ADVANCE

The ADVANCE trial began to recruit nearly 20 years ago and it is apposite to review the trial results in the light of the history and progress made in therapeutics in the treatment of type 2 diabetes since the main results were published in 2008. In this series of articles in this supplement, Zoungas, Knudsen & Cooper, Chalmers & Woodward, and Marre outline the history and the findings on the ADVANCE trial in some detail, covering explicitly the lessons learned about glycaemic control, renal disease, observational studies, and blood pressure (BP) control. Long-term outcome trials are expensive and labour-intensive. They involve years of patient and physician dedication to address explicit problems and the hope is always that the findings will influence our current paradigms of treatment. ADVANCE was no exception, successfully randomising 11 140 patients at 215 centres in 20 countries of the world and studying them for a median of 5 years. It was cleverly conceived as a Latin-square trial, assessing both glycaemic lowering and BP control in the outcomes of mortality, macroand microvascular disease. It used a perindropril (an ACE inhibitor) and indapamide (thiazide-like diuretic) combination as the active arm of the BP control and gliclazide MR (modified release sulphonylurea) in the active glycaemic control arm. The results have been widely disseminated and discussed and the trial has produced more than 50 publications. The unknown at the time of the inception of the trial was the open question of whether, and by how much, such BP and glycaemic control might reduce complications in those with established diabetes—the median duration of diabetes for recruitment into the study being 8 years. There were high hopes based on the UKPDS findings that there would be significant outcomes for macro-, microand all-cause mortality. But also, as in all large trials, not everything progressed as planned, and in particular the glycaemia reduced more slowly than expected leading to the worry that if a difference in glycaemia was not established and maintained then one of the central tenets of the trial would be at risk. But the slow glycaemic response turned out to be one of the most fortuitous outcomes of the trial as the results were announced at nearly the same time as the ACCORD trial. In ACCORD, the investigators had hypothesized that if a small difference were helpful, then a big difference in glycaemia would be very helpful. They used multiple agents—metformin, sulphonylurea, thiazolidinediones and insulin individually and in combination—to reduce glycaemia in their intervention group, achieving a rapid reduction within 9 months from 8.1% median HbA1c to 6.4% in their intensively treated group of patients, only to find that their intervention, far from being advantageous was deleterious, with a significant 22% higher mortality in the intervention group. The ACCORD trial came to an abrupt halt with the intervention of their safety monitoring committee. Meanwhile, in ADVANCE the fortuitous slow decline in glycaemia mediated by gliclazide MR yielded a significant 10% reduction in combined major and microvascular events over the course of the trial. Fortunately ADVANCE rescued the world from the false assumption, adopted by some reading the ACCORD results, that glycaemic lowering was unnecessary or even dangerous—the hypoglycaemia rate in ACCORD was about six times that seen in ADVANCE. Weight increase in ACCORD was also substantial in the intensive group (3.5 kg) compared to a negligible alteration (0.1 kg) in ADVANCE. The important take-home message was that a gradual reduction in glycaemia (Hba1c reducing from 7.5% to 6.5% over 5 years) was associated with a significant reduction in combined events, mainly as a result of significantly lower rates of renal complications, this last being one of the leading signals to the medical community that renal disease was both an important pathological entity (previously recognised but never highlighted) and that it could be prevented. That signal marked a sea-change in the interest of renal physicians and is part of the background to the emerging good news stories about renal protection associated with the SGLT2 inhibitors and the GLP-1 agonists. Received: 3 February 2020 Accepted: 23 February 2020