This page separates reported trial results from statistical interpretation. Numerical trial results on this page are restricted to the ClinicalTrials.gov record for NCT02437318. The registry provides one posted formal statistical analysis for the registered primary endpoint.
1. Trial at a Glance
SOLAR-1 was a randomized, parallel-group, triple-masked, phase 3 treatment trial enrolling men and postmenopausal women with advanced breast cancer that progressed on or after aromatase inhibitor treatment. The registered primary endpoint was progression-free survival in the PIK3CA mutant cohort, analyzed using a log-rank test with a hazard ratio as the effect measure.
| Feature | SOLAR-1 |
|---|---|
| Trial name | SOLAR-1 |
| ClinicalTrials.gov identifier | NCT02437318 |
| Phase | Phase 3 |
| Status | Completed |
| Therapeutic area | Oncology |
| Condition | Breast Cancer |
| Enrollment | 572 |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Triple |
| Primary purpose | Treatment |
| Primary endpoint type | Time-to-event |
| Primary endpoint | Progression-free Survival (PFS) Per Investigator Assessment in the PIK3CA Mutant Cohort |
| Primary statistical method | Log-rank test |
| Effect measure | Hazard ratio |
| Hypothesis type | Superiority |
| Lead sponsor | Novartis Pharmaceuticals |
| Sponsor type | Industry |
2. Clinical Question
The registered trial evaluated alpelisib plus fulvestrant against placebo plus fulvestrant in men and postmenopausal women with advanced breast cancer that progressed on or after aromatase inhibitor treatment. The primary statistical question was whether progression-free survival differed between the randomized treatment groups in the PIK3CA mutant cohort.
Population
Men and postmenopausal women with advanced breast cancer that progressed on or after aromatase inhibitor treatment.
Intervention
Alpelisib plus fulvestrant.
Comparator
Placebo plus fulvestrant.
Primary question
In the PIK3CA mutant cohort, does alpelisib plus fulvestrant improve progression-free survival relative to placebo plus fulvestrant?
3. Trial Design
Alpelisib + Fulvestrant
- Alpelisib
- Fulvestrant
Placebo + Fulvestrant
- Placebo
- Fulvestrant
The ClinicalTrials.gov record identifies the intervention components and the two randomized groups but do not provide randomized allocation ratios or arm-specific enrollment counts. Accordingly, this page does not infer an allocation ratio from the total enrollment.
4. Endpoints
The registry lists one registered primary endpoint, classified as a time-to-event outcome. Its analysis was posted with an estimate, a two-sided 95% confidence interval, and a P-value.
| Endpoint | Registry definition / time frame | Statistical analysis |
|---|---|---|
| Progression-free Survival (PFS) Per Investigator Assessment in the PIK3CA Mutant Cohort | Once approximately 243 PFS events in the PIK3CA mutant cohort had been observed, up to 33.3 months. | Log Rank; hazard ratio; superiority |
Registry definition of PFS
The registry defines PFS as the time from the date of randomization to the date of the first documented progression or death due to any cause. PFS was assessed via a local radiology assessment according to RECIST 1.1.
If a patient did not have an event, PFS was censored at the date of last adequate tumor assessment. The registry states that the PFS distribution was estimated using Kaplan-Meier methodology. The registry-reported endpoint definition also begins its description of progression as "Progression was defined" but the ClinicalTrials.gov record end before the remainder of that definition; this page therefore does not complete or infer that definition.
5. Statistical Analysis Framework
The primary endpoint is a time-to-event outcome, so the analysis is concerned not only with whether progression or death occurred, but also with the amount of time from randomization until that event. This distinction is important because some participants may remain event-free at the time their follow-up ends.
| Statistical element | Registered / posted approach |
|---|---|
| Endpoint type | Time-to-event |
| Primary endpoint | Progression-free survival per investigator assessment in the PIK3CA mutant cohort |
| Analysis population | All randomized subjects with a PIK3CA mutation based on central testing of hotspot mutations in tumor tissue |
| Distribution estimator | Kaplan-Meier methodology |
| Between-group test | Log-rank test |
| Effect measure | Hazard ratio |
| Confidence interval | 95%, two-sided |
| Hypothesis type | Superiority |
| Event-based timing | Once approximately 243 PFS events in the PIK3CA mutant cohort had been observed |
| Maximum listed time frame | Up to 33.3 months |
Why the event target matters
For a time-to-event trial, information accumulates through observed events rather than simply through the number of participants enrolled. The registry specifies that the primary PFS assessment was made once approximately 243 PFS events in the PIK3CA mutant cohort had been observed, with the registered time frame extending up to 33.3 months.
