Sorafenib and Erlotinib or Sorafenib Alone in Advanced Non-Small Cell Lung Cancer Progressing on Erlotinib: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Sorafenib and Erlotinib or Sorafenib Alone in Advanced Non-Small Cell Lung Cancer Progressing on Erlotinib (NCT00609804), a randomized phase 2 study. The primary outcome is Progression-free Survival (PFS).

2. Clinical Question

The clinical question is about outcomes after progression on erlotinib, not about starting treatment in an unselected population with non-small cell lung cancer. The trial compared two assigned regimens: adding erlotinib to sorafenib, or using sorafenib alone. Because participants had already progressed on erlotinib, the study addresses this particular treatment context.

The primary outcome is a time-to-event measure. Its median describes the time by which half of the relevant event-time distribution has occurred, subject to the trial’s outcome definition and follow-up. The registry defines progression as a 20% increase in the sum of the longest diameters of target lesions, a measurable increase in a non-target lesion, or the appearance of new lesions. It does not report a hazard ratio or a formal statistical comparison for the primary outcome.

The study also recorded overall response and treatment-emergent adverse events as secondary outcomes. These provide different kinds of information: response concerns tumor response, while adverse-event measures describe events recorded during treatment. Neither changes the definition of the primary question, and the secondary findings should be read with the limitations of the registry analyses in mind.

3. Trial Design

This was a randomized, open-label, parallel-group phase II study at multiple centers. The registry record lists an enrollment of 53 and describes two assigned regimens: sorafenib with erlotinib and sorafenib alone. The posted primary outcome groups contain 25 participants for the combination and 28 for sorafenib alone. The record lists the trial as completed.

In an open-label trial, participants and investigators are not masked to assigned treatment. The registry lists no masking. This matters particularly when outcomes involve clinical judgment or reporting, although the primary outcome used a stated imaging-based progression definition. The record does not describe procedures used to keep assessment independent of treatment assignment.

The combination arm received sorafenib 400 mg twice daily by mouth and erlotinib 150 mg once daily by mouth. The sorafenib arm received sorafenib 400 mg twice daily by mouth. The population was patients with advanced non-small cell lung cancer progressing on erlotinib. The registry does not report additional eligibility details, such as specific disease characteristics or prior treatments beyond that description.

Random allocation is intended to make assigned groups comparable on average, including for characteristics that were not measured. It does not guarantee that the groups are identical, especially in a phase II study with limited enrollment. The registry record does not report the randomization procedure, allocation concealment, or whether randomization was stratified.

The trial start is listed as 2008-03 and primary completion as 2012-05. The registry gives an 18-month timeframe for the primary outcome and for overall response. It does not report the timing of individual assessments, the length of follow-up for each participant, or how follow-up was handled after treatment discontinuation. These details are relevant to interpreting time-to-event results but cannot be inferred from the endpoint timeframe alone.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Analysis population for the primary outcome, as defined in the registry: All patients on study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Progression-free Survival (PFS), measured over 18 months.

The primary endpoint is progression-free survival, summarized as a median in months. The registry describes it as the length of time patients were alive from their first date of protocol treatment until worsening of their disease. Progression is defined using RECIST v1.0: a 20% increase in the sum of the longest diameter of target lesions, a measurable increase in a non-target lesion, or the appearance of new lesions. The stated timeframe is 18 months.

Time-to-event outcomes are useful when the timing of an event matters and participants may have different lengths of observation. A median gives a central summary of the observed event-time distribution; it does not describe every participant’s experience or the shape of the full distribution. In general, time-to-event methods can account for incomplete observation through censoring, but the registry record does not describe the censoring rules or analysis procedure used for this endpoint.

The reported median was 3.1 months in the sorafenib-plus-erlotinib group, with a 95% confidence interval of 1.7 to 3.7 months. In the sorafenib group, the median was 1.9 months, with a 95% confidence interval of 1.7 to 3.6 months. These are group summaries, not estimates of a treatment effect between groups. The registry record does not report a primary-endpoint hazard ratio, a confidence interval for a between-group contrast, or a p-value.

