← Clinical Trials
Breast Cancer Phase 3 Time-to-Event NCT03155997

MONARCH E: Complete Statistical Analysis of Abemaciclib in Breast Cancer

An independent statistical analysis of the randomized phase 3 MONARCH E trial evaluating abemaciclib plus standard adjuvant endocrine therapy versus endocrine therapy following surgery in participants with breast cancer.

Trial status: Active, not recruiting  ·  Enrollment: 5637  ·  Primary completion: 2020-03-16
Scope of this record

This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.

1. Trial at a Glance

MONARCH E was a randomized, parallel-group, phase 3 trial in breast cancer evaluating abemaciclib given with standard adjuvant endocrine therapy against standard adjuvant endocrine therapy alone. The registry identifies invasive disease-free survival (IDFS) as the single registered primary endpoint and reports a formal hazard-ratio analysis using a log-rank test.

5637
Enrolled
Randomized trial
2
Arms
Parallel design
0.747
IDFS HR
95% CI 0.598–0.932
0.00957
IDFS P-value
Superiority analysis
FeatureMONARCH E
Trial nameMONARCH E
NCT identifierNCT03155997
PhasePhase 3
ConditionBreast Cancer
Brief titleEndocrine Therapy With or Without Abemaciclib (LY2835219) Following Surgery in Participants With Breast Cancer
AllocationRandomized
Design modelParallel
MaskingNone
Primary purposeTreatment
Enrollment5637
InterventionsAbemaciclib; Standard Adjuvant Endocrine Therapy
Trial statusActive, not recruiting
Lead sponsorEli Lilly and Company
Sponsor typeIndustry
Start date2017-07-12
Primary completion2020-03-16

2. Clinical Question

The central statistical question was whether adding abemaciclib to standard adjuvant endocrine therapy changes the time from randomization to invasive disease-free survival events compared with standard adjuvant endocrine therapy alone.

Population

Participants with breast cancer enrolled in the MONARCH E phase 3 randomized trial following surgery.

Intervention

150 mg Abemaciclib plus Endocrine Therapy.

Comparator

Endocrine Therapy.

Primary question

Does the addition of abemaciclib improve invasive disease-free survival relative to endocrine therapy alone?

3. Trial Design

Allocation
Randomized. Participants were assigned to one of two parallel treatment groups.
Masking
None. The registry identifies the trial as unmasked.
Phase
Phase 3, with treatment as the primary purpose.
Enrollment
5637 participants were enrolled in the randomized trial.
01
Randomize 5637 participants
02
Abemaciclib + ET 150 mg Abemaciclib + Endocrine Therapy
03
Control Endocrine Therapy
04
Follow-up IDFS and other outcomes
05
Analysis Time-to-event comparison
INTERVENTION

150 mg Abemaciclib + Endocrine Therapy

  • Abemaciclib
  • Standard adjuvant endocrine therapy
  • Serious adverse events: 344/2791 affected/at risk
COMPARATOR

Endocrine Therapy

  • Standard adjuvant endocrine therapy
  • No masking
  • Serious adverse events: 202/2800 affected/at risk

The available registry data establish a randomized, parallel, unmasked phase 3 comparison. They do not provide a factorial component, crossover procedure, non-inferiority margin, Bayesian analysis, or an interim-analysis rule. Those design features therefore are not part of this statistical reconstruction.

4. Randomization, Stratification, and Analysis Populations

Randomization is the key design feature that creates the basis for comparing outcomes between the two treatment strategies. In a randomized trial, treatment assignment is determined before the outcome is observed, which helps separate treatment assignment from prognostic characteristics on average across the randomized population.

The primary IDFS analysis is explicitly described as a stratified analysis. The registry analysis notes identify stratification by Interactive Web Response Systems (IWRS) Geographical Region, IWRS Prior Treatment, and IWRS Menopausal Status. These factors therefore form part of the reported statistical comparison.

