An Open-label Study of GSK1120212 Compared With Docetaxel in Stage IV KRAS-mutant Non-small Cell Lung Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for An Open-label Study of GSK1120212 Compared With Docetaxel in Stage IV KRAS-mutant Non-small Cell Lung Cancer (NCT01362296), a randomized phase 2 study. The primary outcome is Progression-Free Survival (PFS) as Assessed by the Investigator (INV).

2. Clinical Question

3. Trial Design

The registry describes a phase II, randomized, open-label, multicenter study with parallel treatment groups. Participants were assigned to GSK1120212 or docetaxel in the randomized phase. The registered enrollment was 134. The primary efficacy analysis concerned participants with KRAS mutation-positive non-small cell lung cancer in the modified intent-to-treat population, defined as all randomized participants in that group whether or not they received treatment. The posted endpoint results list 86 participants in the GSK1120212 group and 43 in the docetaxel group.

The study addressed second-line treatment after failure of one platinum-containing chemotherapy regimen in locally advanced or metastatic non-small cell lung cancer. The primary population was defined by KRAS mutation status. A small subset of participants with BRAF, NRAS, or MEK1 mutations was included for exploratory purposes; the primary question and the posted primary endpoint results should not be read as an efficacy assessment for those other mutation groups.

GSK1120212 was given orally once daily, while docetaxel was given intravenously once every 3 weeks. Since the trial was open-label, participants and investigators were not masked to treatment assignment according to the registry description. The primary endpoint was assessed by the investigator. The registry lists a randomized phase and a crossover phase among secondary outcome descriptions, but details of the crossover design and its entry criteria are not reported in the registry record.

The study status is completed. The registry records a start in 2011-09 and primary completion in 2012-09. The rationale for the enrollment target, the original sample-size calculation, and the allocation procedure beyond randomized assignment are not reported in the registry record. These omissions mean the recorded enrollment cannot, by itself, establish what precision or power the investigators expected before the trial began.

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: In the Randomized Phase (RP), participants were treated until disease progression (PD), death, or unacceptable adverse events were experienced. After PD in the RP, participants were given the option of crossing over to the alternative treatment arm in a Cross-over Phase (CP).

Randomized Phase (RP)

Reasons for not completing this period:

Crossover Phase

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified Intent-to-Treat (MITT) Population: all randomized participants with KRAS mutation-positive non-small cell lung cancer (NSCLC), whether or not treatment was administered

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) as Assessed by the Investigator (INV), measured over From randomization (RAN) until the earliest date of documented radiological disease progression (PD) or death (DT) due to any cause (maximum of 10.2 months).

Progression-free survival is a time-to-event outcome. In this trial it was defined as the time from randomization to the earliest of documented radiological disease progression or death due to any cause. Progression was assessed by the investigator using RECIST, version 1.1. The registry’s description defines progression as a sufficiently large increase in the sum of target-lesion diameters, the appearance of a new lesion, or worsening of non-target lesions significant enough to require treatment discontinuation.

For a participant without a recorded progression or death, the registry says progression-free survival was censored at the last adequate assessment. If a participant began subsequent anti-cancer therapy before documented progression or death, censoring was at the last adequate assessment before that therapy began. Censoring means the event time was not observed during the period used for that participant’s analysis; it does not mean that progression or death occurred on the censoring date.

The registry reports the median progression-free survival in weeks for each group, with a 95% confidence interval, and a hazard ratio for the comparison. The median describes the time by which the estimated event-free proportion reaches the midpoint. A hazard ratio summarizes the relative event rates over follow-up under the analysis model; it is not a difference between the two medians. Progression-free survival is used to capture both radiological progression and death in one time-to-event measure, but interpretation depends on assessment and censoring procedures as well as the uncertainty around the estimates.

7. Results Posted in the Registry

• GSK1120212 2 mg in RP vs Docetaxel 75 mg/m^2 in RP: hazard ratio 1.14 (95% CI 0.75–1.75), reported p = 0.520 (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.

