This page separates reported trial results from statistical interpretation. The numerical results presented here are limited to the information reported in the ClinicalTrials.gov record. ClinicalTrials.gov provides the official trial registry record.
1. Trial at a Glance
RIGHT Choice was a randomized, open-label, parallel-group phase 2 trial comparing ribociclib 600 mg with goserelin and hormonal therapy against combination chemotherapy in premenopausal or perimenopausal patients with advanced or metastatic breast cancer. The trial enrolled 222 participants and reported a primary time-to-event endpoint of progression-free survival.
| Feature | RIGHT Choice |
|---|---|
| Phase | Phase 2 |
| Condition | Breast Cancer |
| Population | Premenopausal or perimenopausal patients with advanced or metastatic breast cancer |
| Design | Randomized, parallel, open-label |
| Primary purpose | Treatment |
| Enrollment | 222 |
| Primary endpoint | Progression-free Survival |
| Primary endpoint type | Time-to-event |
| Primary hypothesis | Superiority |
| ClinicalTrials.gov | NCT03839823 |
| Trial status | COMPLETED |
| Lead sponsor | Novartis Pharmaceuticals |
2. Clinical Question
The central question was whether treatment with ribociclib 600 mg plus goserelin and hormonal therapy should be compared with combination chemotherapy for time-to-event outcomes in premenopausal or perimenopausal patients with advanced or metastatic breast cancer.
Population
Premenopausal or perimenopausal patients with advanced or metastatic breast cancer.
Intervention
Ribociclib 600 mg with goserelin and hormonal therapy. The registered interventions include letrozole OR anastrozole and goserelin.
Comparator
Combination chemotherapy. The registered chemotherapy interventions include docetaxel/capecitabine, capecitabine/vinorelbine, and paclitaxel/gemcitabine.
Primary question
Does the ribociclib-based treatment produce a different time-to-event outcome than combination chemotherapy under the trial's prespecified superiority framework?
3. Trial Design
Ribociclib 600 mg
- Ribociclib 600 mg
- Goserelin
- Letrozole OR anastrozole
Combination chemotherapy
- Docetaxel / capecitabine
- Capecitabine / vinorelbine
- Paclitaxel / gemcitabine
4. Analysis Population
The primary and secondary efficacy analyses reported in the registry used the full analysis set, defined as all patients to whom study treatment was assigned by randomization. This is aligned with the intention-to-treat principle: treatment effects are evaluated according to randomized assignment rather than selectively excluding participants because of subsequent treatment experience.
| Population / principle | Registry description and statistical role |
|---|---|
| Full analysis set | All patients to whom study treatment was assigned by randomization; used for the reported efficacy analyses. |
| Intention-to-treat concept | The statistical analyses identify intention-to-treat as an analysis concept and anchor the full analysis set to randomized treatment assignment. |
| Safety population | The ClinicalTrials.gov record does not provide a separate formal definition of the safety analysis population; serious adverse events are reported by treatment arm for on-treatment patients. |
5. Endpoints
| Endpoint | Registry definition / time frame | Type |
|---|---|---|
| Progression-free Survival | Time from the date of randomization to the date of the first documented progression as per local review and according to RECIST 1.1 or death due to any cause. PFS was censored at the date of the last adequate tumor assessment if no PFS event was observed at the time of last patient, last visit (LPLV). Time frame: Up to approximately 34 months. | Time-to-event |
| Time to Treatment Failure | Time frame: Up to approximately 34 months. | Time-to-event |
| Time to Response | Time frame: Up to approximately 34 months. | Time-to-event |
| 3-month Treatment Failure Rate | Time frame: Up to approximately 3 months. | Count / rate |
| Overall Response Rate (ORR) | Time frame: Up to approximately 34 months. | Binary |
| Clinical Benefit Rate | Time frame: Up to approximately 34 months. | Count / rate |
The ClinicalTrials.gov record identifies Progression-free Survival as the sole registered primary endpoint. The other posted outcomes are secondary endpoints in the statistical analyses in the ClinicalTrials.gov record.
