A Study of Sunitinib in Combination With Bevacizumab, Carboplatin, and Paclitaxel in Patients With Advanced Non-Small Cell Lung Cancer (SABRE-L): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Sunitinib in Combination With Bevacizumab, Carboplatin, and Paclitaxel in Patients With Advanced Non-Small Cell Lung Cancer (SABRE-L) (NCT00434226), a randomized phase 2 study. The primary outcome is Best Response.

2. Clinical Question

The question concerns response observed over follow-up, rather than the original planned primary outcome of progression-free survival. The registry states that progression-free survival could not be reported because data were insufficient. It does not report response counts by arm, an estimated between-arm effect, or a statistical test of the response comparison. Accordingly, the clinical question can be described from the trial design and endpoint definition, but the registry record does not establish how the response outcomes differed between the randomized groups.

3. Trial Design

This was a randomized, controlled, open-label, multicenter, parallel-group Phase II study. It enrolled 56 participants and is listed as terminated. The registry gives a start date of 2007-03 and a primary completion date of 2008-04. The sponsor was Genentech, Inc. The registry record does not report the reason for termination.

The two named groups were bevacizumab with carboplatin and paclitaxel plus sunitinib, and bevacizumab with carboplatin and paclitaxel. Randomization assigns participants to treatment groups by chance; the registry does not report the allocation ratio or the number assigned to each group. Open-label means the study was not masked, so treatment assignment was not concealed from participants or study personnel according to the listed design. Further details about how assignment was generated or implemented are not reported in the registry record.

The study focused on safety and preliminary efficacy in patients with advanced non-small cell lung cancer who had not received prior systemic therapy for that cancer. The primary outcome listed in the posted results is best response. The registry description says the original primary outcome was progression-free survival, but insufficient data were available to report it. Safety outcomes listed as secondary include serious adverse events, Grade 3 or higher adverse events, and events leading to changes in sunitinib or bevacizumab treatment. The registry record does not describe the treatment schedules, follow-up visits, or detailed eligibility criteria beyond the disease and treatment-history description.

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: Randomized patients with at least one scan available at baseline and post-baseline

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 Best Response, measured over From randomization until disease progression/recurrence.

The listed primary outcome is best overall response, assessed according to RECIST from randomization until disease progression or recurrence. The definition includes confirmed and unconfirmed responses. The stated population is randomized patients with at least one scan available at baseline and post-baseline. The outcome is expressed as a number of participants, but the registry record does not provide the response counts or category-specific results.

Best response identifies the most favorable response status recorded during the specified observation period. In general, response endpoints can describe whether disease measurements meet prespecified criteria for response, rather than how long a participant remains free of progression. RECIST provides standardized criteria for evaluating response; the registry record does not give additional details about the criteria implementation or how scans were assessed.

The registry notes that progression-free survival was the original primary outcome, but there was insufficient data available to report it. Best response therefore describes a different aspect of the clinical course. It does not, by itself, provide information about the duration of response or the time to progression. The registry record does not report an inferential comparison for best response, so its definition should not be mistaken for evidence of a treatment difference.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does randomization contribute?

Randomization assigns participants to treatment groups by chance. In general, this can reduce systematic differences in measured and unmeasured characteristics at the start of a trial, supporting a fairer comparison of outcomes. It does not guarantee that groups will be identical, particularly in a study with limited enrollment, and it does not resolve missing outcome data or shortcomings in measurement. The registry identifies this study as randomized but does not report the randomization procedure, allocation ratio, or methods used to conceal assignment before enrollment.

What does the open-label design mean?

The trial is listed as open-label with no masking. In general, this means participants and study staff may know the assigned treatment. Knowledge of assignment can influence care, decisions about assessments, or the reporting of subjective symptoms. Standardized assessment criteria such as RECIST can support consistent evaluation of tumor response, but the registry record does not report whether response assessments were performed by reviewers unaware of treatment assignment. It also does not describe safeguards against assessment bias.

How should best response be understood?

