|
Activity Number:
|
505
- Generalized Pairwise Comparisons and Win Statistics (Win Ratio, Win Odds, and Net Benefit): A Ten-Year Journey
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 11, 2022 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #322451
|
|
|
Title:
|
Evidence Synthesis in Oncology Clinical Trials: A Generalized Pairwise Comparison Approach
|
|
Author(s):
|
Bo Huang* and Ying Cui and Gaohong Dong
|
|
Companies:
|
Pfizer Inc. and Emory University and BeiGene
|
|
Keywords:
|
Generalized pairwise comparison;
win ratio;
win odds;
net benefit
|
|
Abstract:
|
The generalized pairwise comparison (GPC) approach for statistical inference and the use of win statistics to quantify treatment benefit are gaining increasing traction in clinical trials of medical products, especially for analyzing multiple prioritized outcomes that contain a fatal outcome. The main advantage of this method is the ability to combine multiple outcomes of various types into a single summary statistic without relying on any parametric model assumptions. It is particularly relevant since health authorities and the pharmaceutical industry are increasingly incorporating structured quantitative methodologies in their benefit risk assessment. We apply the GPC method to oncology clinical trials to assess the overall effect of an investigational treatment by prioritizing the most clinically relevant endpoints in cancer drug development. A simulation study demonstrates that this evidence synthesis approach has substantial advantage over standard methods that analyze each individual endpoint separately. We apply this method to an oncology phase 3 study in first-line renal cell carcinoma for illustration purpose. We also introduce a newly developed R package WINS.
|
Authors who are presenting talks have a * after their name.
