Phase I/II Study of Capecitabine Plus Aflibercept to Treat Metastatic Colorectal Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase I/II Study of Capecitabine Plus Aflibercept to Treat Metastatic Colorectal Cancer (NCT01661972), a non-randomized phase 1, phase 2 study. The primary outcome is Number of Dose-Limiting Toxicities (Phase 1).

2. Clinical Question

3. Trial Design

This was a completed, open-label, non-randomized phase I/II trial in metastatic colorectal cancer. Its sequential structure joined a dose-finding portion to a single-arm phase II portion. The registry record lists enrollment of 63 participants. It gives a start date of 2012-08 and a primary completion date of 2016-06-12.

In phase I, the stated approach was a standard 3+3 dose escalation format. Capecitabine was to begin at 850mg/m2 on days 1-14, with days 15-21 off. If tolerated, the next cohort was to receive 1000mg/m2 on that schedule. Aflibercept was held constant at 6 mg/kg. The phase I objectives were to determine the recommended phase II dose for the combination and describe dose-limiting and non-dose-limiting toxicities.

After the recommended phase II dose was identified, the record says an additional 50 participants with metastatic colorectal cancer would be added to a single phase II arm. The phase II primary objective was to determine progression-free survival. Secondary objectives were response rate, overall survival, and exploration of correlations between clinical outcomes and baseline or on-treatment changes in blood-based angiogenesis biomarkers.

Because treatment allocation was not randomized and masking was absent, this design does not create a concurrent randomized comparison. The record does not report a comparator arm, the timing or details of any interim analyses, or a full statistical analysis plan. It also does not report the results for the phase II efficacy objectives. The design and objectives therefore describe what the trial set out to study, not the observed efficacy of the regimen.

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:

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 Number of Dose-Limiting Toxicities (Phase 1), measured over 28 days.

The posted primary outcome is the number of participants experiencing a dose-limiting toxicity in each phase I cohort, assessed over 28 days. A dose-limiting toxicity is a protocol-defined safety event used to guide dose finding. In this study, the definition included specified severe blood-count abnormalities, prolonged or severe gastrointestinal toxicities despite supportive measures, certain liver-test abnormalities, other qualifying non-hematologic toxicities, treatment delays, treatment-related death, or clinically significant treatment-related hospitalization. The definition also specified exceptions and different thresholds for some toxicities.

In general, a dose-limiting toxicity endpoint helps researchers judge whether a dose can be carried forward for further study. It is not an efficacy endpoint: it does not measure tumor response, progression-free survival, or overall survival. The phase I objective was to identify a recommended phase II dose and characterize toxicity, while the phase II primary objective was progression-free survival. Those are distinct questions, and the phase I endpoint cannot answer the phase II efficacy question.

The registry record gives the endpoint definition and timeframe but does not report the number of participants with dose-limiting toxicities by cohort. Thus, the endpoint’s intended meaning can be described, but its observed result and the resulting dose recommendation are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods determine how a study turns observations into estimates and how much uncertainty to attach to them. In general, a trial’s design, endpoint definitions, analysis population, missing-data handling, and prespecified analysis plan all affect interpretation. The registry record describes some design features and outcome objectives, but many analysis details are not reported in the registry record.

What does the sequential design mean?

In general, a sequential phase I/II design uses an early stage to inform dose selection and a later stage to study outcomes at the selected dose. The record describes a phase I dose-escalation portion followed by a single phase II arm after identifying the recommended phase II dose. It specifies a standard 3+3 format and gives the planned capecitabine dose levels and schedule, with aflibercept held constant at 6 mg/kg. This description explains the intended sequence; it does not provide the observed dose-limiting toxicity counts or confirm the dose selected.

Non-randomized allocation means participants were not assigned at random to treatment and control groups. The phase I and phase II portions also had different roles. As a result, their observed outcomes should not be interpreted as if they were arms of a randomized comparison. Differences between the phase groups could reflect differences in stage, participant mix, exposure, or other factors; those details are not reported in the registry record.

What does the primary endpoint capture?

The phase I endpoint is a participant count: the number experiencing a protocol-defined dose-limiting toxicity in each cohort during 28 days. In general, a count reports how many participants meet a specified event definition, while the definition determines what qualifies as an event. Here, the definition spans several kinds of toxicity and treatment disruption, with criteria involving toxicity grade, duration, supportive measures, and clinical significance. The definition also excludes or treats differently certain conditions.

Counts can be useful for dose finding because they make the observed safety experience within a cohort visible. But counts alone do not describe the full clinical course, the duration of follow-up beyond the stated timeframe, or the relative risk compared with an untreated or alternative-treatment group. The registry record does not report the cohort-specific outcome counts or uncertainty intervals.

How should confidence intervals and uncertainty be understood?