This event-driven structure helps explain why a trial can have a fixed enrollment total but reach its principal analysis at a later point determined by how quickly progression or death events accumulate.
6. Results: Primary Progression-Free Survival Analysis
The ClinicalTrials.gov record contains one formal statistical analysis, corresponding to the registered primary endpoint. The analysis compares the PIK3CA mutant cohort receiving alpelisib plus fulvestrant with the PIK3CA mutant cohort receiving placebo plus fulvestrant.
Hazard ratio for progression or death
95% CI: 0.50–0.85 · P = 0.00065
Primary endpoint: PFS in the PIK3CA mutant cohort.
| Primary endpoint | Alpelisib + Fulvestrant | Placebo + Fulvestrant | Effect estimate |
|---|---|---|---|
| Progression-free Survival (PFS) Per Investigator Assessment in the PIK3CA Mutant Cohort | PIK3CA mutant cohort | PIK3CA mutant cohort | HR 0.65 95% CI 0.50–0.85 P = 0.00065 |
The hazard ratio of 0.65 means that, within the time-to-event framework used for this analysis, the estimated instantaneous rate of progression or death in the alpelisib-plus-fulvestrant group was approximately 65% of the corresponding estimated rate in the placebo-plus-fulvestrant group. Expressed as a relative interpretation, this corresponds to an estimated 35% lower hazard of progression or death.
The HR does not mean that 35% of patients avoided progression, that each individual patient had exactly a 35% reduction in risk, or that the absolute probability of progression or death was reduced by 35 percentage points. A hazard ratio is a relative time-to-event measure, not an absolute risk difference.
The 95% confidence interval of 0.50–0.85 describes statistical uncertainty around the estimated hazard ratio under the analysis framework. It indicates that the estimated relative effect is not represented by a single number alone. The interval does not describe the range of individual patient responses.
The P-value of 0.00065 addresses the statistical evidence against the null hypothesis under the specified testing framework. It does not measure the magnitude of the treatment effect. Effect size is described by the hazard ratio and its confidence interval, while clinical magnitude also requires absolute outcome information.
The PFS definition also includes censoring when an event has not occurred by the date of the last adequate tumor assessment. Consequently, the analysis depends on the handling and assumptions associated with censored observations. The ClinicalTrials.gov record does not provide enough detail to independently assess the censoring mechanism or any proportional-hazards assumption underlying a hazard-ratio interpretation.
Reading HR 0.65 in context
The visual above is a conceptual representation of the reported hazard ratio, not a Kaplan-Meier curve and not an estimate of the proportion of patients who remained progression-free. A hazard ratio summarizes a relative comparison over time; it does not by itself provide a time-specific survival probability.
7. Confidence Interval and Precision
The reported 95% confidence interval extends from 0.50 to 0.85. This interval is useful because it shows the uncertainty surrounding the point estimate of 0.65 rather than presenting the hazard ratio as exact.
The point estimate is below 1, and the entire reported two-sided 95% confidence interval is also below 1. The registry's formal analysis additionally reports P = 0.00065 under a superiority hypothesis.
A confidence interval should not be interpreted as saying that there is a 95% probability that the true hazard ratio lies between 0.50 and 0.85. In the conventional frequentist interpretation, the interval is generated by a procedure designed to contain the underlying parameter in 95% of repeated samples under the model and assumptions used for the analysis.
The interval also does not answer whether a particular patient will benefit. It describes uncertainty in the population-level effect estimate. Individual outcomes can differ substantially from the population-level hazard ratio.
8. Kaplan-Meier Estimation
The registry states that the PFS distribution was estimated using Kaplan-Meier methodology. Kaplan-Meier estimation is particularly useful for clinical-trial time-to-event endpoints because not every participant necessarily experiences progression or death during the observation period.
At each observed event time, the estimated event-free probability is updated according to the number of events and the number of participants at risk immediately before that time.
A participant who has not experienced progression or death when follow-up ends does not simply disappear from the analysis. Instead, that participant can contribute event-free follow-up until the censoring time. The registry specifically states that participants without an event were censored at the date of their last adequate tumor assessment.
