In order to assess whether previous hepatic IR (Hepatic-IRfasting) and beta-cell functionality could modulate type 2 diabetes remission and the need for starting glucose-lowering treatment, newly-diagnosed type 2 diabetes participants who had never received glucose-lowering treatment (190 out of 1002) from the CORonary Diet Intervention with Olive oil and cardiovascular PREVention study (a prospective, randomized and controlled clinical trial), were randomized to consume a Mediterranean or a low-fat diet. Type 2 diabetes remission was defined according to the American Diabetes Association recommendation for levels of HbA1c, fasting plasma glucose and 2h plasma glucose after oral glucose tolerance test, and having maintained them for at least 2 consecutive years. Patients were classified according to the median of Hepatic-IRfasting and beta-cell functionality, measured as the disposition index (DI) at baseline. Cox proportional hazards regression determined the potential for Hepatic-IRfasting and DI indexes as predictors of diabetes remission and the probability of starting pharmacological treatment after a 5-year follow-up. Low-Hepatic-IRfasting or high-DI patients had a higher probability of diabetes remission than high-Hepatic-IRfasting or low-DI subjects (HR:1.79; 95% CI 1.06–3.05; and HR:2.66; 95% CI 1.60–4.43, respectively) after a dietary intervention with no pharmacological treatment and no weight loss. The combination of low-Hepatic-IRfasting and high-DI presented the highest probability of remission (HR:4.63; 95% CI 2.00–10.70). Among patients maintaining diabetes, those with high- Hepatic-IRfasting and low-DI showed the highest risk of starting glucose-lowering therapy (HR:3.24;95% CI 1.50–7.02). Newly-diagnosed type 2 diabetes patients with better beta-cell functionality and lower Hepatic-IRfasting had a higher probability of type 2 diabetes remission in a dietary intervention without pharmacological treatment or weight loss, whereas among patients not achieving remission, those with worse beta-cell functionality and higher Hepatic-IRfasting index had the highest risk of starting glucose-lowering treatment after 5 years of follow-up.