Carboplatin, Paclitaxel, and Everolimus in Treating Patients With Previously Untreated Cancer of Unknown Primary: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Carboplatin, Paclitaxel, and Everolimus in Treating Patients With Previously Untreated Cancer of Unknown Primary (NCT00936702), a na phase 2 study. The primary outcome is Percentage of Participants With Confirmed Tumor Responses.

2. Clinical Question

This question focuses on tumor response during a defined treatment period, rather than directly comparing the regimen with another option. The registry also lists overall survival, progression-free survival, duration of response, and time to treatment failure as secondary outcomes. Their detailed results are not reported in the registry record. Consequently, the central question can be described from the registered endpoint definition, but the record does not establish the observed response frequency or how precisely it was estimated.

3. Trial Design

This was a phase II study of carboplatin, paclitaxel, and everolimus for previously untreated cancer of unknown primary. The registry records a single-group model, allocation as not applicable, and no masking. These features describe a study in which participants received the treatment regimen without a randomized comparison group or masking. The registry lists 46 participants enrolled and marks the study as completed.

The treatment schedule specified carboplatin intravenously over 30 minutes and paclitaxel intravenously over 3 hours on day 1, with everolimus taken orally once daily on days 1, 8, and 15. Courses were scheduled to repeat every 21 days when disease progression or unacceptable toxicity did not prevent further treatment. This description identifies the planned regimen, but does not provide individual treatment exposure, dose modifications, or adherence information.

The primary endpoint was the percentage of participants with confirmed tumor responses during the first 6 cycles, described as an average of 6 months. The primary-endpoint population excluded one participant deemed ineligible because treatment began before registration. The record lists five outcome measures and names four secondary outcomes: overall survival, progression-free survival, duration of response, and time to treatment failure. Specific estimates for those secondary outcomes are not reported in the registry record.

The registry gives the study start as 2009-09 and primary completion as 2012-12. It does not report the statistical analysis plan, the method used to calculate the primary confidence interval, a sample-size justification, or interim-analysis procedures. The absence of a control arm is important when interpreting the findings: outcomes in this group cannot by themselves distinguish treatment effects from other influences on response or follow-up.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All participants except one who was deemed ineligible (treated prior to registration).

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: All participants.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With Confirmed Tumor Responses, measured over First 6 Cycles of treatment (an average of 6 months).

The primary endpoint is the percentage of participants with a confirmed tumor response during the first 6 treatment cycles, described in the registry as an average of 6 months. A response could be a complete response or a partial response. Confirmation required the same objective response status on 2 consecutive evaluations at least 4 weeks apart.

The endpoint uses RECIST criteria as described in the registry. Complete response was defined as disappearance of all target lesions. Partial response was described as a 30% decrease in the sum of the longest diameters of target lesions. The endpoint therefore combines two response categories and applies a confirmation requirement rather than counting a single response assessment.

General: A response proportion offers a direct description of how many people in a study meet a defined response criterion. Requiring confirmation can reduce reliance on an isolated assessment, while a confidence interval communicates uncertainty around the estimated proportion. It does not remove limitations arising from the study design or missing information.

The registry record does not report the observed percentage of participants with confirmed responses or its 95% confidence interval. It also does not give details of assessment timing, handling of missing assessments, or the analysis method. The registered definition explains what the endpoint was intended to measure, but the posted record does not allow readers to quantify the result or its precision.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes the main outcome as a percentage of participants with an event—in this case, a confirmed tumor response. For a binary endpoint, each participant is classified according to whether the specified outcome occurred within the defined assessment period. The response proportion describes the share of the analyzed group meeting that criterion. The registry record does not report the observed primary proportion, so the response frequency cannot be discussed numerically.

The endpoint definition combines complete and partial responses. Complete response was defined as disappearance of all target lesions; partial response was described as a 30% decrease in the sum of the longest diameters of target lesions. A response had to be recorded on 2 consecutive evaluations at least 4 weeks apart to count as confirmed. The specified window was the first 6 treatment cycles, described as an average of 6 months. The record does not provide further detail about assessment schedules or how assessments outside the window were handled.

What does a confidence interval communicate?

General: A confidence interval expresses uncertainty around an estimate. For a response proportion, it helps show how compatible the observed data are with a range of possible underlying response frequencies under the assumptions of the analysis. A narrower interval generally indicates greater precision, while a wider interval indicates less precision. The registry lists a 95% confidence interval as the measure of dispersion for the primary outcome, but the estimate, interval limits, and calculation method are not reported in the registry record.

A confidence interval does not correct for design limitations. In this single-group study, an interval around a response proportion would describe uncertainty in that group’s estimate; it would not supply a comparison with a control group. Nor would it resolve possible issues related to eligibility, missing assessments, or selection of the analysis population. The record does not report enough detail to assess those issues fully.

