This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for A Study of Bevacizumab (Avastin) in Combination With Capecitabine (Xeloda) in Elderly Patients With Metastatic Colorectal Cancer (NCT00484939), a randomized phase 3 study. The primary outcome is Progression-free Survival.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00484939 |
| Phase | Phase 3 |
| Condition | Colorectal Cancer |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 280 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Hoffmann-La Roche |
| Results posted | Yes |
2. Clinical Question
Population
The trial enrolled elderly patients with metastatic colorectal cancer. The primary analysis population was the intent-to-treat population, defined in the registry record as all participants randomized into the study. The record reports enrollment of 280 participants. Further details about the age criteria and other eligibility requirements are not reported in the registry record.
Intervention
Participants in the combination arm received bevacizumab 7.5 mg/kg intravenously on Day 1 of each 3-week treatment cycle, together with capecitabine 1000 mg/m² orally twice daily on Days 1-14 of each cycle. The question was whether adding bevacizumab to capecitabine was associated with a different progression-free survival pattern than capecitabine alone.
Comparator
The comparator arm received capecitabine 1000 mg/m² orally twice daily on Days 1-14 of each 3-week treatment cycle. The registry describes the study as randomized, open-label, and parallel-group. Participants and study personnel were not masked, according to the record.
Primary question
The primary endpoint was progression-free survival, measured from randomization until disease progression or death from any cause, whichever occurred first. The registry reports median progression-free survival of 9.1 months for the combination arm and 5.1 months for capecitabine alone, with 95% confidence intervals for each median. The sponsor-reported log-rank result was p < 0.001. These results describe the recorded time-to-event comparison; they do not, on their own, establish the size or precision of a treatment effect expressed as a hazard ratio, which is not reported in the registry record.
3. Trial Design
This was a randomized, open-label, parallel-group phase III study of bevacizumab plus capecitabine versus capecitabine alone. The registry lists Hoffmann-La Roche as sponsor and describes the treatment as first-line therapy for elderly patients with metastatic colorectal cancer. Enrollment was 280. The study status is completed; the record lists a start of 2007-07 and primary completion of 2013-03.
Participants were assigned to one of two treatment groups. The combination group received bevacizumab 7.5 mg/kg intravenously on Day 1 of each 3-week cycle, as well as capecitabine 1000 mg/m² orally twice daily on Days 1-14. The comparator group received the same capecitabine schedule without bevacizumab. Because the trial was open-label, treatment assignment was not masked to participants or study staff. The method used to generate or conceal the random allocation is not reported in the registry record.
The primary outcome was progression-free survival in the intent-to-treat population. The record defines that population as all randomized participants. The primary endpoint timeframe extended from baseline to the end of the study, up to 5 years 8 months. The registry lists additional outcomes, including response, overall survival, performance status, additional treatment for malignancy, follow-up, and safety measures. Detailed numerical results are not reported in the registry record for every listed outcome.
The registry record does not report a sample-size justification, detailed allocation procedures, an interim-analysis plan, or a complete statistical analysis plan. Those omissions limit what can be concluded about the original design assumptions and prespecified analysis choices.
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
| Stage | Bevacizumab + Capecitabine | Capecitabine |
|---|---|---|
| STARTED | 140 | 140 |
| COMPLETED | 0 | 0 |
| NOT COMPLETED | 140 | 140 |
Reasons for not completing this period:
| Reason | Bevacizumab + Capecitabine | Capecitabine |
|---|---|---|
| Death | 9 | 13 |
| Adverse Event | 22 | 12 |
| Patient Withdrew Consent | 19 | 10 |
| Protocol Violation | 3 | 3 |
| Lost to Follow-up | 0 | 3 |
| Discretion of Investigator or Sponsor | 7 | 3 |
| Disease progression | 67 | 88 |
| Screen Failure | 2 | 2 |
| Reason Not Specified | 11 | 6 |
Analysis population for the primary outcome, as defined in the registry: Intent-to-treat population: All participants randomized into the 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.
Population: Intent-to-treat population: All participants randomized into the study.
| Characteristic | Unit | Bevacizumab + Capecitabine | Capecitabine | Total |
|---|---|---|---|---|
| Participants analysed | 140 | 140 | 280 | |
| Age, Continuous | years | 76.1 (SD 4.18) | 76.5 (SD 3.91) | 76.3 (SD 4.04) |
| Sex: Female, Male: Female | Participants | 56 | 56 | 112 |
| Sex: Female, Male: Male | Participants | 84 | 84 | 168 |
6. Primary Endpoint
The registry lists the primary outcome as Progression-free Survival, measured over Baseline to the end of the study (up to 5 years 8 months).