6. Statistical Methodology
Stratified log-rank test
The primary progression-free survival comparison used a one-sided stratified log-rank test. The log-rank test is designed for comparing time-to-event distributions between treatment groups while accounting for censoring. A stratified version performs the comparison within strata and combines the evidence across them rather than treating all observations as though they belonged to a single unstructured population.
The registry identifies the hypothesis type as superiority and reports a one-sided stratified log-rank test for the primary PFS analysis.
Hazard ratio
The primary effect measure was the hazard ratio (HR). The reported estimate was 0.611, with a two-sided 95% confidence interval of 0.429 to 0.870. The ClinicalTrials.gov record does not identify the specific regression model used to generate the hazard-ratio estimate, so no particular Cox-model specification is assumed here.
An HR below 1 indicates a lower estimated instantaneous event rate in the ribociclib-based group relative to the combination-chemotherapy group under the analysis framework.
Cochran-Mantel-Haenszel test
The registry reports the Cochran-Mantel-Haenszel test for the 3-month treatment failure rate, overall response rate, and clinical benefit rate. This family of methods is useful for comparing categorical outcomes while accounting for stratification when appropriate. The ClinicalTrials.gov record does not provide the individual strata or their weights, so the analysis is described at the method level rather than reconstructed numerically.
Intention-to-treat analysis
The statistical analyses identify intention-to-treat as an analysis concept, and the full analysis set includes all patients assigned treatment by randomization. This preserves the randomized comparison and avoids changing the treatment groups based on later events such as discontinuation or treatment experience.
Kaplan-Meier estimation
Progression-free survival is a time-to-event endpoint with censoring, making Kaplan-Meier estimation a standard descriptive approach for showing the estimated event-free probability over time. The registry statistical-analysis record explicitly reports the stratified log-rank test for the primary PFS comparison but does not separately identify Kaplan-Meier estimation as the posted statistical method. Accordingly, Kaplan-Meier is discussed here as the standard descriptive framework for this endpoint rather than as a claimed registry-specific method.
7. Primary Result: Progression-free Survival
The primary endpoint was progression-free survival, defined as the time from randomization to first documented progression according to local review and RECIST 1.1 or death from any cause. If no PFS event was observed at the time of last patient, last visit, PFS was censored at the date of the last adequate tumor assessment.
Hazard ratio for progression or death
95% CI: 0.429–0.870 · P = 0.003
One-sided stratified log-rank test · Superiority hypothesis
The reported HR of 0.611 means that the estimated instantaneous rate of the PFS event—documented progression or death—was approximately 61.1% as high in the ribociclib-based group as in the combination-chemotherapy group under the reported time-to-event analysis. Expressed as a relative comparison, 1 − 0.611 = 0.389, so the estimate corresponds to an approximately 38.9% lower estimated hazard for progression or death.
The HR does not mean that 38.9% of patients avoided progression, that each patient experienced exactly a 38.9% reduction in risk, or that the absolute difference in progression-free survival probabilities was 38.9 percentage points. A hazard ratio is a relative time-to-event measure, not an absolute risk difference.
The 95% CI of 0.429–0.870 describes uncertainty around the estimated relative treatment effect under the statistical framework. It does not describe the range of outcomes that individual patients might experience. Because the entire interval is below 1, the reported interval is consistent with a lower estimated hazard in the ribociclib-based group.
The P = 0.003 is evidence against the null hypothesis under the reported one-sided stratified log-rank testing framework. It is not a measure of the size of the treatment effect, and it does not mean that there is a 0.3% probability that the null hypothesis is true.
Interpretation also depends on the analysis population, censoring rules, and the assumptions underlying the hazard-ratio summary. A single HR is most straightforward when the relative hazards are reasonably stable over time; the ClinicalTrials.gov record does not provide enough information to independently assess proportional hazards.
| Primary endpoint | Ribociclib 600 mg | Combination Chemotherapy | Reported comparison |
|---|---|---|---|
| Progression-free Survival | — | — | HR 0.611; 95% CI 0.429–0.870; P = 0.003 |
The registry supplies the comparative hazard ratio, confidence interval, and P-value but does not provide median PFS or arm-specific PFS probabilities in the ClinicalTrials.gov record. Those quantities therefore are not added here.