Why does the single-group design matter?

General: Randomized controlled comparisons can help distinguish differences associated with treatment assignment from differences arising from participant characteristics, clinical course, or other factors. A single-group study lacks that concurrent comparison. The registry lists allocation as not applicable and the model as single group. Therefore, even if a response estimate were reported, it would be an observed outcome in the treated group rather than a randomized estimate of a difference between treatment strategies.

The registered treatment schedule specifies the drugs, routes, and days of administration, as well as the 21-day course interval and conditions for repeating courses. Those schedule details describe the intervention, but do not establish that every participant received every scheduled dose. Information on dose intensity, treatment discontinuation, and dose changes is not reported in the registry record.

What can be concluded from the analysis details?

The registry lists five outcome measures and gives the primary endpoint definition, but it does not report an analysis method, the statistical plan, interim analyses, or a sample-size justification. Thus, readers cannot determine from the record whether the primary analysis followed a prespecified method, how missing data were treated, or how secondary outcomes were analyzed. The registry also does not report the primary confidence interval calculation procedure.

General: Over-dispersion matters when data vary more than a model’s assumptions allow. In such cases, a model that assumes less variability may understate uncertainty and produce intervals that are too narrow. This general issue cannot be assessed from the registry record because no model-based analysis or dispersion estimate is reported for the primary binary endpoint.

Planning illustrations based on an observed effect are not evidence of the study’s observed power. The stated assumptions for such illustrations include a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the target effect expressed as a fraction of the observed effect. Time-to-event illustrations use a total number of events; other illustrations use participants per arm. These are planning conventions and do not describe the design of this single-group trial. The registry record does not report the trial’s own sample-size justification, and the illustration assumptions should not be mistaken for the study’s analysis plan.

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("Treatment (Carboplatin, Paclitaxel, and Everolimus")
n      <- c(45)
events <- c(16)
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  Treatment (Carboplatin, Paclitaxel, and Everolimus  45  16  35.6  21.9  51.2  23.2  50.2  23.2  50.2

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary endpoint and names secondary outcomes of overall survival, progression-free survival, duration of response, and time to treatment failure. It does not report a multiplicity strategy, such as whether statistical testing across outcomes was adjusted or whether secondary results were intended to be descriptive. Without that information and without the detailed outcome results, readers cannot assess how the study addressed the chance of false-positive findings across multiple analyses.

The record describes the primary-endpoint population as all participants except one deemed ineligible because treatment began before registration. It also lists 46 participants enrolled. It does not report the numbers treated, assessed for response, or included in each outcome analysis. The exact analysis populations for the secondary outcomes are not reported in the registry record.

The registry does not state how missing response assessments, incomplete follow-up, or participants without an evaluable response were handled. It also does not describe imputation, censoring, sensitivity analyses, or rules for classifying unevaluable participants. These omissions matter because the estimated response proportion can depend on which participants are included and how unavailable outcomes are treated. No assumption about those procedures can be made from the endpoint definition alone.

The registry reports zero analysis records in the available record summary, and does not provide outcome-level numerical results for the listed measures. It also does not report interim-analysis procedures or a statistical analysis plan. Accordingly, the available record supports a description of the intended endpoint and registered analysis population, but not a reconstruction of multiplicity control, missing-data handling, or the detailed analysis populations.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: neutrophil count decreased (32.6%), platelet count decreased (17.4%), hemoglobin decreased (10.9%), leukocyte count decreased (10.9%), diarrhea (6.5%).

The registry lists 46 participants at risk for serious adverse events and 21 participants affected. It also lists 46 participants at risk for other adverse events and 45 participants affected. These are participant counts; the registry does not provide further details about event duration, severity grading, timing, or the way repeated events in the same participant were handled.

The listed serious adverse-event terms and percentages are neutrophil count decreased, 32.6%; platelet count decreased, 17.4%; hemoglobin decreased, 10.9%; leukocyte count decreased, 10.9%; diarrhea, 6.5%; and abdominal pain, 4.3%. The registry record does not clarify the denominator or additional counting details for each term in the available information. The percentages should therefore be read as the registry’s reported figures, not recalculated into participant counts.

No death count is reported in the adverse-event record. The event listings do not by themselves establish causation: they describe safety events recorded during the study, not a comparison of event rates against a control group. The registry record does not give the observation period or detailed attribution information needed to interpret those events more fully.

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

The registry record establishes that a phase II, single-group study evaluated carboplatin, paclitaxel, and everolimus in previously untreated cancer of unknown primary. It specifies the treatment schedule and defines confirmed tumor response as the primary endpoint during the first 6 cycles, described as an average of 6 months. It also records the enrollment count, the primary-endpoint eligibility exclusion, and selected adverse-event information.

