Erlotinib vs. Standard Chemotherapy in Patients With Advanced Non-small Cell Lung Cancer (NSCLC) and Eastern Cooperative Oncology Group (ECOG)Performance Status (PS) 2: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Erlotinib vs. Standard Chemotherapy in Patients With Advanced Non-small Cell Lung Cancer (NSCLC) and Eastern Cooperative Oncology Group (ECOG)Performance Status (PS) 2 (NCT00085839), a randomized phase 2 study. The primary outcome is Progression-free Survival.

2. Clinical Question

The registry describes the trial as randomized and phase II, with parallel groups and no masking. It lists 103 participants as enrolled. The primary outcome results are reported for the Erlotinib and Standard Chemotherapy arms, with the endpoint population defined by treatment receipt and tumor assessments. The registry also lists overall survival and best tumor response as secondary outcomes.

The clinical question is therefore descriptive as well as comparative: what preliminary progression-free survival estimates were observed in each arm, and how uncertain were those estimates? Randomization assigns participants to treatment groups, but the record’s explicit noncomparative purpose matters when interpreting the results. Arm-level estimates can describe the trial experience without establishing a statistically reliable treatment contrast. The record does not report a hazard ratio or other sponsor-reported comparative estimate for this endpoint.

3. Trial Design

This was a randomized phase II study of erlotinib versus standard chemotherapy in patients with previously untreated advanced non-small cell lung cancer and ECOG performance status 2. The registry lists a parallel-group model and no masking. It describes the study as noncomparative and says its purpose was to obtain preliminary efficacy estimates for erlotinib and standard chemotherapy, while also evaluating erlotinib safety in this patient population.

The record lists 103 participants as enrolled. For the primary outcome, it reports 52 participants in the Erlotinib arm and 51 in the Standard Chemotherapy arm. These are the outcome-specific groups reported with the endpoint results; the primary outcome population was defined as participants who received at least one dose of study drug and had both a baseline and at least one on-treatment tumor assessment.

Erlotinib was administered orally as continuous therapy, with a starting dose of 150 mg/day or reduced dose of 100 mg/day. The chemotherapy regimen consisted of paclitaxel 200 mg/m^2 by intravenous infusion over 3 hours and carboplatin AUC 6 mg/mL x min intravenously over 15 - 30 minutes. Both were given on Day 1 every 21 days for 4 cycles.

The registry lists overall survival and best tumor response as secondary outcomes. It records a primary completion date of 2007-03 and a completed status. The record does not report details of sequence generation, allocation concealment, masking procedures beyond NONE, protocol deviations, or a statistical analysis plan. It also does not report the rationale for the enrollment target or a sample-size justification. Those omissions limit assessment of how the design and analysis were planned beyond the descriptive information in the registry.

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

Analysis population for the primary outcome, as defined in the registry: All patients who received at least 1 dose of study drug and who had both a baseline and at least one on-treatment tumor assessment.

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, measured over Until time of disease progression (maximum 5 months).

The primary endpoint is progression-free survival, reported as a median in months. It measures time until disease progression; the registry defines progression as radiological and/or symptomatic progression, or death in the absence of progression. The stated timeframe is until progression, with a maximum of 5 months.

Time-to-event outcomes are useful when the timing of an event matters and when participants may have different lengths of follow-up. Rather than treating every participant as if they had the same observation period, time-to-event methods account for the event times and, when applicable, censoring: a participant may leave observation without a recorded event by the end of follow-up. This is a general explanation of the endpoint type; the registry record does not describe censoring rules or follow-up procedures in detail.

A median is the time by which half of the relevant event-time distribution has occurred, under the endpoint’s analysis method. It is not the average time for every participant, and it does not describe what happened to each individual. The record reports medians and 95% confidence intervals for both arms. The confidence intervals convey uncertainty around each arm-specific estimate; they are not themselves a direct test of the difference between arms.

