A Study Of Palbociclib (PD-0332991) + Letrozole VS. Placebo+ Letrozole For 1st Line Treatment Of Asian Postmenopausal Women With ER+/HER2- Advanced Breast Cancer [PALOMA-4]: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Of Palbociclib (PD-0332991) + Letrozole VS. Placebo+ Letrozole For 1st Line Treatment Of Asian Postmenopausal Women With ER+/HER2- Advanced Breast Cancer [PALOMA-4] (NCT02297438), a randomized phase 3 study. The primary outcome is Progression-Free Survival (PFS) Based on Investigator's Assessment: Up to Primary Completion Date.

2. Clinical Question

The registry describes a multicenter, randomized, double-blind phase 3 study with quadruple masking and a parallel-group model. It was designed to compare the combination of palbociclib and letrozole with placebo and letrozole. The question is specific to the recorded population and endpoint; it does not address other breast cancer populations, treatment combinations, or outcomes not reported in the registry record.

3. Trial Design

PALOMA-4 was a phase 3, randomized, parallel-group study. The registry lists quadruple masking and a planned enrollment of 340. The study status is completed. The registry record gives a start date of 2015-03-23 and a primary completion date of 2020-08-31. It describes the target population as Asian postmenopausal women with previously untreated ER-positive, HER2-negative advanced breast cancer; the brief summary specifies no prior systemic anticancer therapies for advanced or metastatic disease.

Participants were assigned to palbociclib plus letrozole or placebo plus letrozole. Palbociclib and placebo were scheduled orally once daily on Day 1 to Day 21 of each 28-day cycle, followed by 7 days off treatment. Letrozole was scheduled orally once daily continuously. The registry records 169 participants in the palbociclib-plus-letrozole group and 171 in the placebo-plus-letrozole group for the primary endpoint. These are the group counts reported for that analysis, rather than a statement about how many participants completed treatment or follow-up.

The primary outcome was investigator-assessed progression-free survival, measured from randomization through the first documented progression or death in the absence of documented progression. Its timeframe is described as randomization up to 65 months. The registry also lists progression-free survival assessed by blinded independent central review and several response-related outcomes. The original protocol's detailed allocation procedure, stratification factors, treatment discontinuation rules, follow-up schedule, and original sample-size justification are not reported in the registry record. The randomized, masked design helps define the comparison, but the registry entries alone do not establish how every design feature was implemented.

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.

Before assignment: This was a multicenter, randomized (1:1), double blind, placebo controlled, parallel group Phase 3 trial comparing the efficacy and safety of palbociclib in combination with letrozole versus placebo plus letrozole in Asian postmenopausal women with estrogen receptor positive/human epidermal growth factor receptor 2 negative advanced breast cancer.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: This was intent-to-treat (ITT) population, including all participants who were randomized, with study drug assignment designated according to initial randomization, regardless of whether participants received study drug or received a different drug from that to which they were randomized.

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 (PFS) Based on Investigator's Assessment: Up to Primary Completion Date, measured over Randomization up to 65 months.

The primary endpoint, investigator-assessed progression-free survival, measures time from randomization to the first documentation of objective disease progression or death from any cause when progression has not been documented. The registry defines progression using RECIST v1.1: a 20% increase in the sum of diameters of target measurable lesions above the smallest sum observed, with a minimum absolute increase of 5 mm. If the sum does not decrease during therapy, the baseline sum is the reference. The recorded endpoint timeframe is randomization up to 65 months.

Progression-free survival is a time-to-event outcome: it accounts for both whether an event occurs and how long it takes to occur. In general, this can provide a way to compare the timing of progression or death between randomized groups, rather than only comparing the proportion with an event at one fixed time. The median is the time by which the estimated event-free proportion has reached one half, based on the Kaplan-Meier estimate. It is not an estimate of how long every participant lived without progression.

The endpoint is not overall survival: the event combines disease progression with death, and the registry does not report an overall-survival result in the registry record described for this analysis. Progression was assessed by investigators for the primary outcome. The registry also lists a secondary progression-free survival outcome based on blinded independent central review; it is a separate assessment and should not be conflated with the primary investigator assessment.

7. Results Posted in the Registry

• Palbociclib + Letrozole vs Placebo + Letrozole: hazard ratio 0.68 (95% CI 0.53–0.87), reported p = 0.001 (as reported in the registry).

