← Clinical Trial Results
Gastric Cancer Phase 2/3 Completed NCT01216644

FLOT4: Complete Statistical Analysis of FLOT in Gastric Cancer

An independent statistical review of the randomized FLOT4 trial comparing 5-fluorouracil, leucovorin, oxaliplatin and docetaxel (FLOT) versus epirubicin, cisplatin and 5-fluorouracil (ECF) in patients with locally advanced, resectable gastric cancer.

ClinicalTrials.gov identifier: NCT01216644
Scope of this record

This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.

1. Trial at a Glance

FLOT4 was a randomized phase 2/3 treatment trial evaluating perioperative chemotherapy regimens in patients with locally advanced, resectable gastric cancer.

716
Enrollment
2
Arms
2/3
Phase
2 years
Primary endpoint follow-up
FeatureFLOT4
ConditionGastric Cancer
DesignRandomized, parallel assignment
MaskingNone
Primary purposeTreatment
Enrollment716
SponsorKrankenhaus Nordwest
StatusCompleted

2. Clinical Question

Population

Patients with locally advanced, resectable gastric cancer.

Intervention

5-Fluorouracil, leucovorin, oxaliplatin and docetaxel (FLOT).

Comparator

Epirubicin, cisplatin and 5-fluorouracil (ECF).

Primary question

Whether the FLOT regimen should be compared with ECF for the registered primary endpoint of median overall survival.

3. Trial Design

01
Randomization
716 patients
02
Parallel groups
2 arms
03
Treatment
FLOT vs ECF
04
Follow-up
2 years
05
Analysis
Overall survival

FLOT arm

5-Fluorouracil, leucovorin, oxaliplatin and docetaxel.

ECF arm

Epirubicin, cisplatin and 5-fluorouracil.

4. Endpoints

EndpointTime frame
Median overall survival2 years follow-up

5. Planned Analysis

The registry identifies median overall survival as the primary endpoint with a time frame of 2 years follow-up. No statistical analyses are posted in the ClinicalTrials.gov record.

Overall survival analysis

For randomized oncology trials, overall survival is commonly evaluated using time-to-event methods. Typical analyses include Kaplan-Meier estimation to describe survival over time, log-rank testing to compare randomized groups, and Cox proportional-hazards models to estimate relative treatment effects.

S(t) = probability of remaining alive beyond time t

6. Statistical Methodology

Randomization

Randomization creates comparable treatment groups on average and provides the foundation for causal comparison between the assigned interventions.

Time-to-event analysis

Overall survival is a time-to-event endpoint because patients may have different follow-up durations. Censoring allows patients without an observed event at the last follow-up time to contribute available information.

Median survival

The median overall survival is the time at which the estimated survival probability reaches 50%. It summarizes the center of the survival distribution rather than describing every patient's outcome.

7. Statistical Methods Explained

Why is overall survival a time-to-event endpoint?

Death can occur at different times for different patients, so the analysis must incorporate both event times and varying follow-up duration.

Why use Kaplan-Meier estimation?

Kaplan-Meier methods estimate survival probabilities while accounting for patients whose follow-up ends before the event is observed.

What does a hazard ratio measure?

A hazard ratio compares estimated instantaneous event rates between groups over the analyzed follow-up. It is not the same as a ratio of median survival times.

Why does randomization matter?

Randomization reduces systematic differences between groups, allowing observed outcome differences to be interpreted in relation to the assigned treatments.

Why is confidence interval reporting important?

Confidence intervals describe uncertainty around an estimated effect and provide information about precision.

8. Limitations

9. Why This Trial Matters Statistically

ConceptApplication
RandomizationComparison of two treatment strategies in parallel groups.
Time-to-event analysisPrimary endpoint based on overall survival.
CensoringRequired when follow-up differs among patients.
Median survivalSummary measure for survival distributions.

10. Sources