DAPA-CKD: SGLT2 Inhibitors 'The Beta-Blockers for the Kidneys'

Jennie J. Lin, MD, MSc; Leslie S. Gewin, MD


September 04, 2020

This transcript has been edited for clarity.

Jennie J. Lin, MD, MSc: Hi. My name is Dr Jennie Lin. I am an attending physician nephrologist and physician scientist at Northwestern University.

Leslie S. Gewin, MD: Hi. My name is Leslie Gewin, and I'm also an attending physician scientist and nephrologist, at Vanderbilt University Medical Center.

Lin: On August 30, 2020, the highly anticipated, groundbreaking findings from the DAPA-CKD phase 3 trial were presented at the European Society of Cardiology (ESC) digital meeting. Researchers conducted a trial to determine whether dapagliflozin, a sodium-glucose transport protein 2 (SGLT2) inhibitor, conferred clinical benefit to chronic kidney disease (CKD) patients with and without diabetes.

For a bit of background, in 2019, the CREDENCE trial found that canagliflozin given to CKD patients with type 2 diabetes conferred a 30% relative risk reduction in the primary composite endpoint of end-stage kidney disease, doubling of serum creatinine, and renal or cardiovascular death.

Because SGLT2 inhibitors are hypothesized to affect physiology beyond just glucose lowering, a logical question to follow CREDENCE is whether the kidney-protective effects of SGLT2 inhibitors extend to CKD patients without diabetes. The cardiovascular benefits of SGLT2 inhibitors have already been reported in multiple clinical trials. With respect to dapagliflozin, the DAPA-HF trial found that dapagliflozin reduced mortality risk and heart failure progression in patients with and without diabetes. So does dapagliflozin confer clinical and survival benefit to kidney patients with and without type 2 diabetes?

Gewin: The primary endpoint in this study was a composite of a sustained reduction in GFR of greater than 50%, end-stage kidney disease, or renal or cardiac death.

The exciting result from this study was that there was a very statistically significant difference and a relative risk reduction of 39% in the group that was treated with dapagliflozin. This trial was actually stopped early in March because of a signal of a protective effect. What I think is perhaps most exciting is that both patients with diabetic CKD and nondiabetic CKD had a significant decrease in renal failure and the composite of reduction in cardiovascular death or hospitalizations for heart failure, as well as an increased survival benefit.

This drug, which was originally developed as an antiglycemic agent, is shown to have a renal protective effect in patients with both diabetic and nondiabetic CKD.

Although DAPA-CKD showed many positive effects, the one area in which there wasn't a positive finding was in the endpoint of cardiovascular mortality. Unlike in CREDENCE, that finding did not meet statistical significance even though there was an overall mortality benefit.

The DAPA-CKD study looked at patients with CKD who had eGFR ranging from 25 to 75 mL/min/1.73 m2, with a mean eGFR of about 43 mL/min/1.73 m2, which is quite a bit lower than in some of the prior trials that have looked at SLGT2 inhibitors.

Dr Lin, do you want to comment about some of the differences between CREDENCE and DAPA-CKD?

Lin: Sure. You bring up a great point. Because the paper is not published yet (we're anticipating a publication soon), it's a little difficult to get down to the nitty gritty of DAPA-CKD.

One of the main differences is the patient population, particularly the nondiabetic CKD patients in DAPA-CKD. That additional mortality risk with diabetes as an independent risk factor could be affecting that outcome in CREDENCE. DAPA-CKD was stopped early, so maybe there wasn't enough power to detect an effect in a mixed population in terms of diabetes and no diabetes.

The other difference in the numbers is that CREDENCE reported a 30% relative risk reduction for their primary outcome vs DAPA-CKD, which reports 39%. One of the key differences in their composite primary outcome was the inclusion of a doubling of serum creatinine in CREDENCE, which is a much more difficult bar to clear than a 50% decline in eGFR.

Even though you only need a 50% decline in eGFR to meet an FDA indication for the drug, we were impressed that the CREDENCE investigators were overachievers, in some sense, and actually chose a doubling of serum creatinine. It's a slightly different comparison, so I wouldn't say that dapagliflozin is superior to canagliflozin or anything like that. Again, we would need a head-to-head trial to determine that, but both are very strongly positive trials.

One detail that I did not see explicitly explained in the slides was a reduction in albuminuria. In the verbal presentation at the ESC congress, they mentioned a 30% reduction in albuminuria with dapagliflozin, so that would be consistent with the other -gliflozin trials.