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | 95% Confidence Interval |
| Unit | Months |
| Endpoint classification | Time-to-Event Analysis |
Progression-free survival measures the time from randomization until the first recorded disease progression or death from any cause, whichever occurs first. It combines these events into a time-to-event outcome, so a participant’s result is not simply whether progression occurred by a fixed visit. The registry defines progression using changes in target lesions, progression of existing non-target lesions, or the appearance of new lesions.
For target lesions, the protocol description identifies measurable lesions and records their longest diameters. The baseline sum of longest diameters is the sum across the identified target lesions. The record defines progression as an increase of at least 20% in the target-lesion sum, using the smallest sum recorded since treatment began as the reference, unequivocal progression of existing non-target lesions, or one or more new lesions.
Time-to-event outcomes are useful when the timing of an event matters and participants may have different lengths of follow-up. A median gives a central summary of the observed event-time distribution; a confidence interval describes uncertainty around that estimate. The registry reports a median and 95% confidence interval for each arm. A median alone does not describe the full event-time pattern or quantify a relative hazard between groups. The sponsor’s log-rank test supplies a comparison of the time-to-event distributions, but a hazard ratio and its confidence interval are not reported in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry reports median progression-free survival of 9.1 months for bevacizumab plus capecitabine, with a 95% confidence interval of 7.3 to 11.3 months. For capecitabine alone, the median was 5.1 months, with a 95% confidence interval of 4.3 to 6.3 months. The sponsor-reported log-rank test comparing the groups gave p < 0.001. Taken together, the posted results indicate a difference between the observed time-to-event distributions in this trial, with a longer reported median in the combination arm.
The confidence intervals express uncertainty around each arm’s median estimate. They are not confidence intervals for the difference between arms. The log-rank p-value assesses the evidence against a null comparison under the test’s assumptions; it does not state the size of the difference, the probability that a treatment effect is real, or the probability that the null hypothesis is true. The registry does not report a hazard ratio or its confidence interval, so the magnitude of a relative time-to-event effect and its precision cannot be read from a hazard-ratio estimate.
The primary endpoint is defined in the intent-to-treat population, which includes all randomized participants according to the registry description. The arm results each list 140 participants. The record does not describe in detail how missing assessments, censoring, or protocol deviations were handled in the primary analysis. Those details matter because time-to-event estimates rely on follow-up information and on assumptions about participants whose event time is not observed.
The p-value is evidence about the group comparison under the reported log-rank analysis, not a complete account of clinical importance. Interpreting clinical importance would require context beyond statistical significance, including the estimated relative effect, uncertainty, outcome definitions, safety, and applicability to the intended population. A hazard ratio, its confidence interval, and more detailed analysis information are not reported in the registry record.
The result applies to the study’s randomized comparison and its defined population. It does not by itself establish what would happen in other populations, under different treatment schedules, or in usual-care settings. The registry’s open-label design and incomplete reporting of some analytic details also warrant care in interpreting the result. The endpoint and test should be read alongside the reported safety information and the limits of the record.
9. Statistical Methods Explained
The main statistical feature of this outcome is that progression-free survival is a time-to-event endpoint. Such outcomes preserve information about when an event occurs, rather than reducing follow-up to a simple event/no-event outcome at one chosen time. The reported arm summaries are medians, and the sponsor’s comparison uses a log-rank test.
What does a median time-to-event estimate mean?
A median is a central point in the estimated event-time distribution: it marks the time by which half of the population is estimated to have experienced the event, under the analysis method. It is not the average participant’s duration of benefit, and it does not imply that every participant had an event at that time. The registry reports median progression-free survival of 9.1 months in the combination arm and 5.1 months in the capecitabine arm. These arm-specific medians provide an accessible summary, but they do not fully describe how event times are distributed over follow-up.
A median can be useful when event times are skewed or when the mean is not a suitable summary. However, comparing two medians is not the same as estimating a treatment effect over the whole follow-up period. The relative event rate may vary over time, and the medians do not show whether that happens. The registry does not report a hazard ratio or an analysis of whether hazards were proportional.
How should the confidence intervals be read?
A confidence interval represents uncertainty in an estimate under the statistical procedure used to construct it. The registry reports 95% confidence intervals around each median: 7.3 to 11.3 months for the combination arm and 4.3 to 6.3 months for capecitabine alone. These intervals concern the separate medians, not the difference between them. A confidence interval is not a range containing every participant’s event time, and it is not a probability statement about the true value after observing the data.