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Progression-Free Survival (PFS) Based on Blinded Independent Central Review (BICR): Up to Primary Completion Date

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor (Log Rank); it was not recalculated.
• Result:
  • Palbociclib + Letrozole vs Placebo + Letrozole: hazard ratio 0.86 (95% CI 0.65–1.14), reported p = 0.148 (as reported in the registry).
• Interpretation:
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Palbociclib + Letrozole vs Placebo + Letrozole and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants With Objective Response (OR) Based on Investigator Assessment: Up to Primary Completion Date

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Palbociclib + Letrozole: 37.3% (63/169) versus 31.6% (54/171) on control; risk difference 5.7 percentage points (95% CI -4.4 to 15.8), risk ratio 1.18 (0.88–1.59); unadjusted two-proportion z-test p = 0.269.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Palbociclib + Letrozole and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants With Objective Response (OR) Based on Investigator Assessment (Participants With Measurable Disease at Baseline): Up to Primary Completion Date

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Palbociclib + Letrozole: 43.4% (63/145) versus 38.0% (54/142) on control; risk difference 5.4 percentage points (95% CI -5.9 to 16.8), risk ratio 1.14 (0.86–1.51); unadjusted two-proportion z-test p = 0.350.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Palbociclib + Letrozole and the control. The interval gives the range of effects that remain compatible with the data.

The secondary results cover a different range of questions from the primary investigator-assessed progression-free survival endpoint. The central-review progression-free survival analysis points in the same general direction as the primary hazard-ratio estimate, but its confidence interval includes 1 and its reported p-value offers weaker evidence of a between-group difference. This makes the central-review result less conclusive than the primary result; it does not establish that the two assessment methods are equivalent.

The posted objective-response results also have estimates in the direction of a higher response proportion in the palbociclib-plus-letrozole group. Their sponsor-reported comparisons have confidence intervals that include no difference and p-values that do not provide strong statistical evidence for a group difference. The analysis restricted to participants with measurable disease at baseline addresses a narrower population than the broader response analysis. These are response outcomes, not substitutes for progression-free survival.

Secondary outcomes can help describe whether findings are consistent across related measures, but they carry less weight when many outcomes are examined and the registry record does not report a multiplicity-control plan. Without such a plan, nominal p-values should not be treated as if each were the sole prespecified test. The registry lists additional outcome measures, but numerical results for the other outcomes are not reported in the registry record in the registry record used for this interpretation. The overall secondary pattern is therefore limited to the results and does not support broad conclusions across every listed outcome.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Progression-Free Survival (PFS) Based on Blinded Independent Central Review (BICR): Up to Primary Completion Date
• Percentage of Participants With Objective Response (OR) Based on Investigator Assessment: Up to Primary Completion Date
• Percentage of Participants With Objective Response (OR) Based on Investigator Assessment (Participants With Measurable Disease at Baseline): Up to Primary Completion Date
• Percentage of Participants With Objective Response (OR) Based on Blinded Independent Central Review (BICR): Up to Primary Completion Date
• Percentage of Participants With Objective Response (OR) Based on Blinded Independent Central Review (BICR) (Participants With Measurable Disease at Baseline): Up to Primary Completion Date
• Duration of Response (DOR) Based on Investigator Assessment (Participants With Objective Disease Response): Up to Primary Completion Date
• Duration of Response (DOR) Based on Blinded Independent Central Review (BICR) (Participants With Objective Disease Response): Up to Primary Completion Date
• Percentage of Participants With Disease Control/Clinical Benefit Response (DC/CBR) Based on Investigator Assessment: Up to Primary Completion Date

9. Statistical Interpretation

10. Statistical Methods Explained

What does the Kaplan-Meier median describe?

The registry says progression-free survival was estimated using the Kaplan-Meier method. In general, this method estimates the proportion of participants who have not yet experienced the defined event over time, while allowing for participants whose event time is not observed during their available follow-up. Such observations are commonly called censored. The registry entries do not describe the censoring rules, reasons for censoring, or checks of assumptions, so those details should not be inferred.

The median is the time at which the estimated event-free proportion reaches one half. It gives a point on the estimated time-to-event distribution, but it does not describe the full curve or the experience of every participant. The two medians alone do not convey the full comparison, which is why the sponsor-reported hazard ratio and its confidence interval are also important.

