Authors: Agnieszka Rybi Szumińska, Anna Wasilewska, Monika Kamianowska
Categories: Review, children, chronic kidney disease, end-stage kidney disease, protein biomarker
Source: Journal of Clinical Medicine
Doi: 10.3390/jcm12123934
Authors: Agnieszka Rybi Szumińska, Anna Wasilewska, Monika Kamianowska
Chronic kidney disease (CKD) in children is a major concern of medical care and public health as it is related to high morbidity and mortality due to progression to end-stage kidney disease (ESKD). It is essential to identify patients with a risk of developing CKD to implement therapeutic interventions. Unfortunately, conventional markers of CKD, such as serum creatinine, glomerular filtration rate (GFR) and proteinuria, have many limitations in serving as an early and specific diagnostic tool for this condition. Despite the above, they are still the most frequently utilized as we do not have better. Studies from the last decade identified multiple CKD blood and urine protein biomarkers but mostly assessed the adult population. This article outlines some recent achievements and new perspectives in finding a set of protein biomarkers that might improve our ability to prognose CKD progression in children, monitor the response to treatment, or even become a potential therapeutic target.
CKD in children is a progressive condition leading to kidney failure with other dangerous consequences, including high cardiovascular risk, mineral and bone disorder, developmental delay, and growth retardation [1,2]. In the literature, reliable data on the epidemiology of CKD in children are deficient and come mostly from kidney replacement therapy (KRT) registers (ESPN/ERA Registry in Europe or USRDS in the USA) [3,4]. Only a few studies have assessed CKD stages 2–5 in children with the use of estimated glomerular filtration rate (eGFR) according to KDIGO guidelines, and almost none investigates chronicity criterion (3 months of observation) to exclude transient fluctuations in kidney function. Nevertheless, data from the last 3 decades estimate the prevalence of pediatric CKD between 30 and 50 per million age-related population (pmarp) per year [5].
It should be emphasized that the etiology of pediatric CKD differs significantly from the adult population, where it is mainly secondary to hypertension or diabetes. The pattern of causes of progressive kidney impairment in children in developed countries responsible for about two-thirds of all cases is congenital anomalies of the kidneys and the urinary tract (CAKUT), including obstructive uropathy, reflux nephropathy or hypo- and dysplastic kidneys, and glomerulopathies (steroid-resistant nephrotic syndrome and other chronic glomerulonephritides). Similarly, the reduction in the number of nephrons associated with low-birth weight and small-for-gestational-age infants, as well as the rising incidence of pediatric obesity, are very important predisposing factors of the development of CKD [6,7,8].
According to the definition by KDOQI in 2002 and endorsed in 2012 by KDIGO, CKD in children is characterized by persistent kidney damage with normal eGFR (stage 1) or a drop in eGFR below 60 mL/min/1.73m^2^ lasting for at least 3 months [9,10]. Diagnosis of CKD in the pediatric population is still based on the assessment of conventional serum creatinine with eGFR and proteinuria. They are far from the ideal kidney function indicators in children while increasing relatively late in the course of the kidney impairment [11]. It has to be emphasized that despite ongoing injury, the kidney can maintain GFR to some extent by hyperfiltration and compensatory hypertrophy of the glomeruli [12,13]. Each of these traditional indicators has other important limitations. Serum creatinine concentration is affected by the patient’s race, gender, muscle mass, hydration or drugs, and therefore it only partially reflects kidney function [14]. Proteinuria is present in a persistent tubular injury, glomerular hyperfiltration or primary glomerular disorders, inflammation and tubulointerstitial fibrosis [15]. There are several pediatric studies that confirm the association between the intensity of proteinuria expressed as a urinary protein/creatinine ratio and the progression of CKD [16,17,18]; however, none of them indicates proteinuria as an early biomarker of CKD.