Change From Baseline in Heart Rate: Randomized Phase

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Docetaxel 75 mg/m^2 in RP: mean 13.20 versus 10.40 on control; difference 2.80 (95% CI -1.28 to 6.88), standardized effect d = 0.22; z-test p = 0.179.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Docetaxel 75 mg/m^2 in RP and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists 19 posted secondary outcomes, including measures of clinical chemistry and hematology changes in both randomized and crossover phases. It also lists change from baseline in heart rate for the randomized phase. These outcomes broaden the description of laboratory and physiological observations but do not replace the primary progression-free survival question.

The posted registry information does not provide numeric results for the full set of secondary outcomes in the registry record used for this account. Consequently, an overall direction or consistency pattern across those outcomes cannot be established from the information reported in the registry record. Their listed presence should not be mistaken for evidence that all secondary measures favored one group or showed a consistent effect.

The primary endpoint and secondary measures address different questions and may use different analysis populations, time points, and statistical methods. Laboratory change counts, for example, are not directly comparable with a time-to-event hazard ratio. Any relationship between secondary findings and the primary result would require the outcome estimates, uncertainty, and analysis details for those measures. Those details are not available for the full set in the registry record.

General explanation: when many outcomes are assessed, the chance of obtaining at least one apparently unusual result can increase, even if no true effect exists for any of them. This is why unadjusted secondary results generally carry less weight than a prespecified primary endpoint unless the testing strategy and multiplicity handling are clear. The registry record does not report an adjustment plan for multiple secondary comparisons. The secondary outcomes can therefore be treated as descriptive or exploratory evidence, not as independent confirmation of the primary result.

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

• Number of Participants With the Indicated Worst-case On-therapy Change From Baseline in the Indicated Clinical Chemistry Parameters: Randomized Phase
• Number of Participants With the Indicated Worst-case On-therapy Change From Baseline in the Indicated Clinical Chemistry Parameters: Crossover Phase
• Number of Participants With the Indicated Worst-case On-therapy Change From Baseline in the Indicated Hematology Parameters: Randomized Phase
• Number of Participants With the Indicated Worst-case On-therapy Change From Baseline in the Indicated Hematology Parameters: Crossover Phase
• Number of Participants With the Indicated Worst-case On-therapy Change From Baseline With Respect to Normal Ranges in the Indicated Clinical Chemistry Parameters: Randomized Phase
• Number of Participants With the Indicated Worst-case On-therapy Change From Baseline With Respect to Normal Ranges in the Indicated Clinical Chemistry Parameters: Crossover Phase
• Number of Participants With the Indicated Worst-case On-therapy Change From Baseline With Respect to Normal Ranges in the Indicated Hematology Parameters: Randomized Phase
• Number of Participants With the Indicated Worst-case On-therapy Change From Baseline With Respect to Normal Ranges in the Indicated Hematology Parameters: Crossover Phase

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is a time-to-event measure, so the analysis must account for both participants who experience progression or death and participants whose event time is not observed during follow-up. The registry reports medians, confidence intervals, and a hazard ratio, and identifies the comparison method as a log-rank analysis. Details beyond the posted method are not reported in the registry record.

What does the median represent?

In general, the median time-to-event estimate marks the point at which the estimated event-free proportion has reached one half. It is not the average time to progression or death, and it does not mean that every participant experienced an event at that time. A median can be estimated even when some participants have not had an event, provided the observed follow-up supports the estimate. The values reported for this trial are in weeks and are paired with confidence intervals to convey uncertainty in each group’s median.

What does the hazard ratio add?

In general, a hazard ratio compares the event rate at a given time between groups, conditional on participants having remained event-free up to that time. The reported ratio is 1.14 for GSK1120212 compared with docetaxel. A value of 1 is the reference for equal hazards; the interpretation of values above or below 1 depends on which group is placed first in the comparison. The hazard ratio is not a ratio of median times and should not be converted into a percentage improvement or worsening without a valid, clearly specified interpretation.