How should the hazard ratio be read?

The primary comparison reports a hazard ratio from a Log Rank analysis. In general, a hazard ratio compares the event rate at a given time between groups, within the statistical model and its assumptions. A value below 1 indicates a lower estimated event rate in the first-named group relative to the comparator; it is not a ratio of median times and does not directly state the probability that an individual participant will experience an event. The registry reports the hazard ratio and the Log Rank method, but the detailed model specification, covariate adjustment, stratification, and checks of the proportional-hazards assumption are not reported in the registry record.

The Log Rank test evaluates evidence against the null hypothesis that the time-to-event distributions do not differ between groups. Its p-value is not the probability that the null hypothesis is true, nor is it a measure of the size or clinical importance of an effect. Interpretation should combine the estimate, uncertainty interval, endpoint definition, and design rather than rely on a threshold alone.

What do the response outcomes measure?

Objective response is a binary outcome: participants are categorized according to whether they met the response definition during the stated assessment period. The registry reports response percentages and sponsor analyses using Fisher Exact tests and odds ratios. Fisher's exact test is a method for assessing association in a two-group categorical comparison. An odds ratio compares the odds of response between groups; it is not the same quantity as a risk difference or a ratio of response percentages.

The posted unadjusted recalculations for response use the group summaries and are distinct from the sponsor's reported Fisher Exact analysis. They are not adjusted for other factors. The registry record does not report whether the sponsor's response analysis used additional prespecified adjustments or how missing response assessments were handled. The outcome restricted to participants with measurable disease at baseline has a different analysis population from the response outcome that includes the broader randomized group.

How do confidence intervals and p-values work together?

A confidence interval describes a range of values compatible with the observed data under a statistical procedure. In general, a narrower interval indicates greater precision for the particular estimate, while a wider interval indicates more uncertainty. A confidence interval is not a guarantee that the true value lies within its bounds, and its interpretation depends on the method used. The reported confidence level for the primary hazard ratio is 95%.

The p-value describes how unusual data at least as extreme as those observed would be if the null hypothesis and test assumptions were true. It does not measure the probability that the result arose by chance in an everyday sense, and it does not measure clinical importance. A confidence interval and p-value complement each other: the interval gives a range of effect estimates, while the p-value summarizes evidence against a specified null hypothesis.

What can be concluded about the analysis population?

The primary population is intent-to-treat. The registry defines it as all randomized participants, with assignment based on initial randomization regardless of whether the participant received study drug or a different drug. This preserves the randomized comparison in the analysis definition. The registry entries do not explain the handling of missing outcome data, protocol deviations, or participants with incomplete follow-up. Nor do they report a per-protocol analysis for the outcomes.

11. Confidence Intervals and Precision

The primary hazard-ratio estimate is 0.677, with a 95% confidence interval from 0.529 to 0.867. The interval is wholly below 1, consistent with a lower estimated event rate in the palbociclib-plus-letrozole group for the primary endpoint. The corresponding reported p-value is 0.001. These statistics describe evidence and precision for the particular investigator-assessed progression-free survival comparison; they do not quantify uncertainty about all possible outcomes or populations.

The median estimates also have confidence intervals: 16.6 to 24.9 months for palbociclib plus letrozole and 13.7 to 16.6 months for placebo plus letrozole. Each interval expresses uncertainty around its own group median. Comparing the endpoints of separate median intervals is not a substitute for the reported between-group hazard-ratio analysis. The hazard ratio uses time-to-event information across follow-up, whereas a median is a single summary point.

For secondary outcomes, the reported intervals are generally wider relative to their estimates, and some encompass values compatible with no group difference. This indicates that those results are less conclusive, not that the groups are known to be identical. Precision depends on the amount and distribution of outcome information as well as the analysis method. The registry entries do not provide event counts for the primary comparison, detailed censoring information, or a precision target from the original design. A confidence interval also does not account for every possible bias, measurement issue, or multiplicity concern.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning illustrations are not the original trial design calculation. For time-to-event outcomes, the figures indicate total events required, not participants per arm. For other outcomes, the figures indicate participants per arm. The assumptions listed for these illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is expressed as a stated fraction of the observed effect. The original trial's sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the planning number of events or participants per arm under which the specified effect would have an 80% chance of yielding a statistically significant result, if the assumptions hold. The 90% power column describes the corresponding planning target for a higher chance of detecting that specified effect. These are design-planning probabilities, not guarantees about a study's eventual result. For example, the primary time-to-event illustration gives total-event targets, while the objective-response illustration gives participants-per-arm targets; the units should not be mixed.