The ClinicalTrials.gov record does not provide the underlying individual event and censoring times, so this page does not attempt to reconstruct a Kaplan-Meier curve.
9. Log-Rank Test
The formal statistical method reported for the primary endpoint is the log-rank test. The log-rank test compares the event-time experience of two groups across follow-up rather than comparing only a single time point.
At each event time, the basic logic is to compare the observed number of events in each group with the number expected under the null hypothesis of no difference between the groups, conditional on the participants who remain at risk at that time. Contributions from event times are then combined across follow-up.
What it tests
Whether the observed time-to-event experience differs between the randomized treatment groups under the specified log-rank framework.
What it does not provide alone
The log-rank test does not by itself provide an absolute treatment effect. The hazard ratio and confidence interval provide an effect estimate and uncertainty measure.
The distinction matters because a P-value from a log-rank test and a hazard ratio answer different statistical questions. The P-value describes evidence against the null hypothesis, whereas the hazard ratio quantifies the relative event-rate comparison represented by the fitted effect measure.
10. Why Was a Hazard Ratio Used?
PFS is measured from randomization until progression or death, with censoring for participants without an event at their last adequate tumor assessment. A hazard ratio is therefore a natural effect measure for a randomized time-to-event comparison.
For SOLAR-1, the reported HR of 0.65 indicates a lower estimated hazard of progression or death for alpelisib plus fulvestrant relative to placebo plus fulvestrant in the analyzed PIK3CA mutant cohort.
The word instantaneous is important. A hazard is not the same thing as a probability of experiencing an event over the entire study. The hazard ratio therefore should not be read as a direct conversion into a percentage of patients who will remain progression-free.
11. Analysis Population and the PIK3CA Mutant Cohort
The primary analysis was not defined over all 572 enrolled participants without qualification. The posted analysis population was specifically the group of subjects with a PIK3CA mutation based on central testing of hotspot mutations in tumor tissue who were randomized to study treatment.
| Population feature | Posted definition |
|---|---|
| Mutation status | PIK3CA mutation |
| Mutation assessment | Central testing of hotspot mutations in tumor tissue |
| Randomization requirement | Subjects had to be randomized to study treatment |
| Primary outcome | Progression-free survival |
| Assessment | Investigator assessment |
This distinction is statistically important. The enrollment number of 572 describes the trial as a whole, whereas the primary statistical estimate is explicitly tied to the defined PIK3CA mutant randomized cohort. The ClinicalTrials.gov record does not give the number of participants in that analysis population, so no cohort size is inferred here.
12. Safety Results
The ClinicalTrials.gov record reports serious adverse events by randomized treatment arm. These are presented as affected participants divided by participants at risk.
| Safety measure | Alpelisib + Fulvestrant | Placebo + Fulvestrant |
|---|---|---|
| Serious adverse events | 110 / 284 | 54 / 287 |
The affected/at-risk counts are reproduced exactly as reported in the registry. The page does not calculate or report derived percentages because the trial-data instructions restrict numerical reporting to the registry-reported values and do not provide a formal statistical analysis of serious adverse events.
13. Secondary Endpoints and Additional Analyses
ClinicalTrials.gov reports 14 outcome measures for the record, but the ClinicalTrials.gov record contains only one posted formal statistical analysis: the primary PFS analysis in the PIK3CA mutant cohort.
| Registry information | What is available in the ClinicalTrials.gov record |
|---|---|
| Outcome measures posted | 14 |
| Statistical analyses posted | 1 |
| Primary-endpoint analyses posted | 1 |
| Primary analyses with estimate + CI | 1 |
| Formal secondary-endpoint statistical estimates reported here | None |
Accordingly, this page does not invent secondary endpoint estimates, median PFS values, response rates, subgroup hazard ratios, or additional P-values. For an endpoint of the same general time-to-event type, a Kaplan-Meier estimate and an appropriate time-to-event comparison such as a log-rank test would ordinarily be relevant, but the ClinicalTrials.gov record does not establish a formal comparison for any additional endpoint.
14. Statistical Methods Explained
Why was a log-rank test used?
The primary endpoint is progression-free survival, a time-to-event outcome subject to censoring. The log-rank test is designed to compare the event-time distributions of randomized groups across follow-up rather than reducing the outcome to a single fixed time point. The registry identifies the log-rank test as the formal method for the posted primary analysis.
What does a hazard ratio of 0.65 mean?