The primary outcome population required treatment receipt and tumor assessments at baseline and after treatment began. This definition helps specify whose experience contributes to the reported endpoint, but it also means the primary outcome is not defined simply as everyone enrolled. The record does not report additional details about assessment schedules, independent review, or handling of missing tumor assessments.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Progression-free survival is a time-to-event endpoint: the outcome is the time from an origin point to disease progression or death, as defined in the registry. The posted result is a median for each treatment arm, accompanied by a 95% confidence interval. No hazard ratio is reported in the record.

What does a median convey?

A median time-to-event estimate identifies the point in the estimated event-time distribution by which half of the relevant distribution has experienced the event. It is a summary of timing, not a statement that every participant had an event at the median time. It also differs from a mean, which averages observed times and can be more sensitive to very long values. For time-to-event data, the median is often interpretable even when some participants have not had an event during observation, provided the estimation method and data support an estimable median.

The registry reports 1.91 months for Erlotinib and 3.52 months for Standard Chemotherapy. Those values are estimates, not exact descriptions of each participant’s course. The record does not name the estimator used to calculate the medians, describe the event-time origin in detail, or explain how censoring was handled. Those details matter when evaluating reproducibility and assumptions, so they should not be inferred from the endpoint label alone.

What does the confidence interval add?

A 95% confidence interval expresses sampling uncertainty around an estimate under the statistical method’s assumptions. If the same method were applied repeatedly to comparable samples, the resulting intervals would cover the target parameter at the stated long-run rate. The interval is not a probability statement about the fixed parameter after observing one particular interval, and it does not describe the range of outcomes among individual participants.

The reported intervals are 1.28 to 2.69 months for Erlotinib and 1.48 to 4.73 months for Standard Chemotherapy. Each interval concerns its own arm-specific median. It is not a confidence interval for the difference between arms. A formal comparison would need an estimated contrast and a measure of its uncertainty; the registry record does not report one for progression-free survival.

Why is a direct contrast important?

Separate estimates can differ because of a treatment effect, random variation, or both. A direct comparative analysis uses information from both groups to estimate the contrast of interest and its uncertainty. For a time-to-event endpoint, a hazard ratio is one commonly used comparative measure. It compares event rates over time under a model, rather than subtracting medians. Its interpretation depends on the model and assumptions, including whether the relative event rate is sufficiently stable over time. The registry record does not report a hazard ratio or its assumptions for this endpoint.

Comparing confidence intervals by eye is not a substitute for a formal contrast. Overlap or non-overlap does not directly determine whether the treatment groups differ at a specified significance level. Likewise, a difference between two point estimates is not, by itself, evidence that the underlying treatment effect is known precisely. The posted record supports reporting the arm-specific medians and intervals, but not reconstructing an unreported sponsor comparison.

How should the design shape interpretation?

The trial is randomized, parallel-group, and phase II, but it is explicitly described as noncomparative. Randomization can help balance prognostic factors on average, but a randomized allocation alone does not establish that a study was planned or analyzed to make a definitive treatment comparison. The stated purpose was to obtain preliminary estimates for both treatments. The record does not report a sample-size justification or a full statistical plan, so it is not possible to determine the precision or comparative operating characteristics intended by the investigators.

The sample-size illustrations associated with this analysis are planning exercises based on the observed effect, not the trial’s original design. The stated assumptions include a two-sided significance level of 0.05, equal allocation, and a normal approximation; time-to-event illustrations use total events. Such illustrations indicate hypothetical planning requirements under assumptions. They are not observed power, do not change the study’s design, and should not be read as evidence that the trial had a particular probability of detecting an effect.

What does over-dispersion mean in general?

Over-dispersion is a general statistical issue in count data: the observed variability may exceed what a simple model assumes. If unaccounted for, standard errors and confidence intervals can be too optimistic. This concept does not establish that over-dispersion occurred in this trial. The primary endpoint is time-to-event, and the registry record does not report an over-dispersion analysis for it.