Best response is the most favorable response recorded during the specified period, from randomization until progression or recurrence. The registry says the outcome follows RECIST and includes both confirmed and unconfirmed responses. In general, a response measure summarizes status during an observation window; it is not the same as a time-to-event measure. Best response alone does not describe how long a response lasts, when progression occurs, or survival. Scan timing and completeness can affect which response is recorded, but scan schedules and the handling of unavailable assessments are not reported in the registry record.

Why does the analysis population matter?

The stated primary-outcome population consists of randomized patients with at least one baseline scan and one post-baseline scan. This is a scan-available population rather than an explicitly described analysis of every randomized participant. In general, restricting an outcome analysis to participants with observed measurements can affect how representative the analyzed group is of the randomized groups, especially if missing measurements are related to health status or treatment. The registry does not report how many participants met the scan requirement, why scans might be absent, or whether any missing-data method was used. It also does not specify an intention-to-treat analysis population.

What can a confidence interval or p-value tell a reader?

In general, a confidence interval describes the statistical uncertainty around an estimated effect under the assumptions of the analysis. A wider interval indicates less precision, while an interval that includes values consistent with no difference may leave substantial uncertainty about the direction or size of an effect. A p-value describes how compatible the observed data are with a specified null model, assuming the analysis model and its assumptions apply. Neither measure establishes clinical importance by itself. The registry record does not report a confidence interval, p-value, or effect estimate for best response, so these concepts cannot be applied to an observed treatment comparison from the registry entries.

What do the planning assumptions mean?

The registry describes planning illustrations based on the observed effect. Their stated assumptions include a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. In general, a planning illustration asks how much information might be needed to detect an assumed effect under specified statistical assumptions. It is not a report of the trial’s observed power, and it does not show that the assumed effect is real or reproducible. The registry record does not give the observed effect, the illustration results, or a trial-specific sample-size justification. These illustrations should therefore be treated as planning context, not as outcome evidence or confirmation that the trial was adequately powered.

Why are simple comparisons not available?

The endpoint is described as best response, and the registry notes that the original progression-free survival outcome could not be reported because of insufficient data. The registry entries do not include a reduction of the response outcome to a single between-group comparison. A numerical comparison would require the relevant arm-level results and a defined estimand or analysis method. Without those results, calculating an effect from the enrollment total or from the endpoint definition would not be appropriate. The record lists 0 analyses, so the statistical interpretation must remain descriptive of what was planned and reported, rather than inferential about treatment differences.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists best response as the primary outcome and names secondary outcomes concerning serious adverse events, Grade 3 or higher adverse events, and adverse events leading to discontinuation, interruption, or reduction of treatment. It does not report a multiplicity strategy for interpreting results across these outcomes. In general, examining multiple outcomes can increase the chance of a misleading positive finding if each is tested separately; prespecified control methods or a clear hierarchy can help readers interpret such testing. No such method or hierarchy is reported in the registry record.

The primary-outcome population requires randomized participants to have both baseline and post-baseline scans. The record does not state how missing scans were handled, whether outcomes were imputed, or whether participants without the required scans were excluded from other analyses. It also does not provide missingness counts or reasons. These omissions matter because the availability of post-baseline measurements may be related to the outcome or treatment experience.

The registry does not report the analysis population for the listed safety outcomes, the statistical methods used for any outcome, or an approach to missing adverse-event information. It lists 0 analyses and provides no outcome estimates or tests in the registry record for this account. The original progression-free survival outcome is not reported because data were insufficient; the reason and handling of that insufficiency are not reported in the registry record. These gaps limit assessment of both the completeness of outcome reporting and the strength of any potential comparisons.

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

The record establishes that a randomized, open-label Phase II trial compared a regimen containing sunitinib with a regimen without sunitinib, with bevacizumab, carboplatin, and paclitaxel named in both group descriptions. It defines best response as the reported primary outcome and specifies the response assessment period and scan-based analysis population. These details clarify the intended comparison and what the endpoint represents.

