PD0332991 (Palbociclib) in Patients With Advanced or Metastatic Liposarcoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for PD0332991 (Palbociclib) in Patients With Advanced or Metastatic Liposarcoma (NCT01209598), a randomized phase 2 study. The primary outcome is Progression Free Survival at 12 Weeks.

2. Clinical Question

3. Trial Design

This is a randomized, open-label, parallel-arm phase II study sponsored by Memorial Sloan Kettering Cancer Center. The registry record lists an enrollment of 90 and two palbociclib arms, one at 200mg and the other at 125mg. The record gives the arm sizes as 30 and 60, respectively. Both arms are described as evaluating palbociclib in patients with advanced or metastatic liposarcoma.

Randomized allocation means participants were assigned to study arms by a random process, according to the registry’s design classification. The record does not describe the allocation procedure, whether allocation was concealed, or how assignment was implemented. Masking is listed as none, so the study is classified as unmasked. The registry does not report whether endpoint assessment was conducted by assessors unaware of treatment assignment.

The arm descriptions state that a one-stage design was used to assess whether the 12-week PFS rate reached at least 40%. The registry does not report the design’s detailed statistical boundaries, the rationale for selecting the target rate, or a sample-size justification. Enrollment is a registry-level figure; it should not be treated as a substitute for the arm-specific numbers contributing to a particular endpoint.

The registry record lists a start date of 2010-09-23, primary completion on 2016-10-25, and study status as completed. Those dates describe the registered study timeline. They do not, by themselves, indicate when individual participants were followed or establish how long outcome data were available for each participant.

The record identifies Best Response as a secondary outcome. It does not provide a posted analysis entry for a comparison between the arms, and its comparison note states that values were missing or not estimable for a two-arm comparison for the other endpoint. Thus, the randomized design does not imply that a treatment contrast is available in the posted results.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Progression Free Survival at 12 Weeks, measured over 12 weeks.

The primary endpoint is titled “Progression Free Survival at 12 Weeks.” Its timeframe is 12 weeks, and its unit is a percentage of participants. The registry defines PFS for this endpoint as RECIST 1.1 (CR + PR + SD) when treated with palbociclib. In that definition, the listed categories are complete response, partial response, and stable disease. The registry’s wording therefore describes the proportion meeting this specified response-or-stability definition at the assessment time.

A fixed-time percentage can be useful because it communicates the share of participants meeting an outcome definition at a common landmark. It is different from a description of how long each participant remained free from progression over follow-up. The record does not provide the detailed endpoint-assessment schedule, the handling of missing assessments, or a time-to-event analysis method. It is therefore important not to assume that the posted percentages represent a particular survival-curve estimate or a conventional time-to-event calculation.

The arm descriptions identify at least 40% at 12 weeks as the design target. That target is part of the stated design objective; it does not replace the reported estimates or their confidence intervals. The registry reports estimates of 66.0% and 57.2%, each with a 95% confidence interval. The width and placement of those intervals describe uncertainty around each arm’s estimate, not the probability that an individual participant will meet the endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical interpretation depends on distinguishing an estimate within each arm from a comparison between arms. The registry reports a percentage and a 95% confidence interval separately for each palbociclib regimen. Those quantities describe the endpoint in each group. A randomized design can support comparisons between assigned groups, but randomization alone does not create a reported comparison when no contrast or test is posted.

What does a confidence interval describe?

In general, a confidence interval expresses the uncertainty associated with an estimate under a specified statistical procedure. A narrower interval indicates greater precision under that procedure; a wider interval indicates less precision. The interval is not a guarantee that the true value lies within its endpoints, and it is not a range of likely outcomes for individual participants. Here, the record reports 95% confidence intervals for both arm-specific estimates, but does not name the method used to calculate them. That missing detail limits independent assessment of the interval procedure.

For Palbociclib 200mg, the reported estimate is 66.0%, with interval limits of 51.0% and 100.0%. For Palbociclib 125mg, the estimate is 57.2%, with limits of 42.4% and 68.8%. These values should be read as separate summaries. The interval limits are not a confidence interval for the difference between doses. The registry does not report such a difference or its uncertainty.

Why is an arm-level result not a dose comparison?

A between-arm analysis estimates a contrast, such as a difference in endpoint rates, and quantifies uncertainty around that contrast. The result may depend on the chosen statistical model and on the endpoint’s analysis population. A visual or numerical comparison of two separate estimates does not supply the contrast’s standard error, confidence interval, or test. The record identifies the 200mg arm as its control arm in the arm information, but both listed interventions are palbociclib regimens. The designation does not change the fact that the registry does not post a formal dose comparison.

The record reports no analysis entries and notes that values were missing or not estimable for a two-arm comparison for the other endpoint. This means a reader cannot use the posted record to assess a reported p-value, adjusted treatment contrast, or prespecified comparative decision. It does not establish why those values are absent.