Gewin: I agree that it's hard to compare the two drugs, and the results to me seem largely consistent that there's a benefit with both.

This is such an exciting trial for the kidney community. Are there any safety concerns that you have related to this? I know that the earlier studies have shown some signal in regard to fracture and amputations, which was not shown here, which is very reassuring and encouraging. There was no evidence of diabetic ketoacidosis in the treatment group, which is also reassuring. Are there any safety concerns that you have about the drug?

Lin: Not specifically. Again, one of the things about the differences in amputation rate from CANVAS is that we don't have a great explanation for why it was in CANVAS and not in CREDENCE. In DAPA-CKD, with the inclusion of nondiabetics, we also saw a lower event rate.

When I calculated out the absolute percentage of people who ended up with amputations per group, it was about half of what we saw in CREDENCE. Again, there were no differences between groups, so I'm not suggesting that there's an effect of the drug on amputation rate.

We don't have the full table of all of the adverse reactions and events that they may have looked at, including urinary tract infection (UTI), which was a concern with canagliflozin. That is something that would be interesting to see. They did not report anything significant verbally, so there may not have been an effect, but again, we have to wait for the publication.

Gewin: I agree. That's a big concern within the whole class of SGLT2 inhibitors, given the increased glycosuria and increase in UTIs, which can come from that. Theoretically, one might postulate that the UTI rate would be lower in a population with a lower GFR, as the glycosuric effects could be attenuated as GFR drops. Certainly, in the nondiabetic population, this would be less of a concern.

I completely agree that seeing the results in print and looking more closely at the side-effect profile and adverse effects would be really important for these drugs.

Lin: Given all of that, for your clinic patients, do you feel ready to prescribe it at this point?

Gewin: Well, actually, I'm going to clinic this afternoon. Certainly, for my diabetic patients, I feel like there's an abundance of evidence from EMPA-REG, CREDENCE, and DAPA-CKD that this is a great drug for patients with diabetic CKD, with or without proteinuria. I'm ready to prescribe that right away.

I think the bigger question is, what about patients with nondiabetic CKD? I would like to see the paper in print and look more closely at the details before generalizing it to that whole population. It's also important to remember that this patient population did have quite a bit of albuminuria. Whether or not the findings apply to patients who have minimal albuminuria is an open question. I believe they excluded patients with polycystic kidney disease, type 1 diabetes (probably for the concern of diabetic ketoacidosis), lupus, and ANCA-associated vasculitis.

I think there are many questions that remain in terms of which nondiabetic CKD population this drug would best serve.

What about you, Dr Lin? What's your level of enthusiasm for the drug? Are there any populations or subpopulations for which you would hesitate to prescribe this drug?

Lin: I don't have any major reservations at this point. I anticipate the paper being published before we end up changing our practice significantly. One of the things I want to point out for this study is contrasting the enrolled patients with the US kidney population, which has a substantial proportion of people who self-identify as Black. When I was looking at Table 1 for this study, I noticed that fewer than 10% were self-identified as Black. That is one major difference, although I don't anticipate anything in terms of population-specific differences, based on some of the hypotheses for how this drug works.

In terms of who I might have some reservations about using it in, I would include people in whom frequency of urination may be a problem. When you are spilling sugar into your urine, you end up having to go to the toilet more often. There are certain occupations where that may be inconvenient. That is a concern expressed by endocrinologists who don't love the drug because it's not that great of a glucose-lowering drug. Their patients may experience more side effects and less benefit from the glucose-lowering aspect. From our perspective, as nephrologists who may be prescribing this, even to people without diabetes, that risk-benefit ratio may end up favoring the benefits over some of these more mild side effects.

In terms of other populations that I would think about using this in, even more are people with advanced CKD, particularly if you have diabetes and you're at risk for hyperkalemia, and you may end up having hyperkalemia from hyperreninemic hyperaldosteronism.

One thing to note is that this did not replace ACE inhibitors or ARBs. Dapagliflozin was given on top of optimized dosing. In terms of effect on proteinuria and progression, we're not replacing a RAAS blockade with the SGLT2 inhibitors.

A next logical question — I haven't looked at to see this — are they going to do a trial where you actually stop the ACE inhibitor and ARB and replace it with SGLT2 inhibitors and do a head-to-head comparison for people with advanced CKD who are not able to take ACE inhibitors and ARBs because of potassium? Maybe they're not on sodium zirconium cyclosilicate or patiromer at this point. They may benefit, but we don't know that until a study is done. That is one thing that I am thinking about, and we may actually end up hearing a lot of anecdotal stories from our colleagues about that.