It means that the estimated instantaneous rate of progression or death was 0.65 times the corresponding rate in the comparator group under the analysis framework. A simple relative interpretation is an estimated 35% lower hazard. It does not mean a 35% absolute reduction in the probability of progression or death.
Why does the confidence interval matter?
The point estimate alone does not communicate its statistical uncertainty. The reported 95% confidence interval of 0.50–0.85 shows the range generated by the specified confidence-interval procedure around the estimated hazard ratio. It is therefore more informative than reporting HR 0.65 without an interval.
Why doesn't the P-value measure effect size?
The P-value of 0.00065 quantifies statistical evidence against the null hypothesis under the specified testing framework. It is affected by the amount of information in the analysis and does not tell us how large the treatment effect is. The hazard ratio describes the relative effect, while the confidence interval describes uncertainty around that effect.
Why is censoring important?
Some participants can reach the end of available follow-up without experiencing progression or death. The registry states that such participants were censored at the date of their last adequate tumor assessment. Censoring allows those participants to contribute their observed event-free follow-up without treating the absence of an observed event as if it were an event.
Why does the analysis population matter?
The posted primary analysis is restricted to randomized participants with a PIK3CA mutation identified by central testing of hotspot mutations in tumor tissue. Therefore, the reported HR 0.65 is specifically an estimate for that defined analysis population, not automatically an estimate for every participant enrolled in the 572-person trial.
What does the superiority hypothesis mean?
The registry classifies the primary hypothesis as superiority. Conceptually, a superiority analysis asks whether the treatment groups differ in the specified endpoint rather than asking whether a new treatment is merely not unacceptably worse than a comparator. The ClinicalTrials.gov record reports the superiority analysis with a two-sided 95% confidence interval and P = 0.00065.
15. P-Values, Confidence Intervals, and Statistical Evidence
The three principal numerical components of the posted primary analysis answer different questions:
| Quantity | Reported value | Interpretive role |
|---|---|---|
| Hazard ratio | 0.65 | Relative effect estimate |
| 95% confidence interval | 0.50–0.85 | Uncertainty around the estimated effect |
| P-value | 0.00065 | Evidence against the null hypothesis under the specified test |
| Hypothesis type | Superiority | Tests for a treatment difference rather than a non-inferiority relationship |
A common statistical mistake is to treat a small P-value as if it were a measure of clinical importance. That is not what a P-value does. The P-value does not tell us the probability that the treatment works, nor does it quantify the magnitude of benefit.
Similarly, the confidence interval should not be treated as a collection of possible patient-level effects. It describes uncertainty about the population-level effect parameter estimated by the analysis.
16. Proportional-Hazards Considerations
A hazard ratio is a relative time-to-event effect measure. In many survival analyses, its interpretation is especially straightforward when the relative hazard is reasonably stable over time. When hazards vary substantially over time, a single hazard ratio can compress a more complicated pattern into one number.
The registry-reported SOLAR-1 registry data report a hazard ratio but do not provide enough information here to independently evaluate the proportional-hazards assumption or the shape of the underlying PFS curves. This is one reason not to interpret HR 0.65 as though it were a constant 35% reduction in an individual's probability of progression at every point in follow-up.
17. Multiplicity and Interim Analysis
The ClinicalTrials.gov record does not provide a multiplicity strategy, alpha-spending procedure, interim-analysis plan, or detailed sequential-testing framework. They identify the primary hypothesis as superiority and provide one formal primary analysis with a two-sided 95% confidence interval and P = 0.00065.
What is documented
A single posted formal statistical analysis for the registered primary endpoint, using a log-rank test and hazard ratio.
What is not documented in the ClinicalTrials.gov record
The ClinicalTrials.gov record does not specify an interim-analysis boundary, alpha-spending method, or formal multiplicity adjustment for other endpoints.
It would therefore be inappropriate to infer a particular multiplicity or interim-monitoring procedure from the P-value alone. Those design features must be taken from the protocol or statistical analysis plan if they are to be described formally.
18. Missing Data, Censoring, and Analysis Assumptions
The registered PFS definition provides one important piece of information about incomplete follow-up: participants without an event were censored at the date of their last adequate tumor assessment.