The process of CKD begins with an initiating factor causing the chronic injury of affected cells [19,20]. Regardless of the primary cause, the progression of CKD is quite similar and leads to tubulointerstitial disease. Subsequently, nephrons undergo stress resulting in the activation of inflammatory response and maladaptive tissue repair. With time, there is a progressive nephrons loss, developing interstitial fibrosis and compensatory hyperfiltration of healthy glomeruli [16,20]. It increases the permeability of the glomerular membrane, resulting in proteinuria and activation of the renin–angiotensin system with an increase in systemic blood pressure. Those two adverse consequences additionally escalate existing chronic kidney injury [17]. This pathophysiologic model of CKD is accompanied by the presence of different protein markers in plasma or urine that may act as regulators or be products of particular stages of the process.
There is a great need to establish a set of newer kidney biomarkers useful in the early detection and prognosis of the progression of CKD in children. We have to emphasize that CKD in children is not a homogeneous disorder, and proposed biomarkers may also reflect the condition responsible for the CKD.
A biomarker should identify a normal or pathogenic process or a response to therapeutic intervention. Ideally, its measurement should be non-invasive, easy to perform, sensitive and specific, repeatable and low-cost [21,22]. The Food and Drug Administration—National Institute of Health Biomarker Working Group describes types of biomarkers using terms diagnostic, prognostic, predictive, useful for monitoring the disease or for a pharmacodynamic response, safe and assessing risk or disease susceptibility [23]. Protein biomarkers are mainly mediators or byproducts of the processes involved in the pathway of CKD. Some of them are known, but many are still waiting to be detected and tested. We still do not have a newer protein biomarker validated for commercial use to diagnose or assess the progression of CKD in children. There are few candidates for this role.
This paper aims to gather evidence from the literature, especially from the last 5 years, regarding the most common protein biomarkers of CKD in children and present them as a review.
The analysis was performed with the guidelines of the updated PRISMA 2020 statement (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) [24]. The review was based on three PubMed, Web of Science and Embase. All records between January 2018 and January 2023 were included with the search strategy using MeSH (Medical Subject Heading) terms and keywords for the description of population and intervention with the help of the Boolean operators ”or”, “and”. We used the combination “chronic kidney disease” AND “children” AND “protein markers” AND “blood” OR “urine”.
The inclusion criteria were as (1) human research; (2) study group <18 years old and >1 month old; (3) clinical diagnosis of CKD or condition leading to CKD; (4) “new” protein markers obtained in blood or urine (creatinine, proteinuria or albuminuria did not meet the criterion of a “new” marker; (5) cohort and case-control investigations; (6) articles published in English in a peer-reviewed journal. The exclusion criteria were as (1) animal research; (2) newborns or adults included; (3) no specific diagnosis of CKD; (4) presence of other severe diseases or treatment that might interfere with the results of obtained protein kidney markers; (5) no control group to compare samples or control group was not precisely chosen; (6) studies not in English; (7) full text not available.
Three hundred and forty-eight records were found, but only thirty-five were under PRISMA’s guidelines. The excluded studies were duplications or did not meet inclusion criteria; for instance, there was no clear diagnosis of CKD or condition leading to kidney dysfunction; research was performed on animals; study groups were newborns or adults; markers were not proteins; were common, such as creatinine or proteinuria, not novel or obtained in material other than blood or urine. There were also 14 case reports excluded; 2 were not available as full-size text, and 1 was not in English (Figure 1).
We designed a search protocol to include as much research on new potential kidney biomarkers as possible but only in the pediatric population with a clear CKD diagnosis or condition leading to the disease. The methodology of these studies may vary as the authors used different ways of assessing protein markers from the ELISA system when single proteins were involved to proteomics detecting whole panels of molecules. Moreover, examined populations were diverse both in terms of the causes of CKD and the number of participants included. Therefore, direct statistical analysis of these data was not conducted, and we only presented obtained values of the examined molecules and described associations between examined markers and other CKD features (Table 1).