Planning for a smaller fraction of the observed effect requires more information, so the illustrations increase their event or participant requirements as the assumed effect becomes smaller. These calculations are hypothetical planning illustrations based on the observed effect and stated assumptions. They are not observed power, and they do not establish what the trial was originally designed to detect. The registry lists planned enrollment of 340, but the original rationale connecting that enrollment to a target effect, event count, or power is not reported in the registry record.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

d is the total number of events needed (Schoenfeld's formula) with 1:1 allocation and proportional hazards.

Worked calculation for this trial

1. Hazard ratio HR = 0.677, so ln(HR) = -0.3901.
2. d = 4 × (1.96 + 0.842)² / [ln(0.677)]² = 206.3, rounded up to 207 events in total for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
hr <- 0.677                              # observed hazard ratio
events_needed <- function(power, frac = 1) ceiling(round(4 * (z_a + qnorm(power))^2 / log(hr^frac)^2, 6))   # Schoenfeld, 1:1 allocation
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: events = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(events_needed(0.80, f), big.mark = ","), format(events_needed(0.90, f), big.mark = ",")))

# Simulation: given d total events under proportional hazards, events in the treatment arm are Binomial(d, hr / (1 + hr)).
# The score (log-rank type) statistic is (D1 - d / 2) / sqrt(d / 4).
set.seed(2297438)
d    <- events_needed(0.80)
reps <- 20000
D1 <- rbinom(reps, d, hr / (1 + hr))
z  <- (D1 - d / 2) / sqrt(d / 4)
cat(sprintf("Simulated power with %s events: %.3f (%d replicates)\n", format(d, big.mark = ","), mean(abs(z) > z_a), reps))

Output from running the code

100% of observed effect: events = 207 (80% power), 277 (90% power)
 75% of observed effect: events = 367 (80% power), 492 (90% power)
 50% of observed effect: events = 826 (80% power), 1,105 (90% power)
Simulated power with 207 events: 0.796 (20000 replicates)

The simulation uses the fact that, with d events in total and proportional hazards, the number of events in the treatment arm is binomial. It is an approximation that conditions on the total number of events, and it should land close to the target power.

14. Group-Sequential Design Analysis

Many time-to-event trials, especially in oncology, stop or report early when interim data are overwhelming, so a single fixed-sample number understates the real design problem. The calculation below is an illustrative group-sequential design built from this trial's posted control median and hazard ratio. The registry record does not state the planned number of looks or the spending function, so the design is a planning illustration, not a reconstruction of the protocol. Assumed: one-sided α = 0.025, 90% power, three equally spaced looks on information, Lan-DeMets O'Brien-Fleming spending, 24 months of accrual, 60 months in total and a 1% monthly dropout hazard.

Under these assumptions the design needs about 383 participants in total, with the boundaries below.

Fixed design versus group-sequential alternatives

Group-sequential designs cost a small increase in maximum sample size and events compared with the fixed design, in exchange for the chance to stop early for overwhelming efficacy. O'Brien-Fleming spending keeps early boundaries very conservative and the final boundary close to the fixed-design value, whereas Pocock spending spends more alpha early.

Sensitivity to a smaller true effect

Planning for exactly the observed effect is optimistic, because observed effects in published trials tend to be somewhat larger than the true effect. The table repeats the main design (three looks, O'Brien-Fleming spending, 90% power) for smaller true hazard ratios.

For context, this trial enrolled 340 participants.

Cross-check, upper boundaries: gsDesign 3.71, 2.511, 1.993 | rpact 3.71, 2.511, 1.993. Agreement between the two packages confirms the boundary calculation.

