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
FELINE was a randomized, quadruple-masked, parallel-group phase 2 trial evaluating letrozole with ribociclib-containing therapy versus letrozole alone as neoadjuvant treatment for breast cancer. The registered primary endpoint was the rate of PEPI Score 0 at surgery after 24 weeks of therapy, comparing the ribociclib-containing arms combined with the letrozole-alone arm.
| Feature | FELINE |
|---|---|
| Phase | Phase 2 |
| Condition | Breast Cancer |
| Design | Randomized, quadruple-masked, parallel-group |
| Allocation | Randomized |
| Primary purpose | Treatment |
| Enrollment | 121 |
| Number of arms | 3 |
| Status | Active, not recruiting |
| Trial start | 2016-02 |
| Primary completion | 2018-08-27 |
| Lead sponsor | Qamar Khan |
| Sponsor type | Other |
| ClinicalTrials.gov | NCT02712723 |
2. Clinical Question
The central question was whether adding ribociclib to neoadjuvant letrozole would affect the probability of achieving a PEPI Score 0 at surgery after 24 weeks, compared with letrozole alone. The primary comparison combined the two ribociclib-containing treatment arms and contrasted that combined group with the letrozole-alone arm.
Population
Patients enrolled in a phase 2 neoadjuvant study of breast cancer. The registered trial title specifies ER-positive, HER2-negative early breast cancer.
Intervention
Ribociclib-containing therapy in combination with letrozole, represented by two treatment arms: ribociclib 600 mg plus letrozole and ribociclib 400 mg plus letrozole.
Comparator
Placebo plus letrozole.
Primary question
What is the rate of PEPI Score 0 at surgery after 24 weeks when the ribociclib-containing arms are considered together versus letrozole alone?
3. Trial Design
Treatment arms
Placebo + Letrozole
- Placebo
- Letrozole
- Comparator arm for the primary comparison
Ribociclib 600 mg + Letrozole
- Ribociclib 600 mg
- Letrozole
- Ribociclib-containing treatment arm
Ribociclib 400 mg + Letrozole
- Ribociclib 400 mg
- Letrozole
- Ribociclib-containing treatment arm
The primary endpoint was not defined as three independent pairwise comparisons. Instead, the registry specifies a combined comparison of the two ribociclib-containing arms against the letrozole-alone arm. That distinction is important because the scientific question concerns the presence of ribociclib-containing therapy as a combined treatment strategy rather than a separate confirmatory comparison of each ribociclib dose against placebo.
4. Endpoints
| Endpoint | Registry definition | Time frame |
|---|---|---|
| Rate of PEPI Score 0 at Surgery Between Ribociclib Containing Arms (Combined) vs. Letrozole Alone Arm. | The Preoperative Endocrine Prognostic Index (PEPI) score 0 is defined as: pT stage 0, 1, or 2 at surgery after 24 weeks of neoadjuvant therapy; negative axillary lymph nodes at surgery; Ki67 of ≤ 2.7 % at surgery; and tumor is ER+ at surgery. A PEPI score of 0 predicts a good outcome compared to PEPI >0. | 24 weeks |
Why PEPI Score 0 is a binary endpoint
The registered primary endpoint reduces several pathological and biological findings at surgery into a yes-or-no classification: whether the patient satisfies all of the specified criteria for PEPI Score 0. Statistically, that makes the endpoint a binary proportion. Each patient either contributes to the PEPI Score 0 group or does not.
The registry defines the components of the classification, but the record does not provide a numerical PEPI Score 0 rate in the information reported here.
5. Statistical Methodology
Primary comparison of proportions
For a binary endpoint such as PEPI Score 0, the natural statistical target is the proportion of patients achieving the specified outcome in each comparison group. Because the registry specifies the two ribociclib-containing arms as a combined group, the principal analysis would ordinarily compare the PEPI Score 0 proportion in the combined ribociclib group with the corresponding proportion in the letrozole-alone group.
