BAYESIAN SEQUENTIAL MONITORING DESIGNS FOR SINGLE-ARM CLINICAL-TRIALSWITH MULTIPLE OUTCOMES

We present a Bayesian approach for monitoring multiple outcomes in sin gle-arm clinical trials. Each patient's response may include both adve rse events and efficacy outcomes, possibly occurring at different stud y times. We use a Dirichlet-multinomial model to accommodate general d iscrete multivariate responses. We present Bayesian decision criteria and monitoring boundaries for early termination of studies with unacce ptably high rates of adverse outcomes or with low rates of desirable o utcomes. Each stopping rule is constructed either to maintain equivale nce or to achieve a specified level of improvement of a particular eve nt rate for the experimental treatment, compared with that of standard therapy. We avoid explicit specification of costs and a loss function . We evaluate the joint behaviour of the multiple decision rules using frequentist criteria. One chooses a design by considering several par ameterizations under relevant fixed values of the multiple outcome pro bability vector. Applications include trials where response is the cro ss-product of multiple simultaneous binary outcomes, and hierarchical structures that reflect successive stages of treatment response, disea se progression and survival. We illustrate the approach with a variety of single-arm cancer trials, including bio-chemotherapy acute leukaem ia trials, bone marrow transplantation trials, and an anti-infection t rial. The number of elementary patient outcomes in each of these trial s varies from three to seven, with as many as four monitoring boundari es running simultaneously. We provide general guidelines for eliciting and parameterizing Dirichlet priors and for specifying design paramet ers.