In general, a confidence interval gives a range of values compatible with an estimate under the assumptions of the statistical method. Wider intervals indicate less precision; narrower intervals indicate greater precision. A point estimate without an interval can conceal substantial uncertainty, particularly when the amount of information is limited. No confidence intervals for the trial’s efficacy endpoints are reported in the registry record, so the precision of those outcomes cannot be evaluated from this record.

Similarly, a p-value is a measure tied to a specified statistical test and its assumptions; it is not a measure of the size or clinical importance of an effect. The registry record lists 0 analyses. It does not report p-values or the tests that would have generated them. No statistical significance claim about efficacy can therefore be supported from the registry entry.

Why does the absence of randomization matter?

In general, random assignment helps balance known and unknown participant characteristics between treatment groups, supporting a causal comparison. An open-label, non-randomized study does not have that protection. With a single phase II arm, outcomes can be described for participants receiving the regimen, but there is no concurrent randomized control group in the design described. Other factors could influence outcomes, and the registry record does not report an adjustment strategy to address them.

Open-label means masking was absent. In general, awareness of treatment can influence some clinical assessments, reporting, or decisions. That possibility does not establish that bias occurred in this trial, but it is relevant when interpreting outcomes that involve assessment or reporting. The registry record does not report blinded outcome assessment.

What do the planning assumptions mean?

In general, a sample-size calculation estimates the number of participants or events needed under specified assumptions; it is not a statement about what the trial observed. The planning illustration associated with the record assumes a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect stated as a fraction of the observed effect. Time-to-event illustrations use a total event count, while other illustrations use participants per arm. These are planning assumptions, not the trial’s own design justification or observed power. The registry record does not report the trial’s sample-size justification.

The record does not provide the calculated planning values or identify an efficacy estimate from which readers could assess those illustrations. A planning illustration cannot substitute for observed results, and a sample-size target does not show whether an effect was present.

What analysis details remain unknown?

The registry record does not report the statistical analysis plan, analysis populations, missing-data methods, censoring rules for progression-free survival, or procedures for handling multiple outcomes. These choices matter in general because they specify which participants and observations contribute to estimates and how incomplete follow-up or multiple questions are treated. Without these details and the outcome results, readers cannot reproduce or fully evaluate the planned efficacy analyses.

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("Phase 1 Cohort 1", "Phase 1 Cohort 2")
n      <- c(7, 6)
events <- c(1, 2)
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  Phase 1 Cohort 1  7  1  14.3  0.4  57.9  2.6  51.3  3.2  52.7
ARM  Phase 1 Cohort 2  6  2  33.3  4.3  77.7  9.7  70.0  9.9  71.0

11. Multiplicity, Missing Data and Analysis Populations

The trial has several stated objectives and outcomes: phase I dose-limiting and other toxicities; phase II progression-free survival; and secondary questions about response rate, overall survival, and biomarker correlations. When several outcomes or analyses are examined, multiplicity can increase the chance of finding an apparently notable result by chance. In general, a statistical plan may identify a primary endpoint, define the order of testing, or use an adjustment strategy. The registry record does not report such a multiplicity strategy or the statistical analysis plan.

The record identifies progression-free survival as the phase II primary objective, but does not report its result or analysis method. It lists response rate and median survival as secondary outcomes, but does not give their numerical results. There are 5 outcome measures listed in the record, while the number of analyses listed is 0. The relationship between the listed outcomes and any analyses performed is not reported in the registry record.

Missing-data handling and analysis populations are also not reported in the registry record. In general, missing outcomes can affect estimates when follow-up is incomplete, and analysis populations define which participants are included in a given calculation. For time-to-event outcomes, censoring rules determine how incomplete follow-up contributes information. No rules for these issues are described in the record, so it is not possible to assess their influence on the study’s efficacy findings.

These omissions do not prove that no statistical plan or analyses existed. They mean that the relevant methods and results are not reported in the registry record, limiting independent interpretation.

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: death nos (23.1%), abdominal pain (7.7%), colonic obstruction (7.7%), colonic perforation (7.7%), hepatic failure (7.7%).

The safety record lists participants affected by serious adverse events and other adverse events separately for phase I and phase II. It reports serious adverse events for 9 of 13 participants in phase I and 18 of 50 in phase II. Other adverse events are listed for 13 of 13 participants in phase I and 46 of 50 in phase II. These figures are counts of affected participants as listed in the registry, not a randomized comparison between treatments.

The reported serious-event terms include Death NOS at 23.1%, abdominal pain at 7.7%, colonic obstruction at 7.7%, colonic perforation at 7.7%, hepatic failure at 7.7%, and appendicitis at 7.7%. The registry record does not specify the denominator or phase allocation for these term percentages. It also does not report a deaths count in the safety data. The term Death NOS should therefore be read as a recorded serious-event term, not as a complete mortality analysis.

In general, adverse-event counts depend on event definitions, observation periods, and reporting practices. The registry record does not provide enough detail to compare exposure or follow-up between phases. Because the study was non-randomized and sequential, differences between the phase-specific figures should not be interpreted as treatment effects or as evidence that one phase had greater or lesser risk.

