T2DM has become an American Epidemic. Currently 8% of the US population has diabetes and rates may be as high as 33% by the year 2050 (1). Although there are many treatment options for people with T2DM, none have been proven in humans to prevent the defects in insulin secretion (2) and insulin action (3) and beta cell dysfunction (4) that result with very high glucose levels and typically worsen as the disease progresses. Any treatment that could delay the progression of pancreatic beta cell failure (as measured by the need for rescue therapy with oral agents) would be a significant advancement in diabetes treatment.

Insulin therapy is appropriate at any point in T2DM disease progression, but it is commonly only used as a rescue therapy after failure of oral therapies. A number of outpatient insulin titration protocols have been shown to be safe and effective and speed patient's ability to gain glucose control (5-8). Recent studies have shown that initiation of insulin at onset of T2DM is beneficial at achieving early and long-term glucose control (6-9). However these protocols have used intravenous human insulin in the in-patient setting, continuous subcutaneous insulin by insulin pump or older human insulins in the out-patient setting. Many of these protocols are unlikely to be utilized in routine patient care. To date, no "insulin first" studies have been published with analog insulins in an outpatient basal-bolus regimen with patient driven titration.

Insulin, when used as an initial treatment of T2DM, has a great potential to produce glucose control faster than any other treatment regimen. However, it is typically used as the treatment of last resort in T2DM. In this study, the investigators offer a novel approach to use insulin as the initial therapy in new-onset T2DM with the aim of determining its efficacy toward producing lasting glucose control.

Hypothesis: Treating newly diagnosed T2DM patients with insulin therapy versus standard of care for a short period of time will lead to improvement in glycemic control that is durable beyond the length of time taking the insulin and it may improve beta cell function.

Primary endpoints: Time to need rescue therapy, Need for rescue therapy at all time points. A1C change at 3, 6, 9 and 12 months.

Secondary endpoints: Mean glucose and mean fasting glucose at 3, 6, 12 months. C-peptide, HOMA-B, HOMA-IR, A1C the same time points, OGTT at week 12 and 56. Total number of hypoglycemic events (minor and major) and tolerability based on side effects.

Treatment arm: Weight based protocol of insulin Glargine and Glulisine. Control arm: oral medications per ADA 2009 recommended treatment algorithm. Rescue group available for both arms after initial 12 weeks.

Inclusion Criteria:

• Newly diagnosed T2DM (≤ 6 months since diagnosis)
• Drug naïve (less than 2 weeks of insulin and OHAs)
• A1C ≥ 8%
• Age ≥ 18 years
• Normal to high baseline C-peptide (≥ 0.5 ug/dL)
• FBG > 180 mg/dL, A1C > 8%.

Exclusion Criteria:

• Pregnancy
• Clinically evident heart failure
• Nephrotic syndrome
• Allergy to insulin or any of the oral medications in the study
• Presence of anti-GAD antibodies
• Islet cell antibodies
• Anti-insulin antibodies
• Any physical disabilities that would preclude self-administration of injectable insulin.
• Evidence of hypoglycemia during screening phase.
• History of lactic acidosis, allergy to metformin or history of chronic renal disease or a serum creatinine > 1.5 in men or > 1.4 in women