Related Experiment Videos
A comparison of randomized concurrent control groups with matched historical control groups: are historical controls
Summary
Historical control groups in clinical trials do not accurately predict patient outcomes. This study found significant survival differences compared to randomized concurrent control groups, invalidating their use as replacements.
Area of Science:
- Clinical research methodology
- Oncology and hematology
Background:
- Historical control groups are often used in clinical trials to avoid randomization.
- The validity of using historical controls relies on the assumption that they accurately reflect outcomes seen in concurrent randomized groups.
Purpose of the Study:
- To test the assumption that historical control groups approximate the survival and relapse-free survival of randomized concurrent control groups.
- To compare survival and relapse-free survival between matched historical and randomized concurrent control groups.
Main Methods:
- Comparison of survival and relapse-free survival data.
- Matching of historical control groups to randomized concurrent control groups based on disease, stage, and follow-up duration.
- Statistical analysis of discrepancies in outcomes between matched groups.
Main Results:
- Significant variations were observed in 42% of the 43 matched control groups ( >10 percentage points difference).
- 21% of groups varied by over 20 percentage points, and 5% by over 30 percentage points.
- In 17 out of 18 groups with >10 percentage points variation, randomized concurrent controls showed superior survival or relapse-free survival.
Conclusions:
- The assumption that historical control groups can replace randomized concurrent control groups is not supported by this data.
- Differences in survival and relapse-free survival suggest historical controls may not accurately represent contemporary patient populations or treatment standards.