BATTLE Program: Erlotinib in Previously Treated Subjects With Advanced NSCLC: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for BATTLE Program: Erlotinib in Previously Treated Subjects With Advanced NSCLC (NCT00410059), a na phase 2 study. The primary outcome is 8 Week Progression-Free Survival Rate (i.e. Disease Control Rate).

2. Clinical Question

In plain terms, the study was intended to assess whether disease remained controlled over the specified early treatment period among previously treated participants receiving erlotinib. It also aimed to study safety, effects on cells in the body, and overall response, according to the brief summary. The primary endpoint is not a comparison between erlotinib and another treatment: without a comparator arm, the posted design cannot directly estimate a relative treatment effect against concurrent control participants.

The endpoint description defines disease control as being progression-free at 8 weeks after randomization, while the study is recorded as a single-group study with no allocation. That wording does not establish a randomized comparison. How the wording was applied in the conduct of this study is not reported in the registry record. The record also does not give the number of participants who met the endpoint, the resulting rate, or an uncertainty interval. Therefore, the clinical question can be described from the planned endpoint, but its numerical answer cannot be determined from the posted results.

3. Trial Design

This was a Phase II, open-label study sponsored by M.D. Anderson Cancer Center. The registry record describes a single group, with no allocation and no masking. It lists 59 enrolled participants and marks the study as completed. The recorded start is 2006-11 and primary completion is 2016-07. The study title and summary identify previously treated participants with advanced non-small cell lung cancer.

Participants received erlotinib, with the arm description specifying 150 mg by mouth daily for 28 days. The registry record lists one primary outcome measure, the 8-week progression-free survival rate, also called the disease-control rate. Its timeframe is radiographic evaluation after cycle 2, described as 8 weeks of therapy. The endpoint description also says efficacy confirmation occurs 2 cycles after the initial assessment. Whether those confirmation assessments were completed, and how they were incorporated into the reported outcome, is not reported in the registry record.

Because the study is single-group and has no allocation, its design is not a randomized comparison. The absence of masking means the study is open-label, so participants and study personnel were not masked to treatment assignment. The registry record does not report further details about outcome-assessor masking. It also does not report a protocol-level account of recruitment, eligibility criteria, treatment discontinuation, follow-up procedures, or the planned statistical analysis.

The recorded enrollment is not itself the number analyzed for the primary endpoint: the record specifically notes that one participant was not evaluable for that outcome. It does not state the endpoint denominator or give an analysis population definition. Thus, enrollment, evaluability, and the number contributing to any analysis should not be treated as interchangeable. The record identifies a completed study, but completion status alone does not supply an endpoint estimate or demonstrate that every planned assessment was available.

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: One participant was not evaluable for outcome.

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 8 Week Progression-Free Survival Rate (i.e. Disease Control Rate), measured over Radiographic evaluation after cycle 2 (8 weeks of therapy).

The primary endpoint is the 8-week progression-free survival rate, described in the registry record as the disease-control rate. It is defined as the percentage of participants without progression at 8 weeks, assessed radiographically after the second cycle of therapy. In the endpoint wording, a participant is a success or has disease control if progression-free at 8 weeks after randomization. The record also says efficacy is to be confirmed 2 cycles after the initial assessment.

This endpoint focuses on whether progression has occurred by a specified assessment time. As a binary outcome, each evaluable participant would be classified according to whether the stated condition was met; the resulting rate would describe the share meeting that endpoint in the analyzed group. That is a general explanation of a binary endpoint, not a reported result for this study. The registry record gives no count of participants classified as progression-free, no endpoint rate, and no confidence interval.

A fixed-time disease-control endpoint can provide a clinically interpretable snapshot at the scheduled assessment. It does not, by itself, describe the full timing of progression or establish how outcomes compare with another treatment. The endpoint description refers to progression-free survival, but the recorded measure is a count of participants and a rate at 8 weeks, rather than a posted time-to-event analysis. The registry record does not report the operational details used to assess progression, the handling of death or missing radiographic assessments, or how the later confirmation requirement affected classification. One participant was not evaluable for the outcome.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a single-group design estimate?

