The Reference Intervals for Serum C‐Terminal Agrin Fragment in Healthy Individuals and as a Biomarker for Renal Function in Kidney Transplant Recipients

Abstract

Background

C‐terminal agrin fragment (CAF) has been shown to be a promising new biomarker for kidney function. The aim of this study was to verify the reference intervals for CAF in Chinese healthy adults and to assess the efficiency of CAF for monitoring renal function after transplantation.

Methods

Serum samples were collected from 200 healthy adult subjects and 60 living donor kidney recipients before and on day 1, day 2 and at 6 months after transplantation. We measured serum CAF, creatinine, cystatin C and NGAL concentrations at each time. Estimated glomerular filtration rate (eGFR) was evaluated by Chronic Kidney Disease Epidemiology Collaboration (CKD‐EPI) equation. Reference intervals for CAF were determined at 2.5th and 97.5th percentiles.

Results

Serum CAF concentrations were observed to be higher in females of old age groups while no significant differences were discovered in males between age groups. There were significant gender‐related differences in CAF in old age groups (50–64 and ≥65 years). Serum CAF correlated positively with serum creatinine, cystatin C and negatively with eGFR on day 1, day 2 and at 6 months after kidney transplantation. CAF and NGAL fell rapidly into the normal range on the second postoperative day, prior to creatinine and cystatin C.

Conclusions

This study verified the reference intervals for serum CAF. CAF could be a potential new biomarker for kidney function monitoring.

Introduction

Kidney transplantation is the renal replacement therapy of choice for most patients with end‐stage renal disease. Although kidney transplantation may be successful, kidney function is not always restored to the normal state. Therefore, the assessment of kidney function following transplantation is important. Limitations of serum creatinine, the main biomarker to monitor kidney function, have prompted interest in using other biomarkers for measuring kidney function 1, 2. Thus, a series of novel urinary or plasma biomarkers including interleukin 18 (IL‐18), neutrophil gelatinase‐associated lipocalin (NGAL), kidney injury molecule 1 (KIM‐1), liver‐type fatty‐acid binding protein (L‐FABP), YKL‐40, clusterin and cystatin C have been evaluated 3.

Agrin, a large heparan sulfate proteoglycan (HSPG), is expressed in neuronal as well as nonneuronal tissues 4. It is best known for its role in the neuromuscular system, where a special splice form, which is important for the development and organization of neuromuscular junctions, is secreted by motor neurons 5. Literature shows that agrin has a substantial contribution to the HSPG component of the human glomerular basement membrane (GBM) 6. Neurotrypsin, known as a serine protease, cleaves agrin in two homologous sites called α and β site releasing a 22‐kDa C‐terminal fragment (CAF22) at the β site, and the cleavage at the α site generates the 110‐kDa C‐terminal fragment (CAF110) 4, 7. The NTCAF ELISA and NTtotalCAF ELISA are respectively used to detect CAF22 and totalCAF, defined as the sum of CAF22 and CAF110 8, 9. Changes in kidney function may be associated with changes in serum CAF levels in human because CAF is small enough to penetrate the filtration barrier of glomerulus and it is reported that CAF could be a fast biomarker for renal function and also a new tool for the early detection for delayed graft function (DGF) 10.

So far, there has been no report on the reference intervals for CAF among Chinese healthy adults and simultaneous measurements of serum CAF as a biomarker in renal transplant recipients. In this study, we evaluated serum levels of CAF in 200 healthy individuals and 60 patients undergoing living donor kidney transplantation before and on day 1, 2 and at 6 months after transplantation to verify the reference intervals for CAF and to assess whether CAF could be as a biomarker for monitoring kidney function recovery in kidney allograft recipients.

