Study of Fruquintinib in Patients With Metastatic Colorectal Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Fruquintinib in Patients With Metastatic Colorectal Cancer (NCT02196688), a randomized phase 2 study. The primary outcome is Progression Free Survival (PFS).

2. Clinical Question

3. Trial Design

NCT02196688 is a completed randomized phase II study of fruquintinib in patients with metastatic colorectal cancer. Its official title describes a multicenter, double-blind, placebo-controlled comparison of fruquintinib plus best supportive care with placebo plus best supportive care in third-or-later therapy. The record lists colorectal cancer as the condition and Hutchison Medipharma Limited as the sponsor. The study began in 2014-04, and primary completion is recorded as 2015-04.

The design used parallel treatment arms. The treatment arm included 47 participants and the control arm included 24, for a total enrollment of 71. The registry records randomized allocation and quadruple masking. The intervention schedule in both arms was once-daily oral study treatment for three weeks, followed by one week off, within four-week cycles. Both arms also received best supportive care. The treatment arm received fruquintinib 5mg; the control arm received placebo described as 5mg. The arm descriptions are truncated in the record, so they do not fully establish the stopping conditions.

The primary endpoint was progression-free survival. The registry specifies an intention-to-treat analysis for progression-free survival, overall survival, objective response rate, and disease control rate, with participants analyzed according to randomized treatment. Objective response rate, disease control rate, and overall survival are listed as secondary outcomes. Results are posted for the two response-related outcomes, but overall-survival results are not reported in the registry record. The record does not report a detailed statistical analysis plan, sample-size justification, interim-analysis plan, or sponsor-reported comparative analysis for the primary endpoint.

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

Analysis population for the primary outcome, as defined in the registry: The intention-to-treat set will contain all subjects in the Randomized set (RND) set subjects will be classified according to randomized treatment. The intent-to-treat principle is preserved. The Intention to Treat (ITT) will be used for analyses of PFS, OS, ORR and DCR.

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 Progression Free Survival (PFS), measured over From randomization until the date of first documented progression or date of death from any cause, whichever came first..

Progression-free survival measures the time from randomization until the first of documented disease progression or death from any cause. In this study, progression was determined by the investigator using RECIST v1.1 criteria. The registry defines the primary outcome as median progression-free survival in months, with a 95% confidence interval.

In general, a time-to-event endpoint incorporates both when an event occurs and whether follow-up ends before an event is observed. This makes it useful when participants may have different lengths of observation or may not experience progression or death during the recorded period. The registry describes censoring rules: the final tumor evaluation date is used for participants without progression or death by that date; randomization is used for those with no death and no post-baseline tumor evaluation; and a recorded death counts as an event even without a post-baseline tumor evaluation.

The median is the time by which half of the analyzed group has experienced the event, under the endpoint’s event and censoring definitions. It is an arm-specific summary, not a direct estimate of the difference between arms. A comparison requires an appropriate contrast and its uncertainty. A two-arm comparative estimate for progression-free survival is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Progression-free survival is a time-to-event endpoint. Each participant’s time begins at randomization and ends at progression or death, whichever occurs first. The registry defines progression using investigator assessment and RECIST v1.1. Participants who have not had an event by the end of their observed follow-up contribute information up to a censoring date rather than being treated as if an event had occurred. Censoring is necessary because not every participant will necessarily have a documented event during the available follow-up.

What does the reported median describe?

The reported median is an estimate of the time at which the event has occurred for half of the group, accounting for the endpoint’s event and censoring rules. It is a compact summary of a time-to-event distribution. The 95% confidence interval expresses uncertainty around that median estimate: under the statistical procedure used, intervals constructed in this way would cover the underlying quantity at the stated rate over repeated applications. It does not mean that 95% of participants’ event times lie inside the interval.

Here, the treatment-arm median is 4.731 months with a 95% confidence interval of 2.858 to 5.585 months. The control-arm median is 0.986 months with a 95% confidence interval of 0.953 to 1.577 months. Those estimates describe each arm separately. A contrast between arms requires its own model or estimator and uncertainty interval. The registry record says that values are missing or not estimable for a two-arm comparison of this median-time endpoint. It does not report a sponsor-derived hazard ratio or another comparative estimate for progression-free survival. No such estimate can be reconstructed from the two medians alone.

How should the binary-outcome comparisons be read?

Objective response and disease control are posted as participant counts. The accompanying recalculated comparisons use unadjusted risk differences, which express the difference between the observed proportions in the two arms. The objective-response risk difference is 0.021, with a 95% confidence interval from -0.02 to 0.062 and p = 0.472. The disease-control risk difference is 0.472, with a 95% confidence interval from 0.262 to 0.683 and p < 0.001. These are simple unadjusted recalculations from the posted counts, not sponsor-reported analyses; the record lists no reported analyses for these outcomes.

