This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Erlotinib Versus Pemetrexed as Second-Line Therapy in Treating Patients With Advanced Lung Adenocarcinoma (NCT01565538), a randomized phase 2 study. The primary outcome is Progression-Free Survival.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01565538 |
| Phase | Phase 2 |
| Condition | Lung Cancer |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 123 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Si-Yu Wang |
| Results posted | Yes |
2. Clinical Question
Population
People with advanced lung adenocarcinoma described in the trial record as having EGFR wild-type and EGFR FISH-positive disease. The study concerns second-line treatment after standard chemotherapy has failed. The registry record lists lung cancer as the condition and gives no further detail about the primary-outcome analysis population.
Intervention
Erlotinib, described as 150 mg orally once a day continually until progression. The trial was randomized, so treatment assignment was intended to allocate participants between the two study arms.
Comparator
Pemetrexed, described as 500mg/m2 by IV infusion every 3 weeks until progression. The record identifies pemetrexed as the other second-line option being studied.
Primary question
How did progression-free survival compare between the erlotinib and pemetrexed arms? Progression-free survival was defined from randomization to tumour progression or death from any cause. The posted arm-specific medians are 4.1 months for erlotinib and 3.9 months for pemetrexed. No comparative hazard ratio or p-value is reported in the registry record, so the posted figures do not establish a statistical comparison between arms.
3. Trial Design
This was a randomized phase II trial with a parallel-arm design and no masking. The two treatment arms were erlotinib and pemetrexed. The registry record lists enrollment of 123 and reports primary-outcome arm counts of 61 for erlotinib and 62 for pemetrexed. It does not report how allocation was generated or whether any allocation-concealment procedures were used.
The clinical question was framed around second-line treatment of advanced lung adenocarcinoma with EGFR wild-type and EGFR FISH-positive status. The registry summary describes uncertainty about the role of EGFR mutation status and EGFR gene copy number in selecting EGFR-targeted therapy. The trial compared treatment strategies in this specified population; it was not a comparison across all lung cancer types or all EGFR status groups.
Erlotinib was given at 150 mg orally once a day continually until progression. Pemetrexed was given at 500mg/m2 by IV infusion every 3 weeks until progression. The registry record does not describe additional treatment rules, adherence measures, permitted subsequent therapies, or procedures for treatment discontinuation beyond these descriptions.
The record gives a start date of 2008-12, a primary completion date of 2012-05, and a completed status. It lists best tumor response and overall survival as secondary outcomes, alongside progression-free survival as the primary outcome. Specific posted results for the secondary outcomes are not included in the record information. The protocol’s planned sample-size justification and detailed statistical analysis plan are not reported in the registry record.
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
| Stage | Erlotinib | Pemetrexed |
|---|---|---|
| STARTED | 61 | 62 |
| COMPLETED | 61 | 62 |
| NOT COMPLETED | 0 | 0 |
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.
| Characteristic | Unit | Erlotinib | Pemetrexed | Total |
|---|---|---|---|---|
| Participants analysed | 61 | 62 | 123 | |
| Age, Continuous | years | 54.3 (30 to 74) | 55.1 (33 to 75) | 54.7 (30 to 75) |
| Sex: Female, Male: Female | Participants | 21 | 23 | 44 |
| Sex: Female, Male: Male | Participants | 40 | 39 | 79 |
| ECOG PS: 0,1 | participants | 57 | 59 | 116 |
| ECOG PS: 2 | participants | 4 | 3 | 7 |
| Smoking status: Never | participants | 15 | 17 | 32 |
| Smoking status: Former | participants | 7 | 5 | 12 |
| Smoking status: Current | participants | 39 | 40 | 79 |
| Stage: IIIB | participants | 4 | 6 | 10 |
| Stage: IV | participants | 40 | 38 | 78 |
| Stage: Recurrent | participants | 17 | 18 | 35 |
6. Primary Endpoint
The registry lists the primary outcome as Progression-Free Survival, measured over From the date of randomization to the date of tumour progression or death from any cause, assessed until at least 12 months after randomization..
