A Study of Abemaciclib (LY2835219) in Participants With Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Abemaciclib (LY2835219) in Participants With Breast Cancer (NCT02763566), a randomized phase 3 study. The primary outcome is Progression Free Survival (PFS) (Abemaciclib + NSAI & Placebo NSAI).

2. Clinical Question

The registry record frames the study as an evaluation of abemaciclib in this clinical population. The primary endpoint addresses the time until progression or death rather than response alone. The record also includes a separate progression-free survival comparison for the fulvestrant arms and secondary measures of tumor response, survival, symptoms, and pharmacokinetics. Those additional measures broaden the questions addressed, but the primary statistical comparison is the NSAI-arm progression-free survival analysis.

3. Trial Design

This was a phase 3, randomized, double-blind, placebo-controlled, parallel-group study sponsored by Eli Lilly and Company. The registry record gives enrollment as 463. Participants were assigned to one of four combinations: abemaciclib plus anastrozole or letrozole; placebo plus anastrozole or letrozole; abemaciclib plus fulvestrant; or placebo plus fulvestrant. The control for each abemaciclib regimen was placebo paired with the corresponding endocrine treatment.

Abemaciclib and placebo were taken orally every 12 hours on days 1 to 28 of each 28-day cycle. Anastrozole or letrozole was taken orally every 24 hours on those cycle days. In the fulvestrant arms, fulvestrant was given by intramuscular injection on days 1 and 15 of cycle 1, then on day 1 of cycle 2 and beyond. Participants receiving benefit could continue until disease progression, according to the arm descriptions.

The primary endpoint compares progression-free survival in the abemaciclib-plus-NSAI and placebo-plus-NSAI arms. The registry record reports 207 participants in the former and 99 in the latter. A separate progression-free survival outcome compares the fulvestrant arms, with 104 and 53 participants respectively. These are distinct comparisons; the endocrine-treatment partner differs between them.

The registry record lists a study start of 2016-12-05, primary completion of 2019-03-29, and status active, not recruiting. The details of sequence generation, allocation concealment, stratification factors, and the exact operational procedures used to maintain masking are not reported in the registry record. The original sample-size justification and the planned timing of any interim analyses are also not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Participants who died due to any cause or disease progression or alive and on study at conclusion but off treatment are considered as completed.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized participants in Abemaciclib + NSAI \& Placebo NSAI arms. Censored participants: 141 in Abemaciclib + NSAI, 46 in Placebo + NSAI.

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: All randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Progression Free Survival (PFS) (Abemaciclib + NSAI & Placebo NSAI), measured over Randomization to Measured Progressive Disease or Death (up to 26 Months).

The primary endpoint is progression-free survival, measured from randomization until objective progression under RECIST v1.1 or death from any cause. Participants without either event were censored at their last radiographic tumor assessment, if one was available, or at randomization if no post-baseline radiographic assessment was available. The registry record describes follow-up for this endpoint as up to 26 months.

Progression was defined using the sum of target-lesion diameters and changes in non-target lesions, or the appearance of new lesions. Specifically, the description includes at least a 20% increase in the sum of target-lesion diameters relative to the smallest sum on study, with an absolute increase of at least 5 mm; unequivocal progression of non-target lesions; or 1 or more new lesions.

Time-to-event endpoints are useful when the timing of an event matters and participants may have different follow-up durations. They can incorporate information about when progression or death occurred, while accounting for participants who have not had an event by their last assessment through censoring. The median gives a time-point summary of the event-time distribution, while the hazard ratio compares the event rates over follow-up under the model and analysis used. Neither summary alone describes every feature of the survival curves or the experience of every participant.

In the NSAI comparison, the registry record reports a median of 14.73 months for placebo plus NSAI, with a 95% confidence interval of 11.21 to 18.87 months. For abemaciclib plus NSAI, the median estimate and upper confidence limit are not reported in the registry record; the lower confidence limit is 22.32 months. The reported hazard ratio and its uncertainty provide the direct comparative summary.

7. Results Posted in the Registry

• Abemaciclib + Nonsteroidal Aromatase Inhibitor (NSAI) vs Placebo + NSAI: hazard ratio 0.50 (95% CI 0.35–0.72), reported p < 0.001 (as reported in the registry).

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.

