Phase 2 Window Study of SAR439859 (Amcenestrant) Versus Letrozole in Post-menopausal Patients With ER+, HER2- Pre-operative Post-menopausal Primary Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase 2 Window Study of SAR439859 (Amcenestrant) Versus Letrozole in Post-menopausal Patients With ER+, HER2- Pre-operative Post-menopausal Primary Breast Cancer (NCT04191382), a randomized phase 2 study. The primary outcome is Percent Change From Baseline in Ki67 Level at Day 15.

2. Clinical Question

3. Trial Design

NCT04191382 is a randomized, single-masked, parallel-group phase 2 study. It enrolled 105 participants with breast cancer. The registry record lists the study as terminated; the reason for termination is not reported in the registry record. The start date was 2020-02-04, and primary completion was recorded as 2021-04-30.

The three treatment groups received amcenestrant 400 mg, amcenestrant 200 mg, or letrozole 2.5 mg. Each regimen was taken once daily from Day 1 to Day 14. The record identifies the study as single-masked but does not report who was masked. It does not report the allocation ratio or further details of the randomization procedure.

The primary outcome was percent change in Ki67 from baseline to Day 15. The primary analysis included 31 participants in the 400 mg group, 35 in the 200 mg group, and 29 in the letrozole group. The registry describes the analysis population as modified intent-to-treat, with eligibility based on randomization-number allocation, receipt of at least one study drug, and availability of baseline and post-treatment biopsy measurements.

The registry record also lists secondary outcomes concerning the proportion with a Ki67 reduction of at least 50%, estrogen receptor expression measured by H-score, and hematological and clinical chemistry abnormalities. Results for all of these outcomes are not reported in the registry record. The record does not report the reason for termination, the planned sample-size justification, or details of any interim analyses.

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: A total of 105 participants with early breast cancer were randomized in 1:1:1 ratio to receive treatment with amcenestrant 400 milligrams (mg), amcenestrant 200 mg, or letrozole 2.5 mg.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analysis was performed on modified intent-to-treat (mITT) population that included all enrolled participants for whom there was a confirmation of successful allocation of a randomization number by IRT, who had taken at least one study drug, and who had both Baseline and post treatment available biopsies with Ki67 values.

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: Analysis was performed on intent-to-treat (ITT) population that included all enrolled participants for whom there was a confirmation of successful allocation of a randomization number by interactive response technology (IRT) and were analyzed according to the treatment arm assigned at randomization.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change From Baseline in Ki67 Level at Day 15, measured over Baseline, Day 15.

The primary endpoint was percent change from baseline in tumor Ki67 at Day 15. Ki67 expression was assessed as the percentage of positive tumor cells, using tumor tissue collected by core-cut biopsy at baseline and Day 15 and assessed by central reading. The participant-level change was defined as 100*(Ki67pre - Ki67post) / Ki67pre. Under this definition, a positive percent change represents a lower measured Ki67 value after treatment than before treatment.

The endpoint is a biomarker measure of cell proliferation. In general, a short-term change in a proliferation marker can provide information about biological activity during a treatment window. It is not, by itself, a measure of tumor shrinkage, long-term disease control, or survival. The registry record describes the primary objective as comparing antiproliferative activity between the amcenestrant doses and letrozole.

The primary analysis used an analysis of covariance model on the log proportional change, log(Ki67post/Ki67pre), with treatment and log baseline Ki67 as fixed effects. The adjusted geometric least-squares means and their 95% confidence intervals were transformed back using the antilog. The posted results therefore represent model-adjusted group estimates, rather than simple averages of individual percent changes.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the primary estimate represent?

The primary outcome is a percent change in Ki67, calculated for each participant from the baseline and Day 15 measurements. The formula uses baseline Ki67 as the reference and subtracts the post-treatment value from the baseline value. Thus, positive values correspond to a decrease in the measured marker. The reported arm estimates are adjusted geometric least-squares means, not raw arithmetic averages.

The model analyzed the log proportional change, log(Ki67post/Ki67pre), and included treatment and log baseline Ki67 as fixed effects. Including baseline Ki67 in an analysis of covariance model can account for differences in starting values when estimating treatment-group results. Transforming the modeled estimates back by antilog returns results to a ratio-based scale. The registry describes the resulting arm estimates as adjusted geometric least-squares means and reports 95% confidence intervals.

How should the confidence intervals be read?

In general, a confidence interval expresses statistical uncertainty around an estimate under the model and its assumptions. A narrower interval indicates greater precision; a wider interval indicates less precision. The intervals around the primary arm estimates and the reported ratios are not so narrow that the relative effects are pinned down closely. They should be read as uncertainty ranges, not as a list of values that each have the same probability of being the true effect.

For the reported ratios of geometric means, the reference value for no ratio difference is 1. The confidence intervals for both primary comparisons include 1, so the registry results do not rule out that reference value at the confidence level reported. This is not the same as demonstrating equivalence or proving no difference. No equivalence margin or non-inferiority framework is reported.

What does the secondary threshold analysis add?

