A Study of Erlotinib in Participants With Locally Advanced or Metastatic Non-Small Cell Lung Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Erlotinib in Participants With Locally Advanced or Metastatic Non-Small Cell Lung Cancer (NCT01667562), a non-randomized phase 3 study. The primary outcome is Progression-Free Survival as Assessed by Response Evaluation Criteria in Solid Tumors (RECIST) Version 1.1 (v 1.1).

2. Clinical Question

The study also specified secondary outcomes concerning objective response, disease control, EGFR mutations, adverse events, and quality of life measured with the FACT-L instrument. These outcomes address different aspects of tumor response, safety, and participant-reported health; they do not substitute for the primary time-to-event outcome.

The clinical question is therefore descriptive for this treated group rather than a randomized estimate of comparative benefit. Without a concurrent control group, the study cannot isolate the effect of erlotinib from other influences on observed outcomes. The registry record describes the treatment and endpoint framework, but it does not report the primary result needed to characterize the observed progression-free survival duration.

3. Trial Design

This was a phase IIIb, open-label, multi-center study of erlotinib in participants with locally advanced or metastatic non-small cell lung cancer and activating EGFR mutations. The registry lists a planned enrollment of 375 and records the study as completed. It lists a start date of 2012-01-20 and a primary completion date of 2017-09-07.

The design was non-randomized and single-group, with no masking. Participants were screened for EGFR mutations during a diagnostic phase; those without a mutation were excluded from the study. Eligible participants received erlotinib as a single daily oral dose of 150 milligrams, continuing until disease progression, death, or unacceptable toxicity. The record does not describe a concurrent control group or a randomized allocation process.

These design features shape what can be learned from the results. A single-group study can describe outcomes observed among participants who received the treatment, but it cannot make a randomized comparison with another treatment or with no treatment. Open-label treatment means that the record does not describe masking of participants or investigators. For an endpoint based on tumor assessments, the record states that RECIST version 1.1 was used; whether assessments were independently reviewed is not reported in the registry record.

The registry states that the intent-to-treat population included all enrolled participants, and that no patients were excluded from efficacy or safety analysis. This describes the stated analysis population, but the registry record does not report detailed participant flow or the number contributing to each individual outcome. It also does not report a statistical analysis plan, sample-size justification, or details about interim analyses. Those items cannot be inferred from the planned enrollment or from the study’s phase.

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: 375 participants were in the diagnostic population. Out of the 375 participants, only 30 participants had EGFR mutation and were eligible for erlotinib treatment.

Diagnostic Phase

Reasons for not completing this period:

Erlotinib Phase

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The intent-to-treat population included all participants enrolled in the study. There were no patients excluded from analysis either in efficacy and safety analysis.

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.

Population: The intent-to-treat (ITT) population included all participants enrolled in the study.

6. Primary Endpoint

The registry lists the primary outcome as Progression-Free Survival as Assessed by Response Evaluation Criteria in Solid Tumors (RECIST) Version 1.1 (v 1.1), measured over Baseline until disease progression, or death, whichever occurs first (approximately up to 4 years and 9 months).

The primary endpoint was progression-free survival, assessed by RECIST version 1.1. It measures the time from baseline until disease progression or death from any cause, whichever occurs first. Participants who had not progressed or died by the final analysis were censored at their date of last contact, according to the registry description.

For this endpoint, the record defines progression as at least a 20% increase in the sum of target-lesion diameters, relative to the smallest sum observed on study, together with an absolute increase of at least 5 mm. This operational definition helps make the progression event more consistent across participants and assessment visits. Death is also counted as an event, so the endpoint includes both radiographic progression and death rather than tumor measurements alone.

A Kaplan–Meier estimate was used to define the median as the time by which half of the participants had experienced progression or death. The stated dispersion measure was a 95% confidence interval. In general, a confidence interval conveys uncertainty around an estimated quantity; it is not a range containing a known fixed proportion of individual participant outcomes. The registry record gives the endpoint definition and method, but the median progression-free survival estimate and its confidence interval are not reported in the registry record.

