Statistics and Its Interface

Volume 5 (2012)

Number 4

Efficient designs for phase II oncology trials with ordinal outcome

Pages: 463 – 469

DOI: https://dx.doi.org/10.4310/SII.2012.v5.n4.a9

Authors

Anastasia Ivanova (Department of Biostatistics, University of North Carolina, Chapel Hill, N.C., U.S.A.)

Jane Monaco (Department of Biostatistics, University of North Carolina, Chapel Hill, N.C., U.S.A.)

Thomas Stinchcombe (Lineberger Cancer Center, University of North Carolina, Chapel Hill, N.C., U.S.A.)

Abstract

Phase II trials in oncology are usually single arm studies to screen oncology treatments based on tumor response. Treatment response can be categorized into one of four types: complete response, partial response, stable disease and progressive disease. Investigators usually dichotomize responses in phase II trials and use a simple hypothesis regarding that binary outcome. We describe an efficient design to test an intersection hypothesis where the drug is not considered promising if both tumor response (defined as complete or partial response) and disease control (defined as tumor response or stable disease) are low. The new design can be generated using easy-to-use software that is available at http://cancer.unc.edu/biostatistics/program/ivanova/.

Keywords

Simon’s design, Phase II trial, ordinal outcome, tumor response

Published 16 November 2012