A Study of Paclitaxel and Carboplatin in Combination With Bexarotene Oral Capsules in Patients With Advanced Lung Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Paclitaxel and Carboplatin in Combination With Bexarotene Oral Capsules in Patients With Advanced Lung Cancer (NCT00153842), a na phase 1, phase 2 study. The primary outcome is Bexarotene Oral Capsules Safety at Two Dose Levels (300 mg/m2 and 400 mg/m2) in Combination With Carboplatin and Taxol®..

2. Clinical Question

3. Trial Design

The registry describes a Phase I and II study of bexarotene capsules combined with paclitaxel and carboplatin for advanced non-small cell lung cancer. Its summary separates the aims: the Phase I components evaluate safety with two chemotherapy schedules, and the Phase II component evaluates preliminary efficacy with the weekly Taxol schedule and carboplatin. The stated primary outcome is safety at bexarotene dose levels of 300 mg/m2 and 400 mg/m2.

The design is listed as single-group, with allocation not applicable and no masking. Thus, the registry record does not describe random assignment or a blinded comparison. Although the adverse-event record lists Phase IB groups at the 300 mg and 400 mg doses and a Phase II group, these entries do not establish a randomized comparison or a common control. Differences among their observed counts could reflect cohort composition, treatment schedule, follow-up, or other factors not detailed in the registry record; they cannot be attributed to dose or phase alone.

Enrollment is listed as 33. The study start is recorded as 2001-08, primary completion as 2009-03, and status as terminated. The registry does not report why the study was terminated, whether enrollment reached a prespecified target, or whether the intended protocol was completed. The arm description says bexarotene was administered daily beginning on chemotherapy day 1; other treatment-delivery details are not reported in the registry record.

There is one posted primary outcome measure and no analyses recorded. The registry does not report a formal statistical analysis plan, a sample-size justification, or an interim-monitoring plan. Without those details, the intended inferential framework and the relationship between the listed enrollment and any planned precision or decision threshold cannot be assessed.

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

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 Bexarotene Oral Capsules Safety at Two Dose Levels (300 mg/m2 and 400 mg/m2) in Combination With Carboplatin and Taxol®., measured over 36 months.

The primary outcome is the number of participants experiencing initial-dose-limiting toxicity, or IDLT, during the stated timeframe of 36 months. It is a count outcome rather than a continuous measurement or a time-to-event estimate. The registry defines IDLT as a clinical observation that, in the investigator’s judgment, is attributable to bexarotene and requires a dose reduction, suspension, or discontinuation of study drug because of NCI CTC Grade 3 or 4 toxicity. Elevated lipids are excepted from this definition.

This endpoint is used to characterize tolerability in the Phase I components, where the study examines bexarotene at 300 mg/m2 and 400 mg/m2 with chemotherapy. A count can describe how often protocol-defined toxicity led to a treatment change, but by itself it does not describe the full range, timing, duration, or severity of all adverse events. The investigator-attribution and treatment-action requirements also mean that the measure is narrower than a count of every toxicity observed.

The registry description says at least 6 patients will be entered onto each dose level and that dose will not be escalated during an individual patient’s treatment. The primary outcome field does not contain observed IDLT counts or an analysis estimate. Those results are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods concern how an observed outcome is summarized, how uncertainty is represented, and what comparisons the design can support. In this record, the primary endpoint is a count of participants meeting a defined toxicity criterion. The registry does not provide an inferential analysis for that endpoint: the number of recorded analyses is 0, and no primary-outcome result is entered. Interpretation should therefore distinguish the endpoint’s definition from any estimate of its occurrence.

What a count outcome describes

A count states how many participants meet a specified event definition. It is not automatically a rate, a risk estimate, or a comparison. To interpret a count as a proportion, the relevant denominator and the rule for determining evaluability must also be clear. The primary outcome description gives a planned entry statement for each dose level, but does not provide observed IDLT counts or an analysis denominator. The adverse-event record separately includes at-risk figures by group; those entries should not be assumed to be the denominator for the primary IDLT endpoint.