The log-rank test is a general method for comparing time-to-event distributions across groups while incorporating the ordering of observed events and censored observations. It tests a null hypothesis about the groups rather than estimating a clinically meaningful difference by itself. The registry reports the method and a p-value, but does not report details of any additional model adjustments or checks. The estimate and confidence interval remain important because a test result alone does not show the range or direction of effects compatible with the data.

How does censoring enter the analysis?

General explanation: a censored participant contributes information up to the last time the event status is adequately known, but the participant’s later event time is not treated as observed. The registry specifies censoring at the last adequate assessment for participants without documented progression or death, and at the last adequate assessment before subsequent anti-cancer therapy when that therapy began first. These rules help define what follow-up contributes to the analysis. The number of participants censored, reasons for censoring, and further sensitivity analyses are not reported in the registry record.

Time-to-event methods rely on assumptions about how censoring relates to the chance of an event. In general, informative censoring can affect estimates if the participants who leave observation differ systematically in their later event risk in ways not addressed by the analysis. The registry record does not provide enough information to assess that possibility in detail. This is a limitation of what can be concluded from the posted account, not evidence that the censoring rules were necessarily inappropriate.

Why report a confidence interval?

In general, a confidence interval describes statistical uncertainty around an estimate under the analysis method and its assumptions. The hazard-ratio interval from 0.75 to 1.75 includes 1 and ranges across values in both directions. That breadth indicates that the point estimate should not be treated as a precise measurement of the comparative effect. The medians also have their own intervals, which describe uncertainty in each group’s median and should not be mistaken for a direct interval around the difference between medians.

The registry identifies the primary population as modified intent-to-treat: randomized participants with KRAS mutation-positive disease, regardless of whether treatment was administered. That definition is important because it anchors the estimand to randomized assignment rather than only to participants who completed treatment. The registry does not report whether other analysis populations were used for the primary endpoint, nor does it provide further details of analysis-set handling.

Some secondary endpoints are continuous measures and some are counts of participants with laboratory changes; those require different summaries and methods from the primary time-to-event outcome. The registry lists 19 posted secondary outcomes, including laboratory measures in randomized and crossover phases. The primary statistical result should not be treated as a summary of all secondary outcomes. The registry does not report a complete statistical analysis plan, adjustment strategy for multiple testing, or detailed missing-data procedures for the secondary measures.

11. Confidence Intervals and Precision

Precision concerns how narrowly the data estimate an underlying result. The primary median for GSK1120212 is 11.7 weeks, with a 95% confidence interval from 7.0 to 12.4 weeks. The docetaxel median is 11.4 weeks, with an interval from 6.1 to 18.3 weeks. Each interval describes uncertainty in its group’s median; the overlap between two separate intervals is not itself a formal test of the difference between groups.

For the comparative hazard ratio, the estimate is 1.14 and the 95% confidence interval is 0.75 to 1.75. Because the interval includes 1, the data are compatible with no relative hazard difference under the comparison, as well as with relative effects in either direction. The interval spans a meaningful range of possible values, so the point estimate should not be read as a precise estimate of the true relative hazard. The registry’s log-rank result is another part of the inferential record, but it does not remove the uncertainty conveyed by the interval.

Confidence intervals are not guarantees that the true value lies within the particular interval, nor do they represent the range of outcomes for individual participants. In general, they quantify uncertainty arising from sampling under the method’s assumptions. Their interpretation depends on the time-to-event analysis and censoring rules, and the registry does not report every diagnostic or sensitivity analysis that might further characterize those assumptions.

