The registry record lists many secondary outcomes but does not report a multiplicity-control strategy or a testing hierarchy. The reported secondary p-values should therefore be read as individual results, not as proof that the full collection of secondary claims has been confirmed.

The record reports the primary analysis population as randomized participants who took at least one dose of study treatment. It does not report detailed rules for missing outcome data, censoring, withdrawals, protocol deviations, or sensitivity analyses. These details matter for a sustained-response endpoint: if outcome observation ends early, the treatment of that incomplete follow-up can affect a time-to-event estimate. The Kaplan-Meier method accounts for censoring under assumptions, but the registry record does not describe how those assumptions were assessed or what alternative analyses were used.

The primary outcome is described as an estimated cumulative probability through Week 76, while some secondary outcomes are reported at Week 104 or use a time-to-event framework. Interpretation depends on the outcome-specific analysis population and follow-up. The registry record does not provide participant flow or endpoint-specific missingness details in the registry record reported for these results, so it is not possible to assess whether missingness differed between groups or how much it influenced the estimates.

The design and population also constrain interpretation. This was a withdrawal study among participants who had completed earlier treatment and met a response definition, rather than a trial enrolling an unselected population at treatment initiation. Further details about the original sample-size justification, interim analyses, and full statistical analysis plan are not reported in the registry record. Their absence limits assessment of how closely the reported analyses followed the original plan; it is not evidence that a particular procedure was or was not used.

16. 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: myxoid liposarcoma (0.9%), transient global amnesia (0.9%), anaemia (0.9%), breast cancer stage ii (0.9%).

The registry reports serious adverse events for 2 participants in the relugolix plus estradiol and norethindrone acetate group, with 116 at risk, and for 2 participants in the placebo group, with 112 at risk. No deaths are reported in either group. The listed serious adverse event terms are myxoid liposarcoma, transient global amnesia, anaemia, and breast cancer stage II; each is listed as 0.9%.

The registry also reports other adverse events for 29 participants in the combination group and 45 in the placebo group, using the same respective at-risk denominators. These are counts, not comparative risk estimates or tests of a treatment difference. The record does not provide event-specific detail for these other adverse events in the registry record.

Serious adverse event counts should be read as descriptive safety information. They do not establish that a treatment caused an event, and the absence of reported deaths does not establish the absence of other harms. The safety denominators differ from the primary modified intention-to-treat arm denominators, so they should not be substituted for the efficacy population. The registry record does not report further details needed to assess event timing, duration, severity beyond the serious classification, or attribution to study treatment.

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

The posted primary results show a higher estimated probability of maintaining menstrual blood loss below 80 mL through Week 76 in the relugolix, estradiol, and norethindrone acetate group than in the placebo group among randomized participants who had previously met the responder criteria. The sponsor’s treatment-difference estimate has a 95% confidence interval that does not include zero, and the sponsor reports p < 0.001. This is evidence of a between-group difference in the specified endpoint under the study’s analysis.

The result does not mean that every participant maintained the response, or that the same estimate applies to every woman with uterine fibroids and heavy menstrual bleeding. The withdrawal phase enrolled participants who had already completed earlier treatment and met a defined response condition. It therefore addresses maintenance after an initial response rather than response to first starting treatment. The registry results do not establish what would happen in people who did not meet the entry criteria or under treatment schedules not studied.

The endpoint is based on a threshold in measured menstrual blood loss. It does not, by itself, describe the complete pattern of blood loss, symptoms, quality of life, or all other outcomes that might matter to participants and clinicians. Nor do efficacy results alone settle the safety profile. Serious adverse events, other adverse events, and the limited safety details reported in the registry should be considered separately from the primary efficacy estimate.

Finally, statistical significance is not the same as clinical importance, and a confidence interval does not remove uncertainty from design limitations, missing-data handling, or incomplete reporting. The findings describe the posted trial results for this selected population and follow-up. They should not be interpreted as personal medical advice or as a guarantee of an individual outcome.