More broadly, statistical methods depend on assumptions about how observations arise, how missingness and censoring operate, and how outcomes are measured. The registry gives the endpoint definition, analysis population, arm-level medians, and confidence intervals, but it does not report enough detail to assess every assumption or reproduce the analysis. The appropriate reading is therefore descriptive: the record contains two estimates with uncertainty, but not a reported comparative model result.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists overall survival and best tumor response as secondary outcomes, in addition to the primary progression-free survival endpoint. It does not report a multiplicity adjustment or a hierarchy for interpreting several outcomes. When multiple outcomes are tested, the chance of at least one apparently notable result can increase; controlling that risk requires a prespecified approach. This is a general statistical concern, not evidence that a particular adjustment was or was not used outside what the registry records.

The record lists no analyses for the secondary outcomes. It therefore does not provide secondary-outcome estimates or inferential results to interpret. The registry also does not report a p-value or a sponsor-reported comparative estimate for the primary endpoint. The reported results support discussion of the arm-specific medians and their confidence intervals, but do not permit a full account of the analysis plan or of how multiple outcomes were handled.

The primary outcome population is defined as participants who received at least one dose of study drug and had both a baseline and at least one on-treatment tumor assessment. The record does not report how many enrolled participants were excluded from this population, reasons for exclusion, or how missing endpoint information was treated. It also does not report whether imputation was used, how censoring decisions were made, or whether sensitivity analyses were conducted.

These details matter because missing observations can affect an estimate if the availability of data is related to prognosis or treatment experience. The direction and magnitude of any such effect cannot be determined from the record. The stated analysis population should therefore be kept in view when interpreting the results, and no assumptions about an intention-to-treat analysis or missing-data method should be made.

11. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: dyspnoea (7.7%), deep vein thrombosis (5.9%), anaemia (3.9%), febrile neutropenia (3.9%), cerebral haemorrhage (3.9%).

The registry reports serious adverse events for the Erlotinib and Standard Chemotherapy groups. It lists 24 affected among 52 at risk in the Erlotinib group and 13 among 51 at risk in the Standard Chemotherapy group. It separately lists other adverse events as 51 among 52 and 50 among 51, respectively. These counts should be read with their recorded group denominators; the record does not provide a comparative safety analysis or confidence intervals for these figures.

The listed serious adverse event terms and percentages are dyspnoea, 7.7%; deep vein thrombosis, 5.9%; anaemia, 3.9%; febrile neutropenia, 3.9%; cerebral haemorrhage, 3.9%; and hypoxia, 3.8%. The registry record does not specify arm attribution for these listed terms in the information summarized. They should not be assigned to a treatment group without that detail.

Adverse-event counts do not necessarily represent distinct events: a participant may experience more than one event, and event-term totals need not equal the number of participants with any serious event. The record does not report event timing, duration, severity grading details, relatedness assessments, or withdrawals due to adverse events. These summaries describe what the registry records, not a complete comparative assessment of safety.

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

The results provide preliminary arm-specific estimates of progression-free survival in patients with previously untreated advanced non-small cell lung cancer and ECOG performance status 2. The median estimates are 1.91 months for Erlotinib and 3.52 months for Standard Chemotherapy. Their respective 95% confidence intervals communicate uncertainty around each estimate. The reported outcome applies to the population defined by treatment receipt and tumor assessments.

The result does not establish that one treatment is superior. The registry describes the study as noncomparative and does not report a hazard ratio, confidence interval for a between-arm effect, or p-value for this endpoint. Separate medians are not a substitute for a direct comparative analysis. The trial’s randomized design is relevant, but it does not remove the need to interpret results in light of the stated purpose and the analysis that is actually reported.

The endpoint is also limited by its stated timeframe of until progression, with a maximum of 5 months. These results do not describe long-term outcomes, and the registry record does not provide further follow-up details. The record lists overall survival and best tumor response as secondary outcomes, but does not list analyses for them.

The safety counts indicate serious adverse events and other adverse events recorded by group, while the listed serious-event terms provide additional context. They do not, without further details, establish comparative risks or explain causality. The registry record does not report the methods needed to assess event attribution, timing, or the uncertainty around those counts.