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | 95% Confidence Interval |
| Unit | months |
| Endpoint classification | Time-to-Event Analysis |
The primary endpoint was progression-free survival, reported as a median in months with a 95% confidence interval. It measures the time from randomization until tumour progression or death from any cause. The registry specifies that progression was defined using RECIST v1.0: a 20% increase in the sum of the longest diameter of target lesions, a measurable increase in a non-target lesion, or the appearance of new lesions. Assessment was stated to continue until at least 12 months after randomization.
Progression-free survival is a time-to-event endpoint. It can describe how long participants remain alive without documented progression, and it incorporates both progression and death rather than treating either event as a separate competing outcome. The median is the time by which half of the relevant event-time distribution has occurred, as estimated by the analysis method. It is not an average survival time and does not describe every participant’s experience.
The 95% confidence interval expresses uncertainty around each arm’s median estimate. It does not by itself provide a direct test of the difference between arms. The registry record does not report a comparative hazard ratio, a p-value for the primary comparison, or the method used to estimate the medians.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary results give a median progression-free survival of 4.1 months for erlotinib, with a 95% confidence interval from 1.6 to 6.6 months. For pemetrexed, the median is 3.9 months, with a 95% confidence interval from 2.7 to 5.1 months. These are arm-specific estimates. The confidence intervals describe uncertainty in those estimates, rather than the range of outcomes for individual participants.
The figures indicate that the estimated medians are close in the registry’s reported units, but a visual or numerical comparison of medians is not a substitute for a time-to-event treatment comparison. In particular, a median alone does not use the timing of every event in the same way that a hazard-ratio analysis does, and it does not show how the event patterns compare over the full follow-up period. A comparative hazard ratio and its uncertainty are not reported in the registry record. A primary-comparison p-value is also not reported.
The confidence intervals have some overlap. That observation should not be used as a formal significance test: overlap or non-overlap of separate confidence intervals is not itself the reported test of the between-arm contrast. Without a posted comparative estimate and its uncertainty, the registry results do not establish whether the difference between treatment arms is statistically distinguishable from no difference. They also do not establish equivalence or non-inferiority, which require appropriate prespecified margins and analyses; those are not reported in the registry record.
There are 61 participants listed for the erlotinib outcome and 62 for pemetrexed. Those counts provide context for the estimates, but the registry record does not state the number of progression or death events, the number censored, the completeness of follow-up, or the analysis population definition. Without those details, the precision and potential influence of incomplete observation cannot be assessed beyond the reported confidence intervals.
The trial is identified as phase II, randomized, parallel, and unmasked. Randomization is a design feature intended to support a fair comparison, while the absence of masking can matter for outcomes whose assessment or management may be influenced by treatment knowledge. Progression was defined by RECIST v1.0, but the registry record does not report whether response assessments were centrally reviewed or performed by investigators unaware of treatment assignment.
The registry also lists best tumor response and overall survival as secondary outcomes, but results for those outcomes are not reported in the record information available. The primary finding should therefore be read narrowly: it is a pair of median progression-free survival estimates with confidence intervals, not a complete account of all clinical outcomes or a definitive comparative-effect estimate.
9. Statistical Methods Explained
The posted primary endpoint is time-to-event data. Each participant’s time begins at randomization, and the event is the first recorded tumour progression or death from any cause. Participants who have not had a recorded event by their last known assessment may be censored in a time-to-event analysis. Censoring allows participants with different follow-up durations to contribute information up to the point at which their event status is no longer known. The registry record does not report how censoring was handled or the number of censored participants.
What does a median represent?
A median time-to-event estimate marks the point in the estimated event-time distribution at which half of participants are estimated to have experienced the event. It is different from a mean: the median does not add together all observed times and divide by the number of participants. It also does not imply that each participant experienced the event near the median. In these results, the medians are 4.1 months for erlotinib and 3.9 months for pemetrexed.