In general, a confidence interval expresses uncertainty around an estimated quantity under the analysis assumptions. The intervals around separate group medians do not directly provide a confidence interval for the difference between groups. The medians therefore describe the posted results, while the size and precision of any comparative effect remain unreported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Progression-free survival is a time-to-event endpoint. Each participant’s time begins at the first date of protocol treatment, and the registry defines the event as worsening of disease under RECIST v1.0. In general, time-to-event analysis uses both the event time and the amount of follow-up available. A participant whose event has not been observed by the end of follow-up may have a censored observation rather than a known event time. The registry record does not report how censoring was defined or handled for this trial.

What does a median summarize?

A median is the central point of an event-time distribution: half of the distribution lies on either side of it. It is a summary of timing, not a count of participants who experienced an event and not a direct comparison between randomized groups. The posted medians are 3.1 months for sorafenib plus erlotinib and 1.9 months for sorafenib alone. No arithmetic contrast between these values is needed to describe what the registry reports.

In general, a median can be useful when an event-time distribution is skewed or when the mean would be difficult to interpret. However, a median does not describe the full pattern of events over time. Two groups can have similar medians but different event-time patterns elsewhere in their distributions, or different medians without a precise estimated contrast. The registry record does not provide survival curves or a reported hazard ratio for progression-free survival.

How should the confidence intervals be read?

The primary outcome includes a 95% confidence interval for each group median: 1.7 to 3.7 months for the combination group and 1.7 to 3.6 months for sorafenib alone. In general, a confidence interval represents uncertainty in an estimate under the statistical method and its assumptions. It is not a range containing a known proportion of individual participants’ outcomes, and it does not give the probability that the true value lies within that particular interval.

Separate confidence intervals are not the same as a confidence interval for the difference between groups. The registry record reports neither a direct interval estimate for a primary-outcome contrast nor a formal primary comparison. The intervals’ overlap does not by itself determine whether a between-group difference is statistically compatible with the data. A direct comparison method would be needed to quantify that question.

Why are response comparisons different?

Overall response is reported as a number of participants, rather than as a time-to-event measure. The posted counts are 2 participants in the sorafenib-plus-erlotinib group and 1 participant in the sorafenib group. The registry also includes an unadjusted recalculation for this binary outcome. Such a calculation compares observed group summaries without adjustment for other variables. It does not account for the full trial design unless those features are explicitly incorporated, and the registry lists no sponsor-reported analyses for this endpoint.

In general, sparse binary outcomes can produce estimates with substantial uncertainty. A point estimate alone may appear to favor one group, while a confidence interval can still include a range of effects in either direction. The posted unadjusted comparison has a risk-difference estimate of -0.044, with a 95% confidence interval from -0.171 to 0.082 and p = 0.486. These values are recalculations from the posted summary counts, not a reported sponsor analysis. The registry record does not identify a multiplicity adjustment for this comparison.

What are the planning illustrations?

The planning figures are illustrations based on the observed response effect, not the design assumptions used to plan this trial and not observed power. The stated assumptions are a two-sided significance level of 0.05, equal allocation between groups, and a normal approximation. The illustrations consider planning for 100%, 75%, and 50% of the observed effect. For example, the row for planning for 75% of the observed effect gives 844 participants per arm for 80% power and 1,129 participants per arm for 90% power.

In general, such calculations show how a future study’s planned sample size can vary with the effect it aims to detect and the chosen power. They do not show the probability that the completed trial detected a true effect, and they should not be interpreted as evidence that the original trial had a particular power. Because the observed effect is uncertain, planning directly from it can also yield unstable illustrations. The registry record does not report the trial’s own sample-size justification or design assumptions.

What do the adverse-event counts represent?

The safety information gives the number of participants at risk and the number affected by serious adverse events or other treatment-emergent adverse events in each group. In general, these are descriptive counts. They do not establish that an event was caused by treatment, and the registry record does not provide event rates adjusted for duration of exposure. The terms listed as serious events are a separate description; individual terms can overlap within participants, so they should not be added to infer a total number of affected participants.

