Core outcomes in Brain Tumour Trials – The COBra Study Review of Glioma Trial Registration Data

Abstract

Abstract AIMS Trial registration supports unbiased reporting of research studies. Despite inconsistent use globally and sub-optimal completeness, registration is associated with publication of the same outcomes as defined in trial protocols, though these are not necessarily reported in published results. In the core outcomes in brain tumour trials (COBra) study a registry review was undertaken, alongside a systematic review of qualitative literature and semi-structured interviews, to develop a core outcome set (COS) outcome longlist. The study aims to identify all outcomes reported across all glioma trial registration data. METHOD Phase 3 interventional glioma trials involving patients aged 18 and over, registered on clinicaltrials.gov or the ISRCTN registry were included. Two researchers independently searched the databases and registered data and available protocols were retrieved. Two reviewers independently extracted basic trial data including primary and secondary outcomes. Outcomes were categorised according to an ontology developed through literature review. RESULTS 91 trials were identified. Primary outcome categories reported were survival (84%), disease activity (14%), symptoms (4%), adverse events (3%) and resource use (1%). Secondary outcome categories reported were survival (60%), adverse events (52%), quality of life (QoL, 41%), disease activity (30%), symptoms (20%), function (12%), resource use (7%), health status (7%), and economic (3%). The review contributed 15 unique outcomes to the final outcome longlist of 35. CONCLUSION Trial registries represent a pragmatic solution in COS development and enabled identification of outcomes not present in other data sources. Under half of registered glioma studies included QoL outcomes; a small minority included functional outcomes or economic evaluation.