PF-00299804 in Stage IIIB or Stage IV Non-Small Cell Lung Cancer Not Responding to Standard Therapy for Advanced or Metastatic Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for PF-00299804 in Stage IIIB or Stage IV Non-Small Cell Lung Cancer Not Responding to Standard Therapy for Advanced or Metastatic Cancer (NCT01000025), a randomized phase 3 study. The primary outcome is Overall Survival.

2. Clinical Question

The trial addressed treatment after standard therapy for advanced or metastatic disease had failed. Its registry description gives the rationale as investigating whether PF-00299804, which may block enzymes involved in tumor-cell growth, differs from placebo in this setting. That rationale is not itself evidence of a treatment effect; the comparison must be interpreted through the reported outcome estimates and their uncertainty.

The primary question is about the time-to-event outcome overall survival, rather than tumor response alone. The record reports median survival in each group and a hazard ratio comparing the groups over the stated 42 Months timeframe. The ITT label identifies the population specified for the primary outcome, but details of how that population was defined or how missing observations were handled are not reported in the registry record.

3. Trial Design

This was a randomized, phase III, parallel-group, placebo-controlled trial with triple masking. The registry record lists 720 participants as the enrollment and identifies NCIC Clinical Trials Group as the sponsor. It lists Lung Cancer as the condition. The study is marked completed, with a start date of 2009-12-23 and a primary completion date of 2014-01-15.

Participants were assigned to PF-00299804 or placebo. The intervention descriptions specify oral dosing once daily in repeating 28-day treatment periods; in both groups treatment was to continue in the absence of disease progression or unacceptable toxicity. The registry record labels the design triple masked but does not describe which roles were masked or how masking was maintained.

The primary outcome was overall survival, expressed as median months and accompanied by a 95% confidence interval. The stated timeframe is 42 Months, and the primary population is identified as ITT. The posted results give outcome group sizes of 480.0 for PF-00299804 and 240.0 for placebo. These are the group sizes attached to the reported primary outcome; the registry record does not explain the relationship between these outcome counts and enrollment.

The sponsor-reported comparison uses a log-rank method and presents a hazard ratio. Randomization and masking describe the study design, but the registry record does not report details such as the randomization procedure, allocation concealment, masking procedures, or protocol-defined analysis plan. Those details should not be inferred from the design labels alone.

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:

Analysis population for the primary outcome, as defined in the registry: ITT

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 Overall Survival, measured over 42 Months.

Overall survival measures the time until death from any cause. It is a time-to-event endpoint: each participant contributes information about survival time, and participants who have not had the event by the end of observation may contribute censored follow-up. General statistical context: censoring allows an analysis to use the time observed without treating a participant’s later, unobserved outcome as known. The registry record does not describe the censoring rules used in this trial.

The primary outcome reports median overall survival in months for each group, with a 95% confidence interval, and a sponsor-reported hazard ratio from a log-rank analysis. A median is the time by which half of the event-time distribution has occurred under the analysis; it does not describe every participant’s survival time. A hazard ratio compares the event rates over follow-up under the model or analysis that generated it. It is not a difference between the reported medians, and it does not mean that every participant experiences the same relative change in risk.

Overall survival is used in clinical trials because it directly records time to death and does not depend on a judgment that a tumor has progressed or responded. General statistical context: the endpoint can still be affected by follow-up duration, censoring, and subsequent care, so its interpretation relies on the trial’s analysis and follow-up procedures. Those procedural details are not reported in the registry record. The registry identifies the primary timeframe as 42 Months but does not give additional detail about follow-up completeness or endpoint adjudication.

7. Results Posted in the Registry

• PF-00299804 vs Placebo: hazard ratio 1.00 (95% CI 0.83–1.21), reported p = 0.506 (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.

Overall Survival in KRAS-WT Patients

• 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:
  • PF-00299804 vs Placebo: hazard ratio 0.79 (95% CI 0.61–1.03), reported p = 0.043 (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 PF-00299804 vs Placebo and the control. The interval gives the range of effects that remain compatible with the data.

Overall Survival in EGFR-mutant Patients

• 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:
  • PF-00299804 vs Placebo: hazard ratio 0.98 (95% CI 0.67–1.44), reported p = 0.460 (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 PF-00299804 vs Placebo and the control. The interval gives the range of effects that remain compatible with the data.