Analysis populationRegistry description
Primary IDFS populationAll randomized participants, including censored participants.
Abemaciclib + endocrine therapy censored2672 participants were censored.
Endocrine therapy censored2642 participants were censored.
DRFS populationAll randomized participants, including censored participants.
DRFS censored in abemaciclib group2702 participants were censored.
DRFS censored in endocrine-therapy group2677 participants were censored.

The use of censored participants is important in time-to-event analysis. A participant who has not experienced the specified event by the available follow-up can still contribute information up to the point of censoring. Excluding all such participants would discard information that survival methods are specifically designed to use.

5. Primary Endpoint

EndpointRegistry definitionTime frame
Invasive Disease Free Survival (IDFS) IDFS, as defined by the STEEP System, was measured from the date of randomization to the date of first occurrence of one of the following events: ipsilateral invasive breast tumor recurrence, regional invasive breast cancer recurrence, distant recurrence, contralateral invasive breast cancer, second primary non-breast invasive cancer, death attributable to any cause. Baseline to Recurrence or Death from Any Cause (Up to 32 Months)

Statistically, IDFS is a composite time-to-event endpoint. The event is not limited to recurrence in the originally treated breast. The registry definition includes several types of invasive recurrence, contralateral invasive breast cancer, a second primary non-breast invasive cancer, and death from any cause.

Why the definition matters: when a composite endpoint is used, the hazard ratio summarizes time to the first qualifying event under the registered definition. It should not be interpreted as a hazard ratio for one particular component, such as distant recurrence alone.

6. Secondary Endpoint

EndpointDefinition / time frameReported estimate
Distant Relapse-Free Survival (DRFS) Baseline to Distant Recurrence or Death from Any Cause (Up to 32 Months) HR 0.717; 95% CI 0.559–0.920

The registry reports a hazard ratio and a two-sided 95% confidence interval for DRFS, with the comparison stratified by IWRS Geographical Region, IWRS Prior Treatment, and IWRS Menopausal Status. The ClinicalTrials.gov record does not identify a formal statistical method for this secondary endpoint.

Method reporting distinction: the primary IDFS analysis explicitly reports a log-rank method. The ClinicalTrials.gov record does not report a normalized statistical method for DRFS. Although hazard ratios are naturally associated with survival-model methods, the specific DRFS model should not be attributed to the registry unless it is explicitly reported.

7. Statistical Methodology

Time-to-event analysis

Both IDFS and DRFS are time-to-event endpoints. Rather than asking only whether an event occurred, these analyses incorporate when the event occurred and allow participants without an observed event during follow-up to contribute censored information.

Core survival-analysis structure
Time-to-event = time from randomization to first qualifying event or censoring

For IDFS, the event definition is the first occurrence of any event specified by the registered STEEP-based definition. For DRFS, the registered event is distant recurrence or death from any cause.

Kaplan-Meier estimation

Kaplan-Meier estimation is the standard descriptive framework for displaying time-to-event distributions. It estimates the probability of remaining event-free over time while accounting for censoring. The MONARCH E registry analysis identifies the endpoints as time-to-event and reports a hazard ratio analysis; it does not separately list Kaplan-Meier estimation as a method.

Conceptual Kaplan-Meier estimator
S(t) = ∏ti ≤ t (1 − di/ni)

Here, di represents the number of events at event time ti, while ni represents the number at risk immediately before that time.

Log-rank test

The reported primary IDFS comparison used a log-rank test. The log-rank test compares the observed and expected numbers of events between treatment groups over follow-up, taking account of the ordering of event times.

Because the analysis was stratified, the comparison was not simply an unstratified test pooling every participant together. Stratification allows the treatment comparison to account for the prespecified IWRS factors recorded in the analysis.

Hazard ratio

The reported treatment effect is a hazard ratio (HR). An HR below 1 indicates a lower estimated instantaneous event rate in the abemaciclib-plus-endocrine-therapy group relative to endocrine therapy alone, within the time-to-event model underlying the reported estimate.

