Urinary epidermal growth factor as a marker for lupus nephritis: clinical, laboratory, and histopathological study

Lupus nephritis can be seen in up to 60% of all SLE patients with 10–15% of nephritis patients progressing to end-stage renal disease; late diagnosis of lupus nephritis is correlated with a higher frequency of renal insufficiency. The study aim is determination of the value of urinary human epidermal growth factor (urinary EGF) as an early biomarker of lupus nephritis in SLE patients and its relevance to disease activity and renal histopathology. The study included 58 SLE patients and 30 healthy controls; a significant difference was noticed between SLE and controls in urinary protein, creatinine, protein/creatinine ratio, and urinary EGF. The mean level of urinary EGF was less in classes IV and V renal nephritis than in classes I, II, and III. There is a significant difference in urinary EGF (33±29, 27±16, P = 0.04) between class II and class III lupus nephritis, with no significant differences in urinary protein, creatinine, protein/creatinine ratio, and SLEDAI. On the other hand, the comparison between classes II and IV showed significant difference not only in urinary EGF (33±29, 11.7±4.9 m, P=0.003), but also in SLEDAI (37.4±8, 70.5±27, P= 0.007), and protein/creatinine ratio (0.98±0.62, 3±1.8, P=0.006). This study raises the attention to test the sensitivity of urinary EGF in detecting the early and the subsequent changes in renal pathology of SLE patients as an easy, non-invasive, accurate, cheap marker that could help in following up the nephritis progression and adjusting the plan of treatment; also, it can be used to guide the time of biopsy or as an alternative in cases where renal biopsy is contraindicated.

In a longitudinal study by Moroni et al [5], anti-dsDNA, anti-C1q, C3, and C4 all had poor positive predictive values (ranging from 28 to 38%). Although the best multivariate analysis model for renal flare prediction was obtained by combining anti-C1q with C3 and C4, their data clearly showed that anti-C1q antibodies were less reliable in predicting flares in non-proliferative nephritis and flares in the presence of anti-phospholipid antibodies. Furthermore, none of these traditional markers has been shown to possess the ability to predict histology. Clearly, the lack of specificity of our current markers for lupus nephritis and inability to predict histology highlight the pressing need for a true biomarker for lupus nephritis.
Abdou et al. [6] assessed the levels of urinary LXA4 in SLE patients and showed that the urinary LXA4/creatinine ratio levels were significantly lower in cardiovascular and neuropsychiatric manifestations and non-significantly lower in patients with nephritis.
Studies which demonstrated the potential use of urinary biomarkers of LN activity showed correlation with disease activity, renal flare, and histological damage and may help in monitoring the response to immunosuppressive treatment; however, studies about biomarkers in LN still involve relatively few cohorts. The urinary biomarkers are still not superior to renal biopsy, which remains the gold standard to determine LN activity and chronicity [7].
Human epidermal growth factor (urinary EGF), a 6000 molecular weight polypeptide, was first isolated by Cohen and Carpenter in 1975 [8]. EGF is a growth factor that stimulates cell growth, proliferation, and differentiation by binding to its receptor EGFR, some studies found that urinary EGF has a role in the development of body organs such as the brain, lungs, blood vessels, and kidneys [9].
Urinary EGF is locally produced in several tissues, such as Henle's loop and the distal convoluted tubule in the kidney, salivary glands, and duodenum [10]. In the kidney, urinary EGF is involved in the repairing process of renal tissues [11].
High concentrations of urinary EGF can be found in the urine. Based on in vitro experiments, it has been previously suggested that urinary EGF originates from the ultrafiltrate. However, in vivo, it was shown in rats and in humans that the urinary EGF is mainly produced in the kidney itself. Therefore, it is generally accepted that the urinary EGF excretion reflects the renal EGF production [11][12][13]. Reduced concentrations of urinary EGF in the urine have been previously observed in diabetes nephropathy, IgA nephropathy, adult polycystic kidney disease, and children with chronic renal failure [14,15]. Also, the possibility that urinary EGF might serve as a surrogate marker for functional regeneration of the renal tubules, reflecting their ability to respond to future acute or chronic injury, was recently put forward [16].
This study is conducted to determine the value of urinary human epidermal growth factor (urinary EGF) as an early biomarker of lupus nephritis in SLE patients and its relevance to disease activity and renal histopathology progression.

