Hello, my name is Mark Conaway.
I'm a professor in the Division of Translational Research and Applied Statistics
in the Department of Public Health Sciences at the University of Virginia.
This talk is on patient heterogeneity in dose-finding trials.
Specifically, in this part,
I'll talk about designing trials in two groups of patients.
The outline of my talk is as follows: first,
I'll review traditional one-group dose-finding designs,
and it's useful for the material in this lecture to have viewed
Professor Ivanova's talks in this series on adaptive designs for phase one trials,
parts one and two.
Those lectures provide an excellent background for
the material in the two group dose-finding lecture.
So, once I've reviewed traditional dose-finding designs,
I'll talk about generalizing that to dose-finding trials in two groups of patients,
starting with a motivating example.
Then, I'll present a couple of classes of methods.
One class of methods for designing these trials is based on
adding parameters to mathematical or parametric models,
which includes the two-sample CRM and something called the Shift model.
The other class of methods relies on a branch of
statistics known as order restricted inference,
and I'll talk about that branch of statistics in general and
specifically how it applies to dose-finding in two groups.
In traditional single group dose-finding trials,
the primary purpose is almost always to establish
a safe dose and or method of administration of the agent.
There can be other purposes as well such as determining the type of side effect,
maybe getting preliminary evidence of efficacy or
perhaps investigating the clinical pharmacology of the drug.
But in the vast majority of cases,
the primary purpose of the trial is to establish a safe dose.