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 An Observational Study of Tarceva (Erlotinib) in Routine Daily Clinical Practice as Second Line Treatment in Patients With Non-small Cell Lung Cancer (NCT01161173), a registered phase not stated study. The primary outcome is Percentage of Participants With a Best Overall Response of Complete Response (CR), Partial Response (PR), Stable Disease (SD), or Progressive Disease (PD).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01161173 |
| Phase | Phase not stated |
| Condition | Nonsquamous Nonsmall Cell Neoplasm of Lung |
| Allocation | Not Stated |
| Design model | Not Stated |
| Masking | Not Stated |
| Enrollment | 347 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Hoffmann-La Roche |
| Results posted | Yes |
2. Clinical Question
Population
Patients with advanced, recurrent or metastatic non-small cell lung cancer (NSCLC) who had received one course of standard systemic chemotherapy, experienced disease progression, and started second-line erlotinib. The primary endpoint population is described as per-protocol: enrolled participants who started erlotinib and did not violate the study protocol. Only participants with an evaluable response were included in that endpoint analysis.
Intervention
Erlotinib, also known by the brand name Tarceva, was used in routine clinical practice. The investigator determined the dose, guided by the product-characteristic recommendation; the recommended daily dose was 150 mg orally once daily.
Comparator
No comparator group is described. The study was non-interventional and the registry record does not state an allocation method. Its design therefore does not provide a like-for-like control group for comparing erlotinib with another treatment or with no treatment.
Primary question
Among participants treated with second-line erlotinib, what was the best overall response category—complete response, partial response, stable disease, or progressive disease—under RECIST v1.1? The registry record identifies this endpoint and its assessment period, but does not report the category counts, percentages, or confidence intervals. The study also lists outcomes concerning disease progression, progression-free survival, overall survival, lung cancer symptoms, and rash.
3. Trial Design
This was a non-interventional observational study of erlotinib in routine clinical practice, rather than a randomized treatment comparison. The registry record describes patients with advanced NSCLC who had progressed after one course of standard systemic chemotherapy and received erlotinib in the second-line setting. Participants could be followed through third-line treatment if their disease progressed while receiving erlotinib.
The registry lists an enrollment of 347. That enrollment figure should not be treated as the number included in every endpoint analysis: the primary endpoint population was restricted to participants who started erlotinib, did not violate the protocol, and had an evaluable response. The number meeting those criteria is not reported in the registry record.
The investigator determined erlotinib dosing, guided by the recommendation in the Summary of Product Characteristics. The recommended daily dose was 150 mg orally once daily. Because treatment was delivered in routine practice and there is no stated comparator, the record describes experience during treatment rather than an experiment that isolates a treatment effect from other influences.
The study status is completed. Its listed start was 2008-04, and its primary completion was 2012-08. The primary endpoint timeframe was baseline to the end of the study, up to 4 years, 4 months. The registry lists six outcome measures, including the primary response endpoint and secondary outcomes involving time to disease progression, progression-free survival, overall survival, Lung Cancer Symptom Scale scores, and rash. The registry record does not report a randomized allocation, a control arm, or a statistical analysis plan.
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 |
|---|---|
| STARTED | 347 |
| COMPLETED | 0 |
| NOT COMPLETED | 347 |
Reasons for not completing this period:
| Reason | Erlotinib |
|---|---|
| Protocol Violation | 13 |
| Lost to Follow-up | 7 |
| Progressive Disease | 82 |
| Withdrawal of Consent | 5 |
| Unacceptable Toxicity | 21 |
| Death | 194 |
| End of Study | 2 |
| Reasons Not Specified | 23 |
Analysis population for the primary outcome, as defined in the registry: Per-protocol population: All enrolled participants who started erlotinib therapy and did not violate the study protocol. Only participants with an evaluable response were included in the 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: Per-protocol population: All enrolled participants who started erlotinib therapy and did not violate the study protocol.
| Characteristic | Unit | Cohort |
|---|---|---|
| Participants analysed | 334 | |
| Age, Continuous | years | 65 (36 to 86) |
| Sex: Female, Male: Female | Participants | 95 |
| Sex: Female, Male: Male | Participants | 239 |
6. Primary Endpoint
The registry lists the primary outcome as Percentage of Participants With a Best Overall Response of Complete Response (CR), Partial Response (PR), Stable Disease (SD), or Progressive Disease (PD), measured over Baseline to the end of the study (up to 4 years, 4 months).