The record does not report response results, an effect estimate, or uncertainty around an effect. It therefore cannot show whether the groups had different best-response outcomes, how large any difference was, or how precisely an effect was estimated. It also cannot establish that the absence of a reported result means there was no effect. A lack of reported analysis is not equivalent to evidence of no difference.

Progression-free survival, the original primary outcome, is not reported because the registry says there was insufficient data. The record does not explain the reason, so the omission should not be interpreted as a particular trial event or outcome. Listed safety outcomes likewise do not reveal observed adverse-event frequencies without results.

More generally, endpoint definitions, randomization, and planning assumptions do not substitute for observed results. Planning illustrations are not observed power, and the trial’s enrollment figure alone does not establish precision. The registry record supports a description of the study and its reporting limitations, but not a conclusion about comparative efficacy or safety.

12. Why This Trial Matters Statistically

This trial is statistically notable for the distinction between the original primary outcome and the outcome reported in the registry. The original primary outcome was progression-free survival, but the registry states that insufficient data were available to report it. Best response is listed instead, with a scan-based population and an observation window ending at disease progression or recurrence. These are different outcome concepts, so findings for one would not automatically answer the question posed by the other.

The record also illustrates why a clearly defined analysis population matters. Requiring baseline and post-baseline scans makes the availability of assessments part of the endpoint definition. Without counts or missing-data handling details, readers cannot assess how this requirement affected the analyzed population. Similarly, an open-label design and the lack of reported information about blinded response assessment leave important questions about outcome ascertainment unanswered.

Finally, the registry includes planning assumptions but no reported analyses or outcome estimates in the registry record. Planning calculations and observed trial results serve different purposes: planning describes what a future study might need under assumptions, while results describe what was observed and how uncertain the estimate is. Keeping those roles distinct prevents an assumed effect from being mistaken for a finding. The trial is therefore a useful example of how design and endpoint definitions can be reported while the evidence needed for a comparative statistical interpretation remains unavailable.

13. 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 has important trial-specific reporting limitations. It lists 0 analyses and does not give arm-level best-response counts, effect estimates, confidence intervals, or p-values. The primary outcome is described, but its results cannot be evaluated statistically from the reported details. The original primary outcome, progression-free survival, is explicitly unavailable because data were insufficient; no further explanation of that insufficiency is reported.

The primary-outcome population is limited to randomized patients with baseline and post-baseline scans, but the record does not report how many participants met that requirement or how missing scans were handled. The number of randomized participants in each group is also not reported. The registry does not specify the randomization procedure, allocation ratio, response-review process, scan schedule, treatment schedules, or analysis methods.

The study is open-label, and the record does not describe whether response assessments were blinded. Safety outcomes are listed, but their observed results are not available in the registry record. The record also does not report a multiplicity strategy, a sample-size justification, interim analyses, or the reason the study was terminated. These unreported details limit assessment of the design, analysis, and completeness of the evidence; their absence should not be filled by assumptions.

14. Related Tutorials

15. Related Statistical Calculators

16. Sources

• ClinicalTrials.gov: A Study of Sunitinib in Combination With Bevacizumab, Carboplatin, and Paclitaxel in Patients With Advanced Non-Small Cell Lung Cancer (SABRE-L) — NCT00434226.
• Registry record last updated 2009-08-21; status last verified 2009-07.
• Results first posted 2009-08-21.
• 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.

17. Record Summary

NCT00434226 is a randomized, controlled, open-label, multicenter, parallel-group Phase II study sponsored by Genentech, Inc. It enrolled 56 participants with advanced non-small cell lung cancer and is listed as terminated. The registry gives a start date of 2007-03 and a primary completion date of 2008-04.

The comparison was bevacizumab with carboplatin and paclitaxel plus sunitinib versus bevacizumab with carboplatin and paclitaxel. Best response per RECIST is listed as the primary outcome, for randomized patients with baseline and post-baseline scans. The original primary outcome was progression-free survival, which the registry says could not be reported because data were insufficient. The record does not provide response analyses or results, and does not report the reason for termination.