In general, a risk-difference confidence interval describes uncertainty in the estimated difference in proportions. If it includes zero, the data are compatible with no difference as well as with differences on either side of zero at the interval’s confidence level. A p-value measures how incompatible the observed data are with a specified null model under the assumptions of the test. It is not the probability that the null hypothesis is true, nor does it establish the size or clinical importance of an effect. For the disease-control comparison, the interval is wholly above zero and the unadjusted p-value is less than 0.001; that statistical description does not resolve the limitations of an unadjusted secondary analysis.

The record also gives a planning illustration for disease control. In general, such calculations estimate the number of participants per arm that would be needed under specified assumptions for a chosen power and effect size. This illustration assumes a two-sided significance level of 0.05, equal allocation, and a normal approximation, and varies the effect to detect as a fraction of the observed effect. It is a planning exercise based on the observed result, not the trial’s design justification and not observed power. The registry record does not report the trial’s own sample-size justification or its planned target effect.

Why do endpoint definitions and censoring matter?

Time-to-event estimates depend on the event definition, follow-up, and censoring rules. The registry states that participants without progression or death by their final tumor evaluation are censored at that evaluation; participants with no death and no post-baseline tumor evaluation are censored at randomization. A participant recorded as dead is counted as a progression-free-survival event even without a post-baseline tumor evaluation. These definitions clarify how some incomplete follow-up situations were handled, but the registry record does not report the number censored, the completeness of follow-up, or additional sensitivity analyses for missing data.

In general, censoring methods rely on assumptions about the relationship between the reason follow-up ends and the risk of the event. If censoring is related to prognosis in ways not adequately addressed, estimates can be difficult to interpret. The registry’s endpoint description gives operational censoring rules but does not provide enough detail to assess those assumptions or the extent of missing follow-up. The record also does not report a detailed statistical analysis plan, adjusted analyses, or interim-analysis procedures. Consequently, the posted statistics support a descriptive account of the trial but do not establish all the prespecified analysis choices or their execution.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists progression-free survival as the primary outcome and objective response rate, disease control rate, and overall survival as secondary outcomes. It does not report a multiplicity strategy, such as a hierarchy among outcomes or an adjustment to control the chance of false-positive findings across several tests. The binary comparisons available for objective response and disease control are unadjusted recalculations, and the record lists no sponsor-reported analyses. Their p-values therefore do not reflect an adjustment across endpoints. Overall-survival results are not reported in the registry record.

The record also does not report whether interim analyses were planned or conducted, whether any analysis was prespecified beyond the endpoint descriptions, or whether any decisions about stopping or changing the study were based on emerging results. Those details should not be inferred from the completed status. A detailed statistical analysis plan is not reported in the registry record, so it is not possible to verify the complete planned testing strategy or compare it with the posted results.

The primary endpoint description does give an intention-to-treat population: all participants in the randomized set, classified according to randomized treatment. It states that the intention-to-treat principle is preserved and that this population is used for progression-free survival, overall survival, objective response, and disease control. This definition preserves the randomized group assignment in the analysis population, but the record does not give participant-level accounting of missing assessments, exclusions, or deviations from assigned treatment.

Some censoring rules for progression-free survival are specified, including the date used when no event is observed and the treatment of recorded deaths without post-baseline tumor evaluation. The number of censored participants, missing tumor evaluations, and reasons for incomplete follow-up are not reported in the registry record. Sensitivity analyses for missing data are also not reported. These gaps matter because time-to-event summaries depend on follow-up and censoring, while binary endpoints depend on how participants without an assessable outcome are handled. The endpoint descriptions provide useful definitions but do not fully document missing-data handling across all outcomes.

11. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hypertension (6.4%), obstruction intestinal (4.3%), pyrexia (4.3%), blood bilirubin increased (4.3%), functional ileus (4.2%).

The registry reports serious adverse events affecting 12 of 47 participants in the treatment arm and 5 of 24 in the control arm. Other adverse events are recorded for 47 of 47 participants in the treatment arm and 20 of 24 in the control arm. The death counts are 38 in the treatment arm and 19 in the control arm. These are descriptive counts for the recorded groups; the registry figures alone do not establish that an event was caused by study treatment.

The listed serious-event terms include hypertension at 6.4%, obstruction intestinal at 4.3%, pyrexia at 4.3%, blood bilirubin increased at 4.3%, functional ileus at 4.2%, and sudden death at 4.2%. The record does not give a comparative statistical analysis of these terms. The percentages should be read as reported and not treated as evidence of a causal relationship or as a substitute for the full safety record.

In general, adverse-event counts describe how many participants had events recorded under the registry’s definitions. Comparing counts across arms requires attention to denominators, observation time, event definitions, and whether events were considered treatment-related. Those details are not fully. The death counts should not be interpreted as treatment-attributable deaths from the information in the registry record alone.