Progression Free Survival (PFS) (Abemaciclib + Fulvestrant and Placebo + Fulvestrant Arms)

• 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:
  • Abemaciclib + Fulvestrant vs Placebo + Fulvestrant: hazard ratio 0.38 (95% CI 0.24–0.59), 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 Abemaciclib + Fulvestrant vs Placebo + Fulvestrant than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With Best Overall Response (BOR) of Complete Response (CR) or Partial Response (PR) [Objective Response Rate (ORR)]

• 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:
  • Abemaciclib + Nonsteroidal Aromatase Inhibitor (NSAI): 56.0% (116/207) versus 30.3% (30/99) on control; risk difference 25.7 percentage points (95% CI 14.4 to 37.0), risk ratio 1.85 (1.34–2.55); unadjusted two-proportion z-test p = <0.001.
  • Abemaciclib + Fulvestrant: 38.5% (40/104) versus 30.3% (30/99) on control; risk difference 8.2 percentage points (95% CI -4.9 to 21.2), risk ratio 1.27 (0.86–1.87); unadjusted two-proportion z-test p = 0.222.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Abemaciclib + Nonsteroidal Aromatase Inhibitor (NSAI) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Abemaciclib + Fulvestrant and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants Who Exhibit Stable Disease (SD) or CR or PR [Disease Control Rate (DCR)]

• 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:
  • Abemaciclib + Nonsteroidal Aromatase Inhibitor (NSAI): 91.3% (189/207) versus 82.8% (82/99) on control; risk difference 8.5 percentage points (95% CI 0.1 to 16.8), risk ratio 1.10 (1.00–1.22); unadjusted two-proportion z-test p = 0.029.
  • Abemaciclib + Fulvestrant: 92.3% (96/104) versus 82.8% (82/99) on control; risk difference 9.5 percentage points (95% CI 0.5 to 18.5), risk ratio 1.11 (1.00–1.24); unadjusted two-proportion z-test p = 0.040.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Abemaciclib + Nonsteroidal Aromatase Inhibitor (NSAI) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Abemaciclib + Fulvestrant than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary results cover several different questions. The fulvestrant-arm progression-free survival comparison extends the time-to-event evidence beyond the primary NSAI pairing. Objective response and disease control describe tumor-response categories, while overall survival, duration of response, clinical benefit, symptom burden, and pharmacokinetics address other outcomes. Across the posted response and disease-control comparisons, the pattern generally favors the abemaciclib-containing arms over their matching placebo arms. This broad pattern is directionally consistent with the primary progression-free survival finding, but the endpoints are not substitutes for one another.

Some listed secondary outcomes do not have results reported in the registry record. The record lists overall survival, duration of response, clinical benefit rate, quality-of-life symptom burden, and pharmacokinetic measures among the outcomes, but their results are not present in the posted results. Their presence in the outcome list should not be mistaken for evidence of a particular effect.

Secondary endpoints are often numerous and may be analyzed at multiple time points or in multiple groups. Unless the statistical plan specifies how the overall chance of false-positive findings is controlled, isolated nominal p-values can be less persuasive than a prespecified primary result. The multiplicity approach is not reported in the registry record. In addition, some secondary outcomes use binary classifications and others use time-to-event or continuous measures, so their estimates answer different questions and carry different sources of uncertainty.

Accordingly, secondary findings are best treated as supportive or exploratory unless the registry record documents a confirmatory testing strategy. Their consistency with the primary endpoint can add context, but it cannot establish unreported outcomes or resolve uncertainty about the analysis plan. The conclusions should remain tied to the endpoint definitions and posted analyses rather than treating all secondary results as equally decisive.

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

• Progression Free Survival (PFS) (Abemaciclib + Fulvestrant and Placebo + Fulvestrant Arms)
• Overall Survival (OS)
• Percentage of Participants With Best Overall Response (BOR) of Complete Response (CR) or Partial Response (PR) [Objective Response Rate (ORR)]
• Duration of Response (DoR)
• Percentage of Participants Who Exhibit Stable Disease (SD) or CR or PR [Disease Control Rate (DCR)]
• Percentage of Participants With Best Overall Response of CR, PR, or SD With Duration of SD for at Least 6 Months [Clinical Benefit Rate (CBR)]
• Change From Randomization in Symptom Burden on the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-C30 (EORTC QLQ-C30)
• Pharmacokinetics (PK): Area Under the Concentration Curve of Abemaciclib, Its Metabolites (M2 & M20)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is a time-to-event measure. For each participant, the event is the first qualifying progression or death after randomization. A participant who has neither event by the last usable tumor assessment contributes follow-up until censoring. Censoring is not the same as observing no event for the rest of the participant’s life; it means the event time is not known beyond the censoring point. The registry record specifies how censoring was assigned for this endpoint, including the use of randomization when there was no post-baseline radiographic assessment.