Progression-free survival is commonly used in oncology trials because it incorporates the timing of progression as well as death, and can be analyzed even when some participants have not yet experienced either event at the cutoff. Its meaning depends on assessment procedures, follow-up, and censoring. Details beyond the definitions stated in the registry record are not reported.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the median progression-free survival represent?

Progression-free survival is a time-to-event outcome. For each participant, the event is the first of disease progression or death. The registry specifies that progression was assessed using RECIST version 1.1 and gives the lesion-growth criteria used to define it. The analysis uses the time from baseline to the event, rather than simply counting how many participants had progressed at one chosen visit.

The Kaplan–Meier method is a general approach for estimating the distribution of time to an event when some participants have not experienced the event by the end of their observed follow-up. Such participants may contribute information until their last known contact and then be censored. The registry describes censoring participants who had not died or progressed at final analysis at the date of last contact. Censoring does not mean that an event occurred; it means the event time was not observed during the participant’s recorded follow-up.

The median is the point in the estimated time-to-event distribution at which half of participants have experienced the event. It is a group summary and does not imply that each participant’s experience is close to that value. The registry record describes the median and its 95% confidence interval as the primary reporting approach, but reports neither the estimate nor the interval. Without these values, neither the estimated duration nor its statistical precision can be assessed.

How should the confidence interval be understood?

In general, a confidence interval expresses uncertainty arising from sampling and estimation under the statistical method used. A narrower interval indicates greater precision than a wider interval, while an interval’s endpoints describe uncertainty about the group-level estimate rather than a range of individual outcomes. The registry record identifies a 95% confidence interval for the median, but its endpoints are not reported.

Confidence intervals should be considered alongside study design and data completeness. A precise estimate in a single-group study would still be a description of the treated group, not a randomized estimate of the difference between treatments. Precision does not remove confounding or create a comparator that the design did not include.

What can a single-group study estimate?

With one treatment group, outcome summaries can describe what happened among participants enrolled under the study’s eligibility and follow-up conditions. They do not by themselves estimate what would have happened to comparable participants under another treatment. Randomization is a general method for supporting fair comparisons by distributing measured and unmeasured baseline factors across groups; the registry identifies this study as non-randomized and single-group, so that type of comparison is not available.

The open-label design means masking was absent. For a tumor-based endpoint, standardized criteria such as RECIST can help define events consistently, but they do not establish that outcome assessment was blinded or independently reviewed. Whether either safeguard was used is not reported in the registry record. The record also does not describe the timing or frequency of tumor assessments, so the implications of assessment schedules for the recorded event times cannot be evaluated.

What do the planning illustrations mean?

General sample-size calculations depend on assumptions about the effect to detect, the statistical test, allocation, and the desired power. The planning illustrations described for this study use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the target effect is expressed as a stated fraction of the observed effect. For time-to-event planning, the calculations refer to total events, while other calculations refer to participants per arm.

Those calculations are planning illustrations based on the observed effect, not the study’s own design justification and not observed power. Their two-arm equal-allocation assumptions do not describe this study’s non-randomized, single-group design. The registry record does not report the study’s original sample-size justification or a statistical plan, so the illustrations cannot be treated as evidence about the design rationale or the trial’s actual power.

Why is the hazard-ratio note important?

A hazard ratio is a relative comparison of event rates over follow-up, ordinarily requiring groups or another defined reference for comparison. The record’s methodological note says that only sponsor-reported hazard ratios are used for time-to-event analyses. However, no primary hazard ratio or comparison result is included in the registry record for this analysis. A median time and a hazard ratio answer different questions: the median describes a point in one group’s event-time distribution, while a hazard ratio describes a relative rate in a comparison. Neither can be substituted for the other.