Methods
A cross-sectional observational study included 58 patients diagnosed with SLE, fulfilling the 2012 Systemic Lupus International Collaborating Clinics (SLICC) classification criteria of SLE [17], and have signs of renal involvement (hematuria, urinary cast, proteinuria, or histopathologic picture of nephritis); these patients have been admitted to the Department of Rheumatology and Rehabilitation that was during the period between August 2017 and December 2019. The protocol of the study was approved by the local Ethics Committee with number IBR≠S20-135 and conforms to the guidelines of the Declaration of Helsinki.

Patient and public involvement
The researchers explained the study protocol and aim of the work to all the participants, and patients' consent was signed.
Drug-induced lupus, discoid lupus without systemic manifestations, and diabetic patients were excluded. The following data were collected: full medical history; general examination; cardiovascular, chest, abdominal, neurological, and locomotor system examination; age at disease onset (defined at the time of onset of symptoms attributed to SLE); the duration of the disease (defined as the time from disease onset until the date of visit); the clinical features of SLE; routine laboratory; and autoimmune tests for each patient had been done.
For disease activity assessment, we used the SLEDAI index [18].
-Thirty healthy controls matched for age and sex were recruited from the officers in the hospital and volunteers.

Laboratory renal investigations
The patients and healthy volunteers were instructed to collect a spot midstream urine sample in a clean sterile container. For SLE patients, the sample was collected at the day of renal biopsy.
1. Spot urine protein/creatinine ratio (u-P/C ratio): both urinary protein and creatinine concentrations were measured by turbidimetric assay and kinetic colorimetric Jaffé method, respectively, using Cobas c311 Chemistry Analyzer System (Roche Diagnostics, GmbH, Mannheim, Germany). u-P/C ratio in spot urine samples was calculated by dividing the urinary protein concentration in mg/dL by urine creatinine concentration in mg/dL. u-P/C ratio of less than 0.2 mg/mg was considered within normal limits, whereas a ratio in excess of 3.5 was considered as "nephrotic-range" proteinuria [19]. 2. Detection of urinary human epidermal growth factor (EGF): after collection of urine sample in a sterile container, it was centrifuged at the speed of 2000-3000 rpm for 20 min. Supernatant was removed; if precipitation appeared, the urine sample was centrifuged again. The supernatants were collected and were divided into aliquots and stored at -80°C. Urinary EGF was measured in urine samples using an enzyme-linked immunosorbent assay (

Statistical analysis
Data was analyzed by SPSS version 20.0 statistical package; data were presented as number and percent, mean ±SD, or median and range as appropriate. Student's ttest and multivariate analysis were used for comparing means between different groups; Pearson correlation coefficient (r) was used to test the association between quantitative variables. A P-value less than 0.05 was considered statistically significant.

Results
The patients' mean age is 47.2±14.6 years, with 53 females (91.4%), and the mean disease duration in years is 5.4±4.2; SLEDAI mean ±SD was 12.3±9.4, with minimum of 2 and maximum of 22; medications were 100% hydroxychloroquine, 69% mycophenolate mofetil, and 17.2% azathioprine and 86% corticosteroid; the main data of the patients is displayed in (Table 1). The comparison between the patients and the control group was significant in urine protein, urine creatinine, protein/creatinine ratio, and in urine EGF level (Table 2).
We found no significant differences in urinary EGF level between speckled and homogenous ANA pattern and positive and negative anti-dsDNA.
The is no significant correlation of urinary EGF level with urinary P/C ratio, or with any SLE disease parameters and renal laboratory tests except for ESR where there was a negative significant correlation (r= − 0.72, P=0.002).
With classification of renal biopsy according to the International Society of Nephrology/Renal Pathology Society 2003 classification of lupus nephritis [20], we noticed that the mean level of urinary EGF was less in classes IV and V than in classes I, II, and III ( Table 3 and histopathological picture for sample of patients in Fig. 1).
Due to the small number of patients in classes I and V nephritis, we started the comparison with class II.
We further tested the relation of renal biopsy with urinary EGF, protein/creatinine ratio, and SLEDAI through the linear regression analysis which showed only significance with the urinary EGF level (Table 4 and supplementary tables (4a-4b-4c)).
ROC curve has been created to test urinary EGF level as a predictor of lupus nephritis and showed cutoff value at < 40.6 ( Fig. 3).

