Name: Todd

Firstname: Sue

Title: Dr

Institution: The University of Reading

Street: P.O.Box 240, Earley Gate

City: Reading

Zip-Code: RG6 6FN

Country: United Kingdom

Phone: +44 118 931 8917

Fax: +44 118 975 3169

Email: s.c.todd@reading.ac.uk

Authors: Sue Todd

Title: Issues in Monitoring Bivariate Group Sequential Trials

Abstract: When implementing a sequential stopping rule, the situation may arise when two responses should ideally be considered. Examples include the use of an early endpoint which is later replaced by a longer term outcome; simultaneous monitoring of efficacy and safety responses; and the need to demonstrate efficacy for two outcome measures to gain licensing. Sequential designs for bivariate responses have been proposed by Jennison and Turnbull (1993) and Cook and Farewell (1994), who consider the simultaneous evaluation of efficacy and safety responses. Both use a bivariate generalisation of the usual recursive relationship employed in repeated numerical intergration. In this talk the wider range of problems mentioned above is discussed. Emphasis is placed on the monitoring phase of the sequential trial, particularly estimation of correlation, and on issues of analysis.

References: Jennison C. and Turnbull BW. (1993). Group sequential tests for bivariate response: interim analyses of clinical trials with both efficacy and safety endpoints. Biometrics, 49, 741-752.

Cook RJ and Farewell VT. (1994). Guidelines for monitoring efficacy and toxicity responses in clinical trials. Biometrics, 50, 1146-1152.