Several effect measures can describe such a comparison. A risk difference expresses the absolute difference in PEPI Score 0 proportions, while a risk ratio expresses the ratio of the two proportions. A confidence interval around either measure quantifies uncertainty around the estimated treatment contrast.
These are general statistical representations of a binary endpoint. The registry information does not specify which effect measure or inferential test was used for the posted PEPI Score 0 result.
Why the combined-arm comparison matters
The primary endpoint explicitly combines the two ribociclib-containing arms. This changes the unit of the primary hypothesis from three separate groups to two comparison groups: ribociclib-containing therapy versus letrozole alone. Combining the arms can provide a direct estimate of the treatment strategy defined by the primary endpoint, but it also means that the primary comparison does not by itself establish whether the 600 mg and 400 mg regimens have identical effects.
Confidence intervals for a binary outcome
A confidence interval is particularly useful when the endpoint is expressed as a proportion because the observed proportion alone does not indicate how precisely it has been estimated. The interval reflects sampling uncertainty under the statistical procedure used. It is not a range containing the individual outcomes of future patients.
Hypothesis testing
A conventional hypothesis test for the primary binary comparison would evaluate whether the observed difference between the combined ribociclib group and the letrozole-alone group is compatible with a null hypothesis of no treatment-group difference. The p-value from such a test would address compatibility with that null hypothesis; it would not quantify the size or clinical importance of the treatment effect.
For a PEPI Score 0 endpoint, the treatment effect should be expressed using an absolute or relative comparison of proportions, accompanied by a confidence interval. The p-value, if reported, would answer a different question from the effect estimate: it would quantify evidence against the specified null hypothesis rather than describe how large the treatment difference is.
6. Statistical Methods Explained
Why is the primary endpoint a rate?
PEPI Score 0 is defined as a categorical outcome assessed at surgery. Because each patient either satisfies the complete set of criteria or does not, the most direct summary is the proportion of evaluated patients with PEPI Score 0.
Why are the ribociclib arms combined?
The registry defines the primary comparison as the two ribociclib-containing arms combined versus the letrozole-alone arm. Combining the arms therefore follows the structure of the registered primary question. The resulting treatment contrast concerns ribociclib-containing therapy as a group rather than one specific ribociclib dose.
What would a risk difference tell us?
A risk difference would quantify the absolute change in the probability of achieving PEPI Score 0 associated with the treatment comparison. For example, a positive risk difference would indicate a higher observed PEPI Score 0 proportion in the combined ribociclib group. Its magnitude would be expressed in probability units rather than as a relative ratio.
What would a risk ratio tell us?
A risk ratio compares the probability of PEPI Score 0 between the two groups multiplicatively. A value of 1 would correspond to equal observed proportions. Values above 1 would indicate a higher observed PEPI Score 0 probability in the combined ribociclib group, while values below 1 would indicate a lower probability.
Why does the p-value not measure effect size?
A p-value describes evidence relative to a null hypothesis under a specified statistical model. It does not tell the reader whether the difference is small or large, and it does not communicate the absolute probability of PEPI Score 0. Effect estimates and confidence intervals are therefore needed alongside hypothesis tests.
Why does the 24-week time frame matter?
The endpoint is explicitly assessed at surgery after 24 weeks of neoadjuvant therapy. The time point is part of the endpoint definition. An outcome measured at a different time would represent a different statistical question because the patient's pathological and biological status could change with additional treatment or follow-up.
Why is the PEPI definition clinically important to the statistical analysis?
The endpoint is composite in construction even though its final form is binary. A patient must meet the pT-stage, axillary-node, Ki67, and ER criteria simultaneously. Consequently, the PEPI Score 0 rate depends on the joint occurrence of all of those conditions rather than on any single component in isolation.
7. Results
The registry indicates that results have been posted for the study and identifies the PEPI Score 0 endpoint as a results-posted primary endpoint. However, the registry record does not report a numerical PEPI Score 0 rate, confidence interval, p-value, or other formal statistical estimate for that endpoint in the information available here. No formal statistical analyses were posted to ClinicalTrials.gov.