Progression-free Survival

• 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:
  • PF-00299804 vs Placebo: hazard ratio 0.66 (95% CI 0.55–0.79), reported p < 0.001 (as reported in the registry).
• Interpretation:
  • The 95% confidence interval excludes no difference (a ratio of 1), so the hazard of the event is statistically lower for PF-00299804 vs Placebo than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary results extend the survival comparison to molecular subgroups and add progression-free survival. Taken together, they do not provide one consistent pattern across all outcomes. Some estimates are closer to the equal-hazard reference, while another secondary comparison has a hazard ratio below that reference with a confidence interval that does not include it. These findings complement the primary result but do not overturn it: the primary outcome remains overall survival in the ITT population, and secondary comparisons address different questions or subsets.

Subgroup findings require particular restraint. A result within a subgroup does not by itself show that the treatment effect differs between subgroups; that requires a direct interaction analysis. Such an analysis is not reported in the registry record. The record also does not report whether the subgroup definitions and analyses were prespecified, or whether there were additional subgroup comparisons beyond those listed. Those unknowns make it difficult to assess how much weight to place on any isolated subgroup signal.

Multiplicity matters because testing several outcomes or subgroups creates more opportunities for nominally notable p-values to arise by chance. General statistical context: adjustment procedures can control the overall false-positive risk across a family of tests, but they generally require a prespecified definition of that family and method. Neither an adjustment procedure nor a complete testing hierarchy is reported in the registry record. Accordingly, the secondary p-values should not be treated as multiplicity-controlled evidence.

Numerical results for objective response rate and the toxicity outcome listed among the secondary outcomes are not reported in the registry record. The secondary evidence therefore cannot be read as a complete account of every outcome named in the record. The overall pattern is best considered exploratory or supportive unless a documented analysis plan establishes otherwise; the registry record does not report such a plan.

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

• Overall Survival in KRAS-WT Patients
• Overall Survival in EGFR-mutant Patients
• Progression-free Survival
• Objective Response Rate
• Number of Participants With Toxicity as Measured by NCI CTCAE Version 4.0

9. Statistical Interpretation

10. Statistical Methods Explained

What does the hazard ratio describe?

General statistical context: a hazard is the rate at which an event occurs among people who have not yet experienced it, at a given point in follow-up. A hazard ratio compares that rate between groups across the time-to-event analysis. A value of 1.00 is the reference for equal hazards; a value below 1.00 indicates a lower estimated hazard in the first group relative to the comparator, and a value above 1.00 indicates a higher estimated hazard. This is a relative comparison of event rates, not an absolute difference in survival time and not an individual prediction.

For the primary overall-survival analysis, the reported hazard ratio is 1.00. Its 95% confidence interval, 0.83 to 1.21, spans values on either side of 1.00. The estimate is therefore compatible with a range of possible relative hazards, including no difference under the hazard-ratio scale. The registry record identifies the method as log rank and does not provide further detail about the model assumptions or any adjustment for other variables.

What does the log-rank method contribute?

General statistical context: the log-rank test compares the observed pattern of events over follow-up between groups, taking account of the timing of events and censoring. Its p-value measures how unusual the data would be under a specified null hypothesis, subject to the analysis assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. Here the registry reports p = 0.506 for the primary log-rank comparison. That result does not provide strong evidence against equal survival patterns, but it cannot show that the groups are exactly equivalent.

The log-rank p-value and the hazard-ratio estimate answer related but distinct questions. The hazard ratio gives an estimated relative comparison, while the p-value summarizes evidence against a null comparison under the test. The confidence interval communicates uncertainty about the estimated hazard ratio. The registry record does not give the test’s detailed implementation, event counts, censoring distribution, or model diagnostics, so those aspects cannot be assessed from the posted record.

How should medians be read alongside the hazard ratio?

General statistical context: a median time-to-event estimate is a point on a survival distribution, whereas a hazard ratio uses information across follow-up. The median does not capture the full shape of the survival curves. If event patterns change over time, a single hazard ratio can also be an incomplete description of the relationship. The registry record does not report whether proportional hazards were assessed or whether the reported hazard ratio should be understood under a particular time-varying model. It is therefore appropriate to report the sponsor’s stated measure without assuming details not recorded.

