Serum and Urinary Interferon‐Gamma‐Inducible Protein 10 in Lupus Nephritis

Abstract

Background

Lupus nephritis is a serious manifestation of systemic lupus erythematosus (SLE). The objective of this study was to identify the sensitivity, specificity, and cut‐off values of IP‐10 in the serum and urine of patients with lupus nephritis compared to renal biopsy, albumin/creatinine ratio, and serum anti‐dsDNA.

Methods

Thirty female SLE patients were included. SLEDAI was calculated and blood and urine samples were collected. Patients were divided into 10 SLE patients with renal involvement (six active and four inactive), 10 active SLE, and 10 inactive SLE patients. Ten age‐matched healthy (control) were included. Serum and urinary levels of IP‐10 were measured by ELISA. Anti‐dsDNA, urine albumin/creatinine ratio were performed.

Results

Serum and urinary IP‐10 in active SLE patients had significantly increased levels as compared to inactive SLE patients (P = 0.015, P = 0.033, respectively). However, there was no difference in serum and urinary levels between active renal and active non‐renal patients. Albumin/creatinine ratio is a better marker in differentiating between lupus nephritis and SLE with no renal involvement. Any of serum and urinary IP‐10, albumin/creatinine ratio, and anti‐dsDNA did not correlate with the class of lupus nephritis in renal biopsy.

Conclusion

Urinary and serum IP‐10 are useful markers of lupus activity, but not indicative of renal activity. Albumin/creatinine ratio is superior in identifying lupus nephritis and renal activity.

Renal disease is one of the commonest and most devastating complications of systemic lupus erythematosus (SLE) 1. Despite improvement in the medical care of SLE, the prognosis of lupus nephritis remains poor 2. Up to 25% of patients still develop end‐stage renal failure 10 years after onset of renal disease 3.

Conventional laboratory markers for follow‐up of kidney disease in SLE patients such as, urine protein‐to‐creatinine ratio, proteinuria, creatinine clearance, complement, and anti‐dsDNA levels are disappointing. They are neither sensitive nor specific for differentiating renal activity and damage in lupus nephritis. Kidney biopsy is the gold standard for providing information on the histological classes of lupus nephritis and the relative degree of activity and chronicity in the glomeruli 4. However, it is invasive and not always feasible. Moreover, it may not be representative, as only a limited number of glomeruli are sampled. Thus, laboratory biomarkers are necessary to enhance the diagnostic accuracy and sensitivity of lupus nephritis, monitoring of treatment response, and early detection of renal flares 5.

IFN‐γ ‐inducible protein 10 (IP‐10), also known as CXCL10 (chemokine [C‐X‐C motif] ligand 10), is a chemokine secreted by IFN‐γ‐stimulated endothelial cells, fibroblasts, and monocytes. CXCL10 exerts its function through binding to chemokine [C–X–C motif] receptor 3 [CXCR3], IP‐10 promotes migration of T cells to sites of inflammation and is known to play a role in the down‐regulation of angiogenesis 6. Determination of high level of CXCL10 in peripheral fluids is therefore a marker of host immune response, especially T helper 1 (Th1) orientated T cells. In tissues, recruited Th1 lymphocytes may be responsible for enhanced IFN‐γ and tumor necrosis factor‐α production, which in turn stimulates CXCL10 secretion from a variety of cells 7. Blood and urine IP‐10 was demonstrated as potential inflammatory marker in tuberculosis (TB) 8. Cannas and colleagues found that urinary IP‐10 has been detected in patients with lung diseases. Moreover, its level may decline after completion of TB therapy 9.

There is increasing evidence that CXCL10 levels are elevated in the sera and/or the tissues of SLE patients, and that CXCL10 and infiltrating CXCR3‐positive cells may participate in the pathogenesis of a variety of the manifestations of SLE 10, 11. Therefore, we determined the level of CXCL10 in the serum and urine of SLE patients and lupus nephritis patients, and correlating these levels with renal biopsy findings.

This study included 30 female patients with SLE (fulfilling American College of Rheumatology (ACR) 1997 criteria) 12 who were recruited from the outpatient and inpatient services of rheumatology and nephrology departments at the Suez Canal University Teaching Hospitals in Ismailia during the period from May to November 2014. Ten age‐ and gender‐matched healthy controls were also recruited. The faculty of Medicine ethics committee approved the study and subjects' written consent was obtained according to the Declaration of Helsinki.