Primary endpoint: PEPI Score 0 at surgery
Registry-defined primary comparison
Time frame: 24 weeks
Endpoint: Rate of PEPI Score 0 at surgery
The registry defines PEPI Score 0 as pT stage 0, 1, or 2 at surgery after 24 weeks of neoadjuvant therapy, negative axillary lymph nodes at surgery, Ki67 of ≤ 2.7 % at surgery, and an ER+ tumor at surgery. A numerical result for the rate is not reported in the registry information used for this analysis.
The primary endpoint is a proportion, so its treatment effect would ordinarily be communicated using the observed PEPI Score 0 proportions and a corresponding effect measure such as a risk difference or risk ratio, together with a confidence interval.
The absence of a numerical effect estimate in the registry means that the primary endpoint cannot be interpreted quantitatively from the reported record. In particular, there is no reported effect size from which to conclude how much the probability of PEPI Score 0 differed between the combined ribociclib group and the letrozole-alone group.
A confidence interval, when available, would describe the precision of the estimated treatment contrast rather than the range of outcomes for individual patients. A p-value, if available, would address evidence against a specified null hypothesis rather than the magnitude of the treatment effect.
8. Safety Results
The registry reports serious adverse events by treatment arm. The reported measure is the number of patients affected divided by the number at risk for serious adverse events in each arm.
| Treatment arm | Serious adverse events | Interpretation of reported measure |
|---|---|---|
| Placebo + Letrozole | 3/37 | 3 patients affected among 37 at risk |
| Ribociclib 600 mg + Letrozole | 25/41 | 25 patients affected among 41 at risk |
| Ribociclib 400 mg + Letrozole | 20/41 | 20 patients affected among 41 at risk |
The serious-adverse-event counts are descriptive safety results. They should not be interpreted as an adjusted causal comparison without information about the safety analysis population, follow-up, event definitions, exposure duration, and prespecified statistical procedures.
9. Planned Analysis
The registry-defined primary outcome is a binary surgical endpoint measured after 24 weeks. For an endpoint of this type, a typical analysis would first summarize the proportion of patients achieving PEPI Score 0 in the combined ribociclib-containing arms and in the letrozole-alone arm. The principal treatment contrast could then be expressed as an absolute risk difference and/or risk ratio with an appropriate confidence interval.
Step 1: Define the outcome
Classify each patient according to whether all registry-specified criteria for PEPI Score 0 are satisfied at surgery.
Step 2: Define the comparison
Combine the ribociclib 600 mg + letrozole and ribociclib 400 mg + letrozole groups, then compare that combined group with placebo + letrozole.
Step 3: Estimate the effect
Calculate the PEPI Score 0 proportion in each comparison group and summarize the between-group contrast using a suitable binary-outcome effect measure.
Step 4: Quantify uncertainty
Report a confidence interval for the treatment contrast and, where prespecified, a hypothesis test for the null treatment comparison.
Why the primary endpoint is different from a time-to-event endpoint
The PEPI Score 0 endpoint is evaluated at a fixed clinical milestone—surgery after 24 weeks. It is therefore not naturally analyzed as a Kaplan-Meier endpoint. Unlike overall survival or progression-free survival, there is no event-time distribution to estimate for the primary outcome as defined by the registry.
The FELINE primary endpoint belongs to the first category because the registry specifies a surgical assessment at 24 weeks.
10. Interpretation of the Primary Endpoint
What a PEPI Score 0 rate would mean
The PEPI Score 0 rate represents the proportion of patients meeting all of the registry's specified pathological and biological criteria at surgery after 24 weeks of neoadjuvant therapy. It is therefore a treatment-response measure defined at a specific point in the neoadjuvant pathway.
A higher PEPI Score 0 rate would indicate that a larger proportion of patients satisfied the complete PEPI Score 0 definition at surgery. It would not mean that every component of the PEPI definition improved by the same amount, because the endpoint requires all components to be satisfied simultaneously.
