| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
COMMENTARY |
Division of Endocrinology, Diabetes and Metabolic Diseases, Department of Medicine, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania 19107
Address all correspondence and requests for reprints to: Barry J. Goldstein, M.D., Ph.D., Director, Division of Endocrinology and Metabolic Diseases, Department of Medicine, Jefferson Medical College, Suite 349, 1020 Locust Street, Philadelphia, Pennsylvania 19107-6799. E-mail: Barry.Goldstein{at}jefferson.edu.
The long-term goals for the management of type 2 diabetes mellitus (T2DM) center on the prevention of its chronic micro- and macrovascular complications, which requires normalization of glycemia and control of cardiovascular risk factors (1). Growing evidence from large clinical trials supporting earlier and more aggressive therapeutic strategies to manage hyperglycemia in diabetes has been reflected in current guidelines and practice recommendations (2, 3, 4, 5). Recent clinical trials on controlling hyperglycemia have supported more efficacious therapeutic approaches, most of which have clearly shown that double or triple combinations of available and emerging pharmaceutical agents are required to achieve that elusive, but well-recognized, near-normal hemoglobin A1c goal of less than 6.5% (6).
T2DM results from a combination of physiological abnormalities including systemic insulin resistance coupled with inadequate ß-cell insulin secretion (7). The etiology of these defects is not fully understood. Classic studies from the United Kingdom Prospective Diabetes Study (UKPDS) revealed that hyperglycemia in T2DM tends to be progressive due to persistent insulin resistance with gradual worsening of ß-cell insulin secretory function (8). For the majority of patients, it is well-known that maximizing doses of single oral medications does not provide a sustained normalization of glycemia and can provoke unwanted side effects (9). Moreover, diabetes results from a complex interplay among several pathophysiological mechanisms, and it seems doubtful that in the near future a therapeutic "magic bullet" will be identified to effectively "cure" the hyperglycemia of diabetes (7). Thus, successful current therapeutic approaches use early adoption of oral combination therapies, among the variety of medications that (thankfully) are currently available for use.
More than 7 yr ago, before this insight into T2DM was appreciated, "A Diabetes Outcome Progression Trial" (ADOPT) was designed and initiated (10). This large (>4300 patients), double-blind, randomized, controlled, clinical trial evaluated the performance of three major oral medications (rosiglitazone, metformin, and glyburide) as monotherapy in drug naive T2DM patients who were followed for a median of 4.0 yr (11). ADOPT provides an updated view of diabetes management for several reasons. The thiazolidinediones were not available for use during the UKPDS. In addition, patients’ glucose levels that triggered an intensification of therapy in the UKPDS were much higher than modern management goals. The ADOPT investigators used a therapeutic fasting plasma glucose (FPG) goal of less than 140 mg/dl for intensifying monotherapy; the primary outcome was the time to monotherapy failure evidenced by a FPG greater than 180 mg/dl, a test of therapeutic "durability."
The results of ADOPT were not wholly unexpected. Considering the large number of subjects and the duration of blinded therapy in this landmark trial, however, ADOPT provides a superior level of confidence in its assessment of the monotherapy performance of these drugs compared with data from smaller previous studies of shorter duration. The primary results of ADOPT showed a lower failure rate at 5 yr of 15% with rosiglitazone vs. 21% with metformin and 34% with glyburide, a highly significant risk reduction of 32% for rosiglitazone compared with metformin, and 63% compared with glyburide.
The data from ADOPT are consistent with the known mechanism of action of rosiglitazone to improve both insulin resistance and ß-cell function (12). Interestingly, although the apparent durability of the glycemic effect of rosiglitazone in prior studies has largely been attributed to its enhancement of ß-cell function, in ADOPT, this effect of rosiglitazone (estimated by homeostasis model assessment-B analysis), began to wane after 1.5 yr of therapy. More accurate physiological analyses of ß-cell function were also performed during ADOPT and will become available in the future. Rosiglitazone effects on insulin sensitivity, however, continued to improve through 4–5 yr of follow-up. Thus, rosiglitazone monotherapy durability is associated with insulin-sensitizing effects unique to the thiazolidinediones and, perhaps to a lesser degree, enhancement of ß-cell insulin secretion. Interestingly, metformin exhibited reasonable insulin-sensitizing effects (although not as robust as rosiglitazone), an effect that has not been consistently shown in prior metformin trials. Unlike the data from UKPDS, in ADOPT, metformin also enhanced ß-cell function, initially to a similar degree as was seen with rosiglitazone, but this effect returned to baseline by the end of the observation period. The factors that account for this observation in ADOPT are not known. In terms of efficacy, glyburide was clearly the loser, with a steep slope of loss of ß-cell function over time, similar to that seen in the UKPDS.