The primary medians are 6.83 months for PF-00299804 and 6.31 months for placebo. Their confidence intervals are also reported. These descriptive values should be read together with the log-rank hazard-ratio analysis, rather than used alone to infer a treatment effect. A comparison of medians does not reproduce the sponsor’s analysis and does not account for the event timing and censoring used in a time-to-event analysis.

What do the confidence intervals add?

General statistical context: a confidence interval is a range of estimates compatible with the observed data under the method’s assumptions. It conveys precision as well as the point estimate; a wider interval reflects greater uncertainty than a narrower interval on the same scale. A 95% confidence interval is not a guarantee that the true value lies within the particular interval. It is a procedure with a long-run coverage interpretation when its assumptions hold.

For the primary hazard ratio, the 95% confidence interval is 0.83 to 1.21. Because the interval includes 1.00, the data do not exclude equal hazards at that confidence level. It also includes values that could correspond to a lower or higher hazard for PF-00299804. The interval therefore supports caution against interpreting the point estimate alone. The confidence intervals around the group medians describe uncertainty in each median estimate; they are not a confidence interval for the difference between medians.

How should the planning illustrations be understood?

The registry material includes planning illustrations for secondary time-to-event comparisons. Their assumptions are a two-sided significance level of 0.05, equal allocation, a normal approximation, and an effect to detect set to a stated fraction of the observed effect. For time-to-event outcomes, the illustrations give total events needed for 80% or 90% power. These are planning scenarios based on observed effects, not outcomes observed in this trial and not the trial’s own documented design calculations.

General statistical context: planning calculations depend on the assumed effect and analysis conditions. If the effect targeted for a future study is smaller than the observed estimate, more information is generally needed to distinguish it from the null under otherwise comparable assumptions. Such scenarios should not be interpreted as observed power, as a retrospective measure of whether the trial was adequately powered, or as evidence that the registry’s original sample-size planning used these assumptions. The trial’s own sample-size justification is not reported in the registry record.

The registry record states that the primary result is for the ITT population, but does not report how participants with missing outcome data were treated, the censoring rules, or whether additional adjusted analyses were conducted. These omissions limit independent assessment of the time-to-event analysis. They do not change the numbers that the sponsor reported, but they leave important methodological details unresolved.

11. Confidence Intervals and Precision

Precision is reflected by the width of a confidence interval relative to the estimate and the scale being measured. General statistical context: narrower intervals usually indicate less sampling uncertainty under the analysis assumptions, while wider intervals leave more possible values compatible with the data. Precision is not the same as importance, and neither a confidence interval nor a p-value answers whether an effect would matter to patients.

For overall survival, the hazard ratio is 1.00 and its 95% confidence interval is 0.83 to 1.21. This interval includes the equal-hazard reference and allows values in either direction. It is not so concentrated around 1.00 that it establishes equivalence; the registry record does not report an equivalence margin or a non-inferiority plan. The reported p-value, 0.506, is consistent with the lack of strong statistical evidence against the null comparison in the log-rank analysis.

The median estimates also have uncertainty: 6.83 months for PF-00299804, with a 95% confidence interval of 6.08 to 7.49 months, and 6.31 months for placebo, with a 95% confidence interval of 5.32 to 7.52 months. These are separate intervals for each group. Their overlap is not a formal test of the difference between medians, and the registry does not provide a confidence interval for that difference.

The primary outcome group sizes attached to the estimates are 480.0 and 240.0. Group size alone does not determine precision: event numbers, follow-up, censoring, and the distribution of event times also matter. Those details are not reported in the registry record. Consequently, the posted confidence interval is the direct guide to uncertainty in the reported hazard ratio, while further explanation of the interval’s width is limited by missing analysis details.

12. Analyses Reported by the Sponsor

The sponsor-reported primary analysis is a log-rank comparison with a hazard ratio of 1.00, a 95% confidence interval of 0.83 to 1.21, and p = 0.506. These values are the central inferential results for the primary overall-survival endpoint. The posted median estimates provide a descriptive view of each group, but they do not replace the sponsor’s hazard-ratio analysis.

The registry record does not flag a discrepancy between the sponsor analysis and the posted results. No separate recalculated primary estimate is reported in the registry data, so there is no independent recalculated estimate to set against the sponsor result. The sponsor’s own analysis is therefore the result to interpret, rather than an alternative computation from the group medians.