The groups’ posted analysis sizes are 86 and 43. Unequal group sizes can influence the precision of a comparison, although the exact influence depends on event information and follow-up, not only on enrollment. The registry record does not provide the event count for the primary endpoint or the distribution of follow-up time. Accordingly, the confidence interval is the most direct posted indicator of precision for the hazard ratio, and it indicates substantial uncertainty.

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 separate from the trial’s observed analysis. For the time-to-event calculations, the figures are total numbers of events, not participants per arm. The registry’s planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to the stated fraction of the observed effect; the total-event calculations use the Schoenfeld method. The trial’s original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column represents the modeled total event count needed to have an 80% chance of detecting the specified effect under those planning assumptions, if the effect and assumptions used for planning applied. The 90% power column represents the corresponding count for a 90% chance. These are planning quantities, not descriptions of the trial’s achieved power. The registry does not report that these event targets were the basis of the original trial design.

The calculation is repeated for planning for 100%, 75%, and 50% of the observed effect. Smaller effects require more information to detect under otherwise fixed assumptions, so the event totals increase as the effect used for planning gets smaller. The listed total-event requirements are 1,829 and 2,449 for the 80% and 90% columns when planning for 100% of the observed effect; 3,251 and 4,353 when planning for 75%; and 7,315 and 9,793 when planning for 50%. These values are illustrations produced from the observed effect, not observed outcomes. This is not observed power.

General explanation: power is a property of a planned study under specified assumptions, including the effect size, significance level, allocation, and analysis. It is not a retrospective measure of whether a particular result is true. The planning illustration should not be used to infer that the completed trial was designed to detect exactly the displayed effects or that it had a particular chance of success. The original design calculation 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 = 1.140, so ln(HR) = 0.1310.
2. d = 4 × (1.96 + 0.842)² / [ln(1.140)]² = 1,828.7, rounded up to 1,829 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 <- 1.14                              # 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(1362296)
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 = 1,829 (80% power), 2,449 (90% power)
 75% of observed effect: events = 3,251 (80% power), 4,353 (90% power)
 50% of observed effect: events = 7,315 (80% power), 9,793 (90% power)
Simulated power with 1,829 events: 0.805 (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. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparison as a log-rank hazard ratio of 1.14 for GSK1120212 compared with docetaxel, with a 95% confidence interval from 0.75 to 1.75. The posted analysis also identifies a log-rank p-value. The primary result should be interpreted through the estimate, interval, and test together, while recognizing that the interval includes 1 and leaves uncertainty about the direction and size of the relative hazard.

No discrepancy is flagged between the posted sponsor result and the statistical information used for the primary comparison. The primary time-to-event interpretation therefore rests on the sponsor-reported hazard ratio rather than a separate re-calculation from summary statistics. A re-calculation from group medians alone would not reproduce a hazard ratio, because medians do not contain the complete event-time and censoring information used by the time-to-event analysis.

For general context, a re-calculation from summary data can be useful when the required inputs support a simple independent calculation, but it may not match a prespecified analysis that uses participant-level data, censoring, or other model details. No separate recalculated primary hazard ratio is reported. The sponsor’s log-rank result is the relevant comparative estimate in the registry record, and it should not be replaced by an inference drawn only from the closeness of the group medians.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies investigator-assessed progression-free survival as the primary endpoint and lists 19 secondary outcomes. The listed secondary outcomes include clinical chemistry and hematology changes, as well as heart-rate change. However, the registry record does not report a complete statistical analysis plan or a strategy for controlling multiplicity across secondary outcomes. It is therefore not possible to determine from the record whether all comparisons were prespecified, how any hierarchy of testing was defined, or whether adjustment was applied.

General explanation: examining multiple outcomes creates multiple opportunities for results that appear unusual by chance. Adjustment procedures can change how evidence is interpreted when many hypotheses are tested. Without a reported multiplicity plan, isolated secondary findings would generally be considered less conclusive than the primary endpoint. This does not imply that a secondary result is false; it means its evidential strength cannot be judged without the full analysis context.