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

The posted results describe a randomized comparison of fruquintinib plus best supportive care with placebo plus best supportive care in a phase II trial of metastatic colorectal cancer. They report arm-specific median progression-free survival estimates and confidence intervals, alongside secondary response and disease-control counts. The primary endpoint estimates are descriptively different, but the registry record does not report a formal comparative estimate for progression-free survival. Without that contrast and its uncertainty, the arm-specific medians do not provide a complete statistical comparison.

The record also includes an unadjusted recalculation for each posted binary outcome. These results summarize the observed counts under simple comparisons, but they are not sponsor-reported analyses and do not account for multiplicity. The disease-control comparison has a risk-difference interval above zero, while the objective-response interval includes zero. Neither result should be taken as a substitute for a prespecified primary endpoint analysis. Overall-survival results are not reported in the registry record, even though overall survival is listed as a secondary outcome.

In general, randomized allocation helps create comparable groups at the start of a trial, but it does not remove uncertainty in the estimates or fill gaps in reporting. The sample-size illustration based on disease control is a planning exercise using the observed effect and specified assumptions, not evidence about the power achieved by this completed trial. The registry record does not report the trial’s original sample-size justification, detailed analysis plan, multiplicity approach, interim analyses, or full missing-data accounting.

These results therefore support a statistical description of what was posted, not a standalone conclusion about treatment benefit, harm, or clinical use. They should be interpreted with the study’s phase II design, unequal arm sizes, endpoint-specific limitations, and incomplete comparative reporting in mind. No medical recommendation follows from these registry summaries.

13. Why This Trial Matters Statistically

This trial illustrates why a reported endpoint and a reported treatment comparison are not interchangeable. Progression-free survival is the primary outcome, and the registry gives a median and confidence interval for each randomized arm. However, the record does not provide a two-arm estimate for that primary time-to-event outcome. A reader can describe the arm-specific summaries, but should not turn them into an unreported hazard ratio, risk difference, or formal test.

The record also demonstrates how secondary outcomes can add descriptive context while raising separate inferential questions. The unadjusted recalculations for response and disease control use posted counts, but they do not account for testing multiple outcomes. A statistically notable secondary comparison is not automatically confirmatory, especially when the analysis hierarchy and multiplicity approach are not reported. The absence of overall-survival results further limits a full assessment across the listed outcomes.

Another statistical lesson concerns planning calculations. A sample-size illustration based on an observed effect can show how assumptions about the target effect and power influence planning requirements. It cannot explain whether this trial was adequately planned, because the original sample-size justification is not reported, and it is not a measure of observed power. Separating planning assumptions from trial results prevents an illustrative calculation from being mistaken for a property of the completed study.

Finally, the endpoint description makes censoring rules explicit, but the registry does not report the amount of censoring or the extent of missing follow-up. Time-to-event summaries are interpretable only in relation to event definitions, follow-up, and censoring. The combination of clear endpoint definitions and incomplete comparative reporting makes this record useful for statistical reading: it shows both what the posted estimates convey and what remains unreported.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several limitations are specific to the posted record. The most important is the absence of a comparative analysis for the primary progression-free-survival endpoint. The record reports arm-specific medians and confidence intervals, but states that a two-arm comparison is missing or not estimable. It gives no sponsor-reported hazard ratio, comparative confidence interval, or p-value for progression-free survival. The median summaries therefore cannot answer the full comparative question on their own.

The record lists overall survival as a secondary outcome, but does not report its results. It also contains only posted results for objective response and disease control, with no sponsor-reported analyses. The available binary comparisons are unadjusted recalculations, and no multiplicity strategy is reported. This constrains the weight that can be placed on secondary p-values, particularly when several outcomes are listed.

The randomized groups are unequal in size, with 47 participants in the treatment arm and 24 in the control arm. The registry does not report a rationale for this allocation or the trial’s sample-size justification. The planning illustration based on disease control is computed from the observed effect and cannot establish the original design assumptions or observed power.

Although the registry describes intention-to-treat analysis and several censoring rules, it does not report the number of participants censored, missing assessments, reasons for missing follow-up, or sensitivity analyses. The arm descriptions are truncated, leaving the full stopping conditions unclear. A detailed statistical plan and interim-analysis information are not reported in the registry record. These gaps limit assessment of the analysis choices and the completeness of outcome reporting.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Study of Fruquintinib in Patients With Metastatic Colorectal Cancer — NCT02196688.
• Registry record last updated 2020-06-16; status last verified 2020-06.
• Results first posted 2019-08-13.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT02196688 is a completed randomized, double-blind, placebo-controlled, multicenter phase II trial sponsored by Hutchison Medipharma Limited. It enrolled 71 participants with colorectal cancer and compared fruquintinib plus best supportive care with placebo plus best supportive care. The primary outcome was median progression-free survival, defined from randomization to progression or death.

The registry reports arm-specific medians and 95% confidence intervals but no estimable two-arm primary comparison. Objective response and disease control are posted as secondary results; overall-survival results are not reported in the registry record. Binary comparisons are unadjusted recalculations, and a multiplicity strategy, detailed statistical analysis plan, and original sample-size justification are not reported.