Design of sequentially randomized trials for testing adaptive treatment strategies

Abstract

An adaptive treatment strategy (ATS) is an outcome-guided algorithm that allows personalized treatment of complex diseases based on patients' disease status and treatment history. Conditions such as AIDS, depression, and cancer usually require several stages of treatment because of the chronic, multifactorial nature of illness progression and management. Sequential multiple assignment randomized (SMAR) designs permit simultaneous inference about multiple ATSs, where patients are sequentially randomized to treatments at different stages depending upon response status. The purpose of the article is to develop a sample size formula to ensure adequate power for comparing two or more ATSs. Based on a Wald-type statistic for comparing multiple ATSs with a continuous endpoint, we develop a sample size formula and test it through simulation studies. We show via simulation that the proposed sample size formula maintains the nominal power. The proposed sample size formula is not applicable to designs with time-to-event endpoints but the formula will be useful for practitioners while designing SMAR trials to compare adaptive treatment strategies.

Keywords: adaptive treatment strategy (ATS); power; sample size; sequential multiple assignment randomized trial (SMART).

Copyright © 2015 John Wiley & Sons, Ltd.

Figures

Figure 1

Design 1. At entry, patients are randomized to initial treatments A1 and A2. If a patient responds to the initial treatment she is randomized to either B1 or B2, otherwise the patient is randomized to either C1 or C2.

Figure 2

Design 2. At entry, patients are randomized to initial treatments A1 and A2. If a patient responds to the initial treatment she stays on the same initial treatment, otherwise the patient is re-randomized to subsequent treatments: C1 or C2 if she does not respond to A1; C1′ or C2′ if she does not respond to A2.

Figure 3

Design 3. At entry, patients are randomized to initial treatments A1, A2 and A3. If a patient responds to the initial treatment she stays on the same initial treatment, otherwise the patient is re-randomized to subsequent treatments: A2 or A3 if she does not respond to A1; A1 or A3 if she does not respond to A2; A1 or A2 if she does not respond to A3.