What does the hazard ratio compare?

A hazard is the event rate at a particular point in follow-up among participants who have not yet had the event. A hazard ratio compares those rates between groups over the analysis. A value below 1 indicates a lower estimated hazard in the first-named group relative to the comparator, while a value above 1 indicates a higher estimated hazard. The hazard ratio is not a direct comparison of median survival times, and it does not state how many months an individual participant gains. Its interpretation depends on the study’s time-to-event analysis and the assumptions underlying that analysis.

What does the log-rank method contribute?

The registry record identifies the log-rank method for the progression-free survival comparisons. In general, the log-rank test compares observed and expected event patterns across groups over follow-up, taking account of event timing and censoring. The reported hazard ratio gives a magnitude and direction of association, while the p-value from the test summarizes how incompatible the observed evidence is with the test’s null hypothesis. The record does not report additional details such as whether the hazard ratio came from a particular regression model or whether proportional hazards diagnostics were performed.

How should the median be read?

The median event time is the point by which half of the estimated group has experienced the event, according to the survival analysis. A confidence interval around a median expresses uncertainty about that estimated time. For the NSAI placebo group the median and interval are reported. The corresponding median and upper confidence limit for the abemaciclib group are not reported in the registry record, although a lower confidence limit is given. A missing point estimate is not a zero and should not be substituted with a value inferred from another statistic.

What is a confidence interval?

A confidence interval describes the uncertainty associated with an estimate under the method used to construct it. It is not a range that contains a known fixed proportion of individual participants’ outcomes, and it is not a guarantee that the true effect lies within the displayed limits. A narrower interval generally indicates greater precision than a wider interval, when comparing estimates on the same scale and under comparable methods. The interval must be interpreted with its estimator: the reported intervals for hazard ratios are on a relative scale, while intervals for medians are expressed in months.

Why does censoring matter?

Time-to-event analyses can use participants’ observed follow-up even when the event has not occurred before observation ends. This is helpful in trials where follow-up differs among participants. However, standard interpretation relies on assumptions about censoring: in general, censoring should not be related to unobserved event risk in a way that distorts the comparison, after accounting for the analysis. The registry record gives the censoring rule and counts of censored participants in the NSAI arms, but does not report diagnostics or sensitivity analyses that would assess the consequences of censoring.

What does the response analysis measure?

The secondary response outcomes are proportions, not event-time measures. Objective response rate counts participants whose best overall response was complete or partial response. Disease control rate includes stable disease along with complete or partial response. The posted displays include group estimates and confidence intervals, and the sponsor reports Cochran-Mantel-Haenszel tests for these outcomes. In general, this method compares categorical outcomes while allowing for stratification, but the registry record does not report the stratification details needed to explain the specific implementation. Unadjusted recalculations based on posted counts are distinct from the sponsor’s own analysis and should not replace it.

Why do the endpoints answer different questions?

Progression-free survival incorporates event timing and censoring. Response rate and disease control rate classify participants by their best recorded response over the specified period. Overall survival, duration of response, clinical benefit, symptom burden, and pharmacokinetics each address other aspects of outcome or treatment exposure. A result on one endpoint does not automatically establish a result on another. The registry lists these secondary outcomes, but results for several of them are not reported in the registry record.

Finally, a statistically significant p-value does not by itself describe precision or practical importance. The estimate, its confidence interval, event and participant counts, endpoint definition, and analysis population all contribute to interpretation. The statistical plan, including detailed model specifications and any prespecified handling of multiple testing, is not reported in the registry record.

11. Confidence Intervals and Precision

For the primary NSAI comparison, the reported hazard ratio is 0.499, with a 95% confidence interval from 0.346 to 0.719. This interval gives a range of relative hazard estimates compatible with the data and analysis at the stated confidence level. Its limits are both below 1, which is consistent with the reported direction of the comparison. The interval nevertheless allows a range of effect magnitudes; it should not be reduced to a single definitive estimate.

The median for placebo plus NSAI is 14.73 months, with a 95% confidence interval from 11.21 to 18.87 months. The abemaciclib-plus-NSAI median point estimate and upper confidence limit are not reported in the registry record. The lower confidence limit is 22.32 months. Because the group’s full median interval is not available, the medians cannot be compared using a complete pair of reported estimates and intervals. The hazard-ratio interval remains the reported comparative summary.

