Published online June 6, 2008 in the New England Journal of Medicine and presented here at the American Diabetes Association (ADA) 2008 Scientific Sessions, the Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) study is the world's largest diabetes trial, enrolling 11 140 patients with type 2 diabetes at 215 sites in 20 countries.

"Clinicians, diabetes specialists, and the patients whom they look after urgently need guidance as to the level of glycemic control that we can safely aim for," Dr Simon Heller (University of Sheffield, UK), study coauthor and management committee member for ADVANCE, said during an ADA media briefing. "I think this study explains really much more clearly now; it makes it more clear for clinicians and their patients the kind of goals that we should be aiming for."

Patients eligible for the ADVANCE trial were at least 55 years old, had been diagnosed with type 2 diabetes at age 30 years or older, and had a history of or at least one risk factor for vascular disease. Participants were randomly assigned to receive placebo or a fixed dose of perindopril and indapamide to lower blood pressure and to undergo intensive control or standard control of blood glucose (the latter determined by local guidelines).

Results of the blood-pressure portion of the study were previously published in September 2007. Regardless of initial blood pressure, the treatment arm in this population reduced their relative risk of death due to cardiovascular disease by 18% (p=0.03).

Patients randomized to the intensive-glucose-control group (n=5571) received gliclazide (modified release, 30-120 mg/day) and no other sulfonylureas; other glucose-control agents were prescribed by the treating physicians as needed to achieve the ADVANCE target HbA_1c of 6.5% or less. Patients randomized to the standard-treatment group (n=5569) received glucose-lowering agents other than gliclazide as prescribed by their treating physicians, using the current guidelines target for HbA_1c of 7.0%.

The ADVANCE study achieved a mean HbA_1c of 6.5% in the intensive-glucose-control group compared with 7.3% in the standard-treatment group (p<0.001). This decrease was accompanied by a 10% relative risk reduction for major vascular complications (p=0.013) and a 14% relative risk reduction for major microvascular events (p=0.014). No significant difference was found between the treatment groups for major macrovascular events such as MI or stroke (p=0.32).

Heller commented on the absence of macrovascular benefits: "Diabetic nephropathy and, indeed, increased protein excretion in the kidney are powerful predictors of death from cardiovascular disease. But it's important to remember that this study went on for four years or so, and that predicted increase might not be expected until a year or two down the line," he said.

"It's worth just pointing out that right at the end of this study, the lines [showing cumulative incidence of major macrovascular events for both treatment groups] are beginning to diverge a little," Heller noted. "Just a hint that perhaps, if the study had gone on much longer, we might have seen the benefit of macrovascular disease indirectly because of the protection from diabetic kidney disease. But of course that's speculative," he acknowledged.

The greatest relative risk reduction in the intensive-glucose-control group was for new or worsening diabetic nephropathy, which decreased 21% compared with the standard-treatment group (p=0.006). There was also a 9% reduction in new microalbuminuria (p=0.02). The study found no evidence that intensive glucose control decreased new or worsening diabetic retinopathy.

In an interview, Dr Anushka Patel (Royal Prince Alfred Hospital, Sydney, Australia), coauthor and principal clinical coordinator for ADVANCE, discussed the results for nephropathy and retinopathy.

"Certainly the association between HbA_1c and your risk of retinopathy and your risk of nephropathy are very similar," observed Patel. "The confidence interval for the effect on retinopathy . . . overlaps quite a lot with the confidence interval of the effects on the renal outcome. So, they are potentially contingent.

"Having said that, though . . . the event rate for retinopathy was quite a lot lower than expected, and I think the main reason for that is that it is driven by retinal photocoagulation," Patel pointed out. "And retinal photocoagulation is a very specific marker for retinal behavior. You only get it if you have retinal disease. It's not very sensitive. It depends on the availability of treatment."

A substudy is being conducted with 2000 patients in which retinas were examined at the beginning of the study and again at the end. "We'll report later in the year. That will give us a much better idea of what's going on with retinal disease." Patel said.

The reduction in nephropathy remains the most significant finding of the study. As Heller said in the media briefing, "Getting down to [these kinds] of HbA_1c levels—less than 7.0% or even approaching 6.5% as authorities around the world prove—can [protect] patients from diabetic kidney disease and essentially protect one in five over a few years from developing this very bad complication."

"I think [the findings] have to be put into perspective," Dr Harold E Lebovitz (SUNY Health Sciences Center, Brooklyn), chair of the ADA Scientific Sessions Planning Committee and moderator of the ADA media briefing, said about ADVANCE and another recently reported large type 2 diabetes trial, Action to Control Cardiovascular Risk in Diabetes (ACCORD), also published online June 6, 2008 in the New England Journal of Medicine [2].

Several ADA scientific sessions focused on the ADVANCE and ACCORD trials. ACCORD was supported by the National Heart, Lung, and Blood Institute, and it enrolled 10 251 patients with type 2 diabetes. ACCORD was suspended in February 2008 after approximately 3.5 years of follow-up as a result of significantly more deaths in the intensive-glucose-control group than in the standard-therapy group (p=0.04).

There was no evidence of increased risk of death in the ADVANCE intensive-glucose-control group. The investigators attempted to explain factors in these two large studies that would account for their different outcomes.

ADVANCE and ACCORD have several fundamental differences: primary end points of ADVANCE were major microvascular events and/or major macrovascular events, assessed jointly and separately; primary end points for ACCORD were macrovascular events. Mean baseline HbA_1c level in ADVANCE participants was 7.5%; in ACCORD, mean baseline HbA_1c level was 8.3%. Mean body-mass index was 28 kg/m² in ADVANCE participants and 32 kg/m² in ACCORD participants.

Among ACCORD participants, 28% in the intensive-treatment group gained 10 kg or more; 14% in the standard-treatment group gained 10 kg or more. Patel noted: "We were half expecting that we might see some increase in body weight with the intensive-glucose strategy in ADVANCE. In fact, there [was] no change in body weight in the intensive group and. . . . there was a slight reduction in weight in the standard group over the period of follow-up. So by the end of follow-up, there was an average difference between the two groups of only 0.75 kg."

The glucose-lowering strategy in each study was quite different. ADVANCE was pragmatic in trying to achieve tight glucose control, and it took two or three years to separate groups and devise strategies—adding more drugs and encouraging investigators to do more lifestyle modifications. The ACCORD approach was more aggressive, using more agents from the start with multiple injections.

Dr Denise G Simons-Morton (National Heart, Lung, and Blood Institute, Bethesda, MD), project officer for ACCORD, pointed out, "Neither study found a significant reduction in the . . . macrovascular outcomes from intensive glycemia lowering. That's the similarity between the findings." She attributed the problems with ACCORD to a combination of three factors: the strategy used, the goal achieved, and the population in which it was achieved.

"The purpose of [ADVANCE] was to address whether getting very tight control of the glucose, rather than ordinary control, made a difference," Lebovitz said during the media briefing. "We would not conclude that glucose makes no difference, but we would say that attempting to get very very tight glucose control may not give you added benefit for cardiovascular disease."

Lebovitz continued, "On the other hand, remember that 50% of all the people who enter a dialysis program in the West have diabetes as the cause of their kidney disease. So . . . you don't have to get so tight in the control to get a better effect on cardiovascular [events], but you do if you want to really reduce kidney disease, which is a very important end point."