Although rosiglitazone provided superior glycemic durability, the ADOPT safety data were also consistent with the widely appreciated but unfortunate side effects of rosiglitazone to increase lower body adiposity and cause mild peripheral edema. In contrast to the recent DREAM trial, which showed a dramatic efficacy of rosiglitazone to delay glycemic progression in prediabetes that was accompanied by a surprising incidence of congestive heart failure (13), there was no increase in (nonadjudicated) congestive heart failure (CHF) with rosiglitazone compared with metformin in ADOPT. Curiously, the glyburide group had a significantly reduced incidence of CHF. These data fit with the relatively low frequency of CHF with thiazolidinedione use, including when used in combination with sulfonylureas in elderly patients with a high incidence of baseline edema (14). Metformin exhibited its anticipated incidence of unpleasant gastrointestinal side effects, and glyburide was associated with significant hypoglycemia, perhaps the worst performing sulfonylurea in this regard (15). ADOPT also revealed a potential concern with increased bone fractures in women taking rosiglitazone. Cross-sectional data have recently raised the possibility of osteopenia with thiazolidinedione use (16), and until additional studies are completed, it may be prudent to monitor bone density in women taking rosiglitazone in clinical practice who have a high fracture risk.
Tight glucose control might also ameliorate the progressive loss of ß-cell function associated with glucotoxicity (17). Rosiglitazone was more successful in achieving the lowest mean glucose control, and both metformin and rosiglitazone were superior to glyburide, factors that could make a difference in durability. On the other hand, the overall preservation of enhanced ß-cell function was disappointing among all three agents. Further clinical trials will be needed to examine whether tighter control of hyperglycemia will show improved preservation of ß-cell function, beyond that demonstrated in ADOPT.
One of the key practical questions arising from the ADOPT study is how do we achieve tight glucose control in the early course of T2DM? ADOPT, like numerous other monotherapy trials, found a poor achievement rate for aggressive glycemic goals, although the design of the study (targeting FPG less than 140 mg/dl) would not have been expected to control many patients to an A1c level less than 6.5% (9). At the 4-yr evaluation in ADOPT, 40% of the patients in the rosiglitazone group had an A1c of less than 7%, marginally superior to the 36% achieved by the patients in the metformin group (P = 0.03), but clearly superior to the 26% of the patients in the glyburide group (P < 0.001). Many patients were removed from the study according to protocol due to glycemic failure, and overall, only 56–63% of the randomized patients completed the study, although in a balanced manner among the three arms. The mean dose of each monotherapy given to the patients in ADOPT was not reported, and it will be instructive to learn this parameter in the future. Nevertheless, as in clinical practice, we can only surmise that the fraction of patients achieving A1c levels to a goal of 6.5% or less with any monotherapy was unacceptably small.
When initiated several years ago, the ADOPT design to test individual monotherapies appropriately followed a common practice in diabetes therapy at that time. Nowadays, the growing demand for aggressive therapy as reflected in current clinical practice recommendations supports the earlier use of combinations of oral medications (3, 5, 18). The rationale for combination therapy in diabetes practice has become clear: in the majority of patients, any monotherapy is not likely to control glucose levels to near normal. Many current clinical trials with diabetes therapeutics are designed as add-on or initial coadministration of two drugs. The multifactorial nature of T2DM pathogenesis, noted above, is also consistent with a superior clinical response from medication combinations that address several underlying defects at once. In addition, limiting the doses of the available drugs will help minimize their expected adverse effects, especially the gastrointestinal intolerance of metformin and the weight gain and fluid retention associated with thiazolidinediones.
Although we may now accept that aggressive glycemic control is essential to preventing microvascular complications of diabetes, ongoing trials will also help determine whether macrovascular disease is delayed or prevented, as well as restoring and maintaining ß-cell function. To these ends, ADOPT provides clinical trial evidence to support the early use of metformin and rosiglitazone. Prior studies have shown that the combination of these two agents works well and minimizes the side effects of higher doses of each agent without hypoglycemia (19). Sulfonylurea use, especially with aggressive glycemic targets, is not only a poor choice from a durability standpoint, but can be fraught with uncomfortable and potentially dangerous hypoglycemia especially when used in early combinations (20). Thus, translation of the ADOPT data supports a role for early use of metformin and rosiglitazone in combination, a well-tolerated, effective initial therapy that has been approved by the U.S. Food and Drug Administration for this indication (21). New agents that can further enhance glycemic control perhaps by different mechanisms but also avoiding hypoglycemia, including the incretin agents, will be welcome additions to this therapeutic mix. The growing availability of fixed-drug-combination pills of this type will also significantly help to contain costs and simplify patient regimens.
 |
Footnotes |
|---|
Abbreviations: CHF, Congestive heart failure; FPG, fasting plasma glucose; T2DM, type 2 diabetes mellitus.
Received December 22, 2006.
Accepted January 19, 2007.
|
References |
|---|
 Top References |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Z. T. Bloomgarden Exploring Treatment Strategies for Type 2 Diabetes Diabetes Care, October 1, 2007; 30(10): 2737 - 2745. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