The registry does report censoring rules for the primary time-to-event endpoint. Participants without documented progression or death were censored at their last adequate assessment, and participants who began subsequent anti-cancer therapy before documented progression or death were censored at their last adequate assessment before starting that therapy. Further missing-data procedures, the number of censored participants, and sensitivity analyses are not reported in the registry record. No assumptions about additional imputation or alternative censoring rules can be made.

The primary population is defined as the modified intent-to-treat group of randomized participants with KRAS mutation-positive non-small cell lung cancer, whether or not treatment was administered. The registry record does not report additional detail about other analysis populations or how missing endpoint information was handled beyond the censoring description. The original sample-size justification and the full statistical analysis plan are also not reported. These gaps limit the ability to assess how closely the posted analysis followed a prespecified strategy or how robust the conclusions are to alternative analytic choices.

16. 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: febrile neutropenia (50.0%), pneumonia (8.7%), sepsis (8.7%), general physical health deterioration (8.7%), pneumocystis jiroveci infection (4.3%).

For the randomized phase, the registry reports serious adverse events in 32 of 87 participants at risk in the GSK1120212 group and 9 of 43 participants at risk in the docetaxel group. These are participant counts, not event rates per unit of follow-up. A participant may experience more than one serious event, so the number of affected participants does not necessarily equal the total number of serious events.

The registry also reports crossover-phase serious adverse events in 12 of 23 participants at risk in the GSK1120212 group and 1 of 2 participants at risk in the docetaxel group. Those groups belong to the crossover phase and should be kept distinct from the randomized-phase comparison. The registry does not provide a basis for combining these phase-specific counts into a single comparative estimate.

General explanation: adverse event counts describe events observed during the recorded study period and do not, by themselves, establish that a treatment caused an event. The registry lists serious adverse event terms, including febrile neutropenia, but the interpretation of individual terms depends on their definitions and denominators. Death counts are not reported in the registry record. These safety counts should be read as descriptive results, not as a formal adjusted comparison between treatments.

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

The primary result describes investigator-assessed progression-free survival in randomized participants with KRAS mutation-positive non-small cell lung cancer. The reported group medians are 11.7 weeks for GSK1120212 and 11.4 weeks for docetaxel. The sponsor-reported hazard ratio is 1.14, with a 95% confidence interval from 0.75 to 1.75. The interval includes 1 and spans relative effects in both directions, so the comparative estimate is uncertain.

The result does not establish that the treatments are equivalent. Showing equivalence would require a suitable design and a prespecified equivalence margin; such a margin is not reported in the registry record. Nor does the result establish a clear advantage for either group. The point estimate alone is not enough to settle the comparison, and the confidence interval shows that quite different relative hazard values remain compatible with the data.

The endpoint combines radiological progression and death from any cause, and progression was assessed by investigators using RECIST, version 1.1. The registry also specifies censoring rules. These features define what the outcome represents, but the record does not report detailed information on event counts, censoring patterns, or additional analyses. Interpretation should remain tied to the recorded population, endpoint, and method rather than extend to outcomes not measured in this result.

The primary population was KRAS-mutant disease. The small subset with BRAF, NRAS, or MEK1 mutations was included for exploratory purposes, so the primary result should not be generalized to those mutation groups. It also should not be treated as a prediction for an individual participant. The trial’s open-label design and investigator-assessed endpoint are relevant context, while randomization supports a comparison of assigned groups within the defined trial population.

In general, a statistical result is one contribution to understanding a treatment comparison, not a clinical recommendation. The registry data describe estimated outcomes and their uncertainty; they do not determine the preferred treatment for a particular person. The posted results are insufficient to draw conclusions about outcomes beyond the specified endpoint or about mutation groups outside the primary KRAS population.