For the fulvestrant progression-free survival comparison, the hazard ratio is 0.376, with a 95% confidence interval from 0.240 to 0.588. The medians are 11.47 months and 5.59 months in the abemaciclib and placebo groups, respectively; the placebo median interval is 3.65 to 7.69 months, while the upper confidence limit for the abemaciclib median is not reported in the registry record. These intervals describe uncertainty in different estimands and should not be treated as if they were a direct interval for the difference between groups.

Precision is influenced by the amount and timing of observed information, including the number of events and the amount of censoring. The record reports 141 censored participants in the abemaciclib-plus-NSAI arm and 46 in the placebo-plus-NSAI arm. Without further information about follow-up distributions, the pattern of missing assessments, or sensitivity analyses, those counts alone do not establish whether censoring could affect the estimated comparison.

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 figures are illustrations based on the observed effect, not the original trial design. For time-to-event comparisons, the figures are total events needed across the two groups. For other outcomes, the figures are participants per arm. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning scenario asks what sample size would be needed to detect the stated fraction of the observed effect under those assumptions.

The 80% power column gives the number of events or participants associated with an 80% chance of detecting the chosen effect if that effect were the true one and the planning assumptions held. The 90% power column makes the same calculation for a 90% chance. These are design probabilities for hypothetical studies, not probabilities that the reported result is correct. The registry record’s original sample-size justification and its original planning assumptions are not reported in the registry record.

Rows for 100%, 75%, and 50% of the observed effect show how the planning requirement changes when the target effect is smaller than the observed estimate. A smaller target effect is harder to distinguish from the null under otherwise equal assumptions, so more participants or events are needed. For the survival outcomes, the relevant quantity is total events rather than participants per arm; the study may need more participants or longer follow-up to accrue those events. For binary outcomes, the figures are participants in each arm.

These calculations are not observed power. They do not retrospectively establish whether the completed study was adequately designed, and they do not alter the uncertainty in the posted estimates. The planning display should be read as a sensitivity illustration: it shows how hypothetical requirements change with the assumed effect size and chosen power level, not what happened during enrollment or follow-up.

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

d is the total number of events needed (Schoenfeld's formula) with 1:1 allocation and proportional hazards.

Worked calculation for this trial

1. Hazard ratio HR = 0.499, so ln(HR) = -0.6951.
2. d = 4 × (1.96 + 0.842)² / [ln(0.499)]² = 65.0, rounded up to 65 events in total 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
hr <- 0.499                              # observed hazard ratio
events_needed <- function(power, frac = 1) ceiling(round(4 * (z_a + qnorm(power))^2 / log(hr^frac)^2, 6))   # Schoenfeld, 1:1 allocation
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: events = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(events_needed(0.80, f), big.mark = ","), format(events_needed(0.90, f), big.mark = ",")))

# Simulation: given d total events under proportional hazards, events in the treatment arm are Binomial(d, hr / (1 + hr)).
# The score (log-rank type) statistic is (D1 - d / 2) / sqrt(d / 4).
set.seed(2763566)
d    <- events_needed(0.80)
reps <- 20000
D1 <- rbinom(reps, d, hr / (1 + hr))
z  <- (D1 - d / 2) / sqrt(d / 4)
cat(sprintf("Simulated power with %s events: %.3f (%d replicates)\n", format(d, big.mark = ","), mean(abs(z) > z_a), reps))

Output from running the code

100% of observed effect: events = 65 (80% power), 87 (90% power)
 75% of observed effect: events = 116 (80% power), 155 (90% power)
 50% of observed effect: events = 260 (80% power), 348 (90% power)
Simulated power with 65 events: 0.776 (20000 replicates)

The simulation uses the fact that, with d events in total and proportional hazards, the number of events in the treatment arm is binomial. It is an approximation that conditions on the total number of events, and it should land close to the target power.

14. Analyses Reported by the Sponsor

For progression-free survival, the sponsor-reported analysis is the principal comparative result: the registry record gives a log-rank hazard ratio of 0.499, a 95% confidence interval from 0.346 to 0.719, and p < 0.001 for abemaciclib plus NSAI versus placebo plus NSAI. The same record also provides a hazard ratio of 0.376, a 95% confidence interval from 0.240 to 0.588, and p < 0.001 for the fulvestrant comparison. These are sponsor-reported time-to-event analyses, not estimates recalculated from simple group summaries.