The registry record does not report a detailed statistical analysis plan, interim analyses, missing-data methods, or a multiplicity procedure. Without those details, the posted estimates can be described, but the full analysis strategy and its consequences cannot be evaluated.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists one primary outcome and secondary outcomes including overall response and treatment-emergent adverse events. It does not report a multiplicity strategy, such as a hierarchy of tests or an adjustment for testing several outcomes. In general, multiplicity matters when several outcomes or comparisons are examined because repeated testing can increase the chance of a false-positive finding. Without a reported plan, the degree to which the secondary results were intended to support formal claims is unclear.

The record lists zero reported analyses. It does not provide a sponsor-reported hazard ratio or formal test for progression-free survival, and it lists no sponsor-reported analyses for overall response. The unadjusted response comparison is a recalculation from posted summary information. The registry record does not describe whether the original analysis used covariate adjustment, stratification, or other methods.

The primary outcome population is described as all patients on study. Further detail on the analysis population is not reported in the registry record. It does not specify whether all randomized participants were included in the primary analysis, how treatment discontinuation was handled, or whether participants were analyzed according to their assigned group. The record also does not describe missing-outcome procedures, reasons for missing assessments, or sensitivity analyses.

For time-to-event outcomes, incomplete follow-up and censoring can affect estimates. In general, censoring methods rely on assumptions about the relationship between the reason observation ends and the event process. The registry record does not report censoring rules, follow-up completeness, or an assessment of those assumptions for progression-free survival.

The record does not report interim analyses or a statistical analysis plan. It also does not report the trial’s sample-size justification. The posted sample-size illustrations are explicitly planning exercises based on the observed secondary response effect; they are not evidence about the original design or the trial’s observed power. These omissions limit the ability to reconstruct the intended inferential framework or assess how missing data and multiple outcomes affected interpretation.

11. Safety Results

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

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: dehydration (12.0%), diarrhea (12.0%), fatigue (8.0%), anorexia (4.0%), cardiac disorders - other, unspecified (4.0%).

For serious adverse events, the registry records 11 of 25 participants in the sorafenib-plus-erlotinib group and 5 of 28 participants in the sorafenib group as affected. For other treatment-emergent adverse events, it records 24 of 25 and 28 of 28 participants, respectively. These counts describe participants recorded as affected during the study; they are not adjusted comparisons and do not establish that treatment caused an event.

The registry also lists serious event terms with percentages: dehydration, 12.0%; diarrhea, 12.0%; fatigue, 8.0%; anorexia, 4.0%; cardiac disorders - other, unspecified, 4.0%; and cognitive disturbance, 4.0%. The posted information does not attach these term-specific percentages to an individual treatment group. They should not be assigned to an arm or added together to infer the number of participants with serious events. A participant may have experienced more than one event, and the record does not explain overlap among terms.

In general, adverse-event counts do not convey timing, duration, recurrence, severity beyond the serious classification, or exposure-adjusted rates. The registry record does not report deaths in the safety data. These limitations mean the figures are useful as descriptive safety observations, not as a complete comparative safety analysis.

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

The registry results describe the outcomes recorded in a randomized phase II comparison of sorafenib plus erlotinib and sorafenib alone in advanced non-small cell lung cancer progressing on erlotinib. The combination group had a posted median progression-free survival of 3.1 months, with a 95% confidence interval of 1.7 to 3.7 months; the sorafenib group had a median of 1.9 months, with a 95% confidence interval of 1.7 to 3.6 months. These summaries show the observed medians in the two groups.

They do not, on their own, establish the magnitude or precision of a comparative treatment effect. The registry record does not report a hazard ratio, confidence interval for a between-group primary contrast, or p-value for progression-free survival. Separate group medians and their confidence intervals cannot replace such a comparison. The posted record therefore does not support a definitive statistical conclusion about the relative primary outcome.

The secondary results add descriptive information about response and adverse events, but the response analysis is unadjusted and the registry lists no sponsor-reported analyses. The safety counts indicate events recorded in the study, not proof of causation or a full assessment of comparative risk. Neither type of secondary finding resolves the lack of a reported primary comparison.

In general, randomized allocation helps create a fair basis for comparing groups, but interpretation still depends on the quality and completeness of outcome measurement, follow-up, analysis methods, and reporting. The registry provides limited detail on those elements. The open-label design and absent information about the analysis plan, censoring, missing data, and multiplicity further constrain what can be concluded from the posted summaries.