For this record, the primary outcome is specified as a median time-to-event measure, and the key limitation is that its numerical result is absent. General statistical concepts can clarify how the analysis would be interpreted, but they cannot recover the unreported estimate, interval, or comparison.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists several secondary outcomes alongside progression-free survival: objective response, disease control, EGFR mutations, adverse events, and quality of life measured with FACT-L. When a study examines multiple outcomes, the chance of obtaining at least one apparently notable result can rise if each is tested separately without adjustment. Methods for addressing multiplicity may include prespecifying a hierarchy or adjusting statistical thresholds, but the registry record does not report whether any such approach was planned or used. No multiplicity adjustment should be assumed.

The primary outcome description explains that participants without progression or death at final analysis were censored at last contact. This is a stated time-to-event rule, not a complete account of missing-data handling for all outcomes. The registry record does not report the amount or pattern of missing data, reasons for missingness, sensitivity analyses, or how missing quality-of-life assessments were handled. It also does not report details of follow-up completeness.

The stated intent-to-treat population included all participants enrolled, and the record says no patients were excluded from efficacy or safety analysis. This is the registry’s description of the analysis population. It does not establish that every outcome was observed for every participant or that every participant contributed an event-time measurement. Outcome-specific denominators and participant flow are not reported in the registry record.

The record lists 0 analysis entries. It does not describe an analysis plan, interim analyses, or prespecified statistical procedures beyond the primary endpoint definition and Kaplan–Meier median. Consequently, multiplicity control, detailed missing-data methods, and the handling of analysis deviations are not reported in the registry record.

11. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: rash (10.0%), paronychia (3.3%), death (3.3%), haemoptysis (3.3%).

The Erlotinib safety summary lists 30 participants at risk and 6 affected for serious adverse events. For other adverse events, it lists 30 at risk and 22 affected. It also lists 1 death. These figures are reported as separate safety summaries; they should not be added together because the categories may not be mutually exclusive, and the record does not explain their relationship. The relationship between these denominators and the study’s overall enrollment is not reported in the registry record.

The listed serious adverse-event terms are rash at 10.0%, paronychia at 3.3%, death at 3.3%, and haemoptysis at 3.3%. The registry record does not state the denominators for these percentages in the information available for this analysis, nor does it explain whether terms may overlap within participants. The listed terms therefore cannot be used to reconstruct a complete participant-level safety profile.

Adverse-event summaries describe events recorded under the study’s reporting procedures. They do not, by themselves, establish that an event was caused by the treatment. Definitions, attribution procedures, duration, and event-level detail beyond the posted summaries are not reported in the registry record. These limitations are especially important when interpreting the small safety denominators alongside a larger planned enrollment.

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

The registry record establishes that this was an open-label, non-randomized, single-group study of erlotinib in participants with locally advanced or metastatic non-small cell lung cancer and activating EGFR mutations. It specifies a daily oral dose of 150 milligrams, a treatment duration guided by progression, death, or unacceptable toxicity, and a primary endpoint based on progression-free survival under RECIST version 1.1.

The record does not report the numerical median progression-free survival estimate or its 95% confidence interval. It therefore does not support a quantitative statement about how long participants remained free of progression or death, or how precisely that duration was estimated. The absence of these values in the registry record should not be interpreted as a finding of no effect or no events.

Even if outcome estimates were available, the single-group design would limit causal interpretation. There is no randomized comparator in the record, so an observed outcome could not establish what would have occurred under another treatment or without erlotinib. A response or survival summary in one group is not itself a treatment-effect comparison.

The safety summaries provide limited descriptive information, with denominators of 30 for the serious and other adverse-event groupings and listed serious terms with percentages. The record does not explain how these denominators relate to total enrollment or give the detail needed to characterize all safety outcomes. The figures should be interpreted within those limits rather than generalized beyond the recorded summaries.