In short, the posted results are useful as preliminary estimates in the specified population and as a record of reported adverse-event counts. They do not provide a definitive comparative treatment effect, a complete account of longer-term outcomes, or a full analysis of missing data and safety. Conclusions beyond these points would require evidence and methods not reported in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically instructive because it combines random allocation with an explicitly noncomparative purpose. These features are not contradictory: randomization describes how participants were assigned, while the study purpose and analysis determine what conclusions the results can support. The distinction is important when readers encounter arm-specific outcomes in a randomized study and are tempted to treat them as a definitive treatment comparison.

Progression-free survival also illustrates the difference between an endpoint summary and a comparative effect estimate. Medians give a readily interpretable summary of event timing in each group, while confidence intervals show uncertainty around those summaries. Neither set of numbers directly estimates the difference between groups. A treatment contrast needs its own estimate and uncertainty, and the registry does not report a hazard ratio or other direct comparison for the primary endpoint.

The analysis population is another important statistical detail. The primary outcome applies to participants who received at least one dose and had both baseline and on-treatment tumor assessment. That definition helps readers understand which observations contribute, but the record does not report how many participants were outside that population or how missing assessments were handled. Transparent reporting of populations and missingness is essential for evaluating whether an outcome estimate reflects the enrolled group.

Finally, the study’s preliminary purpose and the absence of reported secondary analyses demonstrate why registry records should be read with attention to scope. They provide useful information about the question, design, endpoint definition, and posted estimates, but do not necessarily contain the full statistical plan or every analysis needed to support a comparative conclusion. The result is most informative when the estimate, its uncertainty, and its limits are considered together.

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 statistical analysis plan, the rationale for the enrollment target, or a sample-size justification. It also does not report a sponsor-reported hazard ratio or formal comparative test for progression-free survival. This limits assessment of whether the study was planned to estimate a treatment contrast and prevents interpretation of a direct comparative effect from the posted results.

The primary endpoint population required at least one dose and both baseline and on-treatment tumor assessment. The record does not state how many enrolled participants met that definition, how participants without the required assessments were handled, or whether any sensitivity analyses were performed. Missing-data methods and censoring rules are also not reported.

The endpoint timeframe is stated as until progression, with a maximum of 5 months. The registry does not give further details on assessment timing, follow-up completeness, or whether progression was assessed by independent review. These omissions matter for judging how consistently and completely the event times were established.

Although the registry lists secondary outcomes of overall survival and best tumor response, it records no analyses for them. It also does not report multiplicity procedures. Safety reporting includes group-level counts and a list of serious adverse event terms, but does not give the term-level arm attribution in the information summarized, nor does it report event timing, relatedness, or uncertainty estimates.

These limitations are specific to the details available in the registry record. They do not show that a procedure was absent; they mean that its use is not reported in the registry record. Interpretation should therefore remain close to the arm-specific endpoint estimates, the stated population, and the recorded safety counts.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Erlotinib vs. Standard Chemotherapy in Patients With Advanced Non-small Cell Lung Cancer (NSCLC) and Eastern Cooperative Oncology Group (ECOG)Performance Status (PS) 2 — NCT00085839.
• Registry record last updated 2012-08-09; status last verified 2012-08.
• Results first posted 2011-04-22.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

The registry record describes a completed, randomized, parallel-group phase II study with no masking, comparing erlotinib with standard chemotherapy in previously untreated advanced non-small cell lung cancer and ECOG performance status 2. The stated purpose was to obtain preliminary efficacy estimates for both arms and evaluate erlotinib safety.

The primary endpoint was median progression-free survival, with progression defined as radiological and/or symptomatic progression or death in the absence of progression. The reported medians were 1.91 months for Erlotinib and 3.52 months for Standard Chemotherapy, with 95% confidence intervals of 1.28 to 2.69 months and 1.48 to 4.73 months, respectively. No comparative hazard ratio or formal test is reported for this endpoint.

The record lists overall survival and best tumor response as secondary outcomes but records no analyses for them. It also reports group-level adverse-event counts and serious-event terms. Details of the statistical plan, missing-data handling, censoring, and comparative safety analysis are not reported in the registry record.