In a single-group study, the observed endpoint is described within the participants who received the study intervention. If an endpoint rate is reported, it summarizes the proportion meeting the endpoint under the study’s assessment and analysis rules. It does not directly compare outcomes with a concurrent untreated group or another treatment group. Any comparison with a historical expectation would require a clearly stated benchmark and assumptions; no such benchmark or comparative analysis is reported in the registry record.

A between-group treatment effect is ordinarily built from a contrast between groups. The registry record describes one group, no allocation, and no masking. It therefore does not identify a randomized comparison or a control arm. A single-group result can be useful as a description of observed outcomes, but it cannot alone rule out other explanations for those outcomes. This is a general design principle, not a claim that any particular alternative explanation occurred in this trial.

How should the binary endpoint be understood?

The primary outcome is a binary classification at a defined assessment: progression-free at 8 weeks or not meeting that definition. A rate would summarize the proportion classified as meeting the endpoint among the participants included in the analysis. The record gives the endpoint definition but does not give the numerator, the analysis denominator, or the resulting rate. It also notes one participant was not evaluable, without stating how that participant was handled analytically.

For a binary rate, uncertainty depends in part on how many participants contribute and how the outcome is distributed. A confidence interval is a range of values compatible with the observed data under a specified statistical method; it communicates uncertainty around an estimate rather than guaranteeing that the true value lies in the range. This is general statistical explanation. Since no rate or interval is reported for the primary endpoint, its precision cannot be assessed from the registry record.

What does the assessment time imply?

The primary measure is a fixed-time status at 8 weeks, based on radiographic evaluation after cycle 2. This differs from an analysis that uses the time until progression or another event across follow-up. A fixed-time endpoint asks whether a participant meets the condition at the specified assessment; a time-to-event analysis uses event timing and may account for differing observation periods through its methods. The registry record lists a count-based primary measure and does not report a time-to-event analysis.

The endpoint description also calls for confirmation of efficacy 2 cycles after the initial assessment. Confirmation can be relevant when a classification may be affected by the timing or interpretation of assessments. The registry record does not report how confirmation was operationalized, whether all participants had the later assessment, or whether the confirmed assessment or the initial assessment determined the primary classification. These details matter because different rules can yield different endpoint classifications.

What can be said about statistical planning?

The registry record does not report a protocol-based sample-size justification, a planned precision target, or a formal power calculation for the trial. A sample-size illustration based on an observed effect is not the same as the trial’s original design justification. Such illustrations depend on assumptions about the effect to detect and the statistical procedure. They are planning aids, not observed power and not evidence that the study was designed to detect a particular effect.

The planning assumptions described for the illustration use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect expressed as a stated fraction of the observed effect. This illustration is distinct from the registry’s single-group design. It should not be mistaken for the study’s allocation plan or for a direct analysis of the trial. The registry record does not report the trial’s own sample-size rationale or statistical plan, and no primary endpoint estimate is posted from which to assess the observed effect.

The assumptions also refer to a two-arm allocation, whereas the trial is recorded as single-group with no allocation. That distinction is essential: a planning calculation built around two arms does not describe how participants were assigned in this study. Any calculated participant requirements under such assumptions would remain hypothetical planning quantities, not observed enrollment findings or proof of achieved power.

Why does over-dispersion matter in general?

Over-dispersion is a general statistical issue that arises when observed variability exceeds what a model assumes. If ignored, estimated uncertainty can be too small and intervals or tests can be misleading. It can be relevant for count or rate models in which outcomes vary more than the chosen probability model allows. The primary endpoint in this registry record is described as a binary participant count at a fixed time, and no model or over-dispersion assessment is reported. It would therefore be inappropriate to claim that over-dispersion was present or that a particular adjustment was used.

What information is needed to assess the analysis?

