Parametric Approaches for Optimizing Treatment Schedules in Adaptive Early-Phase Clinical Trials

Thomas Braun
Assistant Professor
University of Michigan

Wednesday, February 20th
3:30pm
Moos T 5-125
Minneapolis Campus

Abstract:

Traditionally, Phase I clinical trials have been designed to determine a maximum tolerated dose (MTD) of an experimental agent based on one initial treatment course. However, in most clinical settings, physicians administer multiple courses of an agent to a patient and monitor long-term toxicity that may be related to the agent's cumulative effects. In addition, some patients may receive one or more dose modifications or delays of their planned treatment course due to low-grade (non-dose limiting) toxicity from previous administrations. To address these situations, I present an outcome-adaptive design that simultaneously optimizes both does and schedule in order to determine a maximum tolerated schedule (MTS) of administration rather than a conventional MTD. The total hazard of any series of administrations is modeled as the sum of the hazards of each administration; the dose of each administration is incorporated either directly into the hazards of each administration or indirectly through a non-mixture cure model. I motivate the methods with two clinical trials of bone marrow transplant recipients and present results of simulation studies in the context of these trials.

A social tea will be held at 3:00 P.M. in A434 Mayo. All are Welcome.
For more details contact 612-624-4655 or see http://www.biostat.umn.edu/seminar_academic.html