8. Secondary Endpoint Results
Time to Treatment Failure
Hazard ratio
95% CI: 0.363–0.680
Superiority hypothesis · Full analysis set
The reported HR of 0.497 indicates a lower estimated instantaneous rate of treatment failure in the ribociclib 600 mg group relative to combination chemotherapy. Numerically, the estimate corresponds to approximately a 50.3% lower estimated hazard because 1 − 0.497 = 0.503.
The 95% CI of 0.363–0.680 remains below 1, indicating that the reported interval is compatible with a lower estimated treatment-failure hazard in the ribociclib-based group. The ClinicalTrials.gov record does not report a formal analysis method or P-value for this endpoint, so no additional inferential conclusion is attached to it.
Time to Response
Hazard ratio
95% CI: 0.546–1.064
Superiority hypothesis · Full analysis set
The estimated HR of 0.762 represents the reported relative time-to-response comparison. The two-sided 95% CI of 0.546–1.064 includes 1, so the interval is compatible with both a lower and a higher hazard of response under the stated model-based comparison. The ClinicalTrials.gov record does not provide a P-value or the exact statistical method used for this endpoint.
3-month Treatment Failure Rate
Between-group test
Cochran-Mantel-Haenszel test · Full analysis set
The 3-month treatment failure rate was compared using the Cochran-Mantel-Haenszel method, with a reported P-value of 0.020 under a superiority hypothesis. The ClinicalTrials.gov record does not provide the arm-specific treatment failure percentages or counts, so the magnitude and direction of the absolute difference cannot be reconstructed from the ClinicalTrials.gov record alone.
Overall Response Rate
Between-group test
Cochran-Mantel-Haenszel test · Full analysis set
Overall response rate was analyzed with the Cochran-Mantel-Haenszel test. The reported P-value was 0.255. The ClinicalTrials.gov record does not include the arm-specific response rates, response counts, or a confidence interval, so the size of the between-group difference cannot be quantified from these data.
Clinical Benefit Rate
Between-group test
Cochran-Mantel-Haenszel test · Full analysis set
Clinical benefit rate was also analyzed with the Cochran-Mantel-Haenszel test. The reported P-value was 0.116. As with overall response rate, the ClinicalTrials.gov record does not provide arm-specific rates or counts or a confidence interval.
| Secondary endpoint | Effect / test | 95% CI | P-value | Method |
|---|---|---|---|---|
| Time to Treatment Failure | HR 0.497 | 0.363–0.680 | Not reported | Not reported |
| Time to Response | HR 0.762 | 0.546–1.064 | Not reported | Not reported |
| 3-month Treatment Failure Rate | Not reported | Not reported | 0.020 | Cochran-Mantel-Haenszel |
| Overall Response Rate (ORR) | Not reported | Not reported | 0.255 | Cochran-Mantel-Haenszel |
| Clinical Benefit Rate | Not reported | Not reported | 0.116 | Cochran-Mantel-Haenszel |
9. Safety Results
The ClinicalTrials.gov record reports serious adverse events by treatment arm for patients on treatment. These figures are reported separately from the efficacy analysis and should not be interpreted as a comparative efficacy measure.
| Safety measure | Ribociclib 600 mg | Combination Chemotherapy |
|---|---|---|
| Serious adverse events, on-treatment | 17 / 112 | 16 / 100 |
The ClinicalTrials.gov record provides affected and at-risk counts rather than a formal statistical comparison of serious adverse-event rates. For that reason, no P-value, confidence interval, relative risk, or odds ratio is presented here. The denominator also differs from the overall enrollment figure because these safety counts are specifically identified as on-treatment.
10. Statistical Methods Explained
Why was a stratified log-rank test used for the primary endpoint?
