Using meta-analysis of Phase II trials to predict Phase III trial results

Danielle Burke, Lucinda Billingham, Alan Girling, Richard Riley

Research output: Contribution to journalArticlepeer-review

140 Downloads (Pure)


Pharmaceutical companies use Phase II trial results to make decisions about proceeding to Phase III. We will show how a meta-analysis of results from multiple Phase II trials is informative toward this decision.

We consider a meta-analysis of nine randomised Phase II trials comparing the efficacy of two therapies for acute myocardial infarction. Results for four outcomes were collected: intracranial haemorrhage, stroke, reinfarction and total mortality.

We apply univariate and multivariate random-effects meta-analysis methods, and use the obtained summary results to derive 95% prediction intervals, which give the predicted treatment effects for the four outcomes in a future trial. The multivariate approach jointly synthesizes all outcomes whilst accounting for their correlation. The methods are applicable in both frequentist and Bayesian frameworks. Predictions calculated are compared to results from subsequent Phase III trials.

The meta-analyses of Phase II trials show that the new treatment is promising for most outcomes. For example, the probability that the odds of stroke will be reduced by >10% in a future trial is 0.67. Importantly, the prediction intervals include the treatment effects that were seen in subsequent Phase III trials. These Phase III results have previously been described as 'contradictory' to the Phase II results, but our prediction intervals reveal this is not the case.

The potential results of a Phase III trial can be informed by 95% prediction intervals derived from a Phase II meta-analysis. Such predictions could help pharmaceutical companies and funding bodies to prioritise interventions for Phase III evaluation.
Original languageEnglish
Article numberO118
Issue numberSuppl 1
Publication statusPublished - 29 Nov 2013
Event2nd Clinical Trials Methodology Conference: Methodology Matters - Edinburgh, UK, United Kingdom
Duration: 18 Nov 201319 Nov 2013

Bibliographical note

Oral presentation. This article is part of the supplement: 2nd Clinical Trials Methodology Conference: Methodology Matters


Dive into the research topics of 'Using meta-analysis of Phase II trials to predict Phase III trial results'. Together they form a unique fingerprint.

Cite this