R code and output

suppressWarnings(suppressMessages({ library(gsDesign); library(rpact) }))
med_control <- 13.9          # control-arm median, months (converted from the posted unit)
hr          <- 0.677          # hazard ratio used for planning
lambdaC     <- log(2) / med_control
eta         <- 0.01               # dropout hazard per month (assumption)
acc_months  <- 24                 # accrual duration (assumption)
study_months<- 60                 # total study duration (assumption)

design <- function(k, sfu, beta, h = hr) gsSurv(k = k, test.type = 1, alpha = 0.025, beta = beta, sfu = sfu,
  lambdaC = lambdaC, hr = h, hr0 = 1, eta = eta, gamma = 1, R = acc_months, T = study_months, minfup = study_months - acc_months)

# Main design: 3 looks, Lan-DeMets O'Brien-Fleming spending, one-sided alpha 0.025, 90% power
x <- design(3, sfLDOF, 0.10)
cat(sprintf("N_TOTAL|%s\n", format(ceiling(sum(x$gamma * x$R)), big.mark = ",")))
for (i in 1:3) cat(sprintf("GS|%d|%.1f|%.0f|%.3f|%.5f|%.5f\n", i, x$T[i], x$n.I[i], x$upper$bound[i], pnorm(-x$upper$bound[i]), cumsum(x$upper$spend)[i]))
cat("\n"); print(gsBoundSummary(x))

# Comparison of designs
cmp <- list(list("Fixed design, one analysis", 1, sfLDOF), list("Two looks, O'Brien-Fleming", 2, sfLDOF),
            list("Three looks, O'Brien-Fleming", 3, sfLDOF), list("Three looks, Pocock", 3, sfLDPocock))
cat("\n")
for (b in c(0.20, 0.10)) for (d in cmp) {
  y <- design(d[[2]], d[[3]], b)
  cat(sprintf("DES|%s|%d|%.0f|%s|%.1f\n", d[[1]], round(100 * (1 - b)), tail(y$n.I, 1), format(ceiling(sum(y$gamma * y$R)), big.mark = ","), tail(y$T, 1)))
}

# Sensitivity: what if the true effect is smaller than the observed one? (3 looks, O'Brien-Fleming, 90% power)
cat("\n")
for (fr in c(1, 0.75, 0.5)) {
  y <- design(3, sfLDOF, 0.10, exp(log(hr) * fr))
  cat(sprintf("SENS|%.2f|%.3f|%.0f|%s\n", fr, exp(log(hr) * fr), tail(y$n.I, 1), format(ceiling(sum(y$gamma * y$R)), big.mark = ",")))
}

# Independent cross-check of the boundaries with rpact
r <- getDesignGroupSequential(kMax = 3, alpha = 0.025, beta = 0.10, sided = 1, typeOfDesign = "asOF", informationRates = x$timing)
cat(sprintf("\nCross-check, upper boundaries: gsDesign %s | rpact %s\n", paste(round(x$upper$bound, 3), collapse = ", "), paste(round(r$criticalValues, 3), collapse = ", ")))

Method: LachinFoulkes 
    Analysis               Value Efficacy
   IA 1: 33%                   Z   3.7103
      N: 312         p (1-sided)   0.0001
  Events: 93        ~HR at bound   0.4627
   Month: 20    P(Cross) if HR=1   0.0001
             P(Cross) if HR=0.68   0.0338
   IA 2: 67%                   Z   2.5114
      N: 384         p (1-sided)   0.0060
 Events: 186        ~HR at bound   0.6915
   Month: 31    P(Cross) if HR=1   0.0060
             P(Cross) if HR=0.68   0.5603
       Final                   Z   1.9930
      N: 384         p (1-sided)   0.0231
 Events: 279        ~HR at bound   0.7874
   Month: 60    P(Cross) if HR=1   0.0250
             P(Cross) if HR=0.68   0.9000



Cross-check, upper boundaries: gsDesign 3.71, 2.511, 1.993 | rpact 3.71, 2.511, 1.993

15. Analyses Reported by the Sponsor

The primary time-to-event comparison is presented using the sponsor-reported hazard ratio and Log Rank result. The displayed primary estimate is 0.677, with a 95% confidence interval of 0.529 to 0.867 and p = 0.001. For time-to-event outcomes, the interpretation uses the sponsor-reported hazard ratio rather than attempting to reconstruct it from the group medians. A hazard ratio cannot be recovered reliably from medians alone.

