Urinary DKK3 as a biomarker for short-term kidney function decline in children with chronic kidney disease: an observational cohort study

Background: Childhood-onset chronic kidney disease is a progressive condition that can have a major effect on life expectancy and quality. We evaluated the usefulness of the kidney tubular cell stress marker urinary Dickkopf-related protein 3 (DKK3) in determining the short-term risk of chronic kidney disease progression in children and identifying those who will benefit from specific nephroprotective interventions.

Methods: In this observational cohort study, we assessed the association between urinary DKK3 and the combined kidney endpoint (ie, the composite of 50% reduction of the estimated glomerular filtration rate [eGFR] or progression to end-stage kidney disease) or the risk of kidney replacement therapy (ie, dialysis or transplantation), and the interaction of the combined kidney endpoint with intensified blood pressure reduction in the randomised controlled ESCAPE trial. Moreover, urinary DKK3 and eGFR were quantified in children aged 3-18 years with chronic kidney disease and urine samples available enrolled in the prospective multicentre ESCAPE (NCT00221845; derivation cohort) and 4C (NCT01046448; validation cohort) studies at baseline and at 6-monthly follow-up visits. Analyses were adjusted for age, sex, hypertension, systolic blood pressure SD score (SDS), BMI SDS, albuminuria, and eGFR.

Findings: 659 children were included in the analysis (231 from ESCAPE and 428 from 4C), with 1173 half-year blocks in ESCAPE and 2762 in 4C. In both cohorts, urinary DKK3 above the median (ie, >1689 pg/mg creatinine) was associated with significantly greater 6-month eGFR decline than with urinary DKK3 at or below the median (-5·6% [95% CI -8·6 to -2·7] vs 1·0% [-1·9 to 3·9], p<0·0001, in ESCAPE; -6·2% [-7·3 to -5·0] vs -1·5% [-2·9 to -0·1], p<0·0001, in 4C), independently of diagnosis, eGFR, and albuminuria. In ESCAPE, the beneficial effect of intensified blood pressure control was limited to children with urinary DKK3 higher than 1689 pg/mg creatinine, in terms of the combined kidney endpoint (HR 0·27 [95% CI 0·14 to 0·55], p=0·0003, number needed to treat 4·0 [95% CI 3·7 to 4·4] vs 250·0 [66·9 to ∞]) and the need for kidney replacement therapy (HR 0·33 [0·13 to 0·85], p=0·021, number needed to treat 6·7 [6·1 to 7·2] vs 31·0 [27·4 to 35·9]). In 4C, inhibition of the renin-angiotensin-aldosterone system resulted in significantly lower urinary DKK3 concentrations (least-squares mean 12 235 pg/mg creatinine [95% CI 10 036 to 14 433] in patients not on angiotensin-converting enzyme inhibitors or angiotensin 2 receptor blockers vs 6861 pg/mg creatinine [5616 to 8106] in those taking angiotensin-converting enzyme inhibitors or angiotensin 2 receptor blockers, p<0·0001).

Interpretation: Urinary DKK3 indicates short-term risk of declining kidney function in children with chronic kidney disease and might allow a personalised medicine approach by identifying those who benefit from pharmacological nephroprotection, such as intensified blood pressure lowering.

Funding: None.