18. Why This Trial Matters Statistically

This trial illustrates why both group summaries and a comparative estimate are needed for time-to-event outcomes. The two reported medians are close, but the hazard ratio and its confidence interval provide a separate comparison that uses the event-time analysis. The interval is broad and includes the reference value of 1, making uncertainty central to interpretation rather than a minor qualification.

It also demonstrates the importance of defining the analysis population. The primary endpoint concerns randomized participants with KRAS-mutant disease, whether or not treatment was administered. The registry separately describes an exploratory subset with other targeted mutations. Keeping those populations distinct prevents the primary result from being applied to groups for which it was not the principal question.

The endpoint definition and censoring rules are essential to understanding what the estimate summarizes. Progression was investigator-assessed under RECIST, version 1.1, and the endpoint included death from any cause. The rules for participants without documented events and for those starting subsequent anti-cancer therapy affect how follow-up contributes to the time-to-event analysis.

Finally, the record shows a difference between what can be reconstructed from summary information and what depends on a sponsor’s time-to-event analysis. A hazard ratio cannot be recovered from the two medians alone. The sponsor-reported log-rank estimate is therefore the relevant comparative result, while the posted confidence interval communicates its limited precision. The original sample-size justification and detailed multiplicity strategy are not reported, underscoring why registry results need to be interpreted alongside the design and analysis information that is actually documented.

19. 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 study was open-label, and the primary endpoint was assessed by the investigator. The registry does not report whether any independent review of progression assessments was performed. This leaves the extent of independent endpoint verification unreported in the registry record; it should not be assumed either to have occurred or not to have occurred.

The primary comparison is limited to the modified intent-to-treat KRAS-mutant population. Although participants with BRAF, NRAS, or MEK1 mutations were included for exploratory purposes, the primary endpoint results do not establish effects in those groups. Details and results for those exploratory populations are not reported in the registry record.

The posted medians and hazard ratio are accompanied by confidence intervals, but the registry information does not include the primary endpoint event count, detailed follow-up distribution, or the number of participants censored. The record specifies censoring rules but does not report sensitivity analyses addressing alternative assumptions. These omissions constrain assessment of the robustness of the time-to-event estimate.

The registry lists 19 secondary outcomes, but a complete numeric result set and multiplicity strategy are not reported in the registry record. Consequently, the overall consistency and evidential weight of secondary findings cannot be established. The crossover phase is mentioned in outcome descriptions and safety reporting, but its design details are not reported in the registry record.

The registered enrollment was 134, while the posted primary analysis lists 86 and 43 participants in the two arms. The registry does not explain this difference in terms of analysis availability or exclusions, so no reason should be inferred. The original sample-size justification and allocation details beyond randomized assignment are also not reported. These limitations restrict conclusions to the posted results and their stated definitions.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: An Open-label Study of GSK1120212 Compared With Docetaxel in Stage IV KRAS-mutant Non-small Cell Lung Cancer — NCT01362296.
• Registry record last updated 2014-07-11; status last verified 2014-06.
• Results first posted 2014-05-26.
• 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.

23. Record Summary

NCT01362296 is a completed, randomized, open-label, parallel phase II study comparing GSK1120212 with docetaxel in second-line locally advanced or metastatic non-small cell lung cancer. The primary efficacy population was randomized participants with KRAS mutation-positive disease, whether or not treatment was administered. GSK1120212 was given orally once daily; docetaxel was given intravenously once every 3 weeks.

The primary endpoint was investigator-assessed progression-free survival from randomization to documented radiological progression or death from any cause. Median progression-free survival was 11.7 weeks for GSK1120212 and 11.4 weeks for docetaxel. The reported hazard ratio was 1.14, with a 95% confidence interval from 0.75 to 1.75. The interval includes 1 and reflects uncertainty in the comparison.

The registry records serious adverse events in both randomized and crossover phases. The original sample-size justification, detailed multiplicity plan, and several details needed to assess censoring and secondary analyses are not reported in the registry record. The posted primary findings should be interpreted for the defined KRAS-mutant population and endpoint.