Hazard-ratio interpretation
HR < 1  →  lower estimated instantaneous event rate in the intervention group

For MONARCH E, the primary IDFS estimate of 0.747 corresponds to an estimated hazard that is 74.7% of the comparator hazard under the reported analysis. The complementary interpretation is approximately a 25.3% lower estimated hazard, calculated directly as 1 − 0.747.

Stratified analysis

The primary analysis was stratified by IWRS Geographical Region, IWRS Prior Treatment, and IWRS Menopausal Status. In a stratified survival comparison, information within the defined strata contributes to the overall treatment comparison while allowing baseline event patterns to differ across strata.

This is different from adjusting for a continuous covariate in a conventional regression model. Stratification preserves separate baseline risk structures for the specified strata rather than forcing a single baseline hazard across them.

Superiority hypothesis

The registered hypothesis type for the primary analysis is superiority. This is conceptually different from non-inferiority testing. A superiority analysis asks whether the evidence supports a difference in favor of the intervention, whereas non-inferiority requires comparison with a prespecified clinically acceptable margin. No non-inferiority margin is reported in the registry-reported MONARCH E data.

8. Results: Invasive Disease-Free Survival

The primary endpoint analysis compared 150 mg Abemaciclib plus Endocrine Therapy with Endocrine Therapy among all randomized participants, including censored participants. The registry reports a stratified log-rank analysis with a hazard ratio as the effect measure.

Primary IDFS treatment effect

HR 0.747

95% CI: 0.598–0.932   ·   P = 0.00957

Comparison: 150 mg Abemaciclib + Endocrine Therapy vs Endocrine Therapy

Primary endpointAbemaciclib + Endocrine TherapyEndocrine TherapyReported analysis
Invasive Disease Free Survival (IDFS) 2672 censored participants 2642 censored participants Stratified Log Rank; HR 0.747 (95% CI 0.598–0.932); P = 0.00957
Clinical Biostats interpretation

The hazard ratio of 0.747 means that, under the reported time-to-event analysis, the estimated instantaneous rate of an IDFS event in the abemaciclib-plus-endocrine-therapy group was about 74.7% of the corresponding rate in the endocrine-therapy group. Expressed as the complementary relative reduction in estimated hazard, this is approximately 25.3%.

The HR does not mean that 25.3% of participants avoided an event, that every participant experienced a 25.3% reduction, or that the absolute probability of an IDFS event was reduced by 25.3 percentage points. A hazard ratio is a relative time-to-event measure, not an absolute risk difference.

The 95% confidence interval of 0.598–0.932 describes statistical uncertainty around the estimated hazard ratio under the analysis framework. It does not describe the range of individual patient responses. Because the interval remains below 1, the reported estimate is consistent with a lower event hazard for the intervention group across the confidence interval.

The P = 0.00957 value addresses evidence against the null hypothesis in the reported statistical testing framework. It is not a measure of the size or clinical importance of the treatment effect. Effect size is conveyed by the hazard ratio and its confidence interval, while the p-value describes evidence against the null under the specified testing framework.

The analysis is based on time-to-event data and includes censored participants. Interpretation of a hazard ratio also depends on the survival-analysis model and, where a proportional-hazards interpretation is invoked, on the adequacy of the proportional-hazards assumption. The ClinicalTrials.gov record does not provide enough information to independently assess that assumption.

9. Results: Distant Relapse-Free Survival

DRFS was reported as a secondary time-to-event endpoint. The registry compared 150 mg Abemaciclib plus Endocrine Therapy with Endocrine Therapy and reports a hazard ratio with a two-sided 95% confidence interval.