The review of the existing literature on protein kidney biomarkers of CKD in children brought a broad spectrum of molecules detected in the urine and blood of affected subjects. The most studied are listed and discussed in the next section. Some molecules may detect the injury of kidney structures, while others predict the progression of the disease. Furthermore, few seem to combine these two roles. There are also markers that might even become a therapeutical target. As there is a wide diversity of these proteins, we decided to divide them according to the role they play in the process of progression of CKD and add commentary concerning their usefulness in the diagnosis of CKD.
Summarized results of studies from the last five years on protein CKD markers in children are organized in Table 1.
NGAL, also known as lipocalin-2, is a 25 kDa protein produced by activated neutrophils as an innate antibacterial factor. However, NGAL seems to have more complex activities than the bacteriostatic effect. It can be found in the proximal and distant tubular cells of the kidney and, to some extent, is produced by epithelial cells of other organs, including the lungs, trachea, prostate or intestine [57]. It is freely filtered in glomeruli and, in normal conditions, undergoes complete reabsorption in the proximal tubule through the megalin, which is an endocytic receptor. Its expression is up-regulated mainly in tubules 2–4 h following acute nephrotoxic and ischaemic kidney injury; therefore, its role as an early marker was especially important in acute kidney damage [58,59,60]. Urinary (uNGAL) as well as serum (sNGAL) increase is also observed in patients with CKD. Older pediatric studies showed the usefulness of uNGAL as a predictor of the severity of lupus glomerulonephritis and hemolytic uremic syndrome [61,62]; however, most research was based on adult populations. According to Smith et al., who assessed baseline uNGAL in adult patients with CKD stage 3 or 4, it was associated with faster progression of CKD within 1 year or end-stage renal disease (ESRD) within 2 years [63]. Similar conclusions were drawn by Liu et al. [64]. There were suggestions from pre- and clinical studies that uNGAL decreases after implementing CKD treatment [65,66]. Multiple recent studies on NGAL in children with CKD of various causes mostly confirm its important role as a diagnostic tool and predictor of disease progression. NGAL seems to be one of the most promising kidney biomarkers, especially in all clinical situations related to the risk of AKI. A recent study by Greenberg et al. [26] examined urinary NGAL and other markers (IL-18, KIM-1, MCP-1, YKL-40) in a cohort of children 5 years after cardiac surgery divided into two groups (AKI vs. no AKI), revealed that uNGAL and uIL-18 were rising within 24 h post cardiopulmonary bypass with a drop in the next days. During the 5 years of follow-up, biomarkers concentrations were not significantly different in a cohort of children with AKI vs. no AKI, as well as with and without CKD or hypertension. However, children with CKD or hypertension had a higher prevalence of abnormally high uNGAL. This finding, although interpreted with caution due to the limitations of the study (relatively small number of CKD children, one cause of CKD, short period of follow-up), may indicate that NGAL might be used for screen and predicting the progression of renal disease. There are a few more studies on NGAL in CKD in children, and most of them have promising results (Table 1).
Urinary IL-18 (uIL-18) is an inflammatory cytokine produced by macrophages and proximal tubular cells as an answer to ongoing injury and subsequent inflammatory reactions involving tubules [67]. It takes part in the ischemia-reperfusion process, infections and autoimmune disorders, and similarly to uNGAL, it is a marker of an early AKI as it increases after 6–24 h post-initiating factor [68]. There are studies indicating that it could also be useful in CKD patients. In a research performed by Zubowska et al. on 85 pediatric oncology patients, IL-18 in urine was assessed to identify the group endangered with subclinical renal dysfunction [69]. After 4.6 years of observation, patients after nephrectomy and chemotherapy had significantly higher levels of uIL-18. Other studies concerning IL-18 in CKD were conducted mainly on adult populations [70,71,72]. For instance, a cohort of 153 transplant recipients showed that higher uIL-18 on the first day of post-transplantation was a predictor of faster deterioration of graft function a year post-transplant [70]. The results of the most recent research (Table 1) on uIL-18 as a predictor of CKD in children are quite conflicting and need further observations on larger and homogenous groups of subjects.
