BAY 43-9006 Plus Cetuximab to Treat Colorectal Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for BAY 43-9006 Plus Cetuximab to Treat Colorectal Cancer (NCT00326495), a na phase 2 study. The primary outcome is Overall Rate of Response.

2. Clinical Question

3. Trial Design

This was a Phase II, completed, single-group study of BAY 43-9006 combined with cetuximab for metastatic colorectal cancer. The registry record lists an enrollment of 51 and states that only 50 participants were on study long enough for assessment of the primary endpoint. The study had no randomized allocation and no masking; a separate control group is not described.

The combination brought together two drugs. BAY 43-9006 was given orally at 400 mg twice a day. Cetuximab was given intravenously at 400 mg/m^2 as a loading dose on week 2, followed by 250 mg/m^2 weekly starting on week 3. The record identifies these as the interventions in the study, but does not report adherence, dose modifications, treatment duration, or the number of participants who received each planned dose.

The primary outcome was an overall response rate defined using complete response, partial response, and stable disease for 4 months. The registry describes RECIST criteria for categorizing tumor response. A secondary outcome was the count of participants with adverse events. The record reports safety information for the combination arm, including serious adverse events and other adverse events.

The study start is listed as 2006-05-10, primary completion as 2014-11-05, and status as completed. The registry record does not report the sample-size justification, recruitment details, detailed eligibility criteria, or a statistical analysis plan. It also does not report an endpoint result for response or an analysis population beyond the note about participants on study long enough for assessment.

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: Although 51 patients were accrued, only 50 were on study long enough for assessment.

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 Overall Rate of Response, measured over 4 months.

The primary endpoint is the overall rate of response, expressed as a percentage of participants. It combines complete response, partial response, and stable disease sustained for 4 months. The registry record states that response was assessed according to RECIST. In that description, a partial response requires at least a 30% decrease in the sum of the longest diameter of target lesions from baseline. A complete response means disappearance of all target lesions. Stable disease is neither sufficient shrinkage for partial response nor sufficient increase for progressive disease, assessed against the smallest sum recorded since treatment started. Progressive disease is defined as at least a 20% increase from that smallest sum or the appearance of one or more new lesions.

A composite endpoint of this kind captures more than tumor shrinkage alone: it also counts stable disease meeting the stated duration criterion. The duration requirement makes persistence relevant to the endpoint, rather than treating a single assessment of stability as sufficient. The endpoint is useful for describing tumor response under the study's stated criteria, but the registry record does not report the actual percentage meeting the definition, the number of participants in each response category, or an uncertainty interval. Consequently, the endpoint definition is available, but its result and precision are not.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical concepts can clarify what the record would need to report for a complete interpretation. These are general explanations; the registry record does not report that each method was used in this study.

What does a response percentage represent?

A response percentage is a proportion: the number of participants meeting the endpoint definition relative to the analysis population used for that endpoint. For this study, the definition includes complete response, partial response, or stable disease for 4 months. Interpreting a percentage therefore requires both a numerator and a denominator, as well as clear rules for determining who is included. The record says 51 patients were accrued and only 50 were on study long enough for assessment, but it does not report the response count, final analysis denominator, or handling of the participant not on study long enough for assessment.

What would an uncertainty interval communicate?

In general, a confidence interval gives a range of values compatible with the observed data under the method used to calculate it. It conveys precision as well as the point estimate: an interval that spans a broad range indicates less precision than a narrower interval. A confidence interval does not show the probability that a particular fixed value is true, and it does not correct for limitations in the study design. No confidence interval for the primary response endpoint is reported in the registry record, so the precision of the response result cannot be evaluated.

What can a single-group study estimate?

In general, a single-group study can estimate outcomes observed among its participants, subject to the endpoint definition, population, follow-up, and missing-data rules. Without a concurrent comparator, it cannot isolate the difference attributable to the study treatment from differences that could arise from participant characteristics, disease course, assessment, or other care. This is a design limitation, not a computational problem that an adjustment to a response percentage can resolve. The study is listed as single-group, allocation NA, and unmasked; a randomized treatment comparison is not reported.

What does the response definition contribute?

RECIST criteria give operational rules for categorizing tumor changes. The registry describes partial response as at least a 30% decrease in the sum of the longest diameter of target lesions from baseline, and complete response as disappearance of all target lesions. It defines stable disease by the absence of sufficient shrinkage for partial response or sufficient increase for progressive disease, using the smallest sum since treatment started as the reference. Progressive disease includes at least a 20% increase from that reference or new lesions. The stated primary endpoint further requires stable disease for 4 months to count. These rules make the outcome more explicit, but the record does not report the assessment schedule, independent review procedures, or handling of unevaluable assessments.