Systemic lupus erythematosus disease activity index [SLEDAI] was calculated 13. Patients were divided into active and inactive lupus based on SLEDAI. Then, patients were divided into three groups (ten subjects per group):

• Group 1: Lupus nephritis patients (six active and four inactive; kidney biopsies were performed in seven). According to International Society of Nephrology/Renal Pathology Society 2003 Classification of Lupus Nephritis 14, the histopathology result was class II in two, class IV in two, class V in two and class VI in one. Active lupus was defined as SLEDAI ≥4. Active renal involvement showed proteinuria ≥500 mg/day or active sediment in urine [Active sediments were defined as presence of hematuria (>5 RBCs/HPF), pyuria (>5 WBCs/HPF), or casts (heme, granular, RBCs)].

• Group 2: Patients with active SLE without renal complications.

• Group 3: Inactive SLE patients.

• Group 4: age‐ and gender‐matched healthy control.

About 4 ml blood was collected from each patient. For each SLE patient, complete blood count (CBC), serum urea, serum creatinine, anti‐double‐stranded DNA (dsDNA) antibody titers, and complements (C3, C4) were analyzed by standard procedures. Serum quantitative anti‐dsDNA was determined using QUANTA Lite dsDNA ELISA (INOVA Diagnostics, inc. San Diago, CA, USA).

Urine was collected as a second catch specimen. Urinalysis and urinary albumin/creatinine ratio were performed. Then, 1 ml was aliquoted and stored at −80°C for IP‐10 assessment.

IP‐10 level was determined in the urine and serum samples using the solid‐phase ELISA, Quantikine human IP‐10 immunosorbent assay according to the manufacturer procedure (R&D Systems, Minneapolis, MN, USA). Urine IP‐10 levels were normalized and reported as a ratio to creatinuria 15.

Gathered data were processed using SPSS version 20 (SPSS Inc., Chicago, IL, USA). The medians and interquartile ranges (IQR) were calculated for continuous measures and Mann–Whitney U test was used to compare median levels between two groups. A probability value (P value) less than 0.05 was considered statistically significant. Receiver operating characteristic (ROC) plots was used to compare the diagnostic performance of IP‐10, albumin/creatinine ratio, and anti‐dsDNA. Likelihood ratios were used to assess the sensitivity and specificity of IP‐10.

Mean age of patients was 30 (18–48) years. The demographic characteristics and laboratory investigations of the four subgroups are presented in Table 1. Active SLE patients had significantly increased serum and urinary IP‐10 as compared to inactive and healthy controls (P = 0.015, P = 0.033, respectively). Serum and urinary IP‐10 correlated significantly with SLEDAI score (r = 0.416; P = 0.02) and (r = 0.448, P = 0.01), respectively. These data are similar to those reported earlier by Lit et al. 16, who found that serum IP‐10 levels were higher among SLE patients than in controls and correlated significantly with SLEDAI score. Another study showed that serial measurement of serum levels of IP 10 correlated with longitudinal change in SLE activity, even at low levels where anti‐dsDNA antibody and complement levels remained unchanged 17. Bauer and colleagues stated that monitoring of IFN‐regulated serum chemokines (IP‐10, MCP‐1, and MIP‐3B) in SLE may be useful in evaluation of disease activity and identifying patients at risk for SLE flares. Moreover, their results provided evidence that therapies targeting the IFN pathway may prove effective in modulating disease activity in lupus 18.

Table 1.

Demographic characteristics and Laboratory investigations of the four subgroups

	Lupus Nephritis	Active non‐renal	Inactive	Healthy controls
	n = 10	n = 10	n = 10	n = 10
Age (years)	29.9 ± 9.3	33.6 ± 10.4	27.1 ± 5	32.6 ± 9.3
18–28 years	5	4	5	4
29–39 years	3	2	5	5
40–50 years	2	4	0	0
51–60 years	0	0	0	1
Mean serum creatinine (mg/dl) ±SD	1.3 ± 0.8	0.8 ± 0.4	0.7 ± 0.4	0.7 ± 0.3
Med. Anti‐dsDNA (mg/dl) (IQR)	806 (174–1404)	380 (42–696)	198 (49–775) a	–
Med. A/C ratio (mg/g) (IQR)	787 (161–4474)	19 (4–159)	13 (6–247) a	10 (8–19)
Med. SLEDAI score (IQR)	11 (3–20)	16 (13–20)	2 (1–3)	–
Med. serum IP‐10 (pg/ml) (IQR)	205 (118–388)	301 (129–536)	124 (90–326)a	96 (77–127)
Med. urinary IP‐10/creatinuria (×10−5 mg/g) (IQR)	24.4 (6.2–84.2)	13.8 (6–36.2)	11 (5.8–24.2) a	3.9 (2.2–9.3)

^a P < 0.05, as compared to active SLE (renal and non‐renal); Mann–Whitney Test; SLE: systemic lupus erythematosus; SLEDAI: SLE disease activity index; A/C: Albumin creatinine ratio; IP‐10: interferon‐γ induced protein 10. IQR: interquartile range (25th to 75th percentile); Med.: Median; SD: standard deviation.

