INTRODUCTION Intensive weight loss can ameliorate, or even induce remission, of diabetes in obese patients. We sought to compare the weight loss efficacy of very low calorie diets (VLCD) comprised of Optifast^® meal replacement versus the anorexigenic properties of glucagon-like peptide-1 analogues (GLP-1) in obese adults with type 2 diabetes. METHODS An observational prospective analysis was undertaken of obese adults with type 2 diabetes at a tertiary hospital outpatient diabetes centre. Patients were initiated on either Optifast^® by a certified dietician (average 2/day) or a GLP-1 analogue with either exenatide (5ug BD, uptitrated to 10ug BD; n=14) or liraglutide (0.6ug daily, uptitrated to 1.2ug daily; n=4) on maximal hypoglycaemic therapy. Glycosylated haemoglobin (HbA1c) and weight were collected at baseline, 1-, 3-, 6- and 12-months. RESULTS Patients were similar between the VLCD (n=10) and GLP-1 (n=18) treatment groups with respect to BMI (37.9 ± 5.2 vs 33.3 ± 9.6 kg/m²), age (57.8 ± 9.6 vs 59.1 ± 12.1 yrs), gender (2/3 males), diabetes duration (11.8 ± 9.3 vs 10.85 ± 7.4 yrs) and glycaemic control (HbA1c 8.2 ± 1.8 vs 8.1 ± 0.7%). There was significant mean percentage weight loss in the VLCD vs GLP-1 groups at 3-months (-9.0 ± 3.7% vs. -2.7 ± 2.0%, p<0.0001) and 6-months (-10.4 ± 2.7% vs. -3.3 ± 2.5%, p<0.0001). In patients who continued, both groups lost similar weight at 12-months (-10.4 ± 4.5% vs. 8.5 ± 3.5%). Mean absolute HbA1c reductions were reduced in both the VLCD (-0.96 ± 1.66%) and GLP-1 (-0.29 ± 1.33%) groups at 3-months as well as at all time points, though statistical significance was not reached due to the small sample size and high drop-out rates. CONCLUSIONS Patients who persist with either VLCD or GLP-1 analogues can achieve significant long-term improvements in weight and glycaemic control, though low patient adherence may make longterm maintenance problematic. Further recruitment of patients to increase the study sample size and assess long-term outcomes is on-going.