This page separates reported registry information from statistical interpretation. The numerical trial facts on this page are restricted to the ClinicalTrials.gov record data. No formal statistical-analysis estimates, confidence intervals, or p-values are reported in the ClinicalTrials.gov record.
1. Trial at a Glance
PARSIFAL was a completed randomized phase 2 trial in metastatic breast cancer. The study enrolled 486 participants and used a two-arm parallel design without masking. Its registered primary endpoint was 1-year progression-free survival, defined as the percentage of patients who are alive and without evidence of tumor progression using RECIST v1.1.
| Feature | PARSIFAL |
|---|---|
| Trial name | PARSIFAL |
| ClinicalTrials.gov identifier | NCT02491983 |
| Phase | Phase 2 |
| Status | COMPLETED |
| Therapeutic area | Oncology |
| Condition | Metastatic Breast Cancer |
| Brief title | Palbociclib in Combination With Fulvestrant or Letrozole in Patients With ER+, HER2- Advanced Breast Cancer |
| Allocation | RANDOMIZED |
| Design model | PARALLEL |
| Masking | NONE |
| Primary purpose | TREATMENT |
| Enrollment | 486.0 |
| Interventions | Palbociclib; Fulvestrant; Letrozole |
| Lead sponsor | MedSIR |
| Sponsor type | OTHER |
| Study dates | Start: 2015-08 · Primary completion: 2020-01 |
2. Clinical Question
The trial addressed the use of palbociclib in combination with endocrine therapy in patients with ER+, HER2- advanced breast cancer. The ClinicalTrials.gov record identifies two interventions used in the study—fulvestrant and letrozole—alongside palbociclib.
Population
Patients with ER+, HER2- advanced breast cancer in the condition category of metastatic breast cancer.
Interventions
Palbociclib, fulvestrant, and letrozole are the registered interventions.
Design
Randomized, parallel-group, two-arm phase 2 study without masking.
Primary question
The registered primary outcome evaluates 1-year progression-free survival.
The ClinicalTrials.gov record does not provide arm-specific enrollment numbers, treatment doses, treatment schedules, baseline characteristics, crossover information, or a formal comparison of fulvestrant versus letrozole. Those details are therefore not inferred on this page.
3. Trial Design
Study timeline
Study start
The ClinicalTrials.gov record lists the study start as 2015-08.
Primary completion
The ClinicalTrials.gov record lists primary completion as 2020-01.
Current registry status in the ClinicalTrials.gov record
The study status is reported as COMPLETED.
4. Endpoints
| Endpoint | Registry time frame | Type | Definition |
|---|---|---|---|
| 1-year Progression Free Survival | One year | Time-to-event | Percentage of patients who are alive and without evidence of tumor progression (defined using RECIST v1.1) |
The ClinicalTrials.gov record identifies 1-year Progression Free Survival as the single registered primary endpoint. The endpoint is expressed as a percentage at one year, while its underlying endpoint type is time-to-event.
5. Planned Analysis
The ClinicalTrials.gov record indicates that results were posted and that 6 outcome measures were posted, but the ClinicalTrials.gov record contains no formal statistical analyses. Consequently, there is no formal treatment-effect estimate, confidence interval, or p-value to reproduce for the primary endpoint.
Primary endpoint reported in the ClinicalTrials.gov record
Time frame: One year
Definition: Percentage of patients who are alive and without evidence of tumor progression, with tumor progression defined using RECIST v1.1.
How this endpoint would typically be analysed
For a time-to-event endpoint such as progression-free survival, the primary analysis would commonly begin with a Kaplan-Meier estimate of the probability of remaining progression-free through time. The one-year estimate can then be reported as the estimated progression-free survival probability at one year.
Here, di represents the number of events at time ti, and ni represents the number at risk immediately before that time. The estimate accounts for patients whose follow-up ends before an event is observed through right censoring.
If a randomized comparison were formally reported, a time-to-event analysis could also use a log-rank test to compare survival curves and a Cox proportional-hazards model to estimate a hazard ratio with a confidence interval. Those formal comparative results are not reported in the ClinicalTrials.gov record, so none are presented here as PARSIFAL results.