To quantify uncertainty in a comparison, a direct treatment-effect estimate with its own confidence interval would be informative. The registry does not report a hazard ratio or a confidence interval for a relative effect. The overlap or separation of separate arm-specific intervals is not a substitute for a formal comparison. The reported log-rank test addresses a group comparison, but its p-value does not give the size or precision of a treatment effect.
What does the log-rank test assess?
In general, a log-rank test compares event-time patterns between groups while accounting for the ordering of observed events and follow-up. Its null hypothesis concerns whether the groups have the same event-time experience under the test. The sponsor-reported result is p < 0.001. This is evidence against that null comparison under the test and analysis conditions; it is not a measure of how large the difference is or how likely the observed result is under a treatment-effect explanation.
The interpretation of a time-to-event test depends in part on how participants without an observed event are handled. In general, those participants may contribute follow-up until their last known event-free assessment, a process commonly called censoring. The registry record does not detail the censoring rules, handling of missing assessments, or related sensitivity analyses for this endpoint. It is therefore not possible to assess those choices from the record alone.
Why is the analysis population important?
The registry says the primary population was intent-to-treat and defines it as all participants randomized into the study. This approach generally preserves the randomized comparison by analyzing participants according to their assigned group, regardless of treatment received after assignment. The record lists 140 participants in each primary-outcome arm. Further details about analysis exclusions, missing outcome data, and adherence to assigned treatment are not reported in the registry record.
For secondary outcomes, the posted calculations include unadjusted re-calculations from summary statistics for additional treatment for malignancy and duration of follow-up. These are descriptive, simple comparisons, not replacements for the sponsor’s analysis plan. The risk-difference interval for the additional-treatment outcome spans values in both directions, and the unadjusted p-value is 0.811. For follow-up, the mean difference is reported with an interval spanning values in both directions, and the unadjusted p-value is 0.214. These results do not provide clear evidence of a group difference in those measures. No calculation from these outcomes should be interpreted as the trial’s primary treatment-effect analysis.
What do planning illustrations mean?
Planning calculations based on observed effects can illustrate how sample-size requirements change when the effect targeted for detection is smaller or larger. They are not observed power and do not show what the trial was designed to detect. The planning assumptions described in the registry record use a two-sided significance level of 0.05, equal allocation, and a normal approximation. The calculations are illustrations based on observed effects, not the trial’s own design justification. The record does not report the original sample-size assumptions or justification.
For time-to-event outcomes, general planning methods often focus on the number of events rather than only the number of participants, because event information drives precision. The registry’s planning note describes total event counts for time-to-event planning illustrations and participants per arm for other outcomes. Such planning figures should be kept separate from the observed trial results and from the sponsor-reported primary analysis.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Log Rank | not reported | < 0.001 |
The registry reports the sponsor’s primary comparison as a log-rank test with p < 0.001. It also reports the median progression-free survival and 95% confidence interval separately for each arm. No sponsor-reported hazard ratio, confidence interval for a relative effect, or other effect-size estimate is included in the registry record.
The posted secondary calculations for additional treatment for malignancy and duration of follow-up are unadjusted re-calculations from the registry’s arm-level summary statistics. They are not the sponsor’s reported analyses: the registry lists no reported comparison analysis for these secondary outcomes. The unadjusted results are therefore descriptive checks based on the posted summaries, not a replacement for a prespecified sponsor analysis.
The record indicates no discrepancy between the available sponsor analysis and the posted recalculations. For the primary endpoint, no two-arm re-calculation is reported; the registry identifies values as missing or not estimable for that type of comparison. Consequently, the sponsor’s log-rank result remains the available formal primary comparison in the record. Neither a secondary unadjusted p-value nor an arm-specific median should be substituted for that analysis. The registry does not provide enough detail to reproduce the sponsor’s full time-to-event analysis independently.
11. Multiplicity, Missing Data and Analysis Populations
The primary analysis population is identified as the intent-to-treat population, defined as all randomized participants. The record reports 140 participants in each arm for the primary endpoint. This provides a stated population basis for interpreting the primary comparison, but the registry does not give a participant-level account of missing assessments, exclusions, treatment discontinuations, or deviations from assigned therapy.
For a time-to-event outcome, incomplete observation is especially relevant. In general, participants who have not experienced an event by their last assessment may contribute information up to that time, but the validity of the resulting estimate depends on how follow-up and censoring are defined. The registry record does not report detailed censoring rules, reasons for loss to follow-up, missing-data procedures, or sensitivity analyses for progression-free survival. It is therefore not possible to evaluate how robust the primary estimate is to alternative assumptions about incomplete follow-up.