The results should not be treated as evidence that either regimen works or does not work. They describe the outcomes recorded for the study groups and the uncertainty that can be read from the posted estimates. A stronger comparative interpretation would require a reported primary analysis and additional information about its methods and completeness.

13. Why This Trial Matters Statistically

This trial is statistically instructive because it reports group medians and confidence intervals for a time-to-event outcome without reporting a direct primary comparison. That distinction is important: a pair of group summaries is not the same as an estimate of the difference between groups. Readers need to know which quantity was estimated and how uncertainty was quantified before making comparative interpretations.

The outcome definition also matters. Progression-free survival depends on when progression is identified under the stated RECIST v1.0 criteria, and time-to-event analysis depends on how participants with incomplete follow-up are handled. The registry states the progression criteria and the outcome timeframe, but does not report censoring rules or the analysis method. That gap makes it harder to assess how the posted medians relate to all observed follow-up.

The study also illustrates why secondary outcomes require context. Overall response is based on sparse counts, and the available comparison is an unadjusted recalculation rather than a sponsor-reported analysis. Multiple outcomes can invite selective emphasis unless the analysis plan and multiplicity approach are clear. The registry does not report those details.

Finally, the planning illustrations should be separated from trial design and from observed power. They use the observed response effect under stated assumptions to show possible future sample sizes. They do not reveal the study’s original sample-size rationale or the probability that the completed study detected a true effect.

These distinctions are useful beyond this trial: descriptive estimates, comparative inference, planning calculations, and safety summaries answer different statistical questions. Keeping them separate helps prevent a numerical pattern from being mistaken for a confirmed treatment effect.

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.

Several limitations are specific to interpreting this registry record. The primary outcome is reported as separate medians and confidence intervals, but there is no reported hazard ratio, direct contrast, or p-value. This prevents assessment of the estimated comparative effect and its precision from the registry results alone. The record also does not report survival curves, which would provide information about the event-time pattern beyond the medians.

The study was open-label. Although progression was defined by RECIST v1.0, the registry does not describe who assessed progression or whether assessment was independent of treatment assignment. It also does not describe scan timing, censoring rules, follow-up completeness, or sensitivity analyses. These omissions leave important aspects of endpoint measurement and analysis unclear.

The registry record gives limited detail on eligibility criteria, baseline characteristics, randomization procedures, and the analysis population. It does not report whether all randomized participants were analyzed as assigned, how missing outcome data were handled, or how discontinuation affected follow-up. These details are necessary for a fuller assessment of potential bias and comparability.

The secondary response finding is based on sparse counts and an unadjusted recalculation; no sponsor-reported analysis is listed. The record does not report a multiplicity plan. Safety information is descriptive, and the listed serious-event terms are not assigned to treatment groups in the posted data. Deaths are not reported in the registry’s safety information.

Finally, the record does not report a statistical analysis plan, interim analyses, or the trial’s sample-size justification. The planning illustrations based on the observed response effect cannot fill that gap. These limitations do not negate the reported observations, but they constrain the strength and scope of conclusions that can be drawn from them.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Sorafenib and Erlotinib or Sorafenib Alone in Advanced Non-Small Cell Lung Cancer Progressing on Erlotinib — NCT00609804.
• Registry record last updated 2016-04-08; status last verified 2016-03.
• Results first posted 2015-06-02.
• 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.

18. Record Summary

The registry record describes a completed, randomized, open-label, parallel-group phase II trial in patients with advanced non-small cell lung cancer progressing on erlotinib. It compares sorafenib plus erlotinib with sorafenib alone. Enrollment is listed as 53.

The primary outcome is median progression-free survival over an 18-month timeframe, using a registry definition of disease worsening based on RECIST v1.0. The posted medians are 3.1 months for the combination group and 1.9 months for sorafenib alone, each with a 95% confidence interval. The registry does not report a formal primary-outcome comparison.

Secondary outcomes include overall response and treatment-emergent adverse events. Response counts and safety observations are posted, but the registry lists no sponsor-reported analyses. Details on missing data, multiplicity, censoring, and the statistical analysis plan are not reported in the registry record.