6. What the One-Year PFS Endpoint Means
The registered endpoint asks for the percentage of patients who are alive and without evidence of tumor progression at one year. This is an absolute time-specific measure. If a value were reported, it would describe the estimated proportion of the study population remaining alive and progression-free at that specified time point.
What it measures
The probability of being alive and without documented tumor progression at one year.
What it does not measure
It is not itself a hazard ratio, median PFS, overall survival estimate, or measure of tumor-response rate.
Why censoring matters
Time-to-event methods allow information from participants whose follow-up ends before progression or death to contribute up to their censoring time.
Why RECIST v1.1 matters
The registered definition specifies that tumor progression is defined using RECIST v1.1, providing a prespecified framework for determining progression.
7. Statistical Methodology
Randomization
Randomization is the central statistical design feature explicitly identified in the ClinicalTrials.gov record. By assigning participants randomly to study arms, the design seeks to balance measured and unmeasured baseline factors in expectation. This allows differences in subsequent outcomes to be interpreted in relation to randomized treatment assignment rather than simply differences in who entered each group.
Randomization does not guarantee that every baseline characteristic will be numerically identical between arms. Its statistical role is instead probabilistic: across repeated randomized trials, allocation is independent of baseline prognostic characteristics in expectation.
Time-to-event analysis
The registered primary endpoint is classified as time-to-event. Unlike a simple proportion observed at a fixed date, time-to-event analysis retains information about when an event occurs and can accommodate right censoring.
For progression-free survival, the event definition combines progression with death. The registry definition describes the one-year outcome in terms of patients who remain alive and without evidence of tumor progression.
Kaplan-Meier estimation
A Kaplan-Meier curve estimates the probability of remaining event-free over time. The one-year PFS endpoint can be obtained from the estimated survival function at one year, provided the underlying event and censoring information are available.
Log-rank testing
A log-rank test is a standard method for comparing two time-to-event distributions in a randomized trial. It evaluates whether the observed event patterns over follow-up are compatible with a common survival experience under the null hypothesis. The registry-reported PARSIFAL data do not report a log-rank statistic or p-value.
Cox proportional-hazards model
A Cox proportional-hazards model is commonly used to summarize a treatment comparison through a hazard ratio. A hazard ratio below 1 would indicate a lower estimated instantaneous event rate in one group relative to the other under the model. The registry-reported PARSIFAL data do not report a Cox model estimate.
A hazard ratio is not the same as a difference in one-year percentages. It is a relative time-to-event measure derived from a model and depends on the event and censoring information used in the analysis.
8. Statistical Methods Explained
Why was randomization important in PARSIFAL?
Randomization provides the statistical foundation for comparing the two study arms. Because treatment assignment is randomized rather than selected according to patient characteristics, differences in outcomes can be attributed to the randomized comparison with substantially less concern that baseline prognostic factors determined treatment assignment.
Why is 1-year PFS a time-to-event endpoint?
Progression-free survival contains a time dimension: participants can progress at different points during follow-up, and some may not have experienced progression or death by the time their observation ends. Treating the endpoint as time-to-event preserves that information instead of reducing every participant to a simple yes/no observation without regard to follow-up time.
What does the one-year percentage represent?
The registered endpoint is the percentage of patients estimated to be alive and without evidence of tumor progression at one year. It is an absolute time-specific quantity. It should not be interpreted as a hazard ratio or as the percentage improvement produced by treatment.
Why would Kaplan-Meier methods be appropriate?
Kaplan-Meier estimation is designed for time-to-event data with right censoring. It allows participants with incomplete event follow-up to contribute information through the time they were observed rather than requiring every participant to have a documented event.
Why might a Cox model be used in addition to Kaplan-Meier estimation?
Kaplan-Meier methods describe the event-free probability over time, whereas a Cox model can summarize the relative event rate between randomized groups through a hazard ratio. The two approaches answer related but different statistical questions.
Why should a missing hazard ratio not be reconstructed from the one-year endpoint?
A single time-point percentage does not contain enough information to recover the full pattern of event times and censoring. Different survival curves can produce the same one-year estimate while having different hazards and different hazard ratios. A formal hazard ratio therefore requires the underlying time-to-event information or an independently reported analysis.
Why should the absence of a p-value be respected?