For binary outcomes, calculations based on posted group counts can produce simple unadjusted comparisons. Those recalculations do not necessarily match a sponsor analysis that uses a specified method such as the Cochran-Mantel-Haenszel test. The registry record reports sponsor p-values for the response and disease-control comparisons, but does not provide all the corresponding sponsor effect estimates and confidence intervals. Therefore, the sponsor’s own analysis takes precedence over any unadjusted recalculation when describing the study’s reported inference.

The registry record marks no discrepancy between the posted results and the calculations summarized for the analyses. That does not make methods interchangeable: the time-to-event interpretation relies on the reported hazard ratios, while simple binary recalculations are supplementary descriptions of posted counts. The original full statistical analysis plan and enough detail to reproduce every sponsor analysis are not reported in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

The registry record lists multiple outcomes and reports 38 analyses. Testing multiple hypotheses increases the chance that at least one result will appear statistically significant by chance if each test is interpreted independently. In general, prespecified approaches such as hierarchical testing or adjustment can manage this issue, but the multiplicity strategy for this trial is not reported in the registry record. It is therefore not possible to determine from the record whether the secondary p-values were adjusted, tested in a prespecified order, or intended to be confirmatory.

The record identifies the primary NSAI-arm progression-free survival outcome and reports its analysis, but does not give the full statistical analysis plan. The number and timing of any interim analyses, stopping boundaries, rules for handling protocol deviations, and detailed model specifications are not reported in the registry record. The record also does not provide enough information to verify whether all analyses were prespecified before outcomes were examined.

The primary endpoint description defines censoring and identifies the population as all randomized participants in the two NSAI arms. It reports 141 censored participants in the abemaciclib-plus-NSAI arm and 46 in the placebo-plus-NSAI arm. The handling of missing tumor assessments beyond the stated censoring rule, sensitivity analyses for censoring, and details of participants excluded from individual secondary analyses are not reported in the registry record.

Different outcomes may have different analysis populations because of endpoint-specific data availability, but the registry does not fully describe those populations for every posted result. The primary outcome’s stated randomized population is useful context, yet it does not establish that every secondary analysis used the same participants. Without the original analysis plan and fuller participant-flow detail, interpretation should distinguish reported results from assumptions about how analyses were conducted.

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: lung infection (2.9%), vomiting (2.0%), pneumonitis (2.0%), embolism (1.9%), pyrexia (1.9%).

The registry record reports serious adverse events by treatment group and gives the number at risk for those summaries. In the abemaciclib-plus-NSAI group, 40 of 205 participants were reported with serious adverse events; in the placebo-plus-NSAI group, 9 of 99 were reported. For abemaciclib plus fulvestrant, the corresponding figures are 16 of 104, and for placebo plus fulvestrant, 4 of 53. The record also reports deaths of 20, 10, 8, and 9 participants, respectively, across those four groups.

These are descriptive counts. They do not by themselves establish that treatment caused an event, and the death counts should not automatically be interpreted as deaths caused by treatment. The registry record lists serious-event terms including lung infection, vomiting, pneumonitis, embolism, pyrexia, and bile duct obstruction, with the posted percentages attached to those terms. The figures should be read with their event definitions and denominators; they do not show the full timing, duration, recurrence, or severity profile for every adverse event.

Exposure duration, person-time summaries, detailed event attribution, and comparative statistical analyses for safety are not reported in the registry record. Consequently, raw counts alone do not support a complete comparative safety assessment.

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

The primary result is a comparative time-to-event estimate for progression or death in the NSAI arms. The reported hazard ratio of 0.499, its 95% confidence interval from 0.346 to 0.719, and p < 0.001 describe the evidence in the registry analysis. The result indicates a lower estimated hazard in the abemaciclib-plus-NSAI group relative to placebo plus NSAI during observed follow-up, subject to the assumptions and limitations of the analysis.

It does not mean that every participant experienced the same course, nor does it provide an individual prediction. A hazard ratio is not a percentage reduction in an individual’s risk, a direct estimate of added months for each participant, or a probability of benefit. The registry record does not provide the median estimate and upper confidence limit for the abemaciclib-plus-NSAI group, so those values should not be inferred from the hazard ratio or from the placebo median.

The separate fulvestrant comparison also reports a hazard ratio below 1, but it involves a different endocrine-treatment partner and a different group of participants. Results from the NSAI and fulvestrant comparisons should be interpreted within their own arm pairs. Likewise, the response and disease-control outcomes classify tumor status and do not answer exactly the same question as progression-free survival.