For this endpoint, an event is counted only when the investigator judges it attributable to bexarotene and it prompts a specified treatment action because of Grade 3 or 4 toxicity, excluding elevated lipids. This makes the outcome clinically anchored to treatment decisions, but dependent on judgment and protocol application. It does not capture every adverse event, and it can differ from an endpoint based only on laboratory thresholds, symptom reporting, or event incidence.

What confidence intervals would contribute

General: a confidence interval expresses uncertainty around an estimate under a statistical model. For an event count converted into a proportion, an interval can show how compatible the data are with a range of underlying event frequencies. Sparse counts and small cohorts commonly yield substantial uncertainty, so a point estimate alone can suggest more precision than the evidence supports. No confidence interval or estimated IDLT frequency is reported in the registry record.

General: hypothesis tests address whether observed data are inconsistent with a specified statistical null model. A p-value does not measure the size or clinical importance of an effect, and a non-significant result does not establish equivalence. No hypothesis test or p-value is reported for the primary outcome or for comparisons among the listed groups.

How design limits comparison

The record identifies a single-group model, allocation as not applicable, and masking as none. General: randomization can help balance known and unknown prognostic factors across assigned groups, while masking can reduce some measurement and reporting biases. Neither feature is described. The groups listed for safety correspond to Phase IB dose cohorts and a Phase II group; they are not identified as randomized, concurrent control groups. A difference in event counts across them could be associated with dose, but could also reflect differences in schedule, enrollment period, participant characteristics, follow-up, or outcome ascertainment. The registry record does not provide information to separate these explanations.

General: over-dispersion matters when count data vary more than a chosen statistical model expects. A model that assumes too little variation can understate uncertainty and produce overly narrow intervals or overly optimistic tests. Assessing that issue requires suitable individual-level or repeated-count data and a specified model. The registry record does not describe such a model or the data needed to assess over-dispersion; it is therefore not appropriate to infer a count-model analysis from the endpoint alone.

Planning illustrations and observed power

The stated planning framework describes a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It defines the target effect as a stated fraction of the observed effect. These are assumptions for planning illustrations, not evidence that this trial used a two-arm randomized design. The record explicitly identifies the trial as single-group and says its arms are not a like-for-like control comparison.

General: a planning sample size estimates how many participants per arm might be needed under chosen assumptions to detect a target effect at a stated power. It depends on the assumed effect, outcome variability, allocation, significance level, and analysis method. If the assumed effect is taken from observed results, the illustration is conditional on that estimate and may be optimistic or unstable, particularly when the underlying study is small. Such planning values do not describe the probability that this completed trial detected an effect; they are not observed power. The registry record does not report a trial-specific sample-size justification or a prospective power calculation.

The stated planning convention distinguishes participant counts per arm from total event counts for time-to-event outcomes. The primary outcome is a count of affected participants, not a time-to-event endpoint. No time-to-event outcome estimate or analysis is reported in the registry record. Any planning illustration should therefore be understood only within its stated assumptions, rather than as a retrospective test of the trial’s success.

Recalculation and analytic provenance

General: recalculated estimates based on posted summary statistics can be useful checks, but they may be unadjusted and may not reproduce a sponsor’s prespecified analysis. The registry record reports no primary analysis and no recalculated estimate for IDLT. No sponsor analysis is available in the record to compare with a recalculation. The absence of an analysis means there is no basis for presenting a reconstructed inferential result as if it were the trial’s official finding.

10. Multiplicity, Missing Data and Analysis Populations

The registry record does not report how multiplicity was handled. Multiplicity arises when several outcomes, dose groups, time points, or comparisons are assessed; repeated testing can increase the chance of a nominally positive finding. The study has Phase I and Phase II components and lists safety groups at different doses and phases, but no testing hierarchy, adjustment method, or prespecified distinction between confirmatory and exploratory analyses is reported. The Phase II summary aim is preliminary efficacy, yet no efficacy outcome result is entered.