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | Percentage of participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint classifies each evaluable participant by their best overall tumor response from baseline to the end of the study, up to 4 years, 4 months. The listed categories are complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD), assessed using RECIST v1.1. The endpoint is expressed as the percentage of participants in each response category.
Under the registry’s definitions, CR means disappearance of all target lesions or all non-target lesions. PR requires at least a 30% decrease in the sum of the longest diameters of target lesions, relative to baseline. SD means the findings meet neither the shrinkage criterion for PR nor the increase criterion for PD. PD requires at least a 20% increase in the target-lesion sum relative to its smallest recorded value since treatment started, or new lesions or unequivocal progression of existing non-target lesions. A CR or PR was considered confirmed if a subsequent RECIST evaluation also showed CR or PR.
In general, a standardized response classification can make tumor assessments more consistent by applying stated criteria rather than relying solely on an unstructured description. The registry record identifies the criteria, but reports neither the response-category estimates nor their 95% confidence intervals. It also does not give the number of participants with evaluable response, so the precision and denominator of the primary endpoint cannot be assessed from the record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies a clinically interpretable response outcome: each evaluable participant’s best overall response was classified using RECIST v1.1. However, it does not report how many participants had CR, PR, SD, or PD, nor does it report the corresponding percentages or 95% confidence intervals. The endpoint’s definition is available, but its observed result is not. No response rate, magnitude of response, or precision estimate can therefore be described from the registry record.
The listed enrollment is 347, but that is not necessarily the denominator for the primary endpoint. The stated analysis population includes only participants who started erlotinib, did not violate the study protocol, and had an evaluable response. The number meeting these conditions, the number excluded, and the reasons for missing or unevaluable responses are not reported in the registry record. Consequently, the enrollment figure alone cannot establish how many participants contributed to the response estimate.
The study is observational and has no comparator described. The findings, if available, would characterize outcomes among people treated with erlotinib in routine practice; without a comparison group and an analysis addressing differences among patients, they would not by themselves separate treatment effects from disease characteristics, prior treatment, other care, or selection into treatment and follow-up. In general, this is a limitation of causal interpretation in non-randomized evidence, not a claim about what happened to any particular participant.
The registry lists serious adverse events affecting 210 participants among 347 at risk. It also lists several serious adverse event terms with percentages, including disease progression at 24.5%, death at 5.5%, general physical health deterioration at 5.5%, rash at 5.2%, diarrhoea at 4.6%, and dyspnoea at 3.5%. These are safety descriptions, not a comparison with another treatment. The registry record does not establish that an event was caused by erlotinib, and the listed term “Death” should not be interpreted as a treatment-attributed death count.
The record lists secondary outcomes but does not report their results in the information available. It also lists zero analyses. Statistical tests, adjustments, missing-data handling, and analysis details beyond the endpoint population are not reported in the registry record. Thus the most defensible reading is descriptive: the study followed patients treated in practice and defined outcomes, but the record does not provide the primary response estimates needed to assess the observed response distribution or its statistical precision.
9. Statistical Methods Explained
The primary outcome is a binary percentage endpoint in the sense that each participant either contributes to a particular response category or does not. The categories themselves—CR, PR, SD, and PD—are mutually distinct classifications of best overall response under the stated RECIST v1.1 definitions. To interpret a percentage, a reader needs both the count in the category and the denominator of participants assessed. The registry record does not report those primary endpoint counts or percentages, nor the number with evaluable response.
