Despite the extensive clinical experience with the use of metformin worldwide, no formal doseranging study has been conducted because the current dosing strategy of metformin was determined empirically, rather than by an understanding of its dose-response relationship in patients with type 2 diabetes. The present study was designed to evaluate the correlation between serum metformin levels and glycemic control, insulin resistance and leptin levels in females newly diagnosed with type 2 diabetes. Sixty type 2 diabetic females were recruited for the study and were allocated into 3 groups; each receiving metformin 1000, 1500 and 2000 mg/day respectively for 3 months. Blood samples withdrawn from each patient at zero time and after 3 months is used to evaluate serum levels of HbA1c, glucose, leptin and insulin, in addition, the measurement of serum level of metformin in blood after 3 months by HPLC. The results demonstrated that all the treated groups with different doses of metformin showed significant improve in all parameters; the use of increased metformin doses was only correlated with plasma leptin levels in the highest dose. In conclusion, serum metformin levels are not good predictors for correlating improvement in clinical and biochemical parameters with increasing the dose in newly diagnosed non-insulin resistant females with type 2 diabetes.