Time-to-event estimates commonly account for censoring through methods such as the Kaplan–Meier estimator. That is a general explanation of a common approach, not a claim about the method used in this trial. The registry record does not name the estimator used for the reported medians. The confidence intervals are reported, but the record does not describe how they were calculated.
How should the confidence intervals be read?
A confidence interval communicates statistical uncertainty around an estimate under the assumptions of the analysis method. The reported 95% confidence interval for erlotinib is 1.6 to 6.6 months, and for pemetrexed it is 2.7 to 5.1 months. These intervals are not guarantees that the true value lies within the limits, nor do they show the spread of individual participants’ times. The interval for an arm-specific median also is not a confidence interval for the difference between treatment arms.
For a randomized time-to-event comparison, a hazard ratio is one commonly used summary. In general, a hazard ratio compares the event rates over time between groups, subject to the model and assumptions used. A hazard ratio is not the same as a ratio of medians. The registry record does not report a hazard ratio or its confidence interval, and no comparative p-value is posted for the primary endpoint. Consequently, there is no sponsor-reported comparative effect to interpret from these registry results.
What does over-dispersion mean in general?
Over-dispersion is a general issue in count-data models: the observed variability in counts exceeds what a simple model assumes. It can make standard errors too small if ignored, leading to confidence intervals that are too narrow and tests that appear more decisive than the data warrant. This concept does not describe the primary endpoint in this trial, which is time-to-event, and the registry record does not report a count-data model for the primary analysis.
The record’s sample-size illustrations are described as planning calculations based on the observed effect, using a two-sided significance level of 0.05, equal allocation, and a normal approximation. For time-to-event outcomes, the illustrations concern a total event count. They are explicitly planning illustrations rather than observed power or a report of the trial’s original design justification. The original sample-size justification is not reported in the registry record. These illustrations cannot show the probability that the trial actually detected an effect, and they do not replace the missing comparative result.
The registry record lists zero analyses in its analysis count. That does not mean that no statistical work was performed; it means the record does not post an analysis entry in that part of the registry. The methods used for the primary analysis, the handling of missing data and censoring, and any adjustments or stratification are not reported in the registry record.
Why do event timing and follow-up matter?
Time-to-event methods use event timing and can incorporate participants whose follow-up ends before an event is recorded. Two trials, or two treatment groups, can have similar medians but different event patterns at other times. Conversely, an apparent difference in medians may not summarize the whole survival experience. The number and timing of events, censoring patterns, and a comparative analysis are therefore important for interpreting a time-to-event result. The registry record does not provide those details beyond the arm-specific medians, confidence intervals, and stated assessment timeframe.
In general, proportional-hazards methods rely on an assumption that the relative event rate between groups is reasonably stable over time. If that assumption is not appropriate, a single hazard ratio may be difficult to interpret. This is general statistical context only: the registry record does not state that a proportional-hazards model was used, whether its assumptions were checked, or whether alternative analyses were planned.
10. Multiplicity, Missing Data and Analysis Populations
The record lists progression-free survival as the primary outcome and best tumor response and overall survival as secondary outcomes. When a study examines multiple outcomes, multiple time points, or multiple analysis choices, the chance of at least one apparently notable result can increase. General statistical approaches may address multiplicity through a prespecified hierarchy, adjustment, or a clearly designated primary analysis. The registry record does not report a multiplicity strategy, an outcome hierarchy beyond the primary and secondary labels, or results for the listed secondary outcomes.
The record gives an assessment timeframe for progression-free survival of at least 12 months after randomization, but it does not report the actual follow-up distribution or the timing of individual events. It also does not state how participants without a documented event were handled, how missing assessments were addressed, or whether sensitivity analyses were conducted. These details are important because incomplete follow-up and missing outcome assessments can affect time-to-event estimates, depending on their pattern and the analysis assumptions.
The analysis population is not specified in the registry record. In particular, the record does not say whether the primary result follows an intention-to-treat approach, a per-protocol approach, or another population definition. The arm counts are reported, but the record does not explain whether all randomized participants were included in the primary analysis or whether exclusions occurred.