A p-value is an output of a specified statistical comparison. The registry-reported PARSIFAL data contain no formal statistical analyses, so this page does not assign a p-value to the primary endpoint.
9. Analysis Population and Missing Information
The ClinicalTrials.gov record identifies the overall enrollment as 486.0, but they do not specify an analysis population definition, arm-specific sample sizes, exclusions, or a separate safety analysis population.
| Analysis detail | Information in the ClinicalTrials.gov record |
|---|---|
| Total enrollment | 486.0 |
| Number of arms | 2 |
| Arm-specific enrollment | Not reported in the ClinicalTrials.gov record |
| ITT population definition | Not reported in the ClinicalTrials.gov record |
| Per-protocol population | Not reported in the ClinicalTrials.gov record |
| Safety analysis population | Not separately defined in the ClinicalTrials.gov record |
| Missing-data/imputation method | Not reported in the ClinicalTrials.gov record |
| Stratification factors | Not reported in the ClinicalTrials.gov record |
These omissions matter because the interpretation of a clinical-trial estimate depends not only on the endpoint but also on who was included in the analysis, how censoring was handled, and how protocol deviations and missing assessments were treated.
10. Safety
The ClinicalTrials.gov record provides serious adverse events by randomized arm as affected participants divided by participants at risk.
| Safety measure | Affected / at risk | Derived proportion |
|---|---|---|
| Arm A serious adverse events | 51/243 | 51/243 |
| Arm B serious adverse events | 72/243 | 72/243 |
The affected/at-risk values are reported exactly as reported in the registry. No percentage has been calculated because the trial-data rule for this page is to report numbers exactly as given rather than recompute or round them.
The serious-adverse-event data provide an arm-specific count of affected participants and the corresponding number at risk. They do not, by themselves, establish a causal difference in safety between the arms. Formal safety comparisons would require the prespecified safety population, endpoint definitions, follow-up, and an appropriate statistical framework.
The fact that both arms have an at-risk denominator of 243 does not establish the full treatment allocation or explain why the registry-reported overall enrollment is 486.0; it simply describes the denominators provided in the safety record.
11. Overall Survival: Registry Caveat
The ClinicalTrials.gov record contains an explicit limitation concerning Overall Survival. The number of deaths was not sufficient to determine the secondary objective of calculating Overall Survival, so this value could not be obtained.
This is an important statistical distinction. Overall survival is another time-to-event endpoint, but its analysis requires sufficient observed deaths to estimate the survival distribution with useful precision. A lack of sufficient events limits what can be estimated, regardless of the total number enrolled.
12. Design Topics Supported by the Trial Data
| Design topic | What the ClinicalTrials.gov record establishes |
|---|---|
| Randomization | Yes. Allocation is reported as RANDOMIZED. |
| Parallel design | Yes. Design model is PARALLEL. |
| Masking | No masking is reported. |
| Two-arm structure | Yes. The trial has 2 arms. |
| Non-inferiority margin | Not reported in the ClinicalTrials.gov record. |
| Crossover | Not reported in the ClinicalTrials.gov record. |
| Factorial design | Not reported; the registry-reported design model is PARALLEL. |
| Multiplicity adjustment | Not reported in the ClinicalTrials.gov record. |
| Interim analysis | Not reported in the ClinicalTrials.gov record. |
| Missing-data or imputation method | Not reported in the ClinicalTrials.gov record. |
| Stratification | Not reported in the ClinicalTrials.gov record. |
| Bayesian methods | Not reported in the ClinicalTrials.gov record. |
The table deliberately distinguishes documented design features from methods that cannot be established from the ClinicalTrials.gov record. In particular, the absence of a reported method is not evidence that investigators did not use one; it means the method is not available in the ClinicalTrials.gov record.
13. Interpretation of Time-to-Event Evidence
The registered primary endpoint is a one-year time-specific measure of progression-free survival. Because the ClinicalTrials.gov record does not contain a formal statistical analysis, there is no reported treatment-effect estimate to interpret quantitatively.
A complete comparative analysis would ordinarily identify the analysis population, number and timing of progression and death events, censoring rules, estimated one-year PFS in each arm, uncertainty around those estimates, and—if prespecified—a comparative time-to-event model or hypothesis test.