Secondary DRFS treatment effect

HR 0.717

95% CI: 0.559–0.920

Time frame: Baseline to Distant Recurrence or Death from Any Cause (Up to 32 Months)

Secondary endpointEstimate95% CIAnalysis population
Distant Relapse-Free Survival (DRFS) HR 0.717 0.559–0.920 All randomized participants, including censored participants

The ClinicalTrials.gov record does not report a p-value or a formal statistical method for the DRFS comparison. It does identify stratification by IWRS Geographical Region, IWRS Prior Treatment, and IWRS Menopausal Status.

An HR of 0.717 corresponds to approximately a 28.3% lower estimated hazard, calculated as 1 − 0.717. This is an interpretation of the reported relative effect and should not be converted into an absolute reduction in distant recurrence or death.

10. Understanding the Censoring in MONARCH E

Censoring is central to interpreting both reported survival analyses. In the primary IDFS analysis, 2672 participants in the abemaciclib-plus-endocrine-therapy group and 2642 participants in the endocrine-therapy group were censored. In the DRFS analysis, the corresponding numbers were 2702 and 2677.

What censoring means

A censored participant contributes information up to the point at which their event-free follow-up is no longer observed for the analysis.

What censoring does not mean

A censored participant should not automatically be interpreted as having remained event-free indefinitely.

Why survival methods use it

Time-to-event methods are specifically constructed to use partial follow-up rather than discarding participants whose complete event history is unavailable.

Interpretive caution

The validity of survival estimates depends on the assumptions governing censoring and follow-up. The ClinicalTrials.gov record does not provide enough detail to evaluate those assumptions directly.

11. Statistical Methods Explained

Why was a log-rank test used for IDFS?

IDFS is a time-to-event endpoint, so a simple comparison of proportions at a single time point would discard much of the available follow-up information. The log-rank test instead compares the event experience of the randomized groups across the observed follow-up period while accommodating censoring.

What does an HR of 0.747 mean?

An HR of 0.747 means that the estimated instantaneous rate of an IDFS event in the abemaciclib-plus-endocrine-therapy group was 0.747 times that of the endocrine-therapy group under the reported analysis. The complementary calculation, 1 − 0.747, gives approximately a 25.3% lower estimated hazard.

That statement is about the hazard, not an absolute event probability. It does not say that 25.3% fewer participants experienced an event.

Why is the confidence interval important?

The 95% confidence interval of 0.598–0.932 shows the statistical uncertainty around the primary HR estimate. A narrower interval would indicate greater precision; a wider interval would indicate less precision. The interval should not be interpreted as a range containing the treatment effect for 95% of individual patients.

Why does the p-value not measure effect size?

The p-value of 0.00957 describes the evidence against the null hypothesis in the reported primary testing framework. It is affected by both the observed data and the amount of information available. It does not tell us whether the treatment effect is large or small. The HR and confidence interval provide the more direct description of the relative effect and its precision.

Why does stratification matter?

The primary IDFS analysis was stratified by IWRS Geographical Region, IWRS Prior Treatment, and IWRS Menopausal Status. Stratification allows the comparison to account for differences in the underlying event experience associated with these predefined strata rather than treating the entire randomized population as one homogeneous risk set.

Does an HR automatically mean proportional hazards?

Not necessarily. A hazard ratio is a relative comparison of event rates under a time-to-event model, but the familiar interpretation of one constant HR over time relies on a proportional-hazards framework. The registry-reported MONARCH E registry data identify the hazard ratio but do not provide sufficient model diagnostics to determine whether proportional hazards hold throughout follow-up.

Why is the secondary DRFS p-value not reported here?

The registry statistical-analysis record reports an HR of 0.717 and a two-sided 95% confidence interval of 0.559–0.920 for DRFS, but it does not report a p-value or a formal statistical method for that analysis. A statistical analysis page should not manufacture a p-value from the confidence interval when the registry does not report one.