What the endpoint does not establish by itself
The PEPI Score 0 endpoint is not itself a direct measure of overall survival, progression-free survival, or another long-term time-to-event outcome. The registry description states that a PEPI Score of 0 predicts a good outcome compared to PEPI >0, but the endpoint remains a classification measured at surgery.
The statistical value of the endpoint lies in its ability to provide a standardized binary treatment-response measure after a defined neoadjuvant interval. Its interpretation should remain anchored to the actual endpoint definition: achieving the complete PEPI Score 0 criteria at surgery after 24 weeks.
11. Three-Arm Design and the Primary Contrast
FELINE has three treatment arms but a primary comparison involving two analysis groups. This is a useful distinction in clinical-trial statistics: the number of randomized arms does not necessarily equal the number of primary hypotheses.
| Arm | Role in primary endpoint |
|---|---|
| Placebo + Letrozole | Comparator group |
| Ribociclib 600 mg + Letrozole | Combined with the 400 mg ribociclib arm for the registered primary comparison |
| Ribociclib 400 mg + Letrozole | Combined with the 600 mg ribociclib arm for the registered primary comparison |
This structure has two statistical consequences. First, the primary effect estimate should correspond to the combined ribociclib-containing population rather than automatically treating the two ribociclib doses as separate confirmatory tests. Second, the combined analysis does not answer a dose-specific question. A separate comparison of the 600 mg and 400 mg arms would be a different statistical question from the registered primary contrast.
12. Randomization and Masking
Randomization is important because it provides the design basis for comparing outcomes across treatment assignments. If treatment assignment is properly randomized, systematic differences in measured and unmeasured baseline factors are expected to be balanced in expectation rather than selected by investigators or patients.
The registry describes the trial as quadruple-masked. Masking can reduce the influence of treatment knowledge on patient management, outcome assessment, and other aspects of trial conduct. Its importance depends on which people are masked and how masking is maintained, but the registry classification establishes that the study was designed with quadruple masking.
Randomization
Creates the basis for an unbiased comparison of randomized treatment groups, subject to chance imbalance and departures from the protocol.
Quadruple masking
Reduces the opportunity for knowledge of treatment assignment to influence trial conduct or outcome assessment.
13. Trial Timeline
Trial start
The registry records the start of FELINE in February 2016.
Primary endpoint assessment
The registered primary endpoint is the rate of PEPI Score 0 at surgery after 24 weeks of neoadjuvant therapy.
Primary completion
The registry records August 27, 2018 as the primary completion date.
Active, not recruiting
The ClinicalTrials.gov record currently identifies the study as active, not recruiting.
14. Serious Adverse Events and Statistical Context
The safety data illustrate an important distinction between descriptive statistics and comparative inference. The registry gives the number affected and the number at risk for serious adverse events in each arm. Those data allow the reader to see the raw distribution of affected patients, but they do not by themselves provide a confidence interval, p-value, adjusted effect estimate, or prespecified safety hypothesis.
| Arm | Affected | At risk | Statistical role |
|---|---|---|---|
| Placebo + Letrozole | 3 | 37 | Descriptive comparator safety data |
| Ribociclib 600 mg + Letrozole | 25 | 41 | Descriptive safety data |
| Ribociclib 400 mg + Letrozole | 20 | 41 | Descriptive safety data |
For a formal comparative safety analysis, the statistical framework would depend on the prespecified safety endpoint and timing of assessment. Binary safety outcomes can be compared using risk differences, risk ratios, odds ratios, or exact methods where appropriate. Repeated or exposure-dependent events may require different methods. The registry information does not specify such an analysis for these serious adverse-event counts.
15. Limitations
- Primary efficacy estimate: the registry does not report a numerical PEPI Score 0 rate, confidence interval, or p-value for the registered primary endpoint.
- Formal statistical analysis: no formal statistical analyses were posted to ClinicalTrials.gov for the primary endpoint.
- Three-arm interpretation: the registered primary comparison combines the two ribociclib-containing arms, so it should not be interpreted as a dose-specific comparison.