That is different from ordinary missing-data imputation. In a survival analysis, censoring preserves the observed follow-up time for a participant who has not experienced the event, rather than assigning that participant an invented event time.
| Issue | What the ClinicalTrials.gov record establishes |
|---|---|
| Event definition | First documented progression or death due to any cause |
| Assessment | Local radiology assessment according to RECIST 1.1 |
| No observed event | Censored at the date of last adequate tumor assessment |
| Distribution estimation | Kaplan-Meier methodology |
| Formal imputation method | Not specified in the ClinicalTrials.gov record |
| Censoring sensitivity analysis | Not specified in the ClinicalTrials.gov record |
The validity of a censored survival analysis depends on assumptions about how censoring relates to the event process. The ClinicalTrials.gov record does not provide sufficient detail to assess those assumptions empirically.
19. Bayesian Methods, Non-Inferiority, and Crossover
Several design features commonly encountered in clinical-trial statistics are not supported by the registry-reported SOLAR-1 data.
| Design topic | Supported by the ClinicalTrials.gov record? | Interpretation |
|---|---|---|
| Non-inferiority margin | No | The primary hypothesis is explicitly classified as superiority; no non-inferiority margin is reported. |
| Bayesian analysis | No | No Bayesian statistical method is reported in the ClinicalTrials.gov record. |
| Crossover | Not specified | The ClinicalTrials.gov record does not report a crossover procedure. |
| Interim analysis | Not specified | No interim boundary or alpha-spending strategy is reported. |
| Multiplicity adjustment | Not specified | No multiplicity procedure is posted on ClinicalTrials.gov for additional endpoints. |
These omissions are intentional. A statistical analysis page should distinguish between methods actually documented for the trial and methods that might commonly appear in similar studies.
20. Trial Timeline
Trial start
The ClinicalTrials.gov record lists July 23, 2015 as the trial start date.
Primary completion
The ClinicalTrials.gov record lists June 12, 2018 as the primary completion date.
Trial status
The registry status in the ClinicalTrials.gov record is COMPLETED.
The registry's primary endpoint time frame specifies an analysis once approximately 243 PFS events in the PIK3CA mutant cohort had been observed, up to 33.3 months. This is an event-driven description of the primary endpoint assessment rather than a statement that every participant was followed for exactly 33.3 months.
21. What the Primary Result Does — and Does Not — Mean
The estimated hazard of progression or death was lower in the alpelisib-plus-fulvestrant group than in the placebo-plus-fulvestrant group in the defined PIK3CA mutant analysis population, with a reported hazard ratio of 0.65.
It does not mean that 35% of patients were protected from progression, that the probability of progression was reduced by exactly 35 percentage points, or that every individual patient experienced the same relative reduction.
The 0.50–0.85 95% confidence interval communicates uncertainty around the point estimate. It is essential for understanding the precision of the reported hazard ratio.
The 0.00065 P-value provides evidence against the null hypothesis under the specified superiority testing framework. It does not measure the size of the effect.
22. Clinical Interpretation vs Statistical Interpretation
Statistical interpretation
The posted primary analysis found a hazard ratio of 0.65 for progression-free survival in the PIK3CA mutant cohort, with a two-sided 95% confidence interval of 0.50–0.85 and P = 0.00065 using a log-rank test.
Clinical interpretation
The statistical result describes a relative time-to-event difference between the randomized treatment groups. Clinical interpretation additionally requires consideration of absolute outcomes, safety, follow-up, patient characteristics, and the broader evidence base; those additional quantities are not all reported in the ClinicalTrials.gov record used for this page.
This distinction prevents a common error in trial reporting: treating a statistically strong result as if it automatically provides every piece of information needed for clinical interpretation. Statistical evidence and clinical context are related but distinct.
23. Important Limitations and Interpretation Issues
- Primary analysis population: the posted HR applies to randomized subjects with a PIK3CA mutation based on central testing of hotspot mutations in tumor tissue, not automatically to all 572 enrolled participants.
- Single formal analysis reported: the ClinicalTrials.gov record contains one statistical analysis, corresponding to the primary PFS endpoint.
- No median PFS reported: the ClinicalTrials.gov record do not report median PFS, so no median treatment difference is presented.
- No time-specific PFS rates reported: the registry data do not provide PFS probabilities at specific time points.
- No subgroup estimates reported: no additional subgroup hazard ratios or interaction tests are included in the ClinicalTrials.gov record.
- Hazard-ratio interpretation: a single HR compresses the time-to-event comparison into one relative measure and should not be treated as an individual-level probability.