Materials and Methods

Study Population

Healthy studies

A total of 5,523 medical staffs attending their annual physical examination throughout July 2010 were enrolled in our study. 369 individuals who had known diabetes, 791 of them had hypertension, 37 of them had coronary heart disease, 866 cases who had known other chronic disease (chronic liver disease, chronic kidney disease, chronic bronchitis, etc.), hypothyroidism or hyperthyroidism, and dyslipidemia were all excluded from our analysis. 3,460 healthy subjects who met the study criteria were chosen through a questionnaire and normal physical examination results including chest X‐ray, type‐B ultrasound and electrocardiogram. Also subjects with albuminuria or leukocyturia were excluded and all were free of the above‐mentioned diseases. 200 healthy individuals who were randomly selected from the 3,460 healthy subjects were enrolled in this study and the sample size was in accordance with the CLSI guidelines, which recommends a minimum of 120 subjects for clinical reference range determination 11. The 200 healthy individuals were divided into four subgroups based on the age in each sex (18–34, 35–49, 50–64 and ≥65 years). The study protocol was approved by Peking University First Hospital Ethics Committee and adheres to the Declaration of Helsinki. The need for informed consent was waived by the committee since all specimens used in the present study were leftover samples.

Kidney transplant recipients

We studied 60 patients (21 women and 39 men) suffering from chronic kidney disease undergoing living donor kidney transplantation in Peking University First Hospital throughout 2012 and 2014. The study protocol was approved by Peking University First Hospital Ethics Committee and adheres to the Declaration of Helsinki. Also, the clinical and research activities follow the Principles of the Declaration of Istanbul. The need for informed consent was waived by the committee since all specimens used in the present study were leftover samples.

In general, delayed graft function (DGF, defined by the need for dialysis in the first week after transplantation), immediate graft function (IGF, defined by serum creatinine level below 3 mg/dl (265.2 μmol/l) on postoperative day 5) and slow graft function (SGF, defined by creatinine level higher than 3 mg/dl (265.2 μmol/l) on postoperative day 5 and not requiring dialysis) are used to describe graft function after transplantation 12. In addition, the fair and poor short‐term graft outcome at 6 months were characterized as eGFR ≥ 60 ml/min/1.73 m² and eGFR < 60 ml/min/1.73 m² respectively.

Sample Collection and Laboratory Analysis

Healthy individuals’ blood samples were collected. Renal transplant recipients’ blood samples were obtained on the day of surgery before transplant and on day 1, day 2 and at 6 months following transplantation. Blood samples of five recipients failed to be collected at postoperative month 6. Samples were centrifuged immediately and then sera were collected and stored at −80°C until the analysis was performed.

The analysis involved measurements of serum CAF, creatinine, cystatin C and NGAL concentrations and then eGFR was calculated by the Chronic Kidney Disease Epidemiology Collaboration (CKD‐EPI) equation 13. CAF concentrations were determined using a commercially available enzyme‐linked immunosorbent assay kit (NTCAF ELISA Kit; Neurotune, Schlieren, Switzerland) 10 according to the manufacturer’s protocol, with an intra‐assay coefficient of variation (CV) of 15.5% at 560.9 pM, 23.9% at 68.7 pM. This assay detects the 22‐kDa C‐terminal fragment generated by β site cleavage (CAF22). Serum NGAL was assayed by the ELISA method (R&D systems, Minneapolis, MN). Serum cystatin C and creatinine were measured by means of the particle‐enhanced immunoturbidimetry method and Jaffe method respectively with Hitachi 7600‐110 system (Hitachi, Tokyo, Japan). For NGAL, the intra‐assay coefficient of variation was 10.6% at 10 ng/ml. And the CV was 9.7% at 0.8 mg/l for cystatin C.

Statistical Analysis

Statistical analyses were performed using SPSS software version 19.0. Continuous data were expressed as mean ± standard deviation (SD) or median (minimum; maximum). Categorical variables were reported in absolute numbers and percentages. The distribution normality of variables was performed using the Kolmogorov–Smirnov test. Reference intervals (RIs) for serum CAF were determined at 2.5th and 97.5th percentiles. Student's t‐test and ANOVA were used for samples with normal distribution, and Mann–Whitney U and Kruskal–Wallis tests for analyses of samples with skewed distribution as appropriate. Differences between concentrations of analyzed parameters in consecutive time points of transplantation were assessed by ANOVA for repeated measurements. Logistic regression analyses were performed to assess the ability of different variables on determination of post‐transplant allograft function. Correlations between CAF and other variables were evaluated by Spearman's test and to adjust for age, partial correlation was performed. For all tests, differences were considered statistically significant at P < 0.05.