Progression-free survival is a time-to-event endpoint in which some participants may not have experienced progression or death by the time their follow-up ends. A log-rank test compares the survival experience of the treatment groups across follow-up while accommodating censoring. Stratification extends that framework when the analysis needs to account for predefined strata. The RIGHT Choice registry specifically reports a one-sided stratified log-rank test for PFS.
What does an HR of 0.611 mean?
An HR of 0.611 is a relative time-to-event measure. It indicates that the estimated instantaneous rate of progression or death in the ribociclib-based group was 0.611 times the corresponding rate in the combination-chemotherapy group under the reported analysis. It does not describe the probability that an individual patient will progress, nor does it directly give an absolute difference in PFS at a particular month.
Why is the confidence interval important?
The point estimate alone does not show how precisely the treatment effect has been estimated. The 95% CI of 0.429–0.870 supplies an interval describing uncertainty around the reported HR. A relatively narrow interval indicates greater precision than a very wide interval, although precision and clinical importance are separate questions.
Why doesn't the P-value measure effect size?
The P-value addresses evidence against the null hypothesis under the specified statistical test. It depends not only on the estimated treatment effect but also on the amount of information in the data and the variability of the estimate. Consequently, a small P-value does not tell you that an effect is large, while a larger P-value does not prove that the groups are equivalent.
What does censoring mean for PFS?
Under the registry definition, if no PFS event was observed at last patient, last visit, PFS was censored at the date of the last adequate tumor assessment. Censoring means that the available follow-up contributes information up to a specified time without being treated as an observed progression or death at that time. Time-to-event methods are designed to incorporate this incomplete event information.
Why should the one-sided primary test and two-sided confidence interval be distinguished?
The primary PFS analysis is explicitly described as a one-sided stratified log-rank test, while the reported hazard-ratio confidence interval is explicitly identified as a two-sided 95% confidence interval. These are different inferential quantities and should be reported exactly as specified rather than silently treating them as interchangeable.
What does the Cochran-Mantel-Haenszel method add to categorical outcomes?
For outcomes such as 3-month treatment failure rate, overall response rate, and clinical benefit rate, the Cochran-Mantel-Haenszel approach can compare treatment groups while accounting for stratification. In RIGHT Choice, the registry explicitly reports this method for those three secondary endpoints. The ClinicalTrials.gov record does not provide the underlying strata or arm-specific counts, so the numerical CMH calculation cannot be independently reconstructed here.
11. Interpreting the Complete Statistical Pattern
The results posted on ClinicalTrials.gov for RIGHT Choice contain several different kinds of statistical evidence. The primary endpoint is a time-to-event comparison with a hazard ratio and confidence interval. Two secondary time-to-event endpoints also have hazard-ratio estimates, while three categorical or rate-based secondary endpoints are summarized through Cochran-Mantel-Haenszel P-values.
| Evidence type | RIGHT Choice result | What it tells us |
|---|---|---|
| Primary time-to-event effect | HR 0.611; 95% CI 0.429–0.870; P = 0.003 | Relative comparison of the rate of progression or death under the primary analysis. |
| Secondary time-to-event effect | Time to Treatment Failure HR 0.497; 95% CI 0.363–0.680 | Relative comparison of the treatment-failure hazard. |
| Secondary time-to-event effect | Time to Response HR 0.762; 95% CI 0.546–1.064 | Relative comparison of the time-to-response hazard; the CI includes 1. |
| Categorical/rate test | 3-month Treatment Failure Rate P = 0.020 | Evidence from a CMH comparison; arm-specific rates are not reported. |
| Categorical test | ORR P = 0.255 | CMH comparison of overall response rate; arm-specific rates are not reported. |
| Categorical/rate test | Clinical Benefit Rate P = 0.116 | CMH comparison of clinical benefit rate; arm-specific rates are not reported. |
This distinction matters because a P-value, a hazard ratio, and a response rate are not interchangeable measures. The strongest interpretation comes from preserving the statistical scale of each endpoint and considering the confidence interval where one is available.