How should the 12-week rate be understood?

In general, a fixed-time rate summarizes the proportion meeting a defined outcome at a specified assessment time. The registry’s definition combines RECIST 1.1 categories CR, PR, and SD. This outcome should be interpreted according to that recorded definition. Although its title uses “Progression Free Survival,” the record does not provide details of a time-to-event analysis, such as how event times or censoring were handled. A reader should not infer those methods from the endpoint title alone.

The arm descriptions state that the design objective was to determine whether the 12-week PFS rate reached at least 40%. A target rate is a benchmark for the design question, not an estimate of the outcome. The observed registry values and confidence intervals are distinct from the target. The registry does not report a formal decision rule or a statement of whether its one-stage design met a prespecified success criterion.

What can planning calculations tell a reader?

General sample-size calculations illustrate how many participants might be needed under stated assumptions to detect an effect with a chosen power and significance level. They are planning exercises, not descriptions of the trial’s actual design unless the record says they were used to plan it. The planning assumptions associated with the site’s calculations specify a two-sided significance level of 0.05, equal allocation, a normal approximation, and effects expressed as fractions of the observed effect. The record explicitly distinguishes these calculations from the trial’s own design. They should not be described as observed power or as evidence that the trial was adequately sized.

The design itself is described as one-stage, but its sample-size justification, operating characteristics, and decision boundaries are not reported in the registry record. No conclusion about the design’s planned precision or ability to detect a particular dose difference can be drawn from the posted endpoint estimates alone.

Why do missing analysis details matter?

In general, interpretation depends on knowing which participants were included, how missing outcomes were handled, and whether the analysis followed a prespecified plan. Different choices can affect an estimate, especially when outcome availability is related to participant status. The registry record does not specify an analysis population, missing-data method, or analysis plan. It also does not report whether analyses were adjusted for other variables. These omissions limit evaluation of the assumptions behind the posted estimates, even though the values and confidence intervals are available.

Statistical methods are most informative when the outcome definition, analysis population, estimation procedure, and uncertainty method are described together. The registry provides the endpoint definition and arm-specific results, but not all of those methodological details. The estimates can be described faithfully while recognizing that the record does not support a complete reconstruction of the analysis.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists Best Response as a secondary outcome and reports a primary endpoint of 12-week PFS. It does not report a multiplicity procedure or explain how statistical error was controlled across outcomes, analyses, or possible comparisons. Multiplicity is a general concern when several questions are tested: considering more tests can increase the chance of finding an apparently notable result by chance. Without a stated plan, it is not possible to determine whether or how that issue was addressed in this study.

The record lists no analysis entries. It does not provide a detailed statistical analysis plan, identify a primary analysis population, or state whether the posted estimates were based on all assigned participants or another group. The population field for the primary outcome is blank. These are limitations in the registry record; they do not establish that the investigators did not define such procedures elsewhere.

Missing outcome information and missing registry information are different issues. The record does not explain how participants without an evaluable 12-week assessment were handled, whether any outcome values were imputed, or how censoring was treated. It is therefore not possible to assess whether the endpoint estimates could be affected by missing assessments or to reproduce the analysis from the registry record.

The study is classified as randomized, but the record does not describe the allocation process or report an adjusted analysis. It also does not report an inferential comparison between the two arms. As a result, the posted arm-specific results should not be treated as a complete analysis of the randomized dose comparison. The registry’s note that comparison values were missing or not estimable applies to the other endpoint and does not provide a substitute for an analysis of the primary endpoint.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: neutrophil count decreased (50.0%), white blood cell decreased (46.7%), lymphocyte count decreased (31.7%), platelet count decreased (30.0%), anemia (21.7%).

The registry reports serious adverse event counts by arm: 24 of 30 participants in the Palbociclib 200mg arm and 55 of 60 in the Palbociclib 125mg arm are listed as affected. For other adverse events, it lists 30 of 30 and 60 of 60 participants as affected, respectively. These are registry counts. The record does not provide enough detail to determine the event-level counting rules, observation period, or whether multiple events in one participant were counted separately.

The listed serious adverse event terms include neutrophil count decreased at 50.0%, white blood cell decreased at 46.7%, lymphocyte count decreased at 31.7%, platelet count decreased at 30.0%, anemia at 21.7%, and fatigue at 6.7%. The registry does not clarify the denominator or detailed coding and grading basis for each displayed percentage. These percentages should therefore be read as reported labels, not as a complete safety profile.

The deaths field is null for both arms. A null value is not evidence that no deaths occurred; the registry record does not report a death count. The adverse event information also does not establish that an event was caused by palbociclib. These figures describe what the registry records and should not be used alone to compare the safety of the regimens.

