At study entry, all oral hypoglycemic agents will be discontinued for 5 days and then metformin (2,550 mg/daily) will be given for 3 months. Fasting plasma glucose will be measured at baseline and 3 months after metformin treatment. Patients will be stratified according to the median value of metformin efficacy as indicated by fasting glucose change after metformin treatment (i.e. baseline fasting glucose minus 3-month fasting glucose).

So, two subgroups of patients will be obtained, defined as relatively "high responders" (individual fasting glucose change > median value) or relatively "low responders" (individual fasting glucose change < median value) to metformin monotherapy.

Metformin, pills, 850 mg three times a day for three months

Changes in fasting glucose levels will be used to evaluate if metformin monotherapy efficacy in type 2 diabetic patients is predicted by mRNA and/or miRNA expression profiles.

Please note that metformin major effect is to decrease hepatic glucose output and, therefore, to lower fasting plasma glucose which is, in fact, the clinical outcome used in this study.

Finally, because of a very short wash-out period (i.e. 5 days) we will not be able to use HbA1c which will be inevitably conditioned by previous oral hypoglicemic therapy.