KIM-1 is a transmembrane protein produced by proximal tubules and undetectable in a healthy kidney but is released and present in plasma and urine after renal injury [73]. Moreover, its expression was confirmed in lymphocytes T to stimulate lymphocyte proliferation and production of other cytokines [73]. Obviously, KIM-1 is detected after 12–24 h post-acute kidney injury (AKI) [74,75], but in several studies in children with different chronic kidney conditions such as diabetic kidney- or steroid-resistant nephrotic syndrome (SRNS), urinary KIM-1 (uKIM-1) was also markedly increased [76,77]. Bienias et al. showed that young patients with SRNS had higher levels of uKIM-1 compared to steroid-dependent patients [76]. Similarly, in the Chronic Kidney Disease in Children (CKiD) cohort, the largest longitudinal study of children with CKD from 54 North American centers, plasma KIM-1 was higher in patients with glomerular etiology of kidney disorder than those with nonglomerular [39,51]. Furthermore, it was independently related to CKD progression in an overall cohort of CKiD (aHR 4.29; 95% CI, 2.49 to 7.38) [39,51]. UKIM-1 was also associated with an increased risk of CKD progression when comparing the highest to the lowest quartile (aHR 3.03; 95% CI, 1.92 to 4,76), and it seemed to have a similar effect with urinary epidermal growth factor (uEGF). When summarising one of the most reliable studies on KIM-1 in children with CKD, both plasma and urinary KIM-1 were associated with the progression of CKD; however, urinary KIM-1, rather than plasma KIM-1, is statistically a better predictor of CKD progression as it improved three independent clinical risk prediction c-statistics, integrated discrimination improvement and continuous net reclassification improvement [39,51]. Other results of the recent studies, in most cases, confirm the important role of KIM-1 as a marker of CKD in children (Table 1).
In humans, EGF is a protein expressed in the thick ascending limb of Henle’s loop and distal tubules [78]. It interacts with the EGF receptor and promotes the regeneration of tubular cells after injury. We can confirm that EGF is a marker of tubular health and plays a protective role in the nephron. Urinary EGF (uEGF) is a good candidate biomarker of CKD as it is highly specific for kidney tissue [78]. In the experimental study in rats with induced bilateral occlusion of the renal arteries, Norman et al. observed that treatment with exogenous EGF attenuated the rise in serum creatinine [79]. There are studies in adults indicating that adding uEGF to conventional markers of CKD (eGFR, albuminuria) improved the prediction of the progression of the disease [80]. One of the most reliable studies in children in the CKiD cohort revealed that patients with lower baseline uEGF developed progression of CKD. Children who had uEGF in the lowest quartile had a 7-fold higher risk of progression of CKD compared to patients with uEGF in the highest quartile (aHR 7.14, 95%CI, 3.45 to 20.0) [74]. These findings are consistent with newer observations on the 4C pediatric cohort, designed to assess cardiovascular comorbidity in 704 children with CKD (GFR 10–60 mL/min/1.73 m^2^) in 14 European countries. In this study, higher uEGF was independently related to decreased risk of progression of CKD [31]. It seems that uEGF, as a highly specific stimulator of tubular regeneration, may serve as a good biomarker of CKD progression; furthermore, it might become a potential target for implementing future therapies. Findings of the most recent studies on EGF in CKD in children are presented in Table 1.
α1-microglobulin is a small protein primarily produced by the liver as a free radicals scavenger, filtered by the glomerulus and reabsorbed by proximal tubular cells. It is present in urine after proximal tubular dysfunction [81]. Increased urinary levels of α1-microglobulin were found in adults with a progression of CKD and cardiovascular disease [82,83]. In the pediatric CKiD cohort, it was higher in patients with progression of CKD in comparison to those with stable kidney disease, but this was shown only in the unadjusted model [51]. This observation does not prove the significance of α1-microglobulin as a reliable biomarker of CKD in children, and further research in this field is necessary.