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

The posted results describe the 12-week endpoint estimates reported for two palbociclib regimens in this phase II study of advanced or metastatic liposarcoma. The estimate for Palbociclib 200mg is 66.0%, and the estimate for Palbociclib 125mg is 57.2%. Each has a reported 95% confidence interval. Those summaries allow readers to see the arm-specific values and their stated uncertainty, while the registry’s endpoint definition clarifies what outcome was counted.

The results do not establish that one dose is superior to the other. No formal comparison, effect estimate for a contrast, or test is posted. The separate confidence intervals do not provide that missing comparison. The record also does not report the prespecified decision rule for the one-stage design, so the posted values alone cannot establish whether a formal design criterion was met.

The results do not show how the estimates would change under a different endpoint definition, analysis population, or handling of missing assessments. The registry does not report those analysis details. Nor do the percentages describe the duration of benefit for an individual or imply that every participant had the same outcome. They summarize a group-level endpoint at a stated assessment time.

Safety figures are also limited to the registry’s recorded adverse event counts and listed terms. The record does not give a death count, a complete account of event severity and timing, or a causal assessment. Neither the efficacy endpoint nor the safety summaries should be interpreted as individualized medical guidance. The registry findings are a description of this study’s posted results, not a guarantee about outcomes in other populations or settings.

13. Why This Trial Matters Statistically

This study is statistically notable for pairing a randomized, parallel-arm design with registry results reported as separate estimates for two active palbociclib regimens. That combination makes it important to distinguish the existence of randomization from the availability of a comparative analysis. Random assignment can be part of a design intended to support a fair comparison, but the registry still needs to report the contrast and its uncertainty for readers to evaluate that comparison.

The primary endpoint is expressed as a percentage at a fixed assessment time, and its definition combines response and stable disease categories under RECIST 1.1. That makes careful attention to endpoint wording essential. The title uses progression-free survival, while the description defines the measure through CR, PR, and SD at 12 weeks. The record does not give further methods that would resolve whether a time-to-event analysis was also performed.

The reported confidence intervals encourage a precision-focused reading rather than a simple ranking of point estimates. Each arm has its own interval, and the registry does not provide an interval for the difference between arms. Recognizing that distinction is central to avoiding unsupported conclusions from arm-level numbers.

The record also illustrates why transparent reporting of analysis populations, missing-data procedures, multiplicity, and decision rules matters. These details are not reported in the registry record, making it harder to assess how the design objective relates to the estimates. The result can still be described accurately, but the boundaries of what can be concluded are determined by the information recorded.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record leaves several trial-specific uncertainties unresolved. It identifies the study as randomized, but does not describe the allocation process, concealment, or implementation. It is classified as unmasked, and the record does not state whether outcome assessments were independently reviewed or conducted without knowledge of assigned regimen.

The primary endpoint’s description gives a RECIST 1.1-based definition using CR, PR, and SD at 12 weeks, but does not provide assessment procedures, rules for equivocal or unavailable assessments, or the handling of participants who did not have an evaluable result at that time. Despite the endpoint title, the record does not report time-to-event analysis details, including event and censoring rules. It also leaves the primary outcome population field blank.

Although the arm descriptions state a one-stage design and a target of at least 40%, the registry does not report the statistical decision boundary, sample-size rationale, or formal success criterion. No between-arm estimate or test is posted, and no analysis entries are listed. The record does not explain why a two-arm comparison for the other endpoint was missing or not estimable.

The safety information is also incomplete for interpretation. The record provides affected counts and selected serious adverse event terms, but does not specify all event definitions, grading details, exposure duration, or causal assessments. A null deaths field does not provide a reported death count. These limitations concern what can be established from the registry record and do not establish what may have been documented elsewhere.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: PD0332991 (Palbociclib) in Patients With Advanced or Metastatic Liposarcoma — NCT01209598.
• Registry record last updated 2017-11-06; status last verified 2017-02.
• Results first posted 2017-11-06.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

The registry record describes a completed, randomized, unmasked, parallel-arm phase II study of palbociclib in advanced or metastatic liposarcoma. It lists an enrollment of 90 and reports two arms: Palbociclib 200mg and Palbociclib 125mg. The primary endpoint is titled Progression Free Survival at 12 Weeks and is defined using RECIST 1.1 categories CR, PR, and SD.

The reported estimates are 66.0% for the 200mg arm, with a 95% confidence interval of 51.0% to 100.0%, and 57.2% for the 125mg arm, with a 95% confidence interval of 42.4% to 68.8%. The arm descriptions state a target of at least 40% at 12 weeks. The registry does not post a formal comparison between regimens, an analysis entry, or detailed methods for the endpoint analysis.

Best Response is listed as a secondary outcome. Serious and other adverse event counts are reported by arm, alongside selected serious adverse event terms. Important details about analysis populations, missing data, multiplicity, and safety assessment are not reported in the registry record.