18. Why This Trial Matters Statistically

This trial is statistically informative because it makes a defined maintenance question the focus of a randomized withdrawal comparison. Rather than comparing treatment initiation in a broad population, it randomizes participants after earlier treatment and a documented response. That design can address whether the response is maintained under continued combination treatment compared with withdrawal to placebo, while also making the selected entry population central to interpretation.

The primary outcome is expressed as a cumulative probability over time, estimated with Kaplan-Meier methods, rather than only as a visit-specific proportion. This illustrates why the statistical method and outcome definition matter: the result concerns sustained status through a specified follow-up point and accounts for time-to-event information under assumptions about censoring. The sponsor also reports an explicit treatment contrast with a confidence interval, which helps readers assess both the estimated difference and its precision.

The posted secondary outcomes broaden the statistical picture by considering later maintenance, amenorrhea, changes in menstrual blood loss, and time to the threshold event. They also highlight an important interpretive principle: a consistent direction across related outcomes can support coherence, but multiple outcomes do not each carry the same evidential weight. Without a reported multiplicity strategy, secondary findings warrant more caution than the primary result.

The record also offers an example of how sponsor-reported analyses and simple unadjusted recalculations can be compared without treating them as interchangeable. The recalculation is a transparent check from summary statistics, while the sponsor analysis is the trial’s reported inferential result. Together with the confidence intervals, that distinction helps keep the interpretation tied to the estimand and method actually reported.

19. 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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The randomized population was selected for an earlier response: participants had completed a parent study and an open-label extension and met the stated menstrual blood-loss criteria. Results therefore have limited direct reach beyond this responder population. The withdrawal design also addresses maintenance after prior treatment, not the effect of starting treatment among people newly presenting with fibroid-associated heavy menstrual bleeding.

The registry record does not report the original sample-size justification or its design assumptions, a full statistical analysis plan, interim-analysis details, or a multiplicity strategy. It also does not report detailed missing-data rules, endpoint-specific participant flow, or sensitivity analyses for the primary outcome. These gaps restrict assessment of how incomplete follow-up and analytic choices may affect the estimates, especially for a time-to-event outcome.

Some secondary outcomes are listed without numerical estimates in the information reported for these results. Accordingly, a broad account of all secondary findings is not possible. The posted safety information is also limited: event counts and selected serious-event terms are available, but detailed timing, duration, attribution, and other clinical context are not reported in the registry record.

The threshold-based primary outcome compresses a continuous measurement into a maintained/not-maintained status. It does not show how far blood-loss values were from the threshold or characterize other aspects of health. The registry describes the measurement method but does not provide the full distribution of measurements or additional outcome information needed to evaluate those questions. These limitations do not negate the reported comparison, but they define what can be concluded from it.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of Relugolix With Estradiol and Norethindrone Acetate in Women With Heavy Menstrual Bleeding Associated With Uterine Fibroids — NCT03751124.
• Registry record last updated 2024-06-25; status last verified 2024-05.
• Results first posted 2024-06-25.
• 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.

23. Record Summary

NCT03751124 is a completed Phase 3 randomized, quadruple-masked, parallel-group, placebo-controlled withdrawal study in women with uterine fibroids and heavy menstrual bleeding. Participants entered the randomized phase after completing earlier treatment and meeting a responder definition. The primary outcome was maintenance of menstrual blood loss below 80 mL through Week 76, estimated using Kaplan-Meier methods in the modified intention-to-treat population.

The registry reports higher estimated maintenance in the relugolix plus estradiol and norethindrone acetate group than in the placebo group, with a sponsor-reported treatment difference of 63.36 and p < 0.001. Related posted secondary results are directionally consistent, though not every listed secondary outcome has numerical results reported in the registry record. Safety reporting includes serious and other adverse-event counts and no reported deaths in either group. Interpretation is specific to this responder withdrawal population and is limited by unreported details on multiplicity, missing data, and the original design assumptions.