The record names multiple secondary outcomes. When many outcomes or comparisons are examined, the chance of at least one apparently notable result can increase. A statistical plan may control this through a hierarchy or multiplicity adjustment, but such a plan is not reported in the registry record. The unadjusted secondary re-calculations should not be treated as confirmatory tests, and the record does not establish which secondary outcomes, if any, were intended to support formal claims.
The registry also does not report a sample-size justification, interim-analysis plan, or full statistical analysis plan. It is not possible to determine from the record whether the sample size was chosen to detect a particular progression-free survival effect, whether interim monitoring was planned, or how the sponsor specified all analysis decisions before seeing results. Planning illustrations based on observed effects do not fill those gaps; they are not the original design rationale or observed power.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Bevacizumab + Capecitabine | 40 | 134 | 29.9% |
| Capecitabine | 42 | 136 | 30.9% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Bevacizumab + Capecitabine | 129 of 134 |
| Capecitabine | 130 of 136 |
Most frequent serious adverse event terms
| Event term | Organ system | Bevacizumab + Capecitabine | Capecitabine |
|---|---|---|---|
| Pulmonary embolism | Respiratory, thoracic and mediastinal disorders | 8 of 134 | 2 of 136 |
| Abdominal pain | Gastrointestinal disorders | 3 of 134 | 4 of 136 |
| Intestinal obstruction | Gastrointestinal disorders | 3 of 134 | 4 of 136 |
| Pyrexia | General disorders | 3 of 134 | 2 of 136 |
| Vomiting | Gastrointestinal disorders | 2 of 134 | 2 of 136 |
| Diarrhoea | Gastrointestinal disorders | 0 of 134 | 4 of 136 |
| Pneumonia | Infections and infestations | 3 of 134 | 1 of 136 |
| Ascites | Gastrointestinal disorders | 0 of 134 | 3 of 136 |
| General physical health deterioration | General disorders | 1 of 134 | 2 of 136 |
| Hypertension | Vascular disorders | 2 of 134 | 1 of 136 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Bevacizumab + Capecitabine | Capecitabine |
|---|---|---|---|
| Palmar-Plantar erythrodysaesthesia syndrome | Skin and subcutaneous tissue disorders | 66 of 134 | 54 of 136 |
| Diarrhoea | Gastrointestinal disorders | 54 of 134 | 48 of 136 |
| Nausea | Gastrointestinal disorders | 32 of 134 | 37 of 136 |
| Fatigue | General disorders | 32 of 134 | 37 of 136 |
| Decreased Appetite | Metabolism and nutrition disorders | 38 of 134 | 31 of 136 |
| Abdominal pain | Gastrointestinal disorders | 31 of 134 | 21 of 136 |
| Asthenia | General disorders | 30 of 134 | 22 of 136 |
| Vomiting | Gastrointestinal disorders | 28 of 134 | 16 of 136 |
| Constipation | Gastrointestinal disorders | 25 of 134 | 19 of 136 |
| Pyrexia | General disorders | 24 of 134 | 16 of 136 |
Serious events with the highest arm-level rate: pulmonary embolism (6.0%), abdominal pain (2.9%), intestinal obstruction (2.9%), diarrhoea (2.9%), pyrexia (2.2%).
The registry reports serious adverse events among 134 participants at risk in the bevacizumab plus capecitabine group and 136 participants at risk in the capecitabine group. Serious adverse events affected 40 and 42 participants, respectively. Other adverse events affected 129 and 130 participants in the corresponding groups. These counts describe the safety data as recorded; they do not establish whether an event was caused by study treatment.
The listed serious adverse-event terms include pulmonary embolism at 6.0%, abdominal pain at 2.9%, intestinal obstruction at 2.9%, diarrhoea at 2.9%, pyrexia at 2.2%, and pneumonia at 2.2%. These are reported event-term percentages, not a comprehensive account of every possible safety outcome. The registry does not report deaths in the adverse-event summary.
Safety comparisons require attention to the definitions used, the observation period, and which participants contributed to each analysis. The registry’s brief summary states that no notable trends or interactions in laboratory values, electrocardiogram, or vital signs suggesting an effect in either direction were observed. That statement is limited to those measures and does not replace a full assessment of adverse events or their severity, timing, and attribution.
13. What the Result Does, and Does Not, Mean
The primary results show a longer reported median progression-free survival in the bevacizumab plus capecitabine arm than in the capecitabine-alone arm, alongside a sponsor-reported log-rank p-value of p < 0.001. This describes what the registry reports for the randomized trial. It is not a guarantee about an individual participant’s outcome, and a median does not predict how long any particular person will remain progression-free.