Results

Reference Intervals for Serum CAF in Healthy Adult Subjects

The mean age of the healthy subjects was 49.1 ± 17.5 years, 103 (51.5%) were females. The dispersal of CAF was found to display normal distribution while creatinine and cystatin C distribution were found to be non‐normal for the healthy individuals (n = 200). Serum CAF concentrations were 147.3 ± 58.0 pM, serum creatinine concentrations were 98.6 ± 15.7 μmol/l and cystatin C concentrations were 0.82 ± 0.21 mg/l.

Table 1 shows an overview of different biomarkers serum levels in various age‐ and sex‐related groups. As a whole group, it was observed that serum CAF values did not show statistically gender‐related differences (P = 0.155) while CAF concentrations were found to have a significant tendency to increase with age (P < 0.001), but the levels of independent subgroups, 18–34 years old vs. 35–49 years old, and 50–64 years old vs. ≥65 years old showed no significant difference (P > 0.05). As shown in Fig. 1A, serum CAF levels did not show statistically differences in males between age groups (P = 0.532) but were observed to be higher in females aged 50–64 and ≥65 years (P < 0.001). And significant gender‐related differences (P < 0.05) were discovered among two groups (50–64 and ≥65 years). These data confirm that the RIs (2.5th, 97.5th) for serum CAF, which could be taken into account are as follows: 45.5–314.0 pM for males and 33.7–209.5 pM, 72.4–306.1 pM for females aged 18–49 and ≥50 years respectively.

Table 1.

Serum Levels of Biomarkers Among Healthy Subjects of Different Age and Sex Groups

Age group	Gender	Number	CAF (pM)	Scr (μmol/l)	Cys C (mg/l)
18–34 years	Male	25	131.2 ± 71.6	110.6 ± 11.8	0.78 ± 0.10
	Female	29	120.3 ± 56.1	89.3 ± 7.0	0.70 ± 0.09
	Both	54	125.3 ± 63.3	99.2 ± 14.3	0.74 ± 0.10
35–49 years	Male	25	138.6 ± 34.8	104.8 ± 13.7	0.77 ± 0.08
	Female	31	117.5 ± 44.6	82.7 ± 9.8	0.65 ± 0.10
	Both	56	126.9 ± 41.5	92.6 ± 16.0	0.70 ± 0.11
50–64 years	Male	25	149.1 ± 49.8	108.6 ± 13.3	0.91 ± 0.22
	Female	19	202.6 ± 42.9	94.7 ± 11.3	0.82 ± 0.15
	Both	44	172.2 ± 53.6	102.6 ± 14.2	0.87 ± 0.20
≥65 years	Male	22	154.1 ± 47.3	112.0 ± 12.6	1.09 ± 0.32
	Female	24	192.3 ± 53.3	91.8 ± 13.7	0.91 ± 0.14
	Both	46	174.1 ± 53.6	101.5 ± 16.5	1.00 ± 0.26

Values are expressed as mean ± SD.

Figure 1.

Serum levels of CAF (A), creatinine (B) and cystatin C (C) in healthy males and females of different age groups. Values are presented as the mean and standard deviation. *Significant difference between males and females (P < 0.05), **significant difference vs. 18–34 and 35–49 years respectively (P < 0.05), ***significant difference between age groups (P < 0.05).

In this study, the predominant characteristics of creatinine and cystatin C levels were that they both demonstrated statistically gender‐related differences whenever as a whole or in different age groups (P < 0.05). Furthermore, cystatin C was found to display higher values in old age groups (50–64 and ≥65 years) in both females and males (P < 0.001) (Fig. 1C). Besides, Serum CAF was discovered to be correlated positively with cystatin C (r = 0.39, P < 0.001) while no significant correlation was observed between CAF and creatinine (P = 0.308).

Comparison of Serum CAF with Creatinine, eGFR, Cystatin C and NGAL in Monitoring Renal Function in Renal Transplant Recipients

Table 2 demonstrated the basic characteristics of kidney transplant recipients and the corresponding donors. Conclusively, 56 (93.3%) recipients had IGF, while 4 (6.7%) recipients had SGF after transplantation. And one patient died on postoperative day 45 with the advent of multiple organ failure (MOF) and another patient was lost to follow‐up just at the first month following transplantation. The median CAF concentrations before transplantation in kidney transplant recipients were 6.3‐fold higher than those in healthy adult subjects (921.7 (618.1, 1508.8) vs. 146.4 (31.9, 326.5) pM).