For secondary outcomes, the sponsor-reported log-rank hazard ratios and p-values are also recorded. The KRAS-WT result has a p-value of 0.043 alongside a 95% confidence interval of 0.61 to 1.03, which includes 1.00. The registry record does not explain this pairing or describe whether the secondary p-values were adjusted for multiple comparisons. The appropriate reading is to report both measures as recorded and retain uncertainty, not to choose one result over the other or infer an unreported analysis plan.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists overall survival as the primary outcome and identifies secondary outcomes that include survival in KRAS-WT and EGFR-mutant patients, progression-free survival, objective response rate, and toxicity measured using NCI CTCAE Version 4.0. Numerical results are posted for the two molecular-subgroup survival outcomes and progression-free survival. Numerical results for objective response rate and the listed toxicity outcome are not reported in the registry record.

The record does not report a multiplicity adjustment, a hierarchy for testing primary and secondary outcomes, or a plan describing which comparisons would be confirmatory. This matters because interpreting several p-values independently at a conventional threshold can overstate the strength of evidence across a set of tests. Without a reported adjustment plan, secondary p-values should be treated as nominal rather than as evidence shown to control the family-wise error rate. Whether any unreported protocol or analysis plan specified such procedures is not reported in the registry record.

The primary population is labelled ITT. Further details about the population definition, exclusions after randomization, and analysis populations for the secondary outcomes are not reported in the registry record. The record also does not state how missing outcome observations, withdrawals, or incomplete follow-up were handled. For time-to-event outcomes, censoring rules and the extent of follow-up are important to interpretation; these details are not reported in the registry record.

Interim analyses and stopping rules are not reported in the registry record. The sample-size justification and original power assumptions are also not reported. The planning illustrations attached to secondary time-to-event comparisons use the observed effects and stated assumptions; they are not evidence of the original trial’s design calculations. These omissions limit assessment of prespecification and missing-data risks, but they do not justify guessing what procedures may have been used.

14. 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: other neoplasms benign, malignant and unspecified (18.0%), lung infection (6.7%), dyspnea (5.9%), diarrhea (3.6%), dehydration (3.4%).

The adverse-event summary reports serious events affecting 191 participants in the PF-00299804 group, among 477 at risk, and 87 participants in the placebo group, among 239 at risk. For other adverse events, the figures are 465 among 477 at risk and 219 among 239 at risk, respectively. These are reported counts, not adjusted comparisons. The registry record gives no deaths count in the adverse-event summary.

The most frequently listed serious-event terms include other neoplasms benign, malignant and unspecified (18.0%), lung infection (6.7%), dyspnea (5.9%), diarrhea (3.6%), dehydration (3.4%), and thromboembolic event (2.5%). These percentages are presented as serious-event terms; the registry record does not provide further detail about their denominator or whether an individual could contribute to more than one term. They should not be summed or interpreted as mutually exclusive categories.

These summaries describe events recorded in the registry, not proof that a treatment caused an event. The safety groups’ at-risk counts differ from the primary outcome group sizes, and the registry does not explain the populations or event-collection rules behind that difference. Without timing, severity breakdowns beyond the serious-event listing, exposure-adjusted rates, or a formal comparative safety analysis, the counts alone do not establish a causal or comparative safety conclusion.

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

The primary overall-survival analysis reports a hazard ratio of 1.00, with a 95% confidence interval of 0.83 to 1.21 and a log-rank p-value of 0.506. In the posted analysis, the estimate is centered at equal hazards and the interval includes that reference. This means the result does not provide clear statistical evidence of an overall-survival difference between PF-00299804 and placebo. It does not prove that there was no difference, establish that the treatments are equivalent, or rule out all potentially meaningful effects.

The group medians, 6.83 months and 6.31 months, are descriptive summaries. They do not show the experience of every participant and do not by themselves give the full comparison over follow-up. The hazard ratio and its confidence interval are the appropriate reported inferential summary for the primary time-to-event analysis. Details that could further inform interpretation, including censoring and missing-data methods, are not reported in the registry record.

The secondary outcomes add information but have important limits. The listed subgroup and progression-free survival results do not replace the primary overall-survival comparison. Their interpretation is constrained by the absence of a reported multiplicity plan and by the missing numerical results for some listed secondary outcomes. The registry record also does not report an interaction test establishing whether treatment effects differ between the molecular subgroups.