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

The study’s design addresses an early development question and a later outcome question. In phase I, it aimed to identify a recommended phase II dose and describe toxicities. In phase II, it aimed to determine progression-free survival for the combination, with additional objectives involving response, overall survival, and blood-based angiogenesis biomarkers. These objectives are not themselves results.

The registry record does show that the trial was completed and that it enrolled 63 participants. It describes the intended dose-escalation approach and reports participant counts affected by serious and other adverse events in the two study phases. It also lists selected serious-event terms. Those are descriptive details about the study record; they do not establish whether the combination improved outcomes or whether any observed safety pattern was caused by treatment.

The record does not report phase II progression-free survival results, response rate results, overall survival results, or the primary phase I dose-limiting toxicity counts. It also does not provide efficacy estimates with confidence intervals or a randomized comparator. Accordingly, the magnitude and precision of clinical outcomes cannot be assessed from the record, and no causal comparison with another treatment can be made.

In general, a single-arm study can describe outcomes among participants who received a regimen, but it cannot by itself separate treatment effects from other influences on those outcomes. Here, even those phase II outcome estimates are not reported in the registry record. The record supports an account of the trial’s objectives, design, and listed safety information, but not a conclusion about efficacy.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of the different questions asked during dose finding and later outcome evaluation. Phase I focused on selecting a dose and characterizing dose-limiting toxicities; phase II focused on progression-free survival. Keeping those endpoints distinct matters: safety observations used to guide dose selection do not answer whether a regimen changes disease outcomes.

The sequential, non-randomized design also illustrates why study structure determines the kinds of conclusions available. The phase I and phase II groups are not randomized comparison arms. Phase II was described as a single arm, so its stated objective was to determine survival associated with the regimen rather than establish a randomized treatment contrast. In general, a treatment-associated outcome in a single arm is not equivalent to a causal comparison against another treatment.

The registry entry further shows why outcome reporting and analysis details matter. It provides a detailed dose-limiting toxicity definition and selected safety counts, but does not report the phase I dose-limiting toxicity results or phase II efficacy results. It also does not report the statistical analysis plan, missing-data approach, analysis populations, or multiplicity methods. Without outcomes and methods, readers cannot assess estimates, precision, or the robustness of conclusions.

The statistical importance of the record is therefore in clarifying both the trial’s intended questions and the limits of what its registry results establish. Study objectives, design descriptions, and adverse-event listings should not be mistaken for complete efficacy evidence.

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.

The main trial-specific limitation is that the registry record does not report the results for the phase I primary endpoint or the phase II primary efficacy objective. It gives the dose-limiting toxicity definition and 28-day timeframe, but not the cohort counts. It states that progression-free survival was the phase II primary objective, but does not provide an estimate, confidence interval, or analysis result. Response rate and median survival are listed as secondary outcomes without numerical results.

The study was open-label, non-randomized, and sequential, with a single phase II arm. There is no randomized control group in the design described, so the study cannot provide a like-for-like randomized treatment comparison. The phase I and phase II groups had different roles and should not be treated as comparable arms. Enrollment was 63, but the record does not give further participant characteristics here, limiting assessment of who contributed to each outcome.

Safety interpretation is also constrained. The record lists phase-specific affected-participant counts and serious-event terms, but does not report the denominators or phase distribution for the term percentages. A deaths count is not reported in the safety data. The follow-up context for the safety listings is not described in the registry record.

Finally, the statistical analysis plan, sample-size justification, analysis populations, missing-data methods, interim analyses, and multiplicity procedures are not reported in the registry record. The stated planning assumptions are not the trial’s design justification and do not replace outcome results. These omissions limit reproducibility and assessment of precision.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Phase I/II Study of Capecitabine Plus Aflibercept to Treat Metastatic Colorectal Cancer — NCT01661972.
• Registry record last updated 2018-10-29; status last verified 2018-10.
• Results first posted 2017-07-11.
• 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

NCT01661972, titled “Phase I/II Study of Capecitabine Plus Aflibercept to Treat Metastatic Colorectal Cancer,” is listed as completed. It was an open-label, non-randomized, sequential phase I/II study with enrollment of 63 participants. The phase I portion used a stated 3+3 dose-escalation approach; the phase II portion was planned as a single arm after selection of a recommended phase II dose.

The phase I primary endpoint was the number of participants with dose-limiting toxicities in each cohort over 28 days. The phase II primary objective was progression-free survival, with secondary objectives concerning response rate, overall survival, and biomarkers. Phase-specific serious and other adverse-event participant counts and selected serious-event terms are recorded.

Phase I dose-limiting toxicity results and phase II efficacy results are not reported in the registry record. The analysis plan, missing-data methods, and multiplicity procedures are also not reported. The entry therefore describes trial design and selected safety information without providing the results needed to assess efficacy or its precision.