The secondary Ki67 outcome counts participants whose percent change reached at least 50%. A binary threshold can be easier to interpret as a proportion meeting a criterion, but it reduces a continuous measurement to categories. Participants with changes on either side of the threshold may be more similar than the binary labels suggest, while participants within a category may have quite different changes. The threshold result is therefore complementary to, not a replacement for, the primary continuous endpoint.

The recalculated risk differences compare the observed proportions without adjustment. Their confidence intervals describe uncertainty in those simple contrasts. The recalculated p-values are unadjusted, so they do not account for testing multiple outcomes or comparisons. They are not the sponsor's reported primary analysis. The planning illustrations based on observed effects are hypothetical design calculations, not observed power and not evidence that the trial was adequately powered for its original objectives.

Why can over-dispersion matter?

In general, over-dispersion occurs when outcomes vary more than a statistical model expects. It can make standard errors too small if the model's variability assumptions are not appropriate, leading to confidence intervals that are too narrow and p-values that are too optimistic. The registry record does not report whether over-dispersion was assessed or whether a correction was used. The endpoint's model description alone does not establish that any particular dispersion issue occurred.

What limits the strength of the statistical conclusion?

The primary endpoint estimates are based on participants with the required treatment and biopsy information, rather than every enrolled participant. The registry record reports the analyzed group counts, but does not describe the amount or pattern of missing data, or the method used to handle missing measurements. If availability of post-treatment biopsies is related to outcomes or treatment experience, the analyzed results might not represent everyone enrolled; the record does not permit assessment of that possibility.

The record does not report a statistical analysis plan, a sample-size justification, an interim-analysis plan, or an adjustment for multiple testing. Without these details, it is not possible to determine which comparisons were designated as confirmatory, how outcome multiplicity was handled, or whether the reported analyses followed a prespecified hierarchy. These omissions matter especially for secondary outcomes and exploratory comparisons.

10. Analyses Reported by the Sponsor

The registry reports sponsor analyses for the primary endpoint as ratios of geometric means: 1.08 for amcenestrant 400 mg versus letrozole, with a 95% confidence interval from 0.72 to 1.63, and 1.42 for amcenestrant 200 mg versus letrozole, with a 95% confidence interval from 0.95 to 2.12. No p-values are reported for these comparisons. Both confidence intervals include 1, so the posted estimates do not clearly distinguish either ratio from the no-difference reference value.

The sponsor's primary analysis is the relevant source for interpreting the prespecified model-based endpoint. Separately recalculated comparisons for the secondary binary Ki67 outcome are simple unadjusted contrasts from posted counts; they are not a substitute for the sponsor's adjusted primary analysis. The registry record does not report sponsor analyses for that secondary outcome. The discrepancy flag is false, so no conflict between a sponsor result and a recalculation is identified in the record. The primary outcome estimates and their reported comparisons should remain distinct from the secondary recalculations.

11. Multiplicity, Missing Data and Analysis Populations

The registry record does not report how multiplicity was handled. It lists a primary endpoint and multiple secondary outcomes, but does not state whether there was a hierarchy among secondary outcomes, whether alpha was divided across tests, or whether any other adjustment was planned. As a result, unadjusted secondary comparisons should not be interpreted as if they were protected from the increased chance of false-positive findings that can arise when several analyses are considered. Whether multiplicity was addressed in a separate analysis plan is not reported in the registry record.

Missing-data handling is also not reported. For the primary endpoint, the described modified intent-to-treat population requires a successful randomization-number allocation, receipt of at least one study drug, and baseline and post-treatment biopsy measurements. The analysis counts are 31 for amcenestrant 400 mg, 35 for amcenestrant 200 mg, and 29 for letrozole 2.5 mg. The record does not state how many otherwise eligible participants lacked measurements, why measurements were unavailable, or whether any imputation or sensitivity analysis was used.

These population requirements matter because the analyzed set is defined partly by having post-treatment tissue available. In general, estimates based on participants with observed measurements may be affected if measurement availability is associated with treatment or outcome. The registry information is insufficient to determine whether such a concern applies.

The record does not report the statistical analysis plan, the sample-size justification, or whether interim analyses were planned or conducted. It also does not report the reason the study was terminated. These omissions limit assessment of the planned analysis, the handling of incomplete data, and the context in which the posted results were generated.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumonia (2.8%), wound infection (2.8%).

The registry separates serious adverse events from other adverse events. Serious adverse events were recorded for 0 of 33 participants in the amcenestrant 400 mg group, 2 of 36 in the 200 mg group, and 0 of 35 in the letrozole group. The listed serious-event terms were pneumonia and wound infection, each reported as 2.8%. The record also reports other adverse events for 11 of 33, 12 of 36, and 14 of 35 participants, respectively.

These are descriptive counts from the registry record. They do not establish that an event was caused by a study drug, and they do not by themselves support a reliable comparison of safety between groups. The record does not provide exposure-adjusted rates, detailed event-level context, or enough information to assess how duration of observation and missing safety data affected the counts. No deaths were reported in any group. These figures should be interpreted as the posted safety summaries, not as a comprehensive estimate of risk.