- Fixed-time endpoint: PEPI Score 0 is assessed at surgery after 24 weeks and therefore does not provide the same type of information as a time-to-event endpoint.
- Composite classification: PEPI Score 0 requires all specified pT-stage, axillary-node, Ki67, and ER criteria to be satisfied simultaneously.
- Safety data: serious adverse events are reported descriptively as affected patients divided by patients at risk; the registry does not provide a comparative statistical analysis of these counts.
- Analysis details: the registry information does not specify the statistical test, effect measure, confidence-interval procedure, or analysis population used for the posted primary endpoint result.
- Long-term outcomes: the registered primary endpoint is a 24-week surgical endpoint, and the registry information does not provide additional numerical long-term efficacy outcomes for interpretation here.
16. Why This Trial Matters Statistically
FELINE is a useful teaching example because its primary endpoint looks simple—a rate of PEPI Score 0—but sits within a more complex three-arm randomized design. The trial illustrates how the statistical question must be defined before selecting an analysis: the registry specifies a combined treatment comparison rather than three independent primary comparisons.
| Concept | How it appears in FELINE |
|---|---|
| Randomization | Patients were randomized among three parallel treatment arms. |
| Masking | The registry describes the study as quadruple-masked. |
| Three-arm design | Placebo + letrozole, ribociclib 600 mg + letrozole, and ribociclib 400 mg + letrozole. |
| Combined treatment comparison | The primary endpoint compares the two ribociclib-containing arms combined against letrozole alone. |
| Binary endpoint | PEPI Score 0 is a yes/no classification at surgery. |
| Fixed assessment time | The primary endpoint is assessed after 24 weeks of neoadjuvant therapy. |
| Composite definition | PEPI Score 0 requires several pathological and biological criteria simultaneously. |
| Effect estimation | Risk difference or risk ratio would be natural measures for a binary treatment comparison. |
| Confidence intervals | Would quantify uncertainty around the estimated treatment contrast. |
| Safety description | Serious adverse events are reported as affected patients among those at risk in each arm. |
17. Statistical Concepts in This Trial
FELINE demonstrates several principles that recur throughout clinical-trial analysis: defining a binary endpoint precisely, distinguishing a randomized treatment comparison from the number of treatment arms, combining prespecified treatment groups, and separating descriptive safety data from formal comparative inference.
Binary outcome analysis
PEPI Score 0 is naturally summarized as the proportion of patients meeting all endpoint criteria.
Risk difference
Provides an absolute measure of the difference in PEPI Score 0 probability between treatment groups.
Risk ratio
Provides a relative comparison of the probability of PEPI Score 0 between the two primary analysis groups.
Composite endpoint
The PEPI Score 0 classification requires several specified conditions to occur together.
18. Sources
- ClinicalTrials.gov: FELINE — NCT02712723.
Continue through Clinical Biostats
Explore additional clinical-trial analyses, statistical tutorials, and practical methods for interpreting randomized clinical-trial evidence.
19. Record Summary
FELINE is a randomized, quadruple-masked phase 2 study with three parallel treatment arms and 121 enrolled patients. Its registered primary endpoint is the rate of PEPI Score 0 at surgery after 24 weeks, comparing the ribociclib-containing arms combined with the letrozole-alone arm. The endpoint is binary and requires the simultaneous presence of pT stage 0, 1, or 2, negative axillary lymph nodes, Ki67 of ≤ 2.7 %, and an ER+ tumor at surgery.
The statistical structure is therefore different from a trial centered on survival or another time-to-event endpoint. The natural treatment contrast is a comparison of proportions, with an absolute or relative effect measure and an accompanying confidence interval. The registry reports serious adverse events descriptively as 3/37 for placebo + letrozole, 25/41 for ribociclib 600 mg + letrozole, and 20/41 for ribociclib 400 mg + letrozole. A numerical PEPI Score 0 result and formal statistical analysis are not reported in the registry information available for the primary endpoint.