Sequential Analysis in Cancer Trials
Author Information
Author(s): A.N. Donaldson, J. Whitehead, R. Stephens, D. Machin
Primary Institution: University of Reading; MRC Cancer Trials Office
Hypothesis
Can sequential designs reduce the sample size and improve the efficiency of cancer clinical trials?
Conclusion
Sequential analysis can significantly reduce the number of patients required and the number of deaths observed to reach conclusions in cancer trials.
Supporting Evidence
- The ECMV treatment group showed better survival outcomes compared to the selective treatment group.
- Cox's regression model identified extent of disease and general condition as significant prognostic factors.
- The median survival was 8 months for the ECMV group compared to 4 months for the ST group.
Takeaway
This study shows that using special methods in cancer trials can help find out if a treatment works faster and with fewer patients.
Methodology
Reanalysis of two completed phase III cancer trials using triangular and double-triangular tests for interim analyses.
Potential Biases
Potential for type I error due to multiple interim analyses.
Limitations
The trials were limited to specific designs and may not generalize to all cancer trials.
Participant Demographics
Patients with small-cell lung cancer, stratified by extent of disease and admitting center.
Statistical Information
P-Value
0.0001
Confidence Interval
(0.30, 0.61)
Statistical Significance
p<0.05
Want to read the original?
Access the complete publication on the publisher's website