Table 2.

Baseline Characteristics of Kidney Transplant Recipients and Donors (N = 60)

Parameter	Details
Recipients
Age (years)	26 (18, 47)
Gender (M/F)	60 (39/21)
BMI (kg/m2)	21.4 (13.4, 35.0)
eGFR pre‐transplantation (ml/min/173 m2)	5.8 (2.6, 11.1)
CAF (pM)	921.7 (618.1, 1508.8)
Serum creatinine (μmol/l)	845.5 (476.0, 1856.0)
Cystatin C (mg/l)	5.49 (1.00, 12.00)
Types of dialysis
Hemodialysis	46 (76.6)
Peritoneal dialysis	10 (16.7)
No dialysis	4 (6.7)
Months on dialysis	7 (0, 60)
Donor
Age (years)	51.5 ± 7.1
Gender (M/F)	60 (17/43)
BMI (kg/m2)	24.0 (19.0, 38.0)
eGFR (ml/min/173 m2)	101.5 (51.0, 127.0)
Serum creatinine (μmol/l)	61.5 (42.0, 115.0)
Hypertension (n)	6
Diabetes	None

Continuous values are expressed as mean ± SD or median (min, max) and categorical values are shown by n (%); BMI, body mass index; eGFR, estimated glomerular filtration rate; Tx, transplantation.

As is shown in Table 3 and Figure 2, we can observe a significant decrease in serum CAF as early as day 1 after kidney transplantation, the same as creatinine, cystatin C and NGAL (P < 0.001). The percent change in CAF concentrations compared to creatinine and cystatin C concentrations did not show any significant difference from pre‐transplant to day 1 following transplantation (P > 0.05), while serum NGAL concentrations decreased significantly faster than CAF concentrations from pre‐transplant to the first postoperative day (76.6% vs. 67.4% P = 0.008). Moreover, CAF levels decreased significantly faster from pre‐transplant to day 2 after transplantation than creatinine levels (82.5% vs. 76.6%, P = 0.004) and cystatin C levels (82.5% vs. 67.5%, P < 0.001) but showed no statistical difference compared with NGAL levels (82.5% vs. 83.9%, P > 0.05). Besides, serum CAF along with NGAL concentrations fell into the normal range on postoperative day 2, while serum creatinine levels were still above the upper limit of the normal range (133 μmol/l) (Table 3). Furthermore, cystatin C concentrations on day 2 were comparable to that of day 1 following transplantation, which still stayed above the upper limit of the normal range (1.03 mg/l) even at 6 months after transplantation.

Table 3.

Time Course Changes in Serum CAF, Creatinine, eGFR (CKD‐EPI), Cystatin C and NGAL in Patients Undergoing Kidney Transplantation

	Before Tx	1 day after	2 days after	6 months after
Serum CAF (pM)	921.7 (618.1, 1508.8)	360.4 (85.9, 1291.3)a	164.1 (6.8, 977.3)a	164.8 (74.3, 338.0)a
Creatinine (μmol/l)	845.5 (476.0, 1856.0)	365.0 (115.0, 1254.0)a	204.5 (80.0, 1275.0)a	144.0 (67.0, 320.0)a
eGFR (ml/min/173 m2)	5.8 (2.6, 11.1)	17.4 (4.2, 58.2)a	35.2 (4.2,8 8.3)a	52.6 (20.1, 121.7)a
Cystatin C (mg/l)	5.49 (1.00, 12.00)	2.00 (1.00, 4.18)a	2.00 (1.00, 4.47)a	1.42 (0.77, 3.60)a
NGAL (ng/ml)	911.0 (305.3, 1783.2)	201.1 (71.0, 654.1)a	158.9 (52.5, 994.4)a	93.1 (11.9, 186.5)a

Values are presented as median (min, max); CAF, C‐terminal agrin fragment; Tx, transplantation; eGFR, estimated glomerular filtration rate; NGAL, neutrophil gelatinase‐associated lipocalin.