- Censoring: participants without an event were censored at their last adequate tumor assessment; the ClinicalTrials.gov record does not provide a detailed assessment of censoring assumptions.
- Proportional hazards: the ClinicalTrials.gov record does not allow an independent assessment of whether a proportional-hazards interpretation is appropriate over the full follow-up period.
- Safety: serious adverse-event counts are reported, but no formal comparative statistical analysis of safety is provided.
- Multiplicity and interim monitoring: no detailed alpha-spending or multiplicity procedure is included in the ClinicalTrials.gov record.
24. Why This Trial Matters Statistically
SOLAR-1 is a useful teaching case because the registry record connects a clinically meaningful time-to-event endpoint with a clearly identified statistical analysis. The primary result illustrates how randomization, biomarker-defined analysis populations, Kaplan-Meier estimation, log-rank testing, hazard ratios, confidence intervals, P-values, and censoring fit together.
| Statistical concept | How it appears in SOLAR-1 |
|---|---|
| Randomization | The trial is registered as randomized with two parallel arms. |
| Triple masking | The registry classifies the study as triple-masked. |
| Time-to-event endpoint | PFS is defined from randomization to progression or death. |
| Kaplan-Meier estimation | The registry states that the PFS distribution was estimated using Kaplan-Meier methodology. |
| Log-rank testing | The posted primary analysis uses a log-rank test. |
| Hazard ratio | The primary effect measure is HR 0.65. |
| Confidence interval | The reported two-sided 95% CI is 0.50–0.85. |
| P-value | The reported P-value is 0.00065. |
| Superiority testing | The primary hypothesis type is superiority. |
| Censoring | Participants without an event were censored at the last adequate tumor assessment. |
| Biomarker-defined analysis | The primary analysis is restricted to the PIK3CA mutant cohort based on central tumor-tissue testing. |
| Safety by arm | Serious adverse events are reported as affected participants over participants at risk. |
25. Statistical Methods Explained: A Worked Reading of the Result
The primary result can be read sequentially rather than as a collection of disconnected numbers.
Step 1 · Define the event
PFS counts the time from randomization until the first documented progression or death from any cause.
Step 2 · Handle incomplete follow-up
If no event occurs, the registry states that PFS is censored at the last adequate tumor assessment.
Step 3 · Estimate the distribution
Kaplan-Meier methodology estimates the PFS distribution while accounting for the observed event and censoring times.
Step 4 · Compare groups
The log-rank test compares the time-to-event experience between alpelisib plus fulvestrant and placebo plus fulvestrant.
The resulting hazard ratio of 0.65 summarizes the relative treatment effect, while the 0.50–0.85 confidence interval describes uncertainty around that estimate and the 0.00065 P-value quantifies statistical evidence against the null under the specified superiority framework.
26. Related Tutorials
Learn more about the methods used in this trial:
27. Related Statistical Calculators
28. Sources
- ClinicalTrials.gov: NCT02437318 — SOLAR-1. Official registry record and source for the trial data summarized on this page.
- PubMed: PubMed record — PMID 39177931.
- PubMed: PubMed record — PMID 38439079.
- PubMed: PubMed record — PMID 33780274.
- PubMed: PubMed record — PMID 33246021.
- PubMed: PubMed record — PMID 32416251.
Continue through the Clinical Biostats statistical pathway
Use the related tutorials and calculators to move from the SOLAR-1 trial result to the underlying survival-analysis concepts.
29. Record Summary
SOLAR-1 provides a focused example of randomized time-to-event analysis in a biomarker-defined population. The trial enrolled 572 participants overall and used a randomized, parallel, triple-masked phase 3 design with two arms. Its registered primary endpoint was progression-free survival per investigator assessment in the PIK3CA mutant cohort, defined as time from randomization to first documented progression or death due to any cause, with censoring at the last adequate tumor assessment for participants without an event.
The registry states that the PFS distribution was estimated using Kaplan-Meier methodology and that the primary comparison used a log-rank test. The posted effect measure was a hazard ratio. The primary analysis reported HR 0.65, with a two-sided 95% CI of 0.50–0.85 and P = 0.00065 under a superiority hypothesis.
The most important statistical lesson is that these quantities should be interpreted together. The hazard ratio describes the relative time-to-event effect; the confidence interval describes uncertainty around that estimate; and the P-value addresses evidence against the null hypothesis. None of these quantities, by itself, supplies an absolute patient-level probability of benefit.