Biomarkers of Kidney Tubule Disease and Risk of End-Stage Kidney Disease in Persons With Diabetes and CKD

INTRODUCTION: Tubulointerstitial damage in diabetes and chronic kidney disease (CKD) is poorly captured by estimated glomerular filtration rate (eGFR) and albuminuria. Urine biomarkers of kidney health may better elucidate disease progression in persons with diabetes and CKD. METHODS: Per case-cohort design, we randomly selected a subcohort of 560 study participants of the REasons for Geographic And Racial Differences in Stroke (REGARDS) study from 1092 adults with diabetes and baseline eGFR <60 ml/min per 1.73 m(2) and registered a total of 161 end-stage kidney disease (ESKD) cases (n = 93 from the subcohort; n = 68 from outside the subcohort) during 4.3 ± 2.7 years mean follow-up. We measured urine biomarkers of kidney tubule injury (kidney injury molecule—1 [KIM-1]), inflammation and fibrosis (monocyte chemoattractant protein—1 [MCP-1]), repair (chitinase-3–like protein 1 [YKL-40]), and tubule function, including reabsorption (alpha-1-microglobulin [α1m]) and synthetic capacity (epidermal growth factor [EGF] and uromodulin [UMOD]). Weighted Cox regression models estimated ESKD risk adjusting for demographics, ESKD risk factors, and baseline eGFR and urine albumin. Least absolute shrinkage and selection operator (LASSO) regression identified a subset of biomarkers most strongly associated with ESKD. RESULTS: At baseline, subcohort participants had mean age of 70 ± 9 years, mean eGFR of 40 ±13 ml/min per 1.73 m(2), and median urine albumin-to-creatinine ratio of 33 (interquartile range 10–213) mg/g. Adjusting for baseline eGFR and albuminuria, each 2-fold higher urine KIM-1 (hazard ratio = 1.43 [95% CI: 1.17–1.75]), α1m (hazard ratio = 1.47 [1.19–1.82]), and MCP-1 (hazard ratio = 1.27 [1.06–1.53]) were independently associated with ESKD. LASSO retained KIM-1 and α1m for associations with ESKD. CONCLUSION: Among adults with diabetes and eGFR <60 ml/min per 1.73 m(2), higher urine KIM-1, α1m, and MCP-1 are independently associated with incident ESKD, providing insight into kidney disease progression in persons with diabetes and CKD.