12. Primary Endpoint vs Secondary Endpoint

EndpointRoleEffect measureFormal method reportedP-value reported
Invasive Disease Free Survival (IDFS) Primary HR 0.747; 95% CI 0.598–0.932 Stratified Log Rank 0.00957
Distant Relapse-Free Survival (DRFS) Secondary HR 0.717; 95% CI 0.559–0.920 Not reported in registry-reported registry analysis Not reported in registry-reported registry analysis

This distinction matters statistically. A primary endpoint is the principal outcome specified for the trial's confirmatory question. A secondary endpoint can provide additional evidence about the treatment effect, but its interpretation depends on the prespecified statistical hierarchy and multiplicity procedures.

The registry-reported MONARCH E data identify the primary hypothesis as superiority, but they do not provide an alpha-allocation scheme, multiplicity adjustment procedure, or endpoint hierarchy beyond identifying IDFS as the registered primary endpoint. Those details should therefore not be inferred.

13. Safety Results

The registry provides serious adverse event counts by treatment arm. These are presented as affected participants over participants at risk.

Safety measure150 mg Abemaciclib + Endocrine TherapyEndocrine Therapy
Serious adverse events 344/2791 202/2800
Serious adverse events: affected / at risk
Abemaciclib + Endocrine Therapy
344 / 2791
Endocrine Therapy
202 / 2800

The reported serious-adverse-event figures are descriptive safety results. They should not be treated as directly interchangeable with the efficacy hazard ratios because they represent a different outcome domain and are expressed as affected participants over participants at risk rather than as a time-to-event treatment effect.

Safety interpretation: the ClinicalTrials.gov record provides the serious adverse event counts by arm but do not provide a formal statistical comparison for these safety data. The figures therefore should be reported as observed affected/at-risk counts rather than converted into an unreported hypothesis test.

14. What the Hazard Ratio Does — and Does Not — Mean

Relative effect

The primary IDFS HR of 0.747 indicates a lower estimated instantaneous rate of the registered IDFS event in the abemaciclib-plus-endocrine-therapy group than in the endocrine-therapy group. The complementary relative-hazard interpretation is approximately 25.3% lower estimated hazard.

Not an absolute risk reduction

An HR of 0.747 does not mean that the absolute probability of an IDFS event decreased by 25.3 percentage points. Absolute risk depends on the baseline event rate and follow-up time. The ClinicalTrials.gov record does not provide time-specific IDFS probabilities, so an absolute risk reduction should not be calculated here.

Not a patient-level guarantee

The HR is a population-level comparative measure. It does not imply that every participant receiving abemaciclib experienced the same proportional reduction in event hazard.

Confidence interval and precision

The 95% CI of 0.598–0.932 quantifies uncertainty around the primary estimate. It does not provide a distribution of individual treatment effects and should not be interpreted as the range of outcomes that patients can expect individually.

P-value and evidence

The p-value of 0.00957 addresses evidence against the null hypothesis in the reported superiority analysis. It does not measure clinical magnitude, probability of treatment benefit, or the probability that the null hypothesis is true.

15. Statistical Interpretation of the Secondary DRFS Result

The DRFS hazard ratio of 0.717 is directionally consistent with the primary IDFS hazard ratio because both estimates are below 1. The DRFS 95% confidence interval of 0.559–0.920 also lies below 1.

However, statistical interpretation should distinguish this descriptive consistency from a formal claim about multiplicity-adjusted confirmation. The ClinicalTrials.gov record does not provide a p-value, formal DRFS method, or multiplicity procedure. Consequently, the most precise statement is that the registry reports a hazard-ratio estimate and confidence interval favoring the abemaciclib-plus-endocrine-therapy comparison for this secondary endpoint.

IDFS

Primary endpoint; HR 0.747; 95% CI 0.598–0.932; P = 0.00957; stratified log-rank analysis.

DRFS

Secondary endpoint; HR 0.717; 95% CI 0.559–0.920; no formal method or p-value reported in the ClinicalTrials.gov record.