For binary response outcomes, simple unadjusted recalculations from posted group summaries are different statistical summaries from the sponsor-reported Fisher Exact odds ratios and p-values. An unadjusted risk difference and an odds ratio have different scales and may use different analysis procedures. They should not be treated as interchangeable estimates or as evidence that one analysis is erroneous. The registry entries do not flag a discrepancy for this trial. If a discrepancy were flagged, the sponsor's own analysis would take precedence over a recalculation from summary statistics.

The registry record does not provide enough detail to reproduce the sponsor's full analysis procedures for every endpoint. The sponsor-reported analyses should therefore be read as the trial's reported inferential results, while the unadjusted recalculations are supplementary descriptions of posted summaries.

16. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, time points, subgroups, or statistical tests are examined. In general, conducting many tests increases the chance that at least one result will appear statistically significant even if there is no true difference for any of the tests. Statistical plans may address this with a prespecified hierarchy or an adjustment to the testing procedure. The registry record does not report whether such a plan was used for the secondary outcomes, how endpoints were prioritized for inference, or whether any multiplicity adjustment was applied. Consequently, secondary p-values should be interpreted cautiously and are not equivalent to a single primary test.

The record identifies the primary outcome and reports the primary analysis, but the detailed statistical analysis plan, interim-analysis procedures, stopping rules, and any alpha-spending approach are not reported in the registry record. Their use or absence should not be inferred from the posted result. The record also does not report the original sample-size justification, the planned event target for the primary endpoint, or all analysis assumptions.

The primary population is described as intent-to-treat and includes randomized participants according to initial assignment. That definition is useful, but it does not resolve how incomplete outcome information was treated. Missing-data methods, censoring rules, sensitivity analyses for missingness, and reasons for loss to follow-up are not reported in the registry record. The detailed handling of protocol deviations and treatment discontinuations is also not reported. These omissions limit assessment of the robustness of the posted estimates, particularly for time-to-event follow-up and response assessment.

For the primary endpoint, the registry states the endpoint definition and Kaplan-Meier estimation, but it does not report the full analysis model or diagnostics. It does not specify whether the hazard-ratio analysis used stratification or covariate adjustment, nor does it report checks of model assumptions. These are limits on independent appraisal of the statistical methods, not evidence that any particular method was or was not used.

17. Safety Results

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

Deaths

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: pleural effusion (1.8%), cataract (1.2%), pyrexia (1.2%), pneumonia (1.2%), urinary tract infection (1.2%).

The registry reports serious adverse events in 34 participants in the palbociclib-plus-letrozole group among 168 at risk, and in 19 participants in the placebo-plus-letrozole group among 171 at risk. It also reports other adverse events in 168 participants among 168 at risk in the combination group and 143 among 171 at risk in the placebo group. These are counts as recorded, not a randomized comparison adjusted for follow-up or exposure time.

Deaths are reported as 102 in the palbociclib-plus-letrozole group and 126 in the placebo-plus-letrozole group. The registry entries do not specify whether these deaths were attributed to adverse events, disease, or other causes, and the figures should not be interpreted as cause-specific outcomes. The record does not provide the detailed definitions, exposure-adjusted analyses, or timing needed to compare safety risks fully.

The listed serious adverse-event terms include pleural effusion at 1.8%, with cataract, pyrexia, pneumonia, urinary tract infection, and ankle fracture each listed at 1.2%. These terms and percentages describe selected reported events; the registry entries do not establish causal attribution. Counts and percentages alone do not establish whether a treatment is safe or unsafe, and they do not replace clinical interpretation of the complete safety record.

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

The primary posted result is a statistically clear difference in investigator-assessed progression-free survival between the randomized groups, with the reported hazard ratio favoring palbociclib plus letrozole. The median estimates provide a summary of time to progression or death, while the hazard ratio uses time-to-event information across follow-up. Both should be understood as group-level estimates with uncertainty, not predictions for an individual participant.

The result applies to the trial's recorded population: Asian postmenopausal women with ER-positive, HER2-negative advanced breast cancer who had not received prior systemic anticancer therapy for their advanced or metastatic disease. It concerns the comparison of palbociclib plus letrozole with placebo plus letrozole and the endpoint assessed by investigators. Generalizing beyond this population, treatment comparison, or endpoint requires evidence not contained in these registry entries.

The result does not establish an effect on overall survival, quality of life, or every secondary outcome. It does not show that every participant experienced the group-average pattern, nor does statistical significance establish the clinical importance of an effect. The central-review progression-free survival result and the response outcomes are less conclusive, and the record does not report enough detail to assess multiplicity, missing-data handling, or all model assumptions.