^a P < 0.001 vs. baseline.

Figure 2.

Changes of serum CAF, creatinine, cystatin C and NGAL over the period of the first 2 days and month 6 after transplantation compared to pre‐transplant levels. The box represents the lower quartile, the median and the upper quartile. The length of the box corresponds to the interquartile range (IQR).

Serum CAF did not show any significant correlation with creatinine, eGFR or cystatin C before transplantation (P > 0.05) while on day 1, day 2 and at 6 months following kidney transplantation, CAF related positively to creatinine (r = 0.43, P = 0.001; r = 0.69, P < 0.001; r = 0.30, P = 0.037), cystatin C (r = 0.64, P < 0.001; r = 0.76, P < 0.001; r = 0.29, P = 0.039) and negatively to eGFR (r = −0.40, P = 0.002; r = −0.62, P < 0.001; r = −0.35, P = 0.014) adjusted for age separately. Contrarily, no significant relations were found between serum CAF levels and NGAL concentrations both before and after renal transplantation in our study (P > 0.05).

In addition, 38 transplant recipients with poor 6‐month graft function (eGFR < 60 ml/min/1.73 m²) exhibited lower donor eGFR (P = 0.019) in comparison to 17 transplant recipients with fair 6‐month graft function (eGFR ≥ 60 ml/min/1.73 m²) in multivariate logistic regression analysis.

Discussion

This is the first study to investigate the reference intervals for CAF in Chinese healthy adult subjects and simultaneously evaluating whether serum CAF could serve as a potential new biomarker for monitoring kidney function after renal transplantation.

In this study, serum CAF concentrations in healthy adult subjects suggested a tendency to increase with age and showed higher mean values in the old groups (50–64 and ≥65 years). Recent studies reported elevated CAF levels in sarcopenic patients and CAF may serve as a potential biomarker for identifying sarcopenia especially in older adults 14, 15, 16. In the circumstance of our study, the higher values of serum CAF in old age groups may be due to the influence of muscle mass caused by degradation of the neuromuscular junction. However, significant differences were discovered only in females’ CAF levels of different age groups and the possible explanation for the case is that older females could be more inclined to suffer osteoporosis or fracture and this could contribute to the elevation of serum CAF levels. The significant gender‐related difference in old groups, probably caused by age, may reflect different hormonal or physical states in elderly females and males. Besides, serum CAF concentration in healthy adult subjects obtained in this study was in disagreement with the discovery from Steubl et al. 10 (147.3 ± 58.0 vs. 56.0 ± 18.5 pM), which raises the possibility that race and age may be the possible influencing factors on serum CAF levels. In addition, Drey et al. highlighted that vitamin D supplementation and physical exercise were associated with a reduction in CAF concentrations 15 while another report from Fragala et al. 17 elucidated that circulating CAF increased by 10.4% in older adults after 6 weeks of resistance exercise training and were significantly related to changes in muscle cross‐sectional area of vastus lateralis (r = 0.544, P = 0.008). Thus, in future researches into reference intervals for serum CAF, larger population, more detailed age stratification and also various factors such as nutrition supplementation, physical states, muscle mass, or hormonal levels should be taken into consideration for further verification.