Detailed statistical methods, including any prespecified interim analyses, are not reported in the registry record. No interim analyses should be inferred from the phase designation or the posted results. The registry record also does not report a plan for adjustment for multiple comparisons or for missing data. These omissions limit assessment of how the reported estimates relate to the complete planned analysis, but they do not by themselves demonstrate that any particular method was or was not used.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Erlotinib | 0 | 61 | 0.0% |
| Pemetrexed | 0 | 62 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Erlotinib | 45 of 61 |
| Pemetrexed | 43 of 62 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Erlotinib | Pemetrexed |
|---|---|---|---|
| Rash | Skin and subcutaneous tissue disorders | 33 of 61 | 4 of 62 |
| Fatigue | General disorders | 12 of 61 | 16 of 62 |
| Nausea | Gastrointestinal disorders | 1 of 61 | 15 of 62 |
| Anorexia | Gastrointestinal disorders | 6 of 61 | 9 of 62 |
| Diarrhea | Gastrointestinal disorders | 10 of 61 | 2 of 62 |
| Insomnia | Psychiatric disorders | 4 of 61 | 3 of 62 |
| Constipation | Gastrointestinal disorders | 0 of 61 | 4 of 62 |
| Infection | Infections and infestations | 3 of 61 | 1 of 62 |
| Headache | Nervous system disorders | 1 of 61 | 3 of 62 |
| Dry skin | Skin and subcutaneous tissue disorders | 3 of 61 | 0 of 62 |
The registry reports zero participants affected by serious adverse events in the erlotinib arm, among 61 at risk, and zero in the pemetrexed arm, among 62 at risk. These are the posted serious-event figures. They should be read as counts of affected participants as recorded, not as proof that no serious events occurred under any definition or outside the registry’s reporting period. The record does not list any serious adverse-event terms.
For the other adverse-event category, the record lists 45 affected participants among 61 at risk for erlotinib and 43 among 62 at risk for pemetrexed. These are counts, not a comparison adjusted for follow-up or participant characteristics. The registry record does not provide event-specific descriptions in the information, nor does it specify the severity threshold or ascertainment procedures for this category.
Deaths in the adverse-event results are not reported in the registry record. The safety counts do not establish whether the treatments differ in safety: a comparative analysis, event definitions, timing, and fuller clinical context are not reported. These figures describe what the registry posts and should not be treated as medical advice or as a complete safety assessment.
12. What the Result Does, and Does Not, Mean
The registry result supports a limited descriptive statement: the reported median progression-free survival was 4.1 months in the erlotinib arm and 3.9 months in the pemetrexed arm, with the respective 95% confidence intervals of 1.6 to 6.6 months and 2.7 to 5.1 months. These estimates apply to the trial’s stated population of advanced lung adenocarcinoma with EGFR wild-type and EGFR FISH-positive status, not automatically to other populations.
The estimates do not establish that one treatment is superior, that the treatments are equivalent, or that either treatment has a particular clinical effect. The registry record does not report a comparative hazard ratio, confidence interval for a between-arm effect, or primary-comparison p-value. The arm-specific medians and their confidence intervals cannot replace those missing comparative results.
The randomized design is relevant because random allocation can reduce systematic differences between treatment groups at baseline. However, the record does not report allocation procedures, baseline characteristics, analysis population, censoring details, or whether outcome assessment was blinded. The trial was unmasked, and the record does not describe independent or blinded review of progression. These factors affect how fully the design and result can be evaluated from the registry information.
The posted safety counts likewise have a narrow interpretation. Zero serious adverse-event-affected participants are recorded in each arm, while other adverse events are recorded for 45 participants in the erlotinib arm and 43 in the pemetrexed arm. The record does not provide enough information to compare safety comprehensively, including event types, exposure duration, or a comparative analysis.
The appropriate conclusion is therefore about the scope of the posted evidence: arm-specific progression-free survival medians and safety counts are available, while a formal comparative effect and key analysis details are not reported in the registry record. Secondary outcomes are listed, but their results are not reported in the registry record information.