In general, a clinical trial record can clarify how a study was designed and which outcomes were intended, while leaving important questions unanswered if numerical results or analysis details are absent. Here, the endpoint definition and design are available, but the primary efficacy estimate, its uncertainty, and detailed analysis methods are not. Conclusions about comparative efficacy, precision of the primary result, and the full safety profile cannot be drawn from the registry information described.

13. Why This Trial Matters Statistically

This study illustrates why trial design and outcome reporting must be read together. A clear endpoint definition is necessary to understand what is being measured, but the definition alone is not a result. The registry describes progression-free survival using RECIST version 1.1 and states that the median was estimated with Kaplan–Meier methods and a 95% confidence interval. Without the estimate and interval, readers cannot assess the observed event-time distribution or its precision.

The single-group, non-randomized design also makes the scope of inference important. Such a design can describe outcomes in the enrolled treatment group, but it does not provide the randomized comparison needed to attribute differences between groups to treatment. The record does not list a concurrent control arm. Recognizing this distinction prevents a descriptive outcome from being mistaken for comparative evidence.

The record further shows why analysis populations and denominators matter. It describes an intent-to-treat population with no exclusions from efficacy or safety analysis, yet the safety summary gives denominators of 30 for its listed groupings. The relationship between those safety figures and the overall enrollment is not reported. Without outcome-specific denominators and participant flow, the scope of the safety summaries is difficult to assess.

Finally, the record identifies secondary outcomes but does not report their results or the approach to multiplicity, missing data, and interim analyses. This leaves the statistical interpretation incomplete. Planning illustrations based on an observed effect are not observed power and do not replace a documented design rationale. Taken together, the record is useful for understanding the intended question and design, while also demonstrating that adequate interpretation depends on posted results and analysis details, not endpoint labels alone.

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 principal limitation for interpreting efficacy is that the registry record does not report the primary median progression-free survival estimate or its 95% confidence interval. Although the endpoint definition and Kaplan–Meier method are described, the outcome’s magnitude and precision cannot be assessed from the record.

The study was non-randomized, open-label, and single-group. It therefore does not offer a like-for-like control comparison, and an outcome observed among participants receiving erlotinib cannot alone identify a treatment effect. The record does not report independent or blinded review of tumor assessments, assessment frequency, or other procedures that could clarify how progression was determined.

Although the record states that all enrolled participants were included in efficacy and safety analysis, it does not report detailed participant flow or outcome-specific denominators. The safety groupings list 30 at risk in each category, while the planned enrollment is 375; how those safety denominators relate to enrollment is not reported in the registry record. The record also does not explain the denominators for the percentages attached to serious adverse-event terms.

Other limitations include the absence of reported numerical results for the listed secondary outcomes, and the lack of information about missing-data methods, multiplicity control, interim analyses, and a detailed statistical analysis plan. The registry does not report the original sample-size justification. These gaps restrict assessment of the completeness and robustness of the reported analyses. They do not establish that any particular procedure was or was not used; those details are not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of Erlotinib in Participants With Locally Advanced or Metastatic Non-Small Cell Lung Cancer — NCT01667562.
• Registry record last updated 2019-12-20; status last verified 2019-12.
• Results first posted 2019-12-20.
• 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

NCT01667562 is a completed phase IIIb, open-label, multi-center, non-randomized, single-group study sponsored by Hoffmann-La Roche. It enrolled participants with locally advanced or metastatic non-small cell lung cancer and activating EGFR mutations. The record lists planned enrollment of 375, a start date of 2012-01-20, and primary completion on 2017-09-07.

Participants received erlotinib as a single daily oral dose of 150 milligrams until disease progression, death, or unacceptable toxicity. The primary endpoint was progression-free survival by RECIST version 1.1, summarized as a Kaplan–Meier median and 95% confidence interval. The median estimate and interval are not reported in the registry record.

The record states that the intent-to-treat population included all enrolled participants, with no exclusions from efficacy or safety analysis. It lists secondary outcomes and limited safety summaries, but does not report numerical results for the primary endpoint or the listed secondary outcomes in the information available for this analysis.