Precision in time-to-event analyses is driven substantially by the number of observed events. The registry's explicit Overall Survival caveat illustrates this principle: enrollment alone does not guarantee that enough events have occurred to support a precise survival estimate.
14. Results Reporting Status
the ClinicalTrials.gov record states that results were posted and that 6 outcome measures were posted. However, no formal statistical analyses were posted and contains no formal analysis records.
| Reporting element | Supplied record |
|---|---|
| Results posted | Yes |
| Outcome measures posted | 6 |
| Formal statistical analyses reported | None |
| Primary endpoint with registry definition reported | 1-year Progression Free Survival |
| Formal effect estimate reported | No |
| Confidence interval reported | No |
| P-value reported | No |
This reporting structure is why the page uses a Planned Analysis framework rather than presenting a conventional numerical Results section. It preserves the distinction between the existence of posted outcome measures and the availability of formal comparative statistical estimates.
15. Why This Trial Matters Statistically
PARSIFAL is useful as a statistical teaching case because its the ClinicalTrials.gov record combines a randomized two-arm design with a time-to-event primary endpoint and a registry-level limitation concerning the availability of an Overall Survival estimate. That combination highlights several principles that recur throughout clinical-trial analysis.
| Concept | How it appears in PARSIFAL |
|---|---|
| Randomization | The allocation is explicitly reported as RANDOMIZED. |
| Parallel-group design | The design model is PARALLEL with 2 arms. |
| Time-to-event endpoint | The registered primary endpoint is classified as time-to-event. |
| Fixed-time survival estimate | The primary endpoint is 1-year Progression Free Survival. |
| RECIST v1.1 | Tumor progression is defined using RECIST v1.1. |
| Censoring | Time-to-event methodology ordinarily requires explicit rules for participants without an observed event by the end of usable follow-up. |
| Event-driven precision | The Overall Survival caveat illustrates the importance of having sufficient observed deaths for an OS analysis. |
| Safety denominators | Serious adverse events are reported as affected participants over participants at risk in each arm. |
| Registry interpretation | Posted outcome measures do not necessarily provide a complete formal statistical-analysis record. |
16. Limitations
- No formal statistical analyses reported: no formal statistical analyses were posted, so treatment-effect estimates, confidence intervals, and p-values cannot be reported from the ClinicalTrials.gov record.
- Arm-specific enrollment is not reported: the overall enrollment is 486.0, but the ClinicalTrials.gov record does not identify the number randomized to each arm.
- Limited endpoint detail: only the registered primary endpoint and its definition are reported in detail. The record states that 6 outcome measures were posted but does not provide the other five outcome definitions in the ClinicalTrials.gov record.
- Overall Survival: the registry explicitly states that the number of deaths was not sufficient to determine the secondary objective of calculating Overall Survival.
- Analysis population: the ClinicalTrials.gov record does not define ITT, per-protocol, or other efficacy populations.
- Statistical procedures: stratification, multiplicity adjustment, interim analysis, missing-data methods, crossover, and Bayesian methods are not reported in the ClinicalTrials.gov record.
- Safety interpretation: serious-adverse-event counts are reported, but the data do not provide a formal statistical comparison or additional safety definitions.
- External validity: the ClinicalTrials.gov record identifies the condition and trial design but does not provide baseline characteristics needed to assess representativeness in detail.
17. Statistical Concepts in This Trial
Learn more about the methods used in this trial:
18. Related Statistical Calculators
19. Sources
- ClinicalTrials.gov: PARSIFAL, NCT02491983.
- PubMed: PMID 36239405.
- PubMed: PMID 34617955.
Continue through the Clinical Biostats statistical library
Explore tutorials and statistical calculators connected to randomized clinical-trial methods and study design.
20. Record Summary
PARSIFAL is a completed randomized phase 2 trial with 486.0 enrolled participants, 2 arms, a PARALLEL design, and no masking. Its registered primary endpoint is 1-year Progression Free Survival, defined as the percentage of patients who are alive and without evidence of tumor progression, with progression defined using RECIST v1.1.
The ClinicalTrials.gov record states that results were posted and that 6 outcome measures were posted, but it contains no formal statistical-analysis records. The registry's explicit Overall Survival caveat further illustrates the importance of sufficient event counts when evaluating time-to-event outcomes.