The result does not provide a reported hazard ratio or confidence interval for a relative treatment effect. The separate confidence intervals around the medians describe uncertainty in each arm’s median, not uncertainty in the difference between arms. Statistical significance and effect size are different concepts: the p-value indicates evidence under a test, while a direct effect estimate and its uncertainty would be needed to characterize the magnitude more fully.
The registry’s secondary results do not replace the primary analysis. The available unadjusted comparisons of additional treatment and follow-up are uncertain and do not supply independent confirmation of the primary result. Other listed secondary outcomes lack detailed numerical results in the registry record. Multiplicity, missing-data procedures, and several design and analysis details are also not reported.
The trial result is specific to the studied population and treatment schedules: elderly patients with metastatic colorectal cancer receiving first-line treatment, with the combination of bevacizumab and capecitabine compared with capecitabine alone. The record does not support conclusions about other populations, regimens, or settings. The open-label design, limited reporting of analytic details, and safety findings are relevant context when interpreting the endpoint. The statistical results describe a group comparison; they do not by themselves determine the overall balance of benefits and harms or make a treatment recommendation.
14. Why This Trial Matters Statistically
This trial illustrates why time-to-event results need more than a single p-value. The registry provides arm-specific medians and confidence intervals, as well as a sponsor-reported log-rank test. Those quantities answer related but distinct questions: the medians summarize each arm’s event-time distribution, the intervals express uncertainty in those summaries, and the test evaluates a group comparison. None alone provides a complete account of the treatment effect.
It also shows the importance of reporting a direct effect estimate. Without a hazard ratio and its confidence interval, readers cannot use the registry record to assess the relative event rate over follow-up or the precision of that relative estimate. A significant test can indicate evidence of a difference while leaving the magnitude less fully characterized.
The distinction between primary and secondary outcomes matters as well. The primary endpoint has a sponsor-reported comparison, whereas the available secondary re-calculations are unadjusted and do not come with sponsor-reported comparisons in the record. When numerous outcomes are listed, uncertainty about multiplicity and analysis hierarchy affects how much weight secondary findings can carry.
Finally, a registry record can report a completed randomized trial and key results while leaving important design and analysis details unavailable. Here, the sample-size justification, interim-analysis plan, full statistical plan, and details of missing-data handling are not reported. Recognizing these boundaries helps readers separate the observed results from assumptions that cannot be checked from the record.
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 record describes the trial as randomized and open-label, but does not report the random-sequence method or allocation concealment. Because masking was absent, knowledge of assignment could potentially influence aspects of care or assessment; the extent of any such influence is not reported. The primary endpoint includes disease progression, and the registry gives a lesion-based definition, but detailed information about assessment schedules and independent review is not reported in the registry record.
Although the primary population is identified as intent-to-treat, details of missing assessments, censoring, withdrawals, and analysis exclusions are not reported. The record also does not include a hazard ratio or confidence interval for the primary comparison, limiting assessment of the relative effect and its precision. The reported arm-specific median intervals do not provide that missing comparison estimate.
The sample-size justification, interim-analysis plan, complete statistical analysis plan, and multiplicity strategy are not reported. This limits assessment of the planned power, whether analysis decisions were prespecified, and how secondary findings should be interpreted. Detailed numerical results for several listed secondary outcomes are also not reported in the registry record.
The registry describes an elderly population with metastatic colorectal cancer, but the specific age eligibility criteria and other details needed to assess representativeness are not reported. Therefore, the applicability of these findings beyond the enrolled population and treatment schedules cannot be determined from the record alone. The safety summary gives counts and selected serious adverse-event terms, but does not provide a complete account of event timing, attribution, or all safety outcomes.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: A Study of Bevacizumab (Avastin) in Combination With Capecitabine (Xeloda) in Elderly Patients With Metastatic Colorectal Cancer — NCT00484939.
- Registry record last updated 2015-01-08; status last verified 2015-01.
- Results first posted 2014-06-10.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
NCT00484939 is a completed, randomized, open-label, parallel-group phase III study sponsored by Hoffmann-La Roche. It compared bevacizumab plus capecitabine with capecitabine alone as first-line treatment for elderly patients with metastatic colorectal cancer. Enrollment was 280, and the registry lists a start of 2007-07 and primary completion of 2013-03.
The primary endpoint was progression-free survival in the intent-to-treat population. The registry reports median values of 9.1 months and 5.1 months, with arm-specific 95% confidence intervals, and a sponsor-reported log-rank result of p < 0.001. A hazard ratio and its confidence interval are not reported. Additional outcome and safety information is available, but the record does not report several design, missing-data, and statistical-plan details needed for a fuller independent assessment.