ROC curve analysis showed that serum and urinary IP‐10 could be used in diagnosing SLE with serum IP‐10 being better (P = 0.011, AUC = 0.772 (0.629–0.915) (95% confidence interval (CI); P = 0.018, AUC = 0.753 (0.606–0.901) (95% CI), respectively (Fig. 1 a). A serum IP‐10 value of 116.5 pg/ml as a cut‐off value is best in differentiating SLE patients from healthy controls with 76% sensitivity and 70% specificity. While urinary IP‐10 value of 37.9 pg/ml is best in indicating SLE with 70% sensitivity and 70% specificity.

Figure 1.

ROC curve showing the ability of (a) serum and urinary IP‐10 in differentiating SLE patients from Healthy controls (b) serum IP‐10 in differentiating active from inactive SLE.

However, there was no difference in serum and urinary IP‐10 levels between active renal and active non‐renal patients. Even in the biopsy‐proved proliferative nephritis patients, urinary IP‐10 was not significantly different compared to inactive and active non‐renal patients.

In another study, Abujam and co‐authors 5 have found higher urinary levels of IP‐10 in active lupus patients with renal involvement as compared to those who were active but without renal involvement. These differences could be attributed to either the relatively small sample size of this study, ethnic, or gender differences where this study was performed on female Egyptians, and the other study 5 was performed on Indianan population where female to male ratio was 5.8 to 1.

In this study, serum IP‐10 shows a significant role in differentiating active from inactive SLE (P = 0.007, AUC = 0.753 (0.594–0.911) (95% CI)) (Fig. 1 b). It is better than urinary IP‐10, albumin/creatinine ratio, and anti‐dsDNA (AUC = 0.654 (0.467–0.841); 0.529 (0.317–0.742); 607 (402–812)), P values were non‐significant). A cut‐off value of serum IP‐10 136 pg/ml is best in differentiating active from inactive SLE with 81% sensitivity and 71% specificity.

Serum and urinary IP‐10 did not significantly differentiate either lupus nephritis from SLE patients without renal involvement (AUC (95% CI) = 0.633 (0.430–0.837), 0.610 (0.397–0.823), P value was non‐significant), nor active renal SLE from active non‐renal SLE (AUC (95% CI) = 0.533 (0.230–0.837), 0.700 (0.431–0.969), P values were non‐significant).

However, albumin/creatinine ratio is a better marker in differentiating between SLE with renal involvement and SLE with no renal involvement (P = 0.008, AUC (95% CI) = 0.803 (0.610–0.995). A value of albumin/creatinine ratio 126 mg/g as a cut‐off value is best for differentiating renal from non‐renal involvement with 90% sensitivity and 75% specificity. It is followed by anti‐dsDNA (P = 0.05, AUC (95% CI) = 0.715 (0.518–0.912)) in this regard.

Albumin/creatinine ratio also can significantly differentiate active renal from active non ‐renal (P = 0.009, AUC (95% CI) = 0.900 (0.744–1.000)) with a cut‐off value of 126 mg/g is best with 100% sensitivity and 80% specificity. Any of serum and urinary IP‐10, albumin/creatinine ratio, and anti‐dsDNA did not correlate with the class of lupus nephritis in renal biopsy (P = 0.658, 0.647, 0.807 and 0.069, respectively).

In conclusion, serum and urinary IP‐10 could be better biomarkers than conventional biomarkers for assessing SLE activity. However, IP‐10 cannot be used to differentiate between lupus nephritis and SLE without renal involvement. Albumin/creatinine ratio could be used to detect renal involvement in SLE and renal activity with high sensitivity and specificity than other biomarkers such as IP‐10 and anti‐dsDNA.

Conflicts of interest

The authors declare that they have no financial or non‐financial conflicts of interest related to the subject matter or materials discussed in the article.