Background: Adults with type 1 diabetes who are treated with multiple daily injections of insulin plus intermittently scanned continuous glucose monitoring (isCGM) can have suboptimal glucose control. We aimed to assess the efficacy of an advanced hybrid closed loop (AHCL) system compared with such therapy in this population. Methods: The Advanced Hybrid Closed Loop Study in Adult Population with Type 1 Diabetes (ADAPT) trial is a prospective, multicentre, open-label, randomised controlled trial that involved 14 centres in three European countries (France, Germany, and the UK). We enrolled patients who were at least 18 years of age, had a type 1 diabetes duration of at least 2 years, HbA1c of at least 8% (64 mmol/mol), and were using multiple daily injections of insulin plus isCGM (cohort A) or real time continuous glucose monitoring (cohort B) for at least 3 months. Here, only results for cohort A are reported. Participants were randomly allocated 1:1 to AHCL therapy or continuation of multiple daily injections of insulin plus continuous glucose monitoring for 6 months with an investigator-blinded block randomisation procedure. Participants and treating clinicians could not be masked to the arm assignment. The primary endpoint was the between-group difference in mean HbA1c change from baseline to 6 months in the intention-to-treat population using AHCL therapy and those using multiple daily injections of insulin plus isCGM. The primary endpoint was analysed using a repeated measures random-effects model with the study arm and period as factors. Safety endpoints included the number of device deficiencies, severe hypoglycaemic events, diabetic ketoacidosis, and serious adverse events. This study is registered with ClinicalTrials.gov, NCT04235504. Findings: Between July 13, 2020, and March 12, 2021, 105 people were screened and 82 randomly assigned to treatment (41 in each arm). At 6 months, mean HbA1c had decreased by 1·54% (SD 0·73), from 9·00% to 7·32% in the AHCL group and 0·20% (0·80) in the multiple daily injections of insulin plus isCGM from 9·07% to 8·91% (model-based difference −1·42%, 95% CI −1·74 to −1·10; p<0·0001). No diabetic ketoacidosis, severe hypoglycaemia, or serious adverse events related to study devices occurred in either group; two severe hypoglycaemic events occurred in the run-in phase. 15 device-related non-serious adverse events occurred in the AHCL group, compared with three in the multiple daily injections of insulin plus isCGM group. Two serious adverse events occurred (one in each group), these were breast cancer (in one patient in the AHCL group) and intravitreous haemorrhage (in one patient in the multiple daily injections of insulin plus isCGM group). Interpretation: In people with type 1 diabetes using multiple daily injections of insulin plus isCGM and with HbA1c of at least 8%, the use of AHCL confers benefits in terms of glycaemic control beyond those that can be achieved with multiple daily injections of insulin plus isCGM. These data support wider access to AHCL in people with type 1 diabetes not at target glucose levels. Funding: Medtronic International Trading Sàrl.