12. Multiplicity and Endpoint Hierarchy
The ClinicalTrials.gov record identifies one registered primary endpoint and five secondary endpoints with statistical analyses posted. The primary endpoint is explicitly associated with a superiority hypothesis. The secondary analyses are also identified as superiority hypotheses.
| Endpoint role | Endpoint | Reported inferential information |
|---|---|---|
| Primary | Progression-free Survival | HR 0.611; 95% CI 0.429–0.870; P = 0.003; one-sided stratified log-rank |
| Secondary | Time to Treatment Failure | HR 0.497; 95% CI 0.363–0.680 |
| Secondary | Time to Response | HR 0.762; 95% CI 0.546–1.064 |
| Secondary | 3-month Treatment Failure Rate | P = 0.020; Cochran-Mantel-Haenszel |
| Secondary | Overall Response Rate (ORR) | P = 0.255; Cochran-Mantel-Haenszel |
| Secondary | Clinical Benefit Rate | P = 0.116; Cochran-Mantel-Haenszel |
The ClinicalTrials.gov record does not specify an alpha-allocation strategy, multiplicity adjustment procedure, hierarchical testing sequence, or interim-analysis framework. These features therefore cannot be inferred from the presence of multiple reported P-values. In particular, the P-values for secondary endpoints should not automatically be interpreted as though they were independently confirmatory tests with no multiplicity considerations.
13. Crossover, Interim Analysis, and Bayesian Methods
Crossover
The ClinicalTrials.gov record does not report a crossover design or crossover results. No crossover adjustment is therefore described.
Interim analysis
The ClinicalTrials.gov record does not report an interim-analysis procedure, stopping boundary, alpha-spending method, or interim efficacy result.
Bayesian methods
No Bayesian analysis or Bayesian decision rule is identified in the ClinicalTrials.gov record.
Non-inferiority
The trial is classified as a superiority hypothesis, not a non-inferiority analysis. No non-inferiority margin is reported.
These omissions are analytically important. Trial design features such as interim monitoring, crossover, non-inferiority margins, and Bayesian decision rules can materially change how a result should be interpreted.
14. Limitations
- Limited registry detail: the ClinicalTrials.gov record provides the key estimates and tests but does not provide every component of the underlying statistical analysis plan.
- No arm-specific PFS summaries: median PFS and time-specific PFS estimates are not included in the ClinicalTrials.gov record, so they are not reported here.
- No subgroup results: the ClinicalTrials.gov record contains no subgroup estimates or interaction tests. No subgroup conclusions can therefore be drawn.
- Secondary endpoint methods: formal statistical methods are not reported for Time to Treatment Failure or Time to Response in the ClinicalTrials.gov record.
- Incomplete categorical results: the CMH P-values for treatment failure rate, ORR, and clinical benefit rate are provided without the corresponding arm-specific counts or percentages.
- Multiplicity: the ClinicalTrials.gov record does not specify how type I error was allocated across the primary and secondary endpoints.
- Hazard-ratio interpretation: a single HR summarizes a relative time-to-event comparison and does not provide an absolute benefit at a specific time point.
- Proportional-hazards considerations: the ClinicalTrials.gov record does not provide sufficient information to assess whether a proportional-hazards assumption is appropriate over follow-up.
- Open-label design: the registry classifies masking as none. Knowledge of treatment assignment can affect some aspects of trial conduct and assessment, although the ClinicalTrials.gov record does not quantify such effects.
- Safety denominators: serious adverse-event counts are reported for on-treatment patients, so these denominators should not be confused with the total enrollment of 222.
15. Why This Trial Matters Statistically
RIGHT Choice is a useful teaching example because the trial combines randomized treatment assignment with a primary time-to-event endpoint and several secondary endpoints on different statistical scales. The same trial therefore illustrates why the choice of statistical method must follow the structure of the endpoint.