The findings do not establish what would happen in populations not described by the registry, or under treatment schedules and clinical circumstances other than those studied. The record concerns postmenopausal women with HR-positive, HER2-negative locoregionally recurrent or metastatic breast cancer, and does not provide detailed subgroup findings in the information summarized. It also does not report enough about missing assessments, multiplicity, or the original analysis plan to resolve every possible uncertainty.

Statistical evidence is one component of interpreting a trial. The estimate, confidence interval, endpoint definition, follow-up, population, and limitations all matter. The posted results should not be expanded into claims about outcomes that were not reported in the registry record.

18. Why This Trial Matters Statistically

This trial illustrates why the endpoint and its statistical scale matter. Progression-free survival uses event timing and censoring, while response outcomes use proportions of participants classified in defined categories. Each can contribute a different kind of evidence, but their numerical summaries are not interchangeable. Reading a hazard ratio alongside the median estimates and confidence intervals helps prevent the comparison from being reduced to a single p-value.

The four-arm structure also shows the importance of matching comparators. Abemaciclib is paired with anastrozole or letrozole in one comparison and with fulvestrant in another, with placebo plus the corresponding endocrine treatment as control. Treating all arms as one pooled comparison would obscure those distinct questions. The primary NSAI comparison and the secondary fulvestrant comparison should therefore remain identifiable in interpretation.

The record provides a useful example of incomplete endpoint reporting: a median estimate and upper confidence limit are absent for the primary intervention group even though a hazard ratio and a lower confidence limit are reported. Statistical summaries answer different questions, and one should not be used to manufacture another missing summary. The distinction between a sponsor analysis and an unadjusted calculation from counts is similarly important, particularly for outcomes where analysis methods may account for design features not represented in a simple comparison.

Finally, the trial illustrates that planning calculations depend on assumed effect size and power, and that time-to-event planning is expressed in events rather than participants. Such illustrations can clarify study-design tradeoffs, but they are not the trial’s original rationale and are not evidence of observed power. Clear separation of reported results, planning illustrations, and unreported methods supports more careful statistical reading.

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.

The registry record does not include the full statistical analysis plan, original sample-size justification, detailed interim-analysis procedures, or a reported multiplicity strategy. These omissions limit the ability to establish how much of the secondary testing was confirmatory and whether every reported analysis followed a prespecified plan.

For the primary NSAI comparison, the abemaciclib-group median estimate and upper confidence limit are not reported in the registry record. The hazard ratio is available, but a complete pair of median estimates and intervals cannot be compared from the record. The registry also gives censoring counts but does not report sensitivity analyses or enough detail to evaluate how censoring and missing tumor assessments could influence the result.

Detailed stratification factors and implementation details for the reported categorical analyses are not reported in the registry record. Nor does the record provide complete results for every listed secondary outcome. The status active, not recruiting and the listed primary completion date describe registry fields, but do not in themselves explain the maturity of each endpoint or the duration of follow-up for every participant.

Safety summaries are counts without the full exposure-adjusted context or detailed event-level information in the record. The listed serious-event terms and death counts cannot alone establish causation or support a comprehensive comparison of safety. Finally, these findings are specific to the described population, arm pairings, and treatment schedules; the registry record does not provide enough subgroup or external-validity information to extend them confidently beyond those circumstances.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study of Abemaciclib (LY2835219) in Participants With Breast Cancer — NCT02763566.
• Registry record last updated 2026-02-05; status last verified 2026-01.
• Results first posted 2020-03-30.
• 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

NCT02763566 is a phase 3, randomized, double-blind, placebo-controlled, parallel-group study sponsored by Eli Lilly and Company, with enrollment of 463. It compares abemaciclib plus anastrozole or letrozole with placebo plus the same NSAI treatment, and separately compares abemaciclib plus fulvestrant with placebo plus fulvestrant, in postmenopausal women with HR-positive, HER2-negative locoregionally recurrent or metastatic breast cancer.

The primary endpoint is progression-free survival in the NSAI arms. The registry reports a hazard ratio of 0.499, with a 95% confidence interval from 0.346 to 0.719 and p < 0.001. It also reports a separate progression-free survival comparison in the fulvestrant arms, plus secondary response and disease-control outcomes. Several listed secondary outcomes do not have results reported in the registry record. Details of the original sample-size justification, multiplicity plan, and full statistical analysis plan are not reported in the registry record.