New Directions in Diabetic Kidney Disease Trials

Vlado PerkovicVlado Perkovic, MBBS, PhD, is Executive Director of The George Institute, Australia, Professor of Medicine at UNSW Sydney, and a staff specialist in nephrology at the Royal North Shore Hospital.

Search for other papers by Vlado Perkovic in
Current site
Google Scholar
Full access

The outlook for people diagnosed with type 2 diabetes and chronic kidney disease today is more hopeful than it has ever been. A broad array of treatments are available, and the last decade has seen an explosion of evidence from high-quality, properly powered, randomized trials that have defined the benefits and risks of many of these treatment options.

The 2008 decision by the U.S. Food and Drug Administration (FDA) and other regulatory agencies to require the conduct of cardiovascular safety trials for all new diabetes medications (1) has directly led to the generation of evidence that can guide treatment. We now know which agents reduce the risk of cardiovascular disease, kidney disease, or both, as well as lowering glucose levels. These trials have also taught us much about the effects of these agents on both common and uncommon adverse events, and have driven new areas of basic research, as we try to understand the mechanisms underpinning the clinical effects observed. The decision to mandate these trials will allow more effective and efficient use of glucose-lowering treatments, and has directly improved outcomes for people with diabetes.

Over the coming years, a number of additional placebo-controlled outcome trials of novel glucose-lowering therapies will report (Table 1), providing further richness to the available evidence. But a number of factors suggest that the landscape of trials going into the next decade are likely to look quite different from those completed over the past 10 years.


One reason for this is that the proven benefits of existing treatments must be taken into account in designing new trials. Previous trials looking at clinical renal outcomes in diabetes and CKD have required most or all participants to be receiving renin-angiotensin system blockade. Clear benefit for canagliflozin was demonstrated in people with diabetes and very high albuminuria in the CREDENCE trial (2), and there is growing evidence of renal benefits for SGLT-2 inhibitors across the spectrum of diabetes and kidney disease (3). Rapid increases in the use of these agents by nephrologists and other practitioners is therefore appropriate and will need to be taken into account for future trial design. While it would be ideal to test future treatments on top of SGLT inhibitors, many people may not have access to them for financial reasons, or be able to tolerate SGLT-2 inhibitors. So some degree of pragmatism will be required, particularly as uptake is (unfortunately) likely to take some time.

Slower kidney function loss in diabetes with proven new treatments is obviously a great outcome. But it may also make it more difficult to demonstrate benefits on existing renal outcomes. Event rates will be lower in treated participants, so that larger sample sizes will be required to demonstrate realistic effects on these outcomes. In this light, the recent initiatives by the National Kidney Foundation, the U.S. FDA, and the European Medicines Agency to explore the role of changes in kidney function (eGFR slope) as an outcome for future trials may be critical (4). Slope-based outcomes are likely to make reliable demonstration of benefit easier, but separate attention to collecting adequate safety data will also be required. Slope-based outcomes may also facilitate the development of more efficient approaches to the conduct of trials, using platform approaches and adaptive methodologies (5), particularly as new targets are identified through modern ‘omics’ approaches.

Another difference will be a growing need to understand the absolute and relative effects of combinations of therapy. Incomplete uptake of SGLT-2 inhibitors in future trials will allow assessment of effects in people with and without this treatment. As more renoprotective therapies (hopefully) are identified, the assessment of different combinations is likely to become more important.

Perhaps most important, the development of a growing number of proven renoprotective therapies poses a new challenge. Use of RAS blockade among people in whom it is indicated is still likely to be suboptimal, almost two decades after the benefits were proven. Vast numbers of people who could have benefited from this treatment are likely to have reached kidney failure prematurely as a result of implementation failure. The challenge for us going forward will be to make the development and testing of implementation strategies a research focus, so that we can translate research findings much faster, for the benefit of people with kidney disease.

We have achieved much in diabetic kidney disease, and the rich tapestry of ongoing research suggests we are likely to achieve much more over the coming years. But we will need to adapt our questions, our approaches, and our goals if we want to achieve the best possible outcomes for our patients into the future.