Plasma NGAL, suPAR, KIM-1 and GDF-15 for Improving Glomerular Filtration Rate Estimation in Older Hospitalized Patients

Accurate glomerular filtration rate (GFR) estimation is crucial for diagnosing kidney disease and prescribing renal risk medications [1-3]. In clinical practice, estimated GFR (eGFR) is typically determined by plasma creatinine levels adjusted for age and sex (eGFR_cre) [4]. However, creatinine is an imperfect metric of GFR, particularly in older (age ≥ 65 years) hospitalized patients, due to age-related changes in non-GFR factors such as muscle mass and nutritional status that affect creatinine level [5]. The addition of cystatin C to creatinine (eGFR_comb) improves the accuracy of GFR estimates for older hospitalized patients, but there remains a large and unpredictable margin of error compared with measured GFR (mGFR) [5].

mGFR is not feasible for large-scale use due to its complexity, underscoring the need for alternative biomarkers to improve GFR estimation. We previously reported that the addition of β-trace protein (BTP) and β2-microglobulin (B2M) (eGFR_panel) did not substantially improve accuracy beyond eGFR_comb, so we considered other novel biomarkers that may reflect kidney function. We are currently evaluating the utility of neutrophil gelatinase–associated lipocalin (NGAL), soluble urokinase plasminogen activator receptor (suPAR), kidney injury molecule-1 (KIM-1) and growth differentiation factor-15 (GDF-15) as early diagnostic and prognostic tools for acutely hospitalized patients. Elevated levels of these biomarkers have been associated with accelerated GFR decline [6-9]; therefore, we hypothesized that they may improve the accuracy of GFR estimates for older hospitalized patients.

The study was carried out in compliance with the guidelines set by the Basic & Clinical Pharmacology & Toxicology policy for experimental and clinical research [10]. This is a post-hoc analysis of data from a previously described study of GFR performance [5, 11] based on data from the umbrella ‘OptiNAM’ clinical trial [12]. The study included older Caucasian patients admitted to the Emergency Department at Hvidovre Hospital in Denmark. All patients provided informed written consent, and the clinical trial was approved by the Regional Ethical Review Board (H-18023853) and the Danish Data Protection Agency (H-18023853) and registered at ClinicalTrials.gov (NCT03741283). Additional details, including eligibility criteria, are described elsewhere [5]. Importantly, patients in this study were excluded because of dialysis, clinical signs of ascites or edema, prior amputation, use of immunosuppressants or acute kidney injury [5].

In total, data was analysed for 106 older hospitalized patients (median age 78 years, 43% male). The median body mass index was 26.0 kg/m², the median mGFR was 62.9 mL/min/1.73 m², 49% of patients had hypertension and 32% had diabetes.

Plasma levels for each novel biomarker, the association between novel biomarkers and mGFR and the effect of novel biomarkers on the association between eGFR and mGFR are given in Table 1. Association with mGFR was highest for NGAL (r² of 0.3590) and GDF-15 (r² of 0.2937) and lowest for suPAR (r² of 0.1563) and KIM-1 (r² of 0.0035). The association between mGFR and eGFR_cre (r² of 0.6883) increased most with the addition of NGAL (adjusted r² of 0.7345, change of 0.0463) or GDF-15 (adjusted r² of 0.7255, change of 0.0372). Compared with eGFR_cre, the association with mGFR was notably higher for eGFR_comb (r² of 0.8160) and eGFR_panel (r² of 0.8193), and these associations were largely unaffected by the addition of a novel biomarker. The largest improvement was observed for NGAL, which increased the r² for eGFR_comb or eGFR_panel by 0.0047 or 0.0032, respectively.

In this study, we evaluated whether the novel biomarkers NGAL, suPAR, KIM-1 or GDF-15 could improve the association between eGFR and mGFR. We found that the addition of NGAL or GDF-15 increased the association for eGFR_cre. However, this association remained smaller than that of eGFR_comb, which was largely unaffected by the novel biomarkers. This might indicate some informational overlap between cystatin C and NGAL, suPAR, KIM-1 and GDF-15, despite the fact that these biomarkers reflect distinct biochemical pathways. More work is needed to identify new biomarkers that can improve the accuracy of GFR estimates across various health and disease states [4]. Until then, eGFR_comb appears to be the most accurate metric of kidney function for older hospitalized patients. mGFR should be used in cases where a more accurate GFR assessment is needed, ideally with a minimally invasive setup such as dried blood spots [13].

Concept, design and methodology: M.B.H., A.L.N, A.B.W., O.A., E.I., J.T. and H.G.J-L. Collection of data: M.B.H., L.W.S.C, R.L.N., A.A., M. D, E.I., J.T. and H.G.J-L. Data analysis and interpretation: M.B.H., A.L.N. and H.G.J-L. Statistical analysis: M.B.H., A.L.N. and H.G.J-L. Funding: M.B.H. Writing—original draft preparation: M.B.H. and A.L.N. Writing—review and editing: M.B.H., A.L.N, A.B.W., L.W.S.C, R.L.N., A.A., O.A., M.D., E.I., J.T. and H.G.J-L. All authors have read and agreed to the published version of the manuscript.

The authors declare no conflicts of interest.