Despite effective monotherapy for diabetes, approximately 50% of patients require additional medications after 3 years to achieve target glycosylated hemoglobin (A1C) < 7%. Three new agents, each the first in its therapeutic class with a unique mechanism of action, have been approved for the treatment of type 2 diabetes by the U.S. Food and Drug Administration: pramlintide in March 2005, exenatide in April 2005, and sitagliptin in October 2006. To review the efficacy and safety of 3 new agents for type 2 diabetes (exenatide and pramlintide by subcutaneous injection and sitagliptin by oral administration) and to define their place in therapy given their relatively high cost and unknown long-term safety and efficacy. A MEDLINE search (1950 to June 2007) for English-language articles of studies in human subjects was conducted using these search terms: type 2 diabetes, exenatide, pramlintide, and sitagliptin. This database was supplemented by systematic reviews and meta-analyses through December 2007 and reference citations from the articles identified in the MEDLINE search. Exenatide, pramlintide, and sitagliptin have all been shown to have a modest effect on reducing A1C. In several relatively short-term trials (generally 15-30 weeks in duration), exenatide injection has been shown to be safe and effective for patients with type 2 diabetes who are either at the maximum doses of or cannot tolerate metformin, sulfonylurea, and/or thiazolidinedione therapy and need to further decrease A1C by at least 0.5% to 1%. While weight loss of 1.5 kg to 2.5 kg associated with exenatide is modest, this effect is of obvious value in many patients with type 2 diabetes. Nausea is the most notable side effect with exenatide, occurring in up to 50% of patients within the first 8 weeks of therapy but decreasing to 5% to 10% by week 24. In addition, the risk for hypoglycemia increases 4- to 5-fold when used in combination with sulfonylureas. Like exenatide, pramlintide injection reduces A1C by approximately 0.5% to 1%, carries the advantage of modest weight loss (1.5 kg over 1 year), and has a high incidence of nausea. Pramlintide can also result in severe hypoglycemia because of its ability to enhance the effects of insulin, a concern given that it is only indicated for use in combination with insulin. Sitagliptin is an oral agent that can be used alone or in combination with other oral hypoglycemic agents and has been shown to reduce A1C by 0.5% to 0.7%. It has only been studied in short-term studies, to date, so the long-term safety and efficacy are unknown. There is potential for severe allergic and dermatologic reactions with sitagliptin. The 3 new agents for the management of type 2 diabetes have been shown to reduce A1C by no more than 1.0%, modest by comparison with insulin and the older oral agents. The 3 newer agents have either modest positive effects on body weight or are weight neutral. The longterm safety and efficacy of the 3 newer agents are unknown, and their cost is considerably higher than the first-line agents, metformin and sufonylureas, which are available by generic name. The newer agents offer treatment options in select patients, although their use should be reserved for patients who are not adequately managed by agents with known longterm efficacy and safety, which are often available at a lower cost.