Biomarkers in Lupus Nephritis




Biomarkers have the potential to be useful tools for noninvasively evaluating and managing patients with lupus nephritis. Many candidate biomarkers have been identified, but they require validation in larger cohorts. It is likely that combinations or biomarker profiles, rather than individual markers, will emerge to help better predict the severity of inflammation, the extent of fibrosis, degree of drug responsiveness, and other variables. This approach has the potential to reduce the use of the renal biopsy, improve therapeutic efficacy, and limit toxicity. We predict algorithms based on genotype and biomarkers combined with clinical presentation will emerge to help guide physicians in management. Assays that show the most potential include serum erythrocyte bound complement C4d, interleukin 17, interleukin 23, interferon score/chemokine score ratio, and anti-C1q antibodies. Such urinary biomarkers as fractional excretion of endothelial-1, monocyte chemoattractant protein–1, vascular cell adhesion molecule–1, and TWEAK (tumor necrosis factor–like weak inducer of apoptosis) may also be useful but require validations.


Although biomarkers have high utility for the diagnosis of systemic lupus erythematosus (SLE), current serologic and urinary markers do not correlate well with nephritic activity. This has led to a search for better indicators of renal flare (or lack thereof). In this context, assessment of renal functional activity (eg, markers of glomerular filtration rate [GFR]), such as serum creatinine, has limitations. For example, the GFR may be still preserved while there is severe inflammation thus making it difficult to assess its true changes. For example, in a 25-year-old, 50-kg woman with lupus nephritis, an increase in the serum creatinine from 0.6 mg/dL to 0.9 mg/dL (estimated GFR change from 114 to 75 mL/min), with both levels in the normal laboratory range, represents a 35% reduction in function. Although this example seems obvious, in practice less evident changes are often missed. Such a delay in recognition may lead to a delay in treatment and irreversible scarring in the kidney. Urine protein quantization also has its own limitations. For instance, resolution of proteinuria may take weeks to months to normalize, or not normalize at all, irrespective of immunologic or inflammatory activity. Immunosuppressive therapy in this situation will not be effective.


In other situations, even with the pathology in hand, it is difficult to ascertain the extent of disease activity. Distinguishing the relative extent of ongoing inflammation from chronic fibrotic disease may be especially difficult. Thus, it would be extremely useful to identify either inflammatory or other renal functional signals in blood or urine that are indicative of inflammation and ongoing autoimmune activity (ie, with flares). Similarly, better markers of irreversible damage would also be helpful. To address this problem, researchers began looking for better urinary and serum biomarkers.


For purposes of this discussion, the potential use of urine and serum markers will be considered separately. Although the American College of Rheumatology criteria and other criteria (eg, systemic lupus erythematosus disease activity index 2000 [SLEDAI-2K], systemic lupus activity measure [SLAM]) have diagnostic utility, they do not accurately assess either severity or changes in nephritic activity, and serologic evaluations alone are insufficient. Ideally, biomarkers should indicate the severity of nephritis and guide therapy at various stages of disease.


Biomarkers are defined as a genetic, biologic, biochemical, or event, whose alterations correlate with disease pathogenesis or manifestations, and can be evaluated qualitatively and/or quantitatively in laboratories. Biomarkers should (1) be biologically active and pathophysiologically relevant, (2) be simple to use in routine practice, and (3) accurately and sensitively change with disease activity. In lupus nephritis, biomarkers should identify patients at risk for flare so that therapy can be tailored to individual situations and duration of treatment can be precisely determined. Biomarkers should also provide surrogate end points for evaluating the efficacy of new therapies. In the paragraphs below, serum and urine biomarkers are considered. The authors do not consider the utility of renal biopsy, although it is anticipated that combining information from biomarkers and renal pathology may lead to a more precise determination of the most rationale therapy for an individual patient. The authors’ focus is on the use of noninvasive biomarkers that complement the use of pathologic evaluation during the course of disease ( Box 1 ). It is anticipated that in some situations, fluctuations in activity in one or more of these parameters will lead to renal biopsy, whereas in others, the levels alone may obviate the procedure.



Box 1





  • Overall disease activity



  • Genes




    • PTPN22 (protein tyrosine phosphatase 22)



    • IRF-5 (interferon regulatory factor–5)



    • STAT-4



    • Type I interferon



    • IFIT 1



    • OAS1



    • LY6E



    • ISG15



    • MX1



    • FCγIIa polymorphism




  • Interleukins




    • IL-22



    • IL-6



    • IL-10



    • IL-12



    • IL-18



    • IL-2 receptor α




  • Chemokines




    • RANTES (regulated on activation, normal T expressed and secreted)



    • CXCL-11 (C-X-C chemokine ligand 11)



    • CCL-19 (C-C chemokine ligand 19)



    • MCP-1 (monocyte chemoattractant protein–1)



    • CXCL-13 (C-X-C chemokine ligand 13)



    • IP-10 (interferon inducible protein 10)




  • Other molecules




    • CD27



    • Reticulocyte-C4d



    • BLyS (B-lymphocyte stimulator)




  • Lupus diagnosis




    • E-C4d (erythrocyte-bound complement-activation product C4d)



    • Anti-DsDNA (anti–double-stranded DNA)



    • ANA (antinuclear antibody)



    • Antinucleosome





  • Organ specific



  • Renal involvement




    • Serum



    • Antinucleosome



    • Anti-Clq



    • α-actinin



    • Anti–α-actinin



    • Adrenomedullin




  • Urine




    • Endothelial-1



    • Lipocalin-2 (neutrophil gelatinase-associated lipocalin)



    • U-MCP-1 (urinary monocyte chemoattractant protein–1)



    • Migration inhibition factor



    • Adiponectin



    • VCAM-1 (vascular cell adhesion molecule-1)



    • P-selectin



    • CXCL-16 (C-X-C chemokine ligand 16)



    • FOXP3 (forