New study uses biopsy, machine learning to identify pathways in diabetic kidney disease
Click Here to Manage Email Alerts
Diabetic kidney disease is one of the most frequent complications in both type 1 and type 2 diabetes and the most common cause of chronic kidney disease in the United States.
As part of an effort to reduce the rate of people with diabetic kidney disease (DKD) who develop renal failure, researchers in the Perelman School of Medicine at the University of Pennsylvania are spearheading an industry-academic collaboration to better understand the development and progression of the disease.
Investigators at more than 20 institutions nationwide will participate in the Transformative Research in Diabetic Nephrology (TRIDENT) study. The study, which is supported by about $5 million in grants, aims to identify a combination of clinical, molecular or histopathologic features that characterize people with DKD and rapid renal function decline. Investigators are collecting kidney tissue samples from people with diabetes to perform comprehensive genetic, genomic and epigenetic profiling studies. The goal, according to researchers, is to identify the molecular fingerprints of progressive renal failure in people with DKD and pinpoint drug targets.
Healio spoke with Katalin Susztak, MD, PhD, professor of medicine in renal-electrolyte and hypertension and genetics in the Perelman School of Medicine and TRIDENT’s principal investigator, about the need for therapeutics in DKD, the possibility of preventing progression to kidney transplant or dialysis, and the promise of genetics and precision medicine.
What are some of the challenges in developing effective therapeutics for diabetic kidney disease?
Susztak: In the United States and most of the western world, roughly 50% of all chronic kidney disease is caused by diabetes. It has been very difficult to develop specific therapeutics.
Most of the available drugs for DKD were not initially approved for DKD. SGLT2 inhibitors were developed for type 2 diabetes, and angiotensin-converting enzyme inhibitors and angiotensin receptor blockers were developed for hypertension. It just so happened that these agents demonstrated a benefit in DKD.
There may now be potential with the SGLT2 inhibitor empagliflozin (Jardiance, Boehringer Ingelheim and Eli Lilly), which the FDA announced received fast track designation for an investigation to reduce the risk for renal disease and CV death in people with chronic kidney disease. However, even though empagliflozin seems to be effective, it is not fully clear how it works. The effect size of the SGLT2 inhibitors generally seems to be quite good. That means they could be beneficial for 20% to 30% of patients with DKD, which is, in our domain, a fairly large number. But that still leaves 70%. When we have 20 million people with kidney disease — 10 million of them with diabetes — then we’re talking about 7 million people in need of targeted therapy. Even though these drugs are a sign of great progress in our field, they are not going to be sufficient. These agents slow disease progression. They do not stop disease progression. It just extends the time before a person with kidney disease needs transplant or dialysis.
There are several reasons we do not have more drugs to treat DKD. One is we need to better understand the genetics of disease. For reasons we don’t fully understand, the genetics of DKD is poorly understood. Animal models are generally not good tools for drug development in DKD. In addition, in the past, kidney biopsy was often not performed on people with DKD.
What led to the development of the TRIDENT study?
Susztak: Due to all the deficiencies in modeling diabetic kidney disease, we feel that the best option is to study people with DKD. It has been quite difficult to start unbiased studies. TRIDENT is unique in that it developed as a private-public partnership. We approached industry, who are interested in developing drugs for kidney disease in general, to form a large consortium, which now includes four pharmaceutical partners. They had to come to an agreement that this is a discovery study and they will share all the findings, which was quite a significant undertaking. Right now, we have roughly 20 institutions in the U.S. that are able to participate because they have the infrastructure.
That provided the groundwork to develop a proposal. We want to enroll people with diabetes and DKD on biopsy. Each person enrolled in TRIDENT will undergo a kidney biopsy using the ideas of “omics” profiling established in cancer research. We will then perform genetic studies, including whole exome sequencing and epigenetic studies, to understand the heritability of DKD. The primary outcome for this study is the rate of renal decline — essentially, who is going to end up on dialysis? That is the most important factor, not just who will develop DKD, but who will progress to dialysis or needing transplant. We hope to use all of the data from this comprehensive study and run algorithms to identify pathways that drive disease progression and also understand patient subgroups who may be eligible for more targeted therapy.
How will this trial help you and your colleagues gain a better understanding of the pathogenesis of DKD?
Susztak: The diagnosis of DKD or diabetic nephropathy is frequently established on clinical parameters. This disease course is heterogeneous; some patients will have stable kidney function, while others will have rapid deterioration of kidney function. TRIDENT will recruit people with diabetes who are undergoing clinically indicated kidney biopsies at one of our recruitment centers. We will enroll patients that meet the histologic gold standard and diagnostic criteria of DKD from their biopsy and follow the participants to characterize changes in kidney function.
Once we have generated these datasets, we will run machine-learning algorithms and identify key pathways that essentially drive disease progression in patients or among a subgroup of patients. After that, we hope to learn several things. Some of these pathways could be directly targeted by drugs. Others might need to be modeled in animal models for us to gain further proof.
We also are trying to take TRIDENT to the next level — TRIDENT 2.0. The 2.0 study would serve as an umbrella study for early drug development. The idea is, if we identify a subgroup of patients where there is a specific pathway that drives progression, we would like to treat those patients if a drug is available, and see if this is impactful and could potentially change the course of disease. That is a more complicated analysis. We would need to find a specific pathway, find specific patients and then find specific drugs. The initial phase was aiming to enroll 200 patients with complete follow-up for 18 months, but we are hoping to increase the number of people enrolled.
When can we expect the learn more?
Susztak: We have one of the first multi-omic datasets being analyzed in the lab, so we are sifting through the new genes and pathways as we speak. We have to see what we find. Hopefully, we will have news soon. – by Regina Schaffer
For more information:
Katalin Susztak, MD, PhD, can be reached at ksusztak@pennmedicine.upenn.edu; Twitter: @KSusztak.
Disclosures: Boehringer Ingelheim, Gilead Sciences, GlaxoSmithKline, JDRF and Regeneron Pharmaceuticals are providing grant support for the TRIDENT study. Susztak reports no relevant financial disclosures.