Interpretation of a reported endpoint would require the observed count, the analysis denominator, the handling of the unevaluable participant, the outcome-assessment rules, and an uncertainty estimate with its method. It would also be useful to know whether the analysis followed a prespecified plan and how deviations or missing assessments were treated. These are general considerations for evaluating a clinical trial analysis. The registry record does not report these details for the primary endpoint, nor does it list any primary-outcome analysis. Without them, the record supports description of the intended measure but not an assessment of its estimated magnitude or statistical precision.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Erlotinib")
n      <- c(58)
events <- c(20)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Erlotinib  58  20  34.5  22.5  48.1  23.6  47.3  23.6  47.4

11. Multiplicity, Missing Data and Analysis Populations

Multiplicity concerns arise when several outcomes, comparisons, time points, or analyses are examined. When many statistical tests are conducted, the chance of at least one apparently notable result can increase unless the analysis plan accounts for the testing structure. This is a general statistical principle. The registry record lists one primary outcome measure, but it does not report a statistical analysis plan, a hierarchy of outcomes, multiplicity adjustments, or whether any interim analyses were conducted. The number of listed primary measures alone does not establish the full set of analyses performed.

Missing data also affect interpretation. The registry record says one participant was not evaluable for the primary outcome, but it does not state why, whether that participant had a missing assessment, or how the participant was treated in analysis. It does not report the number of other missing or incomplete assessments, if any. No imputation method, sensitivity analysis, or missing-data assumption is reported. The endpoint’s later confirmation requirement adds another assessment step, but the record does not describe missing confirmation information or its effect on classification.

The analysis population is not defined in the posted results. The enrollment count is 59, but the registry record does not give the denominator used for the primary endpoint or identify an intention-to-treat, per-protocol, or other analysis population. These labels have specific meanings and should not be assigned without information about eligibility, treatment received, follow-up, and analysis rules. The record also does not state whether exclusions occurred beyond noting one participant was not evaluable.

These omissions limit reproducibility and the ability to judge possible bias. They do not establish that an inappropriate method was used; they mean the method and its consequences are not reported in the registry record. No inference about missingness, multiplicity control, or analysis-population choices should be made without those details.

12. Safety Results

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

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: melena (1.7%).

The registry record lists 59 participants at risk in the erlotinib group for the serious-event summary and records 1 participant with a serious adverse event. Melena is listed among the leading serious adverse event terms at 1.7%. The record also lists 59 participants at risk for other adverse events and 57 affected participants. These are descriptive counts and a reported percentage, not a comparison with another treatment group.

A participant count with an event is not the same as an event count: one participant may experience more than one event, depending on how events were collected and summarized. The posted figures do not give event-level detail, timing, severity grades, relatedness assessments, or the complete set of terms. The single listed serious term should not be treated as a full description of the safety profile. The registry record does not report deaths in this summary.

Because the study has one group and no comparator, these figures cannot establish whether event frequencies differ from those under another treatment or from background risk. They describe the recorded safety information for the study group only. The record does not provide the additional detail needed to assess duration of exposure or how adverse events were attributed to treatment.

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

The record establishes that a completed Phase II, open-label, single-group study enrolled 59 previously treated participants with advanced non-small cell lung cancer and evaluated erlotinib. The planned primary measure was disease control at 8 weeks, defined through progression-free status at radiographic assessment after cycle 2. One participant was not evaluable for that outcome. These details describe the study and its intended endpoint; they do not establish how many participants met the endpoint.

The primary endpoint result is not reported in the registry record. There is no posted disease-control rate, count of participants meeting the definition, confidence interval, or analysis result. As a result, the magnitude and precision of the primary outcome cannot be assessed. Completion status does not fill that gap, and the enrollment count cannot be used as a substitute for an outcome estimate.

Even if a rate were available, the single-group design would limit what it could establish. An observed rate would describe outcomes in this study group under the recorded treatment and assessment conditions. Without a concurrent comparator, it would not by itself show how outcomes would have differed under another treatment or no study treatment. This is a general interpretation of uncontrolled designs, not a conclusion about the efficacy of erlotinib.

The safety record gives descriptive counts, including 1 participant with a serious adverse event and a leading serious term of melena at 1.7%. Those entries do not provide a complete safety analysis or a comparative risk estimate. More detailed information about event timing, attribution, duration of exposure, and the full range of events is not reported in the registry record.