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

The results show short-term changes in a tumor proliferation biomarker among participants included in the primary analysis. The positive Ki67 percent-change estimates indicate lower measured Ki67 at Day 15 than at baseline under the endpoint's definition. The reported comparisons between each amcenestrant dose and letrozole have confidence intervals that include 1, leaving uncertainty about the relative biomarker effects.

This does not establish that the treatments are equivalent, nor does it prove that a difference is absent. Equivalence requires a defined margin and an analysis designed to test that question; neither is reported in the registry record. The estimates also do not establish which regimen would have a larger effect in a broader population, because the intervals allow a range of relative results.

Ki67 is a biomarker endpoint, not a direct measure of patient-centered clinical benefit. These data do not establish effects on tumor response, recurrence, survival, quality of life, or other clinical outcomes. The treatment window ended at Day 14, and the primary measurement was at Day 15; the registry results therefore address an early biological response rather than long-term outcomes.

The analysis population required both baseline and post-treatment biopsy information, and the registry record does not explain missing-data handling. The study is listed as terminated, with the reason not reported. Taken together, the findings are best viewed as estimates of a short-term biomarker response with considerable statistical uncertainty and incomplete reporting about analysis decisions. They should not be generalized beyond the enrolled population and study setting without additional evidence.

14. Why This Trial Matters Statistically

This study illustrates how a biomarker endpoint can be analyzed and interpreted in a randomized, parallel-group clinical trial. The primary outcome is a continuous percent change, estimated with a model that accounts for baseline Ki67 and reports adjusted geometric means. Comparing this with a thresholded secondary outcome highlights a common statistical trade-off: a threshold can give a simple proportion, but loses information carried by the underlying continuous measurements.

The registry also illustrates why point estimates need to be read alongside uncertainty intervals. The primary arm estimates are all positive, but the reported ratios have confidence intervals that include the no-difference reference value. The point estimates alone do not settle the comparative question. The width of the intervals and the analyzed group counts are central to judging how precisely the relative biomarker effects were estimated.

Another important issue is the distinction between a prespecified sponsor analysis and an unadjusted recalculation from summary counts. They may address related questions but use different methods and should not be conflated. Recalculated planning illustrations can help communicate how much information a future study might require under stated assumptions; they are not a measure of power observed in this trial and do not show that the original design was adequate.

Finally, a registry record's omissions affect interpretation. Incomplete reporting of multiplicity, missing data, sample-size justification, interim analyses, and the reason for termination limits the ability to evaluate the analysis process. The study is therefore useful statistically not because it resolves the treatment comparison, but because it shows how design, endpoint definition, analysis population, precision, and reporting completeness jointly shape what can be concluded.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The study tested a short treatment window, with dosing from Day 1 to Day 14 and the primary measurement at Day 15. The results therefore do not describe sustained biomarker changes or long-term clinical outcomes. The endpoint was based on biopsy tissue, and the primary analysis included only participants meeting the stated criteria for treatment and available baseline and post-treatment measurements.

The registry record identifies the study as single-masked but does not state who was masked. It does not provide details of the randomization procedure or allocation ratio. It also does not report the reason for termination, the planned sample-size justification, or whether interim analyses occurred. These gaps make it difficult to assess the context for enrollment, stopping, and analysis decisions.

The primary comparison intervals are broad, and the registry reports no p-values for them. The record does not provide enough information to evaluate missing-data mechanisms or methods, and it does not describe sensitivity analyses. Only part of the listed secondary outcome set has results in the record. Multiplicity handling and the status of secondary analyses as confirmatory or exploratory are not reported.

Safety summaries are descriptive, with no exposure-adjusted rates or detailed account of event timing and attribution. The posted summaries alone cannot establish comparative safety. Finally, because the primary outcome is a biomarker and not a clinical endpoint, the results cannot be taken as evidence about long-term patient outcomes. These limitations are specific to the available trial description and posted results; additional information not reported in the registry record could change how fully the study can be assessed.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Phase 2 Window Study of SAR439859 (Amcenestrant) Versus Letrozole in Post-menopausal Patients With ER+, HER2- Pre-operative Post-menopausal Primary Breast Cancer — NCT04191382.
• Registry record last updated 2025-09-18; status last verified 2025-09.
• Results first posted 2022-06-29.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT04191382 was a randomized, single-masked, parallel-group phase 2 study in post-menopausal patients with ER-positive, HER2-negative primary breast cancer. It compared amcenestrant 400 mg, amcenestrant 200 mg, and letrozole 2.5 mg, each taken once daily from Day 1 to Day 14. The trial enrolled 105 participants and is listed as terminated.

The primary endpoint was percent change in tumor Ki67 from baseline to Day 15. Adjusted geometric least-squares mean changes were 75.9% for amcenestrant 400 mg, 68.2% for amcenestrant 200 mg, and 77.7% for letrozole. The reported ratios of geometric means had 95% confidence intervals that included 1. The results describe short-term biomarker changes and do not establish clinical outcomes or equivalence. The registry record does not report the reason for termination, missing-data methods, or multiplicity handling.