16. Design Features That Are Not Reported

The available MONARCH E data are sufficiently detailed to reconstruct the primary time-to-event comparison, but several statistical-design topics requested for some trial records are not present in the ClinicalTrials.gov record.

Design topicAvailable information
Non-inferiority marginNot reported; the registered hypothesis is superiority.
CrossoverNot reported in the ClinicalTrials.gov record.
Factorial designNot reported; the design model is parallel.
Multiplicity adjustmentNot reported in the ClinicalTrials.gov record.
Interim analysisNot reported in the ClinicalTrials.gov record.
Missing-data or imputation methodNot reported in the ClinicalTrials.gov record.
Bayesian methodsNot reported in the ClinicalTrials.gov record.
Primary statistical methodLog-rank test, with stratified analysis and hazard ratio as the effect measure.

These omissions are methodological boundaries rather than evidence that the corresponding procedures were absent from the underlying protocol or statistical analysis plan. The ClinicalTrials.gov record simply do not provide those details, so this page does not infer them.

17. Why the Primary Analysis Population Matters

The registry specifies that the IDFS analysis used all randomized participants, including the censored participants. This is important because randomization establishes the treatment comparison before outcomes are observed.

Using the randomized population helps preserve the comparability created by random assignment. In contrast, restricting the analysis only to participants who remained on treatment could introduce selection because continued treatment can itself be related to tolerability, disease status, adherence, or other post-randomization factors.

Core principle
Randomized assignment → treatment comparison → time-to-event outcome

For the primary IDFS result, the registry explicitly anchors the analysis to all randomized participants and accounts for censored participants within the survival analysis.

18. Reading the Primary Result in Context

There are three separate statistical questions embedded in the reported IDFS result.

1. Direction

Is the estimated relative event hazard above or below 1? The reported HR of 0.747 is below 1.

2. Precision

How uncertain is the estimate? The 95% CI is 0.598–0.932.

3. Statistical evidence

How much evidence is reported against the null? The reported two-sided p-value is 0.00957.

4. Clinical meaning

What does the effect mean for patients? That requires absolute event probabilities, follow-up-specific estimates, adverse-event context, and other clinical information not contained in the ClinicalTrials.gov record.

Keeping these questions separate prevents a common statistical error: treating a small p-value as though it were itself a measure of treatment magnitude. The p-value and the effect estimate serve different purposes.

19. Primary vs Secondary Endpoint Interpretation

IDFS and DRFS are both time-to-event outcomes, but they answer somewhat different questions because their event definitions differ. IDFS includes several invasive recurrence and death components, whereas DRFS focuses on distant recurrence or death from any cause.

FeatureIDFSDRFS
RolePrimarySecondary
Time frameBaseline to Recurrence or Death from Any Cause (Up to 32 Months)Baseline to Distant Recurrence or Death from Any Cause (Up to 32 Months)
Effect measureHR 0.747HR 0.717
95% CI0.598–0.9320.559–0.920
P-value0.00957Not reported
Formal method reportedLog RankNot reported
StratificationIWRS Geographical Region, IWRS Prior Treatment, IWRS Menopausal StatusIWRS Geographical Region, IWRS Prior Treatment, IWRS Menopausal Status

The two estimates should not be treated as interchangeable endpoints. Their definitions differ, and the primary endpoint carries a different confirmatory role from the secondary endpoint.

20. Limitations

21. Why This Trial Matters Statistically

MONARCH E is a useful statistical teaching case because the registry data bring together randomization, stratification, censoring, a composite time-to-event primary endpoint, a stratified log-rank test, and hazard-ratio interpretation in a single phase 3 comparison.

