Scientists from Sanford Burnham Prebys and the Chinese University of Hong Kong have developed a computational approach that can predict the likelihood of kidney disease development in individuals with type 2 diabetes. The research, published in Nature Communications, presents DNA methylation markers as predictors of kidney function and the progression of diabetic kidney disease. The findings have the potential to aid doctors in preventing or effectively managing kidney disease in individuals with type 2 diabetes. The study highlights the future potential of predictive diagnostics and the integration of clinical data with advanced technology to optimize treatment strategies for type 2 diabetes patients and reduce the risk of kidney complications. Diabetes is the leading cause of kidney failure globally, accounting for a significant proportion of cases of end-stage kidney disease and dialysis. In the United States, 44% of these cases are attributed to diabetes, while in Asia, the number rises to 50%.

Co-senior author Ronald Ma, a professor at the Chinese University of Hong Kong, highlights the progress made in developing treatments for kidney disease in individuals with diabetes. However, assessing an individual's risk of developing kidney disease based solely on clinical factors can be challenging. Therefore, there is a crucial clinical need to identify those at the highest risk of diabetic kidney disease. The computational approach developed by the researchers provides a promising solution to address this need and optimize the management of kidney disease in people with type 2 diabetes.

Figure .1 Making Use of a New Algorithm to Predict Diabetic Kidney Disease

Figure 1 shows in a prospective cohort study using the Hong Kong Diabetes Register, researchers identified epigenetic markers associated with kidney function in individuals with type 2 diabetes. They conducted two independent epigenome-wide association studies, analysing DNA methylation patterns in 1,271 participants to identify markers associated with baseline estimated glomerular filtration rate (eGFR) and subsequent decline in kidney function (eGFR slope).

The study revealed 40 CpG sites (including 30 previously unidentified) significantly associated with baseline eGFR and eight CpG sites (all previously unidentified) associated with eGFR slope. Additionally, a multisite analysis method identified 64 CpG sites for baseline eGFR and 37 CpG sites for eGFR slope. The identified CpG sites were located near genes related to kidney diseases and showed associations with renal damage.

The findings highlight the potential of DNA methylation markers in risk stratification for kidney disease in individuals with type 2 diabetes. The algorithm developed in this study relies on DNA methylation, which can provide valuable information about gene activity and can be easily measured through blood tests.

Model built using certain data

The computational model developed by the researchers can utilize methylation markers obtained from a blood sample to predict both current kidney function and the future functioning of the kidneys. This means that the model could be readily integrated into existing methods for assessing a patient's risk of developing kidney disease. The use of methylation markers offers a promising approach for improving the evaluation and prediction of kidney disease in clinical settings.

The development of the computational model involved analysing detailed data from over 1,200 patients with type 2 diabetes in the Hong Kong Diabetes Register. To validate their approach and ensure its applicability in diverse populations, the researchers also tested the model on a separate group of 326 Native Americans with type 2 diabetes.The study highlights the significant value of the Hong Kong Diabetes Register and its potential to drive further discoveries in understanding diabetes and its complications. The long-term follow-up of patients within the register, which spans over two decades, provides a comprehensive view of how health can evolve in individuals with diabetes.

The researchers emphasize the Hong Kong Diabetes Register as a valuable scientific resource that offers insights into the long-term effects and changes in health among people with diabetes. The utilization of this extensive dataset enhances the robustness and applicability of the developed computational model.The scientists are presently attempting to improve their model. They are also extending the use of their methodology to consider additional issues relating to health and disease, such as figuring out why some cancer patients don't respond well to specific therapies.[1]

References:

1. https://www.genengnews.com/news/predicting-diabetic-kidney-disease-with-the-use-of-a-novel-algorithm/

Cite this article:

Janani R (2023),Making Use of a New Algorithm to Predict Diabetic Kidney Disease, Anatechmaz, pp.291