We have previously shown that plasma TNFR1, TNFR2, and KIM-1 combined with clinical variables provides effective risk stratification for kidney outcomes in individuals with diabetic kidney disease (DKD) stages 1-3. With the recommended use of a race-free eGFR and a refinement of our progression endpoint, we sought to develop and independently validate a version 2.0 of KidneyIntelX assay in a previously untested, independent, contemporary cohort of subjects from BioMe Biobank (Mount Sinai Health System, NYC, NY). The primary outcome was a composite of ≥40% sustained decline in eGFR or kidney failure (sustained eGFR<15 ml/min/1.73 m2) within 5 years from enrollment. In a training cohort of 573 patients with DKD from the Penn Medicine Biobank (PMBB), 15.4% experienced the progressive decline in kidney function over a median of 3.1 years. The following features were selected in a training model utilizing optimized random forests to predict the kidney outcome: TNFR1, TNFR2, KIM-1, baseline UACR, HbA1c, and blood urea nitrogen. The external validation cohort (BioMe, n=657) had an 11.7% event rate over a median of 3.8 years. Based on risk categorization cutoffs derived from training, 57%, 35% and 8.3% of participants were classified as low, intermediate and high-risk, respectively. The cumulative event incidence probability in the low, intermediate, and high-risk groups were 5.9% (95% CI: 3.6-9.4%), 21.2% (95% CI: 15.6-28.3%) and 66.9% (95% CI:, 49.3-83.5%). The unadjusted HRs for high vs. low and intermediate vs. low were 18.4 (95% CI: 9.4 -34.6) and 4.2 (95% CI: 2.4-7.4), respectively. After adjustment for age, sex, race, baseline eGFR, uACR, SBP, and HbA1c, the adjusted HRs were 7.7 (95% CI: 3.0-19.6) and 3.7 (95% CI: 2.0-6.8), respectively. In summary, we developed and independently validated a version 2.0 KidneyIntelX with previously established biomarkers combined with selected clinical variables and demonstrated robust prediction of progressive decline in kidney function, independent of common risk factors.

Disclosure

G.N.Nadkarni: Advisory Panel; Renalytix, Consultant; Renalytix, Other Relationship; Renalytix, Speaker's Bureau; Daiichi Sankyo, GlaxoSmithKline plc., Menarini Group, Stock/Shareholder; Renalytix. F.Fleming: Employee; Renalytix. M.J.Donovan: Employee; Renalytix. S.Coca: Advisory Panel; Renalytix, Bayer Inc., Boehringer Ingelheim International GmbH, Consultant; Renalytix, Nuwellis, 3ive Labs, Reprieve Cardiovascular, Vifor Pharma Management Ltd., Stock/Shareholder; Renalytix.

Funding

Renalytix

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.