Abstract AIMS Early phase cancer clinical trials have become more complicated and patients often misunderstand their nature and purpose. CONSENT (NCT04407676) is a randomised controlled trial testing whether enhanced informed consent for patient education can improve comprehension – since patients with glioblastoma multiforme (GBM) have a higher rate of baseline cognitive impairment, we studied this group separately. METHOD GBM patients (from the ICE-CAP Phase 1 study - NCT03673787), underwent the schedule for the standard CONSENT arm - full length trial PIS, Quality of Informed Consent Questionnaire Parts A and B (QuIC-A and QuIC-B), experimental intervention (2 page study aid and 10 educational videos), and a repeat QuIC-A and QuIC-B. The primary endpoint for this subgroup was the difference in QuIC-A scores before and after the intervention using a paired t-test. RESULTS 6 patients with GBM were recruited - 3 did not complete any study questionnaires. The three pre intervention QuIC-A scores were 70, 81, 88, with a mean of 75 (unit reference 76). The three pre intervention QuIC-B scores were 69, 62, 75 with a mean of 69 (unit reference is 91). Only one patient completed the post intervention questionnaire - their QuIC-A score moved from 88 to 100. CONCLUSION This study demonstrates the significant difficulties in studying comprehension in patients with GBM considering early phase trials. There is a need for creative multi-modality solutions to provide information to GBM patients considering clinical trials, and novel tools to assess the effectiveness of these solutions.