In addition, this study involved 60 recipients undergoing living donor kidney transplantation. First, at each time point after transplantation, we found that serum CAF was significantly correlated positively to serum creatinine, cystatin C and negatively to eGFR but not related to serum NGAL, which should be considered as a potential early marker of kidney impairment 3. Meanwhile, Goldberg et al. 18 reported that mice lacking agrin exhibited normal glomerular structure and normal renal function including glomerular function. Moreover, recent studies 19, 20 have demonstrated that CAF was likely cleared from circulation by glomerular filtration and subsequent endocytosis in the proximal tubule in their animal experiments. Thus, we could preliminarily infer that changes in serum CAF concentrations could most likely to be explained by extra‐renal source and decreased renal clearance. However, no significant correlations were found between serum CAF and the above mentioned three biomarkers before renal transplantation, which was slightly inconsistent with the existing studies 10, 21. Data showed that as high as 50% of patients with chronic kidney disease (CKD) would have sustained a fracture prior to initiating dialysis 22. And Jamal et al. 23 also reported that impaired neuromuscular function was related to fracture in hemodialysis patients. Therefore, the possible explanations for this finding were that neuromuscular injury in patients with end‐stage renal disease and decreased renal clearance may make conjunct effects on the elevation of serum CAF concentration before transplantation. Furthermore, serum CAF in healthy subjects was found to be correlated positively with cystatin C but not with creatinine while after transplantation the serum CAF correlated with both creatinine and cystatin C in those recipients. A possible explanation for this discrepancy was that serum creatinine may display relatively higher intra‐individual variation (5.95% vs. 5.0%) and inter‐individual variation (14.7% vs. 13.0%) compared to serum cystatin C due to the impact of a number of additional factors 24. And to date, no studies upon the correlations between serum CAF and other conventional biomarkers among healthy population have been reported. Conclusively, a larger sample size should be needed explicitly in future researches.

Second, we observed that serum CAF concentrations decreased significantly as early as 1 day following renal transplantation, which was analogous to serum creatinine, cystatin C and NGAL concentrations. In addition, serum CAF, the same as NGAL, decreased faster and finally fell into the normal range on the second postoperative day, prior to a fall in serum creatinine and cystatin C, which demonstrated that serum CAF may be an early potential biomarker monitoring post‐transplant renal function among kidney transplantation recipients in comparison to creatinine and cystatin C. Nevertheless, we found that serum CAF levels before and on day 1, 2 after transplantation were not associated with the fair or poor renal function at 6 months after transplantation in the present study (data not shown). Interestingly, Steubl et al. 25 highlighted that early postoperative serum CAF appeared to be a useful tool for assessment of long‐term outcomes in renal transplant recipients, with a median follow‐up time of 3.1 years. Hence, further investigation in larger population especially with longer follow‐up time will be needed to verify serum CAF’ utility in monitoring allograft function in renal transplant recipients. Besides, the NTCAF ELISA Kit used for our study may merit further evaluation with respect to its performance in future researches.

Third, in our study, cystatin C was observed to still stay above the upper limit of the normal range at 6 months after renal transplantation. It is acknowledged that serum cystatin C gets some advantages over creatinine when estimating the glomerular filtration function in recent publications. Our previous study 26 has shown that after follow‐up of 9 months, serum cystatin C concentrations were still stayed above the upper limit of the normal range. Furthermore, Knight et al. 27 demonstrated that some factors such as inflammation, use of immunosuppressant may be correlated with the high levels of cystatin C after transplantation, which may support our discovery that the continuous high concentrations of serum cystatin C after transplantation.

Moreover, reports upon serum CAF as a potential biomarker for renal function are much fewer than those on biomarkers such as creatinine, cystatin C, NGAL and several other markers. Steubl et al. first published their findings that CAF could serve as a new biomarker for kidney function and also a new tool for the early detection of delayed graft function in renal transplant recipients 10 and subsequently they reported that CAF appears to be a promising biomarker for the early postoperative risk assessment of long‐term allograft outcome 25. Furthermore, serum CAF concentration was observed not to be influenced by conventional high‐flux hemodialysis using Fx60 membrane, which displayed its potential advantage of dialysis‐independence over conventional parameters as a promising biomarker for evaluation of kidney function 28, 29. In addition, another investigation showed that serum CAF, possibly prior to conventional biomarkers, provided a robust serum biomarker for residual renal function in peritoneal dialysis patients undergoing automated peritoneal dialysis 21, which also exhibited the potential probability of CAF for evaluation of renal function.

However, our study does have certain limitations. First, due to the high cost of ELISA kit, only a relatively small number of healthy adult subjects were enrolled in this study and also the effects of more factors such as muscle mass or hormonal levels on serum CAF are not known. Second, compared with the previous publication 9, the complicated operations of this kit based on its protocols result in relatively high intra‐assay coefficient variations through our continuous improvement and therefore more efforts will be definitely needed for the verification of the properties in future studies. Furthermore, long‐term follow‐up of the 60 cases of living donor kidney transplant recipients is still ongoing.