Protein profiles and associated biological pathways among different aetiologies of chronic kidney disease: new insights from DAPA-CKD.

BACKGROUND

We investigated circulating protein profiles and molecular pathways among various chronic kidney disease (CKD) aetiologies to study its underlying molecular heterogeneity.

METHODS

We conducted a proteomic biomarker analysis in the DAPA-CKD trial recruiting adults with and without type 2 diabetes with an estimated glomerular filtration rate of 25-75 ml/min/1.73 m2 and a urine albumin:creatinine ratio of 200-5000 mg/g. A total of 2926 proteins were analysed by Olink Explore in plasma samples collected at baseline. Proteins associated with diabetic kidney disease (DKD), glomerulonephritis (GN) and hypertensive nephropathy were identified using linear models to compare each CKD aetiology against the others.

RESULTS

Of 4304 randomized participants, baseline plasma proteomic profiles were available for 2485 participants (57.7%). Compared with patients without DKD, those with DKD had higher plexin B2 (PLXNB2; log2 fold difference 0.34, P < .001), vascular adhesion protein-1 (VAP-1; log2 fold difference 0.34, P < .001) and kidney injury molecule-1 (KIM-1; log2 fold difference 0.79, P < .001), suggesting involvement of immune activation, oxidative stress and tubular damage. Based on these molecular pathways, interleukin-6 was identified as an upstream regulator in DKD. In an external database of the Kidney Precision Medicine Project, PLXNB2 and VAP-1 were also elevated in DKD compared with non-DKD (log2 fold difference 0.41 and 0.42, respectively). In GN, levels of nephrin, a key slit diaphragm protein, were lower (log2 fold difference -0.27, P < .001). Furthermore, receptor for advanced glycation end products (log2 fold difference 0.28, P < .001) and heparan sulphate 6-O-sulfotransferase 2 (log2 fold difference 0.2, P < .05) were higher, suggesting glycocalyx disruption and inflammation. In hypertensive nephropathy, epiregulin, a member of the epidermal growth factor family, was higher compared with other CKD aetiologies.

CONCLUSION

Our study highlights distinct proteomic differences in patients with DKD, GN and hypertensive nephropathy. These findings may aid in unravelling the heterogeneous underlying molecular phenotype of CKD pathophysiology.