Why does the difference between accrued and assessable participants matter?

When some accrued participants cannot be assessed, the denominator and analysis rule can affect a response rate. General approaches may include excluding participants without an assessment or counting them as not having a response, but the chosen approach must be reported rather than assumed. The registry note says only 50 of 51 accrued patients were on study long enough for assessment; it does not state the primary analysis rule or reasons for the difference. Consequently, no alternative rate should be reconstructed from the enrollment alone.

What does the safety reporting describe?

The adverse-event information is presented as counts of participants at risk and affected, along with selected serious-event terms and percentages. These summaries can describe the frequency of reported events in the study arm, but they do not by themselves describe severity over time, exposure-adjusted rates, causality, or comparative risk. General safety interpretation also depends on how events were defined, collected, graded, and attributed. Such details are not reported in the registry record. The posted counts should therefore be read as descriptive safety information rather than as a controlled estimate of excess risk.

What statistical analysis was actually reported?

The registry record lists 0 analyses. It does not report a model, hypothesis test, confidence interval, adjustment strategy, or analysis plan for the primary endpoint. The trial's sample-size justification is also not reported in the registry record. Statistical methods used in other trials or commonly used for response outcomes cannot be attributed to this study without documentation. This distinction matters because the endpoint definition and design are reported, while inferential claims about the result are not.

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("BAY 43-9006 & Cetuximab")
n      <- c(50)
events <- c(8)
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  BAY 43-9006 & Cetuximab  50  8  16.0  7.2  29.1  8.3  28.5  8.4  28.6

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 2 outcome measures: the primary overall rate of response and a secondary count of participants with adverse events. It does not report an analysis plan explaining whether multiple outcomes or any additional comparisons were considered, or whether any adjustment for multiplicity was planned or used. The record also lists 0 analyses. Therefore, no claim about multiplicity control or the statistical treatment of the outcomes can be made from the posted information.

Missing-data handling is also not reported in the registry record. The record states that 51 patients were accrued, but only 50 were on study long enough for assessment of the primary endpoint. It does not state why the other participant was not on study long enough, whether an endpoint assessment was available, or how that participant was treated in the response analysis. The response denominator, any rules for unevaluable participants, and any sensitivity analyses are not reported.

The distinction between accrued participants and those assessed is relevant because the endpoint is a percentage. Its interpretation depends on who is counted and how missing assessments are handled. Neither the denominator nor the analysis population can be inferred safely from enrollment alone. The record also does not report whether all participants who started treatment were included in the safety summaries, beyond listing 50 participants at risk for each adverse-event grouping.

Overall, the information does not establish a prespecified statistical plan, multiplicity approach, missing-data method, or full analysis population. These omissions limit assessment of how the endpoint was analyzed and whether the reported summaries, if available elsewhere, would correspond to an intended analysis.

12. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: small intestinal obstruction (12.0%), fever (10.0%), abdominal distension (4.0%), anemia (4.0%), bladder infection (4.0%).

The registry record lists 50 participants at risk for serious adverse events, with 22 affected. It also lists 50 participants at risk for other adverse events, with 50 affected, and 2 deaths. These are descriptive counts for the single study group. The record does not report a comparator arm, so the counts cannot be used to make a controlled comparison of event frequency.

Selected serious adverse-event terms are listed with percentages: small intestinal obstruction, 12.0%; fever, 10.0%; abdominal distension, 4.0%; anemia, 4.0%; bladder infection, 4.0%; and Death NOS, 4.0%. These percentages are registry-listed figures and should not be interpreted as evidence that an event was caused by a particular study drug. The record does not specify attribution, timing, duration, overlap between event categories, or whether the selected terms represent all serious events.

Safety summaries depend on how events were defined, collected, and coded, as well as on how long participants were observed. Those details are not reported in the registry record. The counts therefore describe the posted safety experience without establishing comparative risk, treatment causation, or the overall balance of benefits and harms.

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

The study record documents a Phase II investigation of BAY 43-9006 combined with cetuximab in metastatic colorectal cancer. It specifies the combination, dosing schedule, single-group design, and a RECIST-based primary endpoint. It also reports safety counts and selected serious adverse-event terms. Those details describe what the study set out to evaluate and what safety information is posted.

The primary result itself is not reported in the registry record. No response percentage, response counts, confidence interval, or test result is given. Thus the record does not establish how often participants met the endpoint or how precisely a response rate was estimated. That missing result cannot be interpreted as zero responses, nor can it be replaced by a result inferred from the endpoint definition or from the study rationale.