These results are statistical evidence from a randomized trial, not medical advice. Treatment decisions require clinical context and information beyond the registry record. The findings should be read alongside the endpoint definition, uncertainty intervals, analysis population, safety data, and limitations in the available record.

19. Why This Trial Matters Statistically

This trial illustrates why a time-to-event result is not captured by comparing medians alone. The registry reports both group medians and a sponsor-reported hazard ratio, allowing readers to distinguish a summary of the estimated event-free time distribution from a relative event-rate comparison across follow-up. The confidence interval adds information about how precisely that relative estimate is determined.

It also shows how endpoint assessment can matter. Investigator-assessed progression-free survival is the primary outcome, while blinded independent central review provides a separate secondary assessment. Their results are not identical in inferential strength, so interpretation depends on knowing which assessment was prespecified as primary and how each was conducted. The registry entries do not provide all procedural details needed to explain the difference.

The response outcomes demonstrate another statistical distinction: binary response proportions, risk differences, odds ratios, and time-to-event hazard ratios answer different questions and use different scales. Recalculations from summary data can provide descriptive context, but they are not replacements for the sponsor's reported analysis. Planning illustrations also need careful reading: event targets for time-to-event outcomes differ from participants-per-arm targets for other outcomes, and neither should be confused with the trial's observed results.

Finally, this record underscores that a statistically significant primary result does not remove uncertainty about secondary outcomes, safety, missing information, or generalizability. Clear interpretation depends on the endpoint, analysis population, design, precision, and the limits of what the registry reports.

20. 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 gives the primary endpoint definition and reported result, but it does not report several details needed for a full independent statistical appraisal. The original sample-size justification and design assumptions are not reported in the registry record. The planning illustrations based on observed effects do not reconstruct those original design decisions. Detailed methods for allocation, stratification, covariate adjustment, censoring, missing outcomes, and sensitivity analyses are also not reported.

The primary endpoint relies on investigator assessment of disease progression. A secondary outcome used blinded independent central review, but the record does not describe adjudication procedures, agreement between assessors, or reasons for differences between the assessments. The registry defines progression using RECIST v1.1, but does not provide the full details of assessment timing or handling of unevaluable scans.

The registry lists many outcome measures, while the registry entries available for this analysis provide detailed numerical results for only a subset. The multiplicity plan is not reported in the registry record, which limits how confidently secondary p-values can be interpreted collectively. The trial's completed status does not by itself establish that every planned outcome was reported in the available record.

Safety figures are counts, and the record does not provide the exposure-adjusted comparisons or attribution details needed to interpret them fully. Follow-up duration is described as up to 65 months for the primary and listed secondary outcomes, but individual follow-up distributions are not reported. These limitations do not negate the primary reported result; they define what can and cannot be concluded from the registry information.

21. Related Tutorials

22. Related Statistical Calculators

23. Sources

• ClinicalTrials.gov: A Study Of Palbociclib (PD-0332991) + Letrozole VS. Placebo+ Letrozole For 1st Line Treatment Of Asian Postmenopausal Women With ER+/HER2- Advanced Breast Cancer [PALOMA-4] — NCT02297438.
• Registry record last updated 2026-11-05; status last verified 2026-11.
• Results first posted 2021-11-18.
• 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.

24. Record Summary

PALOMA-4 (NCT02297438) was a completed, randomized, quadruple-masked, parallel-group phase 3 study with planned enrollment of 340. It compared palbociclib plus letrozole with placebo plus letrozole in Asian postmenopausal women with ER-positive, HER2-negative advanced breast cancer who had not received prior systemic anticancer therapy for advanced or metastatic disease.

The primary endpoint was investigator-assessed progression-free survival, defined from randomization to documented progression or death in the absence of documented progression. The sponsor reported a hazard ratio of 0.677 with a 95% confidence interval of 0.529 to 0.867 and p = 0.001. The result indicates a difference between groups for this endpoint in the direction of the combination group. Secondary findings and safety counts require separate, qualified interpretation.

The registry record does not report the original sample-size justification, detailed multiplicity plan, missing-data methods, or all analysis assumptions. Those omissions limit assessment beyond the reported estimates and their stated uncertainty.