Tumor necrosis factor α (TNFα) is a multifunctional cytokine that plays a pivotal role in the process of inflammation. It regulates cell proliferation, differentiation and apoptotic death [84]. TNFα interacts with two membrane receptors, TNFR1 and TNFR2. TNFR1 is primarily found in glomeruli and peritubular endothelium, while TNFR2 can be transcriptionally induced in the kidney but is mostly expressed in lymphocytes. Stimulation of TNF receptors increases macrophage infiltration in the interstitium and interstitial fibrosis [84,85]. Animal models of obstructive uropathy showed that TNFR2 was increased in affected animals, and TNFR 2 deficient mice had significantly less tubulointerstitial fibrosis [86]. There are several studies in adults with different causes of CKD, including diabetic nephropathy, obstructive nephropathy, lupus or ANCA-induced nephritis, and all of them showed these receptors are predictors of the progression of CKD [86,87,88]. Serum TNFR1 was higher in children with reflux nephropathy versus healthy controls [89]. A relatively new study from Brazil, although performed on a small patient group of 34 children with CKD, showed higher blood levels of TNFR1 and TNFR2 when compared to healthy controls. Both receptors inversely correlated with eGFR (TNFR1 r = −0.853 and TNFR2 r = −0.729; p < 0.01 for both) [90]. In the recent study of Greenberg et al. on the CKiD cohort, plasma TNFR1 and TNFR2 were strongly associated with the progression of CKD [39] (Table 1).
UPAR is a membrane receptor protein expressed in podocytes, regulating their migration, adhesion and apoptosis. It is also found in endothelium and immature myeloid cells [91]. Activated uPAR is cleaved, and the soluble form, suPAR, is present in the blood [92]. First investigations on suPAR concerned FSGS. In the study by Wei and al. on 70 participants with a histopathologic diagnosis of FSGS (children and adults), a decrease in suPAR blood concentration was related to a reduction in proteinuria and achieving remission of nephrotic syndrome [93]. SuPAR was evaluated in the two cohorts of CKD in the ESCAPE study (originally designed for assessment of blood pressure control and use of angiotensin-converting-enzyme inhibitors on CKD progression in pediatric patients) and the 4C trial (The Cardiovascular Comorbidity in Children with CKD). The results showed that in a group of patients with CKD and GFR >40 mL/min/1.73 m^2^, log-transformed serum suPAR concentrations were related to a higher risk of CKD progression (HR 5.12; 95% CI, 1.56 to 16.7; p < 0 01) [94]. In the CKiD pediatric cohort, two researchers assessed suPAR blood concentration using different methods and produced distinctive results. Widemann et al. used enzyme-linked immunoabsorbent assay (ELISA) and observed in regression models with adjustment for eGFR, proteinuria or other demographic variables that patients with plasma suPAR in the highest quartile had a 33% faster progression of CKD than individuals with suPAR in the lowest quartile [40]. Greenberg et al. measured plasma suPAR with electrochemiluminescence multiplex assay with the Meso Scale Discovery platform. They showed that suPAR was a predictor of CKD progression; however, after adjustment for other variables, the significance of this relation was lost [39] (Table 1). It should be pointed out that validation and use of approved, similar tests are essential to achieve reliable and repeatable results of the concentrations of the biomarkers.
MCP-1 is a chemotactic factor for monocytes promoting their transformation into macrophages. It activates memory T lymphocytes, NK cells and basophils [95]. It is found in podocytes, endothelial and mesangial cells, and mononuclears, especially in the presence of inflammation [95]. Lupus nephritis severity correlated with urinary MCP-1 (uMCP-1) levels in pediatric patients [96]. UMCP-1 was higher in children with CKD compared to healthy controls. Furthermore, the glomerular disorder was related to even higher urinary levels of MCP-1 [97]. In the CKiD cohort, MCP-1 blood concentration was not related to CKD progression, opposite to its urinary levels [39] (Table 1). After adjustment for eGFR, hypertension and proteinuria, children with uMCP-1 in the fourth quartile had a significantly higher risk of CKD progression in comparison with those having uMCP-1 in the lowest quartile. This relation was evident in patients with autosomal recessive polycystic kidney disease (ARPKD) [51]. It is worth noticing that uMCP-1 is highly correlated with uKIM-1 (r = 0.7, p < 0.01) [51]. Other results of recent studies on MCP-1 in pediatric CKD are included in Table 1.