What does a confidence interval convey?
In general, a confidence interval expresses the statistical uncertainty around an estimate under the assumptions of the method used. For a percentage, its width depends in part on the number of observations contributing to that estimate and on the observed proportion. A narrower interval indicates greater precision under that method; it does not establish that an estimate is unbiased, nor does it resolve limitations in study design. The primary endpoint specifies a 95% confidence interval, but the registry record does not report the estimated percentages or intervals. Their width and the precision of the observed response results therefore cannot be evaluated.
How should response categories be understood?
RECIST v1.1 supplies operational criteria for classifying tumor response. For PR, the record specifies at least a 30% decrease in the sum of the longest diameters of target lesions from baseline. For PD, it specifies at least a 20% increase from the smallest recorded sum since treatment started, or the appearance of new lesions or unequivocal progression of existing non-target lesions. SD is defined by not meeting either the shrinkage criterion for PR or the increase criterion for PD. CR requires disappearance of all target lesions or all non-target lesions. These categories summarize assessed disease status; they are not, by themselves, a comparison of treatment strategies.
The registry also states that a CR or PR was “confirmed” when a subsequent RECIST evaluation also showed CR or PR. This definition makes confirmation dependent on a later assessment. The record does not give the timing or completeness of assessments, the number of confirmed responses, or whether any participants lacked the subsequent evaluation needed for confirmation. Those details are not reported in the registry record.
What does the observational design imply statistically?
In general, an observational study records care as it occurs rather than assigning treatment through randomization. Results can describe outcomes in the treated group, but comparisons across treatment groups—if groups exist—may be distorted by differences in prognosis, disease status, care choices, or follow-up. No comparator is described in this record, so there is no stated between-group treatment contrast to interpret. The dose was selected by the investigator, guided by the product-characteristic recommendation, which is consistent with treatment in clinical practice rather than a fixed dose assigned by the study.
Without a comparator, an observed response percentage would not show what the same participants’ outcomes would have been under another treatment or no treatment. In general, that unobserved alternative is the counterfactual needed for a causal treatment-effect interpretation. The registry record does not report an adjusted causal analysis or a statistical plan that would address such a question.
What can be inferred from the listed analysis information?
The registry lists zero analyses. This means the record does not provide reported analysis results in the analysis listing; it does not establish that no analyses were performed outside the registry record. No hypothesis tests, p-values, model specifications, adjustment variables, interim analyses, or multiplicity procedures are reported in the registry record. The primary endpoint’s stated 95% confidence interval is a measure of intended reporting, not evidence that a particular interval was calculated or available.
The registry describes the primary endpoint population as per-protocol: participants had to start erlotinib and not violate the protocol, and only those with evaluable response were included. In general, restricting analysis to evaluable participants can make the estimate dependent on who remains assessable. If assessment availability is related to health status or response, the resulting percentage may not represent all enrolled participants. Whether that occurred, and how missing response assessments were handled, are not reported in the registry record.
How should planning illustrations be interpreted?
In general, sample-size calculations describe the amount of information a hypothetical study might need under specified assumptions; they do not report an observed treatment effect or demonstrate power achieved by a completed study. The registry’s planning assumptions describe a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and a target effect expressed as a stated fraction of an observed effect. Those assumptions are illustrations and do not describe this study’s design, which has no comparator arm stated. The registry record does not report a protocol-based sample-size justification. Any sample-size values based on such planning assumptions should therefore be read as planning scenarios, not as an observed-power result or evidence that this observational study was designed as a two-arm trial.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists six outcome measures: the primary best overall response endpoint and secondary outcomes concerning time to disease progression, progression-free survival, overall survival, change from baseline in Lung Cancer Symptom Scale scores, and the percentage of participants who developed rash. The record identifies these outcomes, but does not report their analysis results in the information available.