The defensible conclusion is therefore limited: the registry records the study design, treatment regimen, endpoint definition, and selected safety figures, but it does not report the primary endpoint analysis. No statement about whether treatment controlled disease, or about how outcomes compare with another option, follows from the available result information.

14. Why This Trial Matters Statistically

This study is statistically instructive because its endpoint sounds comparative in places, while the registered design is single-group. The description defines success as being progression-free at 8 weeks after randomization, yet the record lists no allocation and a single group. The language should be read as part of the endpoint wording, not as evidence that participants were randomized. Distinguishing endpoint language from design information is necessary before interpreting any reported result.

It also illustrates the difference between an endpoint definition and an endpoint finding. The registry record precisely identifies an 8-week assessment and a progression-free criterion, but does not provide the observed count or rate. A clear measurement plan does not replace the result needed to estimate disease control, and neither enrollment nor study completion can supply that estimate.

The unevaluable participant highlights why denominators and analysis populations matter for binary rates. A rate requires knowing which participants were eligible for classification and how missing assessments were handled. The registry record reports one participant as not evaluable but does not specify the analysis denominator or the handling rule. A reader therefore cannot reconstruct the endpoint estimate, even if the definition is understood.

Finally, the planning assumptions described for an illustration refer to equal allocation between two arms, while the registered study is single-group. That contrast is a useful reminder that planning calculations are conditional on their assumptions and are not descriptions of the trial’s actual design. Sample-size planning also does not establish observed power or substitute for a posted effect estimate with an uncertainty interval.

These points matter beyond this study. Clear design labels, analysis denominators, endpoint results, and uncertainty estimates are all needed to distinguish what a clinical trial intended to measure from what its data show and how precisely they show it.

15. 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 offers limited information for evaluating the trial’s primary result. Although the primary endpoint is defined, no result or analysis is reported. The record also does not give a confidence interval, analysis denominator, or endpoint count. One participant was not evaluable, but the reason and handling of that participant are not reported in the registry record. These omissions prevent reconstruction of the primary rate and assessment of its precision.

The design is single-group and open-label, with no allocation and no masking. Without a concurrent comparator, the study cannot directly estimate a comparative effect from the registered design. Open-label conduct may also affect some study procedures or assessments in general, though the record does not report whether outcome assessors were masked or whether any such influence occurred. No specific bias should be presumed from the design alone.

The endpoint language includes a reference to randomization even though the study is registered as single-group with no allocation. The registry record does not explain this wording. It also does not report how the required efficacy confirmation 2 cycles after the initial assessment was handled, or whether confirmation information was complete. The relationship between the initial and confirmatory assessment is therefore unclear.

Further limitations arise from unreported methods: eligibility criteria and recruitment detail, the statistical analysis plan, interim analyses, multiplicity handling, missing-data methods, analysis populations, and protocol deviations are not reported in the registry record. The listed adverse-event information is also limited and does not provide full event-level details or a comparator. These are reporting limitations; they do not prove that the corresponding procedures were absent from the conduct of the study.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: BATTLE Program: Erlotinib in Previously Treated Subjects With Advanced NSCLC — NCT00410059.
• Registry record last updated 2017-08-02; status last verified 2017-07.
• Results first posted 2017-08-02.
• 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.

19. Record Summary

The registry record describes NCT00410059, a completed Phase II, open-label, single-group study sponsored by M.D. Anderson Cancer Center. It enrolled 59 previously treated participants with advanced non-small cell lung cancer. The intervention was erlotinib, 150 mg by mouth daily for 28 days. The recorded start is 2006-11 and primary completion is 2016-07.

The primary endpoint was the 8-week progression-free survival rate, also described as disease-control rate, assessed radiographically after cycle 2. One participant was not evaluable. The registry record does not report the endpoint result or an analysis. Its safety summary lists 1 participant with a serious adverse event, melena at 1.7% as a leading serious term, and 57 participants affected by other adverse events. The design has no concurrent comparator, and the record does not report a primary endpoint estimate or uncertainty interval.