|
Activity Number:
|
77
- Contributed Poster Presentations: Biopharmaceutical Section
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #312872
|
|
|
Title:
|
Statistical Impact of Delayed Treatment Effect: A Simulation Study
|
|
Author(s):
|
Yuanbo Song* and Bharani Bharani-Dharan
|
|
Companies:
|
Novartis Pharmaceuticals Corp and Novartis Oncology
|
|
Keywords:
|
Delayed treatment effect;
non-proportional hazard ;
log-rank test
|
|
Abstract:
|
Delayed treatment effect is a phenomenon observed in quite a few immuno-oncology studies. Immunotherapy stimulates the patient's immune system to observe an anti-tumor response. The time needed to observe this response inserts the delay in clinical treatment effect. Conventional statistical approaches to trial design and statistical analysis require careful consideration in the presence of delayed treatment effects as proportional hazard assumption no longer holds under delayed treatment effect. In such cases, traditional log-rank test and cox regression needs to be re-evaluated. Using a case study as an example, we conducted a simulation study to explore the impact of different design factors such as lag time, median and enrollment rate on power loss and hazard ratio in the presence of delayed treatment effect. Some alternative methods such as weighted log-rank test to analyze the data under delayed treatment effect will be presented and discussed.
|
Authors who are presenting talks have a * after their name.