When several outcomes or comparisons are assessed, the chance of at least one apparently unusual result can increase if each is tested separately. In general, multiplicity procedures can control or describe that issue, but the appropriate approach depends on the questions and analysis plan. The registry record does not report whether a multiplicity strategy was planned or used, how outcomes were prioritized beyond identifying the primary endpoint, or whether any interim analyses took place. The registry lists zero analyses; this does not establish whether analyses occurred outside the record.
The primary response population is described as per-protocol and limited to participants with an evaluable response. The registry does not report the number in this population, the number without an evaluable response, reasons for missing assessments, or any imputation or sensitivity analyses. The extent and potential consequences of missing endpoint data therefore cannot be assessed. For the listed secondary outcomes, analysis populations and missing-data procedures are also not reported in the registry record.
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 | 210 | 347 | 60.5% |
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 | 0 of 347 |
Most frequent serious adverse event terms
| Event term | Organ system | Erlotinib |
|---|---|---|
| Disease progression | General disorders | 85 of 347 |
| Death | General disorders | 19 of 347 |
| General physical health deterioration | General disorders | 19 of 347 |
| Rash | Skin and subcutaneous tissue disorders | 18 of 347 |
| Diarrhoea | Gastrointestinal disorders | 16 of 347 |
| Dyspnoea | Respiratory, thoracic and mediastinal disorders | 12 of 347 |
| Pneumonia | Infections and infestations | 7 of 347 |
| Anorexia | Metabolism and nutrition disorders | 7 of 347 |
| Hypoxia | Respiratory, thoracic and mediastinal disorders | 5 of 347 |
| Fatigue | General disorders | 4 of 347 |
Serious events with the highest arm-level rate: disease progression (24.5%), death (5.5%), general physical health deterioration (5.5%), rash (5.2%), diarrhoea (4.6%).
The registry lists 347 participants at risk for serious adverse events and records 210 as affected. These figures describe the safety information recorded for the erlotinib group; they do not provide a comparison with an untreated or alternative-treatment group. The record also lists serious adverse event terms and their reported percentages: disease progression, 24.5%; death, 5.5%; general physical health deterioration, 5.5%; rash, 5.2%; diarrhoea, 4.6%; and dyspnoea, 3.5%.
These term-specific percentages should be read as registry-reported safety summaries, not as proof that erlotinib caused the events. In particular, a listed “Death” term does not establish that the deaths were attributed to treatment. The record does not report exposure time, event timing, severity details beyond the serious-event classification, or a comparator-based safety contrast. It also does not explain how overlapping event terms were counted, so the term percentages should not be added together. In general, adverse-event summaries describe events recorded during a study; causal attribution requires evidence beyond the presence of an event after treatment.
12. What the Result Does, and Does Not, Mean
The registry record establishes that a completed non-interventional study followed patients receiving erlotinib as second-line treatment for advanced NSCLC in routine practice. It specifies the primary endpoint, the RECIST v1.1 response categories, a per-protocol evaluable-response population, and a set of secondary outcomes. It also records serious adverse event information. These elements describe the questions and data structure of the study, but they do not supply the primary outcome estimates.
Without reported counts, percentages, or confidence intervals for CR, PR, SD, and PD, the registry record does not show the observed response distribution or how precisely it was estimated. Enrollment of 347 does not substitute for the missing endpoint denominator, because the primary analysis includes only participants who started treatment, did not violate the protocol, and had an evaluable response. The number who met those conditions is not reported.
The observational design also limits what an outcome can establish. In general, outcomes among treated patients do not by themselves show what would have happened to comparable patients under a different treatment. No comparator is described, and the record does not report an analysis designed to address confounding or estimate a causal effect. The study can be understood as documenting experience in routine care, rather than as a randomized comparison of treatment options.
