Activity Number:
|
37
- Statistical Issues Specific the Therapeutic Areas - 1
|
Type:
|
Contributed
|
Date/Time:
|
Sunday, July 29, 2018 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Biopharmaceutical Section
|
Abstract #330039
|
Presentation
|
Title:
|
Statistical Evaluation of Oncology Drug Trial Portfolios and the Potential for Inappropriate Regulatory Approval: a Simulation Study
|
Author(s):
|
Renee Gennarelli* and Peter Bach and Mithat Gonen
|
Companies:
|
Memorial Sloan Kettering Cancer Center and Memorial Sloan Kettering Cancer Center and Memorial Sloan Kettering Cancer Center
|
Keywords:
|
clinical trial;
simulation ;
family-wise error rate;
type 1 error rate;
immunotherapy
|
Abstract:
|
It is becoming increasingly common in drug development to have a large portfolio of trials. For example, there are currently more than 800 trials evaluating checkpoint inhibitors, a new class of immunotherapy drugs. Type I errors are no longer controlled at the nominal levels when trials are selected from a large portfolio. This is particularly relevant in oncology where indications for a drug are typically approved following a single pivotal trial. In an effort to quantify the potential risks of this practice, we developed a simulation to replicate a portfolio of existing clinical trials and assess the type 1 error rate and family-wise error rate. The simulation allows for customization such as the implementation of stopping rules, two-sided or one-sided testing and randomization schema as well as generating survival times off of a Weibull distribution based on an input median survival time for each arm. This presentation will feature an overview of the features and applications of the simulation. Issues regarding error rates when evaluating large portfolios of trials for drug approval will also be discussed.
|
Authors who are presenting talks have a * after their name.