Missing-data methods are also not reported in the registry record. It does not state whether outcome status was collected for every enrolled participant, how incomplete follow-up was handled, or whether any participant was excluded from an analysis. The record gives at-risk counts for the serious and other adverse-event summaries, but does not explain how those counts relate to enrollment or to the IDLT analysis population. The primary endpoint description does not supply an observed denominator.

Analysis populations are not defined in the registry record. There is no stated intention-to-treat, per-protocol, safety, or evaluable-population rule for the posted endpoint. General: analysis-population definitions matter because exclusions or incomplete outcome ascertainment can alter both the denominator and the meaning of an estimate. Without those definitions, readers cannot determine whether the reported adverse-event entries represent all treated participants or a selected subset, and cannot assess the impact of missing outcome information.

No interim-analysis plan, stopping rule, or adjustment for interim looks is reported in the registry record. The status is listed as terminated, but the reason for termination and its relation to any planned monitoring are not reported. It would be speculative to infer that an interim safety or efficacy analysis affected study conduct.

11. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 3% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: hypertriglyceridemia (28.6%), myelosuppression (7.1%).

The serious-adverse-event figures are descriptive counts organized by the registry’s Phase IB and Phase II group labels. In the 300mg Phase IB group, 1 of the 14 participants listed at risk is recorded as affected by a serious event; the other-event count is 0 and deaths are 0. In the 400mg Phase IB group, serious events, other events, and deaths are each listed as 0, with 12 at risk for each event category. In Phase II, 2 of 7 are listed as affected by a serious event; the other-event count and deaths are 0.

The listed serious terms are hypertriglyceridemia at 28.6% and myelosuppression at 7.1%. The record does not make clear how these percentages map to the group counts, nor does it provide event-level timing or participant-level detail. They should therefore be read as recorded terms and percentages, not as group-specific risks or evidence of a dose relationship.

These counts do not establish that an event was caused by study treatment, and the absence of recorded deaths or other events in a group does not establish that risk was absent. The groups are not randomized comparison arms, and the record does not report an adjusted safety analysis or confidence intervals. Serious-event summaries also are not interchangeable with the primary IDLT endpoint, which has its own investigator-attribution and treatment-action definition.

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

The registry record establishes the study’s stated question and design: Phase I safety assessment and preliminary Phase II efficacy assessment of bexarotene with paclitaxel and carboplatin in advanced non-small cell lung cancer. It specifies an IDLT definition, the dose levels of 300 mg/m2 and 400 mg/m2, and a 36-month primary-outcome timeframe. It also records serious-adverse-event counts for the Phase IB dose groups and Phase II group.

It does not establish the observed frequency of the primary IDLT endpoint because no primary-outcome count or analysis is posted. It also does not report an efficacy estimate, uncertainty interval, or inferential test. The serious-event counts and listed terms are descriptive registry entries; they are not a substitute for the missing primary endpoint result or an efficacy analysis.

The record cannot support a causal comparison between the dose groups or between Phase IB and Phase II. The study is listed as single-group, allocation is not applicable, and the groups are not described as like-for-like controls. Differences among their event counts should not be interpreted as treatment effects. The registry also does not report enough detail to judge the completeness of follow-up, the analysis populations, the handling of missing outcomes, or whether observed events were associated with treatment.

General: a result’s interpretation depends on its estimate, uncertainty, endpoint definition, and design. A count without a clearly reported denominator and analysis may be informative descriptively but cannot answer a comparative question on its own. In this record, the primary outcome definition is more complete than its results reporting. The reason for termination, protocol adherence, and the relation between the listed enrollment and analyzed outcomes are not reported in the registry record.