| Concept | How it appears in RIGHT Choice |
|---|---|
| Randomization | The trial used randomized allocation with a parallel two-arm design. |
| Intention-to-treat analysis | The full analysis set comprised all patients assigned treatment by randomization. |
| Time-to-event endpoint | Progression-free survival was the registered primary endpoint. |
| Stratified log-rank test | The primary PFS analysis used a one-sided stratified log-rank test. |
| Hazard ratio | PFS, Time to Treatment Failure, and Time to Response were reported using HRs. |
| Confidence interval | The primary HR had a two-sided 95% CI of 0.429–0.870. |
| Cochran-Mantel-Haenszel test | Used for 3-month Treatment Failure Rate, ORR, and Clinical Benefit Rate. |
| Endpoint hierarchy | One registered primary endpoint was distinguished from multiple secondary outcomes. |
| Censoring | The PFS definition specifies censoring at the last adequate tumor assessment when no PFS event was observed. |
| Safety analysis | Serious adverse events were reported by arm using on-treatment denominators. |
16. Clinical Biostats Interpretation of the Primary Finding
The reported primary PFS analysis produced an HR of 0.611, with a two-sided 95% CI of 0.429–0.870 and a one-sided stratified log-rank P = 0.003. Under the stated superiority framework, this provides statistical evidence of a difference in PFS between the randomized treatment groups, with the estimated relative hazard below 1.
The HR does not establish an absolute number of additional months without progression, because the ClinicalTrials.gov record does not include median PFS or time-specific PFS probabilities. It also does not establish that every patient experiences the same relative treatment effect. Those questions require additional descriptive and, where appropriate, subgroup information.
The secondary endpoints operate on different statistical scales. Time to Treatment Failure has an HR of 0.497, Time to Response has an HR of 0.762, while the three categorical/rate endpoints are represented by CMH P-values. These results should not be combined into a single numerical measure of treatment benefit.
Serious adverse events occurred in 17 of 112 on-treatment patients in the ribociclib 600 mg group and 16 of 100 in the combination-chemotherapy group. These counts provide descriptive safety information, but the ClinicalTrials.gov record does not include a formal statistical comparison. Efficacy and safety therefore require separate interpretation.
17. Statistical Methods Explained in More Depth
Why does the PFS definition include both progression and death?
Progression-free survival treats the first occurrence of documented progression or death from any cause as the event. This creates a composite time-to-event endpoint: a participant does not need to experience both events. The definition also specifies how observations without a PFS event are censored.
Why is randomization important for interpreting the hazard ratio?
Randomization assigns participants to treatment groups using a design intended to balance measured and unmeasured prognostic factors in expectation. Consequently, the treatment-group comparison is anchored to randomized assignment rather than to a comparison of patients who independently chose or received different therapies.
Why isn't an HR of 0.497 automatically a 50.3% improvement?
The HR of 0.497 corresponds to an approximately 50.3% lower estimated hazard because 1 − 0.497 = 0.503. But the word "improvement" can obscure what was actually measured. The endpoint is Time to Treatment Failure, and the HR describes the relative instantaneous event rate, not a percentage increase in a patient's probability of remaining on treatment.
Why does the Time to Response confidence interval matter?
The Time to Response estimate is HR 0.762 with a 95% CI of 0.546–1.064. Because the interval extends above 1, the reported uncertainty includes the possibility that the relative response-event hazard is not lower. This illustrates why a point estimate should not be interpreted without its confidence interval.
Why can't the ORR P-value tell us the size of the ORR difference?
A P-value indicates the evidence generated by the specified test, not the magnitude of the underlying difference. The registry-reported RIGHT Choice data reports P = 0.255 for ORR but does not provide the two arm-specific response rates. Without those rates, the magnitude of the ORR difference cannot be calculated from the ClinicalTrials.gov record.
Why is the CMH test different from the log-rank test?
The log-rank test is designed for time-to-event data, where both event timing and censoring matter. The Cochran-Mantel-Haenszel test is used for categorical or rate-based comparisons and can account for stratification. RIGHT Choice uses these methods for different endpoint structures rather than applying a single test indiscriminately to every outcome.