13. Why This Trial Matters Statistically
This trial raises several statistical issues that recur in randomized clinical studies. Its primary endpoint is time-to-event, so interpreting the result requires attention not only to the medians but also to event timing, censoring, follow-up, and the comparative analysis. Reporting arm-specific medians with confidence intervals gives useful descriptive information, yet it does not fully answer the randomized comparison question without an effect estimate for the contrast and its uncertainty.
The trial also illustrates the distinction between a design label and a complete analysis plan. Randomized phase II status describes important features of the study, but it does not reveal how allocation was implemented, how many events accrued, what analysis population was used, or whether a multiplicity strategy was specified. Those details are not reported in the registry record and should not be inferred from the trial’s phase or completed status.
The population definition matters statistically as well as clinically. The study focused on advanced lung adenocarcinoma described as both EGFR wild-type and EGFR FISH-positive. Results from this selected group cannot be assumed to apply to other biomarker groups or broader lung cancer populations without evidence. The registry summary itself notes uncertainty about the role of EGFR mutation status and gene copy number as markers in this setting.
Finally, the posted results show why precise reporting matters. Medians and confidence intervals can be interpreted descriptively, but a comparative hazard ratio, its uncertainty, event counts, and analysis details would be needed to assess the treatment contrast more completely. Planning illustrations based on the observed effect are not the same as the trial’s original sample-size justification and are not observed power. Keeping those distinctions clear prevents descriptive summaries from being overstated as definitive comparative evidence.
14. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The registry record does not report a comparative hazard ratio or a primary-comparison p-value for progression-free survival. It also does not give the number of progression or death events, the number censored, or detailed follow-up information. These omissions prevent a fuller assessment of the time-to-event comparison and the amount of information supporting each estimate.
The primary analysis method, confidence-interval method, censoring rules, analysis population, and handling of missing assessments are not reported in the registry record. The record does not describe allocation procedures, baseline balance, adherence, treatment exposure, or subsequent therapies. As an unmasked trial, it also leaves unanswered whether progression assessments were centrally reviewed or assessed without knowledge of treatment assignment.
The registry lists best tumor response and overall survival as secondary outcomes, but their results are not reported in the record information. Serious adverse-event terms are not listed, and deaths in the adverse-event results are not reported. The other adverse-event counts are not accompanied by event-specific details or a comparative safety analysis.
The record’s planning illustrations are based on the observed effect and are not a substitute for the trial’s original sample-size justification. That justification and a detailed statistical analysis plan are not reported in the registry record. The result should therefore be interpreted as a limited registry report of arm-specific outcomes, rather than a complete publication-level account of trial conduct and analysis.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Erlotinib Versus Pemetrexed as Second-Line Therapy in Treating Patients With Advanced Lung Adenocarcinoma — NCT01565538.
- Registry record last updated 2014-09-15; status last verified 2014-09.
- Results first posted 2014-09-15.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT01565538 is a completed, randomized, unmasked, parallel-arm phase II trial of erlotinib versus pemetrexed as second-line therapy for advanced lung adenocarcinoma described as EGFR wild-type and EGFR FISH-positive. Enrollment was 123. The registry lists a start date of 2008-12 and primary completion in 2012-05.
The primary endpoint was progression-free survival from randomization to tumour progression or death from any cause, assessed until at least 12 months after randomization. Median progression-free survival was 4.1 months (95% confidence interval, 1.6 to 6.6) for erlotinib and 3.9 months (95% confidence interval, 2.7 to 5.1) for pemetrexed. A comparative hazard ratio and primary-comparison p-value are not reported in the registry record.
The registry lists best tumor response and overall survival as secondary outcomes. It records zero serious adverse-event-affected participants in each arm and reports other adverse events for 45 of 61 participants receiving erlotinib and 43 of 62 receiving pemetrexed. Detailed analysis methods, analysis populations, and several outcome and safety details are not reported in the registry record.