Accordingly, the posted information documents a clinical research effort and some safety summaries, but does not permit a firm statistical conclusion about the primary safety endpoint or preliminary efficacy. That is a statement about what the registry record reports, not a conclusion about the treatment’s clinical effects.

13. Why This Trial Matters Statistically

This trial is statistically notable because its record combines distinct study aims within a single-group design. The Phase I components focus on safety at two bexarotene dose levels, while the Phase II component is described as a preliminary efficacy assessment with a weekly chemotherapy schedule. Those aims call for different interpretations: a safety count can describe observed toxicity, while an efficacy analysis needs a defined outcome and an estimate with appropriate uncertainty.

The registry identifies initial-dose-limiting toxicity with a detailed operational definition. The definition links attribution to bexarotene with a treatment action and a toxicity grade, while excluding elevated lipids. This illustrates how an endpoint’s meaning depends on its rules for deciding which events count. It also makes clear why a general serious-adverse-event count should not be treated as the same endpoint.

The design also illustrates why cohort labels do not create a valid comparative experiment. Phase IB groups at different doses and a Phase II group are listed, but the registry does not describe random allocation or a concurrent control. Comparisons across such groups may be confounded by differences in phase, schedule, follow-up, or participant mix. Descriptive counts can still be reported, but causal claims require more than numerical contrasts.

Finally, the record highlights the importance of complete results reporting. It includes a primary outcome definition and safety summaries, but no posted primary endpoint result, inferential analysis, or efficacy result. Without denominators, analysis populations, and uncertainty estimates for the primary endpoint, readers cannot judge precision or compare the findings with a prespecified threshold. The distinction between protocol aims and results actually entered in a registry is central to sound 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 primary endpoint has a definition but no posted observed result. The registry therefore does not report how many participants experienced IDLT, the denominator used for any estimate, or uncertainty around an estimate. Although the timeframe is listed as 36 months, the record does not describe individual follow-up or the completeness of outcome ascertainment.

The study is listed as single-group with allocation not applicable and no masking. The safety entries for different dose levels and Phase II cannot be treated as randomized comparisons. The registry does not report the treatment schedules in sufficient detail to establish whether group differences in outcomes would be attributable to bexarotene dose rather than phase or schedule.

The serious-event record contains group counts and top serious terms, but does not give enough detail to link the terms to specific groups or explain the denominators behind their percentages. It also does not report event timing, recurrence, participant characteristics, or treatment attribution for each event. These limitations constrain interpretation of the safety summaries.

Other important details are not reported in the registry record: eligibility criteria beyond the broad disease description, baseline characteristics, protocol deviations, analysis populations, missing-data handling, statistical methods, multiplicity procedures, sample-size justification, interim analyses, and the reason for termination. Enrollment is listed as 33, but the relationship between that figure and the group-specific safety records or an outcome analysis is not explained. These gaps prevent a detailed assessment of representativeness, precision, and possible bias.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of Paclitaxel and Carboplatin in Combination With Bexarotene Oral Capsules in Patients With Advanced Lung Cancer — NCT00153842.
• Registry record last updated 2019-03-06; status last verified 2019-02.
• Results first posted 2019-03-06.
• 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 describes a Phase I and II, single-group study of bexarotene with paclitaxel and carboplatin in advanced non-small cell lung cancer. The Phase I aim is safety assessment at 300 mg/m2 and 400 mg/m2; the Phase II aim is preliminary efficacy with the weekly Taxol schedule and carboplatin. The primary outcome is the count of participants experiencing IDLT within 36 months.

Enrollment is listed as 33. The study started in 2001-08, reached primary completion in 2009-03, and is listed as terminated. Serious-adverse-event counts are entered for Phase IB groups at the 300mg and 400mg doses and for Phase II. The record lists hypertriglyceridemia at 28.6% and myelosuppression at 7.1% among top serious terms.

No primary IDLT result, efficacy result, or statistical analysis is reported in the registry record. The reason for termination and key analytic details, including missing-data handling and analysis populations, are also not reported.