18. Data Completeness and What Is Not Reported Here
A statistically careful trial summary should distinguish between information that was not reported and information that was negative or absent in the trial itself. The registry-reported RIGHT Choice data contains six statistical analyses, including one primary analysis, but it does not contain several commonly reported clinical-trial summaries.
| Potential result or design feature | Status in the ClinicalTrials.gov record |
|---|---|
| Median PFS | Not reported in the ClinicalTrials.gov record. |
| Time-specific PFS rates | Not reported in the ClinicalTrials.gov record. |
| Arm-specific ORR | Not reported in the ClinicalTrials.gov record. |
| Arm-specific clinical benefit rate | Not reported in the ClinicalTrials.gov record. |
| Arm-specific 3-month treatment failure rate | Not reported in the ClinicalTrials.gov record. |
| Subgroup analyses | Not reported. |
| Formal analysis method for Time to Treatment Failure | Not reported. |
| Formal analysis method for Time to Response | Not reported. |
| Multiplicity adjustment | Not reported. |
| Interim analysis | Not reported. |
| Non-inferiority margin | Not applicable to the reported superiority hypothesis; no margin is reported. |
| Bayesian methods | Not reported. |
| Crossover | Not reported. |
This is why the page does not infer median survival, subgroup treatment effects, forest plots, or additional design procedures from the existence of the trial. The goal is to preserve the distinction between documented evidence and general statistical knowledge.
19. Trial Timeline
Trial start
The registered trial start date was 2019-02-25.
Primary completion
The registered primary completion date was 2022-04-12.
Trial status
The ClinicalTrials.gov record identifies the trial status as COMPLETED and reports results for 11 outcome measures, including 6 posted statistical analyses.
20. Overall Statistical Interpretation
RIGHT Choice provides a compact example of how a randomized oncology trial can generate evidence across several endpoint types. The primary progression-free survival analysis is the most formally developed result in the ClinicalTrials.gov record: it uses a one-sided stratified log-rank test and reports an HR of 0.611 with a two-sided 95% CI of 0.429–0.870 and P = 0.003.
The secondary results add context but require endpoint-specific interpretation. Time to Treatment Failure has an HR of 0.497 with a 95% CI of 0.363–0.680, while Time to Response has an HR of 0.762 with a 95% CI of 0.546–1.064. The 3-month Treatment Failure Rate, ORR, and Clinical Benefit Rate are summarized using Cochran-Mantel-Haenszel P-values of 0.020, 0.255, and 0.116, respectively, without the arm-specific rates in the ClinicalTrials.gov record.
The central statistical lesson is that the primary endpoint, effect measure, confidence interval, hypothesis test, analysis population, and censoring rule must be interpreted together. A hazard ratio describes a relative time-to-event effect; a P-value describes evidence under a specified test; and a confidence interval describes uncertainty around an estimate. None of these quantities, by itself, supplies the complete clinical picture.
21. Related Tutorials
Learn more about the methods used in this trial:
22. Related Statistical Calculators
23. Sources
- ClinicalTrials.gov: RIGHT Choice, NCT03839823.
- Linked publication: PubMed record for PMID 38771995.
Continue through Clinical Biostats
Explore statistical tutorials, calculators, and additional clinical-trial analyses that connect trial endpoints to the methods used to evaluate them.
24. Record Summary
RIGHT Choice is a randomized phase 2, open-label, parallel-group trial with 222 enrolled participants and a primary time-to-event endpoint of progression-free survival. The primary analysis used a one-sided stratified log-rank test and reported an HR of 0.611, a two-sided 95% CI of 0.429–0.870, and P = 0.003. Secondary analyses included hazard-ratio comparisons for Time to Treatment Failure and Time to Response and Cochran-Mantel-Haenszel comparisons for 3-month Treatment Failure Rate, Overall Response Rate, and Clinical Benefit Rate.
The statistical interpretation should remain tied to the information actually reported. The PFS result provides a relative time-to-event estimate with quantified uncertainty; the secondary endpoints provide additional but differently scaled evidence; and the serious adverse-event counts provide descriptive safety information. Median PFS, subgroup estimates, multiplicity procedures, interim-analysis rules, crossover information, and other details are not reported in the ClinicalTrials.gov record used for this page and are therefore not inferred.