Gewin: Great point. I think that over 97% of the patients in this study were on some type of RAAS blockade, either ACE inhibitor or ARB. I look at it as this great add-on therapy, which probably has a mechanism that's independent of RAAS blockade, that's adding a benefit.

It would be hard to have a study comparing them directly because RAAS blockade is such a standard of care for diabetic and proteinuric CKD. I agree that it's a great opportunity for patients (and we have several in the clinic) who have type 4 renal tubular acidosis and can't tolerate RAAS blockade because of the hyperkalemic side effects. To have a drug that can have an impact without those effects is really a great opportunity for the nephrology community.

One thing in terms of the safety profile that was notable is that the nondiabetic population did not have any hypoglycemic episodes. It doesn't seem like this drug, by itself, can induce hypoglycemia. I do think with my diabetic patients who are on either insulin or sulfonylureas, which independently have a risk for hypoglycemia, that I would probably have a conversation with their endocrinologist and coordinate adding on this medication, because it may potentiate those hypoglycemic effects. It seems otherwise that hypoglycemia is not as much of a concern.

The other thing that was reassuring was that there was no statistically significant difference in hypovolemia events, which some have raised concerns about previously.

Lin: Yes, absolutely. People have been speculating about how this drug works, and the nephrology community, we love to argue, as you know. Many people are talking about tubular glomerular feedback vs other cellular mechanisms. The question is, do we need to investigate this further? I would argue yes, but I would be curious to know about your thoughts.

Gewin: I agree 100% that we need to understand better the mechanisms of action for many reasons. I think that it will help us better define subpopulations of CKD patients who may benefit from these drugs. In addition, it can also point to future antifibrotic drugs that we may be able to bring to the market.

The mechanism of action is a really fascinating area. Obviously, there are antiglycemic and natriuretic effects of these drugs. It's been pretty well demonstrated that the protective effect we're seeing here in trials like DAPA-CKD are not solely due to the effects on blood glucose or on blood pressure.

The other effect that's been well publicized is the effect on the transglomerular pressure. Reducing glomerular hyperfiltration can help protect against CKD progression. Certainly, the initial decrease in GFR seen with patients on this drug would point to that as a part of the mechanism.

I think there may be part of the story that goes even beyond those mechanisms of action. What do you think?

Lin: I agree. I've always been a proponent of studying what's going on at the cellular level with this drug because we all know that the proximal tubule requires a lot of energy to function. I like referring to this class of drug as the beta-blockers for the kidney, in the sense that we're reducing the workload by blocking glucose reabsorption, but it may also point to benefits in terms of mitochondrial function, suppressing oxidative stress, and restoring autophagy. These are some of the basic science mechanisms that have been investigated in the endocrinology and cardiology spheres.

From what we know about genome-wide association studies and other genetic studies, those are pathways that are important in the development and progression of CKD. If we can identify which specific parts of those pathways the SGLT2 inhibitors affect, we could end up getting even more specific therapies with fewer adverse events, including not having to worry about UTIs in our diabetic patients or the amputations, even though there wasn't a difference in the groups. It's still of concern to the small number of people who may end up with those adverse effects.

Gewin: It's really exciting. There's a nice article by Milton Packer about how these SGLT2 inhibitors may be causing a state of nutrient deprivation that activates AMPK and transcription factors like sirtuin 1, which can then have more systemic effects on inflammation, potentially oxidative stress, and, as you mentioned, autophagy, which can really reprogram some of the metabolism in the body and lead to widespread effects not just in the kidney but also in the heart.

Some of these more systemic effects make sense from a global perspective of metabolic reprogramming. Even though I think the proximal tubule is very important to renal injury and I would like to say that all of the effects are due to the proximal tubule, I think it's probably much more widespread and systemic.

Lin: This could be our drug to fix cardiorenal syndrome, right? It's something that we've always been struggling with. Any last thoughts, Dr Gewin?

Gewin: I just think this is such an exciting moment for the nephrology community. We've really needed some positive studies looking at prevention of CKD progression. To have this drug in our armamentarium for the CKD population is really exciting, and the idea that it can be extended potentially beyond diabetic CKD is also really great news for the nephrology community.

Lin: Absolutely. I agree. Thank you for watching. We hope that you found this discussion to be informative and helpful.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.