Those details do not reveal the primary response result. The response percentage and its 95% confidence interval are not reported in the registry record. As a result, the record cannot show how many participants met the response criterion, how precise an estimate was, or how response findings varied across participants. The treatment schedule and the endpoint’s definition cannot substitute for those missing outcome data.

Even a reported response proportion from this design would describe outcomes in the single treatment group, not a randomized difference against a comparator. The study has no control group in the registry record, so it cannot directly establish how outcomes would compare under another treatment strategy. The record also does not report the analysis plan, missing-data procedures, detailed analysis populations, or secondary-outcome estimates.

Safety counts and listed serious-event percentages offer a limited description of events recorded among participants. They do not, by themselves, establish causality or permit a controlled comparison. The registry does not supply detailed timing, attribution, or denominator information for the listed event percentages in the available record.

Overall, the registry supports a clear account of the study’s design, regimen, response definition, and certain safety figures. It does not support a numerical conclusion about the primary response outcome or a comparative claim about efficacy. Those boundaries are important when distinguishing what the registered study set out to measure from what its posted results allow readers to evaluate.

14. Why This Trial Matters Statistically

This trial illustrates why a registered endpoint definition and a reported endpoint result are different kinds of information. The registry provides a detailed primary response definition, including the RECIST-based response categories, confirmation on consecutive assessments at least 4 weeks apart, and an assessment window covering the first 6 cycles. Yet the response percentage and its 95% confidence interval are not reported in the registry record. Without those values, neither the observed frequency nor its precision can be evaluated.

The design also makes the role of comparison clear. A single-group phase II study can describe outcomes among participants who received a regimen, but it has no concurrent control group for estimating a randomized contrast. This distinction matters when translating an observed outcome into a claim about what the treatment caused. The record does not provide a basis for such a comparative estimate.

The registry’s population description is another important statistical detail: the primary-endpoint population excluded one person deemed ineligible because treatment began before registration. The record gives an enrollment of 46, but does not describe the analyzed denominator or handling of unevaluable responses. This shows why population definitions and missing-data rules are needed alongside endpoint definitions to interpret a proportion.

Finally, selected safety counts and percentages can inform a limited description of recorded events, but they cannot answer every safety question. Event timing, attribution, and detailed counting rules are not reported. The statistical lesson is that endpoint definitions, study design, analysis populations, uncertainty estimates, and reporting completeness all shape what readers can responsibly conclude.

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.

Several trial-specific limitations follow from the registry record. The study used a single-group design with no comparator, so outcomes cannot be contrasted with a concurrent control group. Allocation is listed as not applicable and masking as none. These features limit causal interpretation and make it important to distinguish descriptive outcomes from comparative treatment effects.

The most consequential reporting limitation is the absence of the primary response estimate and its 95% confidence interval. The endpoint is defined, but its observed result and precision are not reported in the registry record. Numerical results for the listed secondary outcomes are also not reported. Therefore, the registry does not permit evaluation of the response frequency, time-to-event outcomes, or duration of response.

The primary-endpoint population excluded one participant deemed ineligible because treatment began before registration. Although the registry lists 46 enrolled participants, it does not report the exact analysis denominator or explain how missing, incomplete, or unevaluable response assessments were treated. The analysis populations for secondary outcomes are not reported in the registry record.

The record also does not report the statistical analysis plan, sample-size justification, interim-analysis procedures, multiplicity approach, or method used to calculate the confidence interval. These omissions prevent assessment of prespecification and analytical choices. For safety, the record provides participant counts and selected serious-event percentages, but does not report detailed timing, attribution, or the counting basis for each percentage. These limitations concern the information available in the registry record and do not establish that the study lacked procedures that were not reported.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Carboplatin, Paclitaxel, and Everolimus in Treating Patients With Previously Untreated Cancer of Unknown Primary — NCT00936702.
• Registry record last updated 2017-03-21; status last verified 2017-02.
• Results first posted 2017-02-16.
• 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

NCT00936702 is a completed phase II, single-group study of carboplatin, paclitaxel, and everolimus in previously untreated cancer of unknown primary. The registry lists 46 participants enrolled, no masking, and allocation as not applicable. The primary endpoint was the percentage of participants with confirmed complete or partial tumor responses during the first 6 cycles, described as an average of 6 months. The primary-endpoint population excluded one participant deemed ineligible because treatment began before registration.

The registry lists overall survival, progression-free survival, duration of response, and time to treatment failure as secondary outcomes. It reports 21 participants affected by serious adverse events and 45 affected by other adverse events, with selected serious-event terms and percentages. The primary response estimate, its 95% confidence interval, numerical secondary-outcome results, and detailed analysis methods are not reported in the registry record.