ConceptHow it appears in MONARCH E
RandomizationThe allocation is randomized across two parallel treatment groups.
Time-to-event analysisIDFS is measured from randomization to the first qualifying event; DRFS is measured from randomization to distant recurrence or death.
CensoringThe primary analysis includes censored participants; 2672 and 2642 participants were censored in the two IDFS groups.
Stratified analysisThe primary analysis is stratified by IWRS Geographical Region, IWRS Prior Treatment, and IWRS Menopausal Status.
Log-rank testThe reported formal method for the primary IDFS comparison.
Hazard ratioHR 0.747 for IDFS and HR 0.717 for DRFS.
Confidence intervalThe primary IDFS 95% CI is 0.598–0.932; the DRFS 95% CI is 0.559–0.920.
Superiority testingThe primary hypothesis type is reported as superiority.
Composite endpointIDFS includes multiple invasive recurrence, second-primary-cancer, and death events.
Safety analysisSerious adverse events are reported separately by treatment arm.

22. A Practical Workflow for Analyzing MONARCH E

01
Define Identify IDFS event definition
02
Randomize Preserve treatment assignment
03
Follow Record events and censoring
04
Compare Stratified log-rank analysis
05
Interpret HR + CI + p-value

This workflow illustrates why clinical-trial statistics should not begin with the p-value. First define exactly what counts as an event, establish the randomized comparison, account for censoring, specify the comparison method, and then interpret the effect estimate together with its uncertainty.

23. What the Registry Data Support — and What They Do Not

QuestionSupported by the ClinicalTrials.gov record?
Was the trial randomized?Yes.
Was the design parallel?Yes.
Was the trial masked?No; masking is reported as none.
What was the primary endpoint?IDFS.
What was the primary statistical method?Stratified log-rank analysis.
What was the primary HR?0.747.
What was the primary 95% CI?0.598–0.932.
What was the primary p-value?0.00957.
Was DRFS reported?Yes.
Was a DRFS HR reported?Yes, 0.717.
Was a DRFS p-value reported?No, not in the ClinicalTrials.gov record.
Were serious adverse events reported by arm?Yes.
Were median survival times provided?No.
Were subgroup results provided?No.
Was a non-inferiority margin provided?No.
Was a Bayesian analysis provided?No.
Was an interim-analysis procedure provided?No.

This distinction is important for reproducible statistical reporting. A complete analysis should be detailed where the registry supplies evidence and deliberately restrained where the registry does not provide the necessary information.

24. Related Tutorials

Learn more about the methods used in this trial:

25. Related Statistical Calculators

26. Sources

Continue through the Clinical Biostats knowledge graph

Connect this trial's endpoints and statistical methods to deeper tutorials and practical statistical tools.

27. Record Summary

MONARCH E provides a clear example of randomized time-to-event analysis in a phase 3 oncology trial. The primary endpoint, invasive disease-free survival, was defined from randomization to the first occurrence of a broad set of invasive recurrence, second-primary-cancer, or death events. The primary comparison used a stratified log-rank test and reported a hazard ratio of 0.747 with a 95% confidence interval of 0.598–0.932 and a p-value of 0.00957.

The secondary distant relapse-free survival analysis reported an HR of 0.717 with a 95% confidence interval of 0.559–0.920, although the ClinicalTrials.gov record does not identify a formal statistical method or p-value for that analysis. Serious adverse events were reported as 344/2791 in the abemaciclib-plus-endocrine-therapy group and 202/2800 in the endocrine-therapy group.

The most useful statistical interpretation therefore combines the hazard ratio, confidence interval, reported p-value, endpoint definition, randomized analysis population, censoring, and stratified analysis structure. Just as importantly, the interpretation should stop where the ClinicalTrials.gov record stop: no unreported subgroup estimates, median event times, interim rules, multiplicity procedures, or model specifications are added.

Clinical Biostats methodology: A trial-results page should distinguish reported evidence from statistical interpretation. For MONARCH E, the primary IDFS analysis provides the strongest statistical detail in the ClinicalTrials.gov record, while the DRFS and safety results are presented only to the extent supported by the reported data.