The single-group design also limits what any observed outcome could establish. Without a concurrent control group, outcomes in the treated participants cannot by themselves identify the effect of the combination relative to another treatment or a different course of care. The absence of masking may also matter for outcome assessment, although details of assessment procedures and review are not reported.

The 51 accrued participants and the note that only 50 were on study long enough for primary endpoint assessment indicate that the accrued population and endpoint-assessable population were not identical. Since the analysis denominator and missing-data rules are not reported, the response analysis population cannot be confirmed. Safety figures are similarly descriptive and lack a control comparison.

In statistical terms, the posted information is adequate to characterize important aspects of design and endpoint definition, but not to quantify the primary outcome or assess its precision. It does not support a conclusion about whether the treatment combination changes outcomes compared with an alternative. The registry record also does not report a statistical analysis plan, sample-size justification, or the handling of multiplicity and missing data.

14. Why This Trial Matters Statistically

This trial illustrates why a clearly defined endpoint and a reported endpoint result are separate parts of statistical reporting. The registry describes a response outcome based on complete response, partial response, or stable disease for 4 months, using RECIST criteria. Such definitions specify what is to be counted. They do not reveal how many participants met the definition, the denominator used, or the uncertainty around the resulting estimate. The response percentage and its precision are not reported in the registry record.

The difference between accrual and endpoint assessment is another central statistical detail. The record states that 51 patients were accrued and only 50 were on study long enough for assessment. Without an explicit analysis population and missing-data rule, readers cannot know how the participant not on study long enough was handled in any response calculation. This is a general lesson in reading proportion outcomes: denominators and analysis rules are part of the result, not incidental details.

The single-group design demonstrates the distinction between describing outcomes and estimating a comparative treatment effect. A response rate in a treated group can be descriptive, but there is no randomized concurrent comparator in this study. That means the design cannot separate treatment-associated outcomes from other influences as directly as a controlled comparison can.

Finally, the contrast between available safety summaries and the absent primary outcome shows that different forms of reporting can be incomplete in different ways. Safety counts and selected event percentages are posted, while the primary response result and inferential analyses are not. Statistical interpretation must be bounded by what is actually reported, rather than filling gaps with assumptions about planned analyses or expected effects.

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 limitations specific to this record constrain interpretation. The primary response result is not reported in the registry record, despite the endpoint being defined. No response count, percentage, confidence interval, or test result is available. The record therefore does not permit evaluation of the estimated response level or its precision.

The study is single-group, with allocation listed as NA and no masking. There is no concurrent control arm for a like-for-like comparison. The study can describe outcomes among participants receiving the combination, but its design does not support a randomized comparison of the combination against another treatment.

The enrollment count is 51, while the primary endpoint population note says only 50 were on study long enough for assessment. The record does not explain this difference, identify the primary analysis denominator, or describe missing-data handling. It also does not report the analysis population for safety beyond listing 50 participants at risk.

Other relevant details are not reported in the registry record, including a sample-size justification, a statistical analysis plan, assessment schedule, handling of unevaluable assessments, and procedures for reviewing response classifications. For adverse events, attribution, event timing, duration, grading, and collection methods are not reported. These gaps mean that the posted safety counts cannot be interpreted as comparative or causal estimates.

The registry lists 0 analyses, so methods for estimation, uncertainty, multiplicity, or sensitivity analysis are not documented. These limitations should be kept distinct: some concern the study design, while others concern absent reporting. Neither category can be resolved by assuming methods that are not described.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: BAY 43-9006 Plus Cetuximab to Treat Colorectal Cancer — NCT00326495.
• Registry record last updated 2017-07-24; status last verified 2017-06.
• Results first posted 2015-08-07.
• 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 a completed Phase II, single-group study of BAY 43-9006 with cetuximab in metastatic colorectal cancer. Enrollment was 51; only 50 participants were on study long enough for primary endpoint assessment. The primary endpoint was the percentage with complete response, partial response, or stable disease for 4 months under RECIST criteria. BAY 43-9006 was given orally at 400 mg twice a day, and cetuximab intravenously at 400 mg/m^2 as a loading dose on week 2, followed by 250 mg/m^2 weekly starting on week 3.

The record lists a secondary outcome of adverse events and provides safety counts for 50 participants at risk, including 22 affected by serious adverse events, 50 affected by other adverse events, and 2 deaths. The primary response result, statistical analysis, sample-size justification, multiplicity approach, and missing-data rules are not reported in the registry record.