The result is about this randomized trial’s comparison in patients described as having stage IIIB or stage IV non-small cell lung cancer after standard therapy had failed. It should not be extended automatically to other populations, treatment settings, or outcomes. It is a statistical interpretation of the posted registry data, not individual medical advice and not a claim that a treatment works or does not work.

16. Why This Trial Matters Statistically

This trial illustrates why a point estimate should not be read without its uncertainty. The primary hazard ratio is 1.00, but the confidence interval extends from 0.83 to 1.21. The estimate alone might appear to settle the comparison at equal hazards; the interval makes clear that the data remain compatible with a range of relative effects. Conversely, the interval’s inclusion of 1.00 does not prove equivalence. The registry record reports no equivalence margin or non-inferiority framework.

It also illustrates the difference between a time-to-event summary and a median. The reported medians describe each group’s survival distribution at a particular point, whereas the hazard ratio and log-rank analysis use information across follow-up. Readers should not substitute an arithmetic comparison of medians for the sponsor-reported analysis.

The secondary results show why multiplicity and prespecification matter. A subgroup p-value may appear notable, but its weight depends on the role assigned to the analysis, the number of comparisons considered, and whether an adjustment strategy was planned. Those details are not reported in the registry record. The primary result remains the anchor, while secondary estimates add context with greater interpretive uncertainty.

Finally, registry records often leave analytical details unresolved. Here the ITT population is named, but censoring, missing-data handling, interim analyses, and the original sample-size justification are not reported. Recognizing these limits is part of sound interpretation: it distinguishes what the posted estimates support from what cannot be established from the record.

17. 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.

Several trial-specific limitations affect interpretation of the registry results. The record identifies the primary analysis population as ITT but does not explain how that population was defined or how participants with missing outcome data were handled. Censoring rules, follow-up completeness, and details of outcome ascertainment are also not reported. These are important for evaluating a time-to-event analysis, because incomplete follow-up and censoring assumptions can influence estimates.

The registry describes the study as triple masked, randomized, and parallel, but does not report the allocation procedure, concealment method, masking roles, or steps used to maintain masking. It does not report whether proportional hazards were assessed or whether additional adjusted analyses were performed. The posted log-rank result is therefore interpretable as the sponsor’s stated comparison, but the underlying analytical assumptions cannot be fully examined from the record.

The primary result is accompanied by subgroup and additional outcome analyses, yet the registry does not report a multiplicity plan, test hierarchy, interaction analysis, or whether the secondary analyses were prespecified. Numerical results for objective response rate and the listed toxicity outcome are not reported. The absence of those details limits assessment of the completeness and confirmatory status of the secondary evidence.

The safety summary gives event counts and selected serious-event terms but does not report event timing, exposure-adjusted comparisons, or detailed attribution. The safety at-risk counts differ from the primary outcome group sizes, with no explanation in the registry record. Finally, the record does not report the trial’s sample-size justification, interim analyses, stopping rules, or detailed follow-up plan. These limitations do not determine the direction of any effect, but they constrain how fully the posted results can be evaluated.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: PF-00299804 in Stage IIIB or Stage IV Non-Small Cell Lung Cancer Not Responding to Standard Therapy for Advanced or Metastatic Cancer — NCT01000025.
• Registry record last updated 2023-08-22; status last verified 2020-04.
• Results first posted 2014-10-27.
• 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.

21. Record Summary

NCT01000025 is a completed randomized, phase III, triple-masked, parallel-group placebo-controlled trial of PF-00299804 in stage IIIB or stage IV non-small cell lung cancer after standard therapy for advanced or metastatic disease had failed. NCIC Clinical Trials Group is listed as sponsor. The record lists enrollment of 720 and dates of 2009-12-23 for study start and 2014-01-15 for primary completion.

The primary outcome is median overall survival in the ITT population over a stated timeframe of 42 Months. The sponsor-reported log-rank hazard ratio is 1.00, with a 95% confidence interval of 0.83 to 1.21 and p = 0.506. Secondary results include survival subgroup and progression-free survival analyses; other listed secondary outcomes lack numerical results in the registry record. The record does not report multiplicity procedures, missing-data methods, or the original sample-size justification.