YKL-40 participates in tubular epithelium repair after ischemia-reperfusion injury [98]. There are not many studies assessing YKL-40 in kidney diseases in children, and those reviewed do not give promising results (Table 1).
TGF-β1 is a pro-fibrotic growth factor produced by tubular epithelial cells, fibroblasts and different inflammatory cells in the condition of persistent inflammation with irreversible tissue damage. Few studies on CKD in adults confirm increasing urinary levels of TGF-β1 with the progress of CKD [99,100]. In children with nephrotic syndrome, urinary levels of TGF-β1 were significantly higher when the etiology was FSGS compared to minimal change disease (MCD) [101]. Another pediatric study revealed higher urinary TGF-β1 in individuals with obstructive than non-obstructive uropathy [102]; however, the results of previous research do not indicate its great value as a predictor of CKD (Table 1).
BMP-7 is an antagonist of TGF-β1 and prevents inflammation and subsequent fibrosis. In the study by Musial et al., children with CKD had higher concentrations of BMP-7 when compared to healthy individuals [103]. Some trials show that BMP-7 could be used in the therapy of kidney diseases as recombinant BMP-7 infusions used in the pre-clinical phase decreased fibrosis and inflammation occurring in obstructive uropathy [104] or slowed GFR decline in mice with lupus nephropathy [105]. However, we lack studies on bigger populations of children with CKD.
Matrix metalloproteinase (proteases) family plays an important role in the degradation of the extracellular matrix. This effect is also crucial in the development and progression of CKD. MMP-2 and MMP-9 belong to gelatinases and are produced under normal conditions by the mesangial cells and tubular epithelial cells in the human kidney but at low levels [68]. However, in the process of renal fibrosis, their upregulation is very rapid due to the interactions of multiple signaling pathways and further abnormal activation [106]. MMP-2 and 9 interact with TNFs and MCP-1 in the development of CKD. MMP-2 and 9 can be detected both in blood and urine. CKD children had significantly higher plasma MMP-9 concentrations than healthy participants [107]. Similarly, pediatric patients with FSGS showed higher urinary MMP-9 in comparison with children with MCD or the control group [108]. In the recent study by Stabouli et al., MMP-2 rather than MMP-9 correlated with the progression of CKD in children (Table 1) [102].
PIIINP, as a byproduct of collagen 3 deposition, is another marker of fibrosis. In a study on children with ureter obstruction, urinary PIIINP was related to worsening kidney blockage [109]. It also correlated with more intense interstitial fibrosis in the biopsies of the kidney with CKD [109] and was a marker of CKD progression in renal transplant recipients [110]. The study by Taranta-Janusz et al. showed that children with solitary functioning kidneys (SFK) had significantly higher urinary PIIINP than healthy controls [33].
Several studies in our research included markers that do not fit the preceding categories. Fibroblast growth factor 23 (FGF23) is somewhere between the “old-fashioned” kidney function indicators and newer markers. FGF23 is a phosphaturic factor as it inhibits both phosphate reabsorption in the kidney and vitamin D activation. It is crucial for the development of secondary parathyroidism in the course of CKD [111] and rises in the early stage of CKD [112]. In the CKiD pediatric cohort, Portale et al. revealed that FGF23 was associated with CKD progression, even after adjustment for the age of the patients, mineral metabolites and other CKD-specific factors [112]. Another protein that might be a potential marker of kidney function is liver-type fatty acid-binding protein (L-FABP). It is a cytoplasmic protein existing mainly in adipose tissues and is expressed not only in the liver or pancreas but also in the kidney [113]. L-FABP presented in the proximal tubules can likely be used for the detection and monitoring of AKI or CKD [114]. Assessing urinary L-FABP seems to be less susceptible to being influenced by other conditions such as heart failure or metabolic syndrome. In the recent study by Lipiec et al., children after HUS had significantly higher urinary and serum L-FABP values when compared to healthy controls (results are shown in Table 1) [29].