The serious adverse event summaries likewise describe recorded events, not treatment causation. The registry does not establish attribution for the listed terms and does not provide a comparator for judging relative safety. Missing information about exposure duration, event timing, endpoint completeness, and statistical procedures further limits interpretation. The appropriate conclusion is therefore about the scope of the record: it defines outcomes and reports some safety counts, while the primary efficacy estimates and many analysis details are not reported in the registry record.
13. Why This Trial Matters Statistically
This study is statistically informative as an example of the distinction between defining an endpoint and reporting an estimate for it. The registry specifies a best overall response endpoint, its RECIST v1.1 criteria, a stated confidence-interval type, and a per-protocol population. Yet the response-category counts, percentages, and confidence intervals are not reported in the registry record. A well-defined endpoint is essential for reproducibility, but without its observed estimate and denominator, readers cannot evaluate the result’s magnitude or precision.
The record also illustrates why analysis populations matter. Although enrollment is listed as 347, the response analysis was restricted to participants who started erlotinib, had no protocol violation, and had an evaluable response. Since the size and composition of that subset are not reported, the enrollment figure cannot be used as the response denominator. In general, the relationship between the enrolled population and the analyzed population affects how broadly an estimate can be interpreted.
Finally, the non-interventional design clarifies the difference between describing outcomes and estimating a treatment effect. In general, routine-practice data can document what was observed in a treated population, while causal comparisons require a credible comparator and methods suited to addressing differences between groups. No comparator is described. The registry’s safety summaries are similarly descriptive and do not establish causation. Together, these features make the record a useful case for reading endpoint definitions, analysis populations, uncertainty, and study design as separate parts of statistical interpretation.
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 main trial-specific limitation is that the registry record does not report the primary response-category results, even though it defines the endpoint and specifies a 95% confidence interval. Without the observed estimates, their denominators, and interval limits, the response findings cannot be evaluated quantitatively. The record’s enrollment figure of 347 does not resolve this gap because the endpoint population was limited to protocol-compliant participants who started erlotinib and had an evaluable response.
The number of participants in that analysis population, exclusions, and reasons for unavailable or unevaluable assessments are not reported in the registry record. The completeness and timing of RECIST assessments, including follow-up assessments for confirmation of CR or PR, are also not reported. These omissions prevent assessment of how outcome ascertainment or missing data may have influenced the response summary.
The study is observational, and no comparator group or allocation method is described. Investigators determined the dose, guided by the product-characteristic recommendation. As a result, the record does not offer a randomized contrast, and it does not report methods for adjustment for prognostic differences or other potential confounding. The registry also does not report the statistical analysis plan, tests, model details, multiplicity procedures, interim analyses, or secondary outcome estimates.
For safety, the registry reports participants at risk and affected and gives selected serious adverse event terms, but does not report exposure time, timing, attribution, or comparative risks. These limitations mean that the recorded events cannot be interpreted as treatment-caused or compared against an alternative treatment using this record alone.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: An Observational Study of Tarceva (Erlotinib) in Routine Daily Clinical Practice as Second Line Treatment in Patients With Non-small Cell Lung Cancer — NCT01161173.
- Registry record last updated 2016-03-28; status last verified 2016-02.
- Results first posted 2016-03-28.
- 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
NCT01161173 is a completed, non-interventional study of erlotinib (Tarceva) as second-line treatment for advanced NSCLC in routine clinical practice. The registry lists enrollment of 347, a start of 2008-04, and primary completion in 2012-08. Investigators determined the dose, guided by the recommended daily dose of 150 mg orally once daily.
The primary endpoint was the percentage of participants in each best overall response category under RECIST v1.1, among protocol-compliant participants who started treatment and had an evaluable response. The registry record does not report response-category estimates, confidence intervals, or the size of this analysis population. It lists secondary outcomes involving disease progression, survival, symptoms, and rash, but does not report their results in the information available.
Serious adverse events were recorded for 210 participants among 347 at risk, with selected event terms and percentages listed. The observational design has no comparator described, and the record does not establish event causation or report a statistical analysis plan.