Discussion
Renal biopsy with histological study of kidney tissue is an esteemed tool for diagnostic classification and prognostication in lupus nephritis patients, but we cannot deny the accompanied significant morbidity with the procedure of renal biopsy that why it is not usually performed serially. Furthermore, with an essentially "blind" needle biopsy, there can be a question of how representative is the limited number of glomeruli usually obtained of kidney activity and chronicity [1].
We seriously need a noninvasive, easily obtainable, and accurate marker that can be followed serially in monitoring lupus patients. Pathologic studies provide limited information because patients are not biopsied frequently, and clinical measures provide limited information since they do not reflect intrarenal injury very well. In the previous studies, many laboratory markers have been used, which include serological determination of serum anti-double-stranded (ds)DNA antibodies and complement levels, and those can be helpful clinically, but the correlation between them and lupus renal disease is lacking. Sensitivity and specificity for active lupus nephritis among all SLE patients are different with different studies and tests used [5,21,22].
Our study included 58 SLE adult patients, 53 (91%) of them are females, the mean disease duration in years is 5.4± 4.2, and the patients showed significant differences with the healthy control in the renal lab tests (urine protein-protein/ creatinine ratio and urinary EGF level) which matches with the results of many other studies [11,13,18,23].
Interestingly, we did not find significant correlations between urinary EGF level and urinary P/C ratio, despite that urinary protein is known to be generally a simple marker for detecting renal glomerular disease activity, and that could be explained by the earliest change in the level of urinary EGF than p/c ratio, similar observations were present in studies that tested the urinary chemokines' correlation with the u-P/C ratio [24,25].
Other clinical parameters such as u-P/C ratio and urinary EGF level did not show any significant correlation with the SLE disease parameters and renal laboratory tests, but the ESR showed negative significant correlation (r = − 0.72, P=0.002).
In the study of Worawichawong et al. [25], they tested the combined use of urinary biomarkers with opposing actions such as EGF and MCP-1 (monocyte chemoattractant protein-1) to offer additional information compared to either cytokine alone. Previously, in the ratio of urinary biomarkers, they found that urinary EGF/ MCP-1 ratio is independently associated with tubulointerstitial severity in primary glomerulonephritis. However, they did not address the benefit of EGF/MCP-1 ratio over EGF alone at discriminating renal histological grade when the additional costs of the MCP-1 assay is considered. By contrast, neutrophil gelatinase-associated lipocalin (NGAL) appeared to be strongly associated with proteinuria, and less useful as a biomarker of tubulointerstitial disease severity compared to EGF. They recommended for further prospective studies to support and evaluate role of EGF or EGF/MCP-1 as candidate biomarkers to guide to therapy in various types of GN.   It was essential in this study to use the most approved tool for detecting the renal affection in SLE nephritis patients, it is the histological study of renal biopsy, and that work showed significant negative correlation with only urinary EGF level (r= − 0.55, P= 0.008) and ESR (r= − 0.56, P=0.03). This significant negative correlation with urinary EGF has been translated by the noticeable differences in its level among the renal histology classes, not only this, but it gave significant differences between two close classes (II and III), while widening the interclasses comparison (classes II and IV) the significant differences between the urinary EGF levels increased, and the other parameters appeared to give significant differences (SLEDAI, u-P/C ratio); this observation raise the attention to test the sensitivity of urinary EGF level for detecting the subsequent changes in renal pathology in SLE patients and soon, can be an excellent, noninvasive, accurate, cheap marker for following up the nephritis progression and adjusting the plan of treatment, and to support our finding, linear regression analysis of renal histological biopsy with the urinary EGF, protein/creatinine ratio, and SLEDAI was done and again has supported the urinary EGF importance by being significant only with it, and ROC curve has shown a cutoff value for urinary EGF < 40.6 with a sensitivity 90.4% and specificity 83.3%.
According to our knowledge this study is one of the few studies which tested the level of urinary EGF in adult systemic lupus nephritis and discussed its relationship with disease activity index and histological study in Arab SLE patients with determining cutoff value.

Conclusion
SLE renal biopsy histopathological results were parallel to the decrease in urinary EGF level. Urinary EGF is a simple urine test that showed significant reduction at the very early stage of lupus nephritis (class I) and showed a significant correlation as well with renal nephritis changes among classes.
Urinary EGF is a noninvasive, cheap, and easy way to follow the progression of lupus nephritis and could help in the early management and monitoring of progression.
The urinary EGF owns the advantage as a LN biomarker by having a good correlation with kidney activity and damage, being useful for serial monitoring, being superior to conventional clinical or laboratory parameters, possessing the ability to assess the severity of renal involvement, being cost-effective, and being easy to perform and available in most clinical laboratories

Limitation of the study
Our study is cross-sectional, while we recommend a future longitudinal study in order to provide more information and accuracy for its course and use as a biomarker for follow-up of SLE nephritis. The study did not include SLE patients without nephritis as a comparative group, which is a step forward will be done in an extension study of this one.