The next protein that should be mentioned is retinol-binding protein (RBP), a transporter molecule for vitamins synthesized in the liver [115]. RBP is filtered by the glomerulus, reabsorbed and degraded in proximal tubules; therefore, urinary RBP can be used for monitoring renal tubular dysfunction [68]. Together with vitamin D-binding protein (VDBP) and heat shock proteins (hsp27 and HSF1) are markers of tubular function and apoptosis that accompany CKD from the very beginning. In the study by Musiał et al. [25], urinary fractional excretion (FE) of RBP4, VDBP, hsp27 and HSF1 was assessed in 70 children with CKD stages 1–5 treated conservatively and 12 age-matched healthy peers. It appeared that FE values of all parameters exceeded 1% in CKD st. 2 and raised significantly versus the control in stage 2, it was RBP4 and HSF1; in stage 3, it was VDBP; and in stage 4, it was Hsp27 (Table 1) [25].
A few other molecules were investigated as potential kidney biomarkers of CKD in clusterin, osteopontin, N-Acetyl-β-d-amino Glycosidase (NAG), symmetric and asymmetric dimethylarginine (SDMA, ADMA), or Trefoil Factor3 (TFF3) and results of recent studies were presented in Table 1. None of them has an unequivocal position of a useful diagnostic indicator, and further research on this field is necessary.
In this review, we concentrated on protein markers that may serve as a diagnostic tool for CKD. Nevertheless, it is worth noticing that also metabolomic profiling may bring us the discovery of potential metabolite associations with CKD in children. In the study of Denburg et al. [54], the authors evaluated metabolomics quantification of plasma samples from 645 children (median age 12 years) with chronic kidney disease (CKiD cohort participants). Metabolites were standardized and logarithmically transformed. The main issue was examining the association between 825 nondrug metabolites and progression to the composite outcome of KRT or 50% decline of eGFR adjusting for age, sex, race, body mass index, hypertension, glomerular vs. nonglomerular diagnosis, proteinuria, and baseline eGFR. The median time of follow-up was 4.8 years. It was shown that among participants with baseline eGFR ≥ 60 mL/min/1.73 m^2^, two-fold higher levels of seven metabolites were significantly associated with higher hazards of KRT/halving of eGFR three of purine or pyrimidine metabolism (N6-carbamoylthreonyladenosine, dihydrouridine, pseudouridine), two amino acids (C-glycosyltryptophan, lanthionine), the tricarboxylic acid cycle intermediate 2-methylcitrate/homocitrate and gulonate. Among those with baseline eGFR < 60 mL/min/1.73 m^2^, a higher level of tetrahydrocortisol sulfate was associated with a lower risk of progression.
In this review, we provided insight into the most recent research on protein kidney markers with the potential for becoming a useful diagnostic indicator of CKD or even a target for future therapies. Despite there being so many studies in this field, we still do not have consistent conclusions about which molecules have the best utility as markers of CKD. One of the most promising seems to be NGAL; however, not when it is alone. Combination with EGF, which is highly specific for kidney tissue, might improve the identification of the risk of CKD progression. Nevertheless, “old type” markers such as creatinine, albuminuria or proteinuria, despite their limitations, are still used in the pediatric CKD population, and we cannot underestimate their importance in clinical practice [116]. There is a great need for molecules that react early in kidney deterioration and are cheap, easy to assess and independent from other factors. It would be perfect to use them together with those “old and tried” markers. Nevertheless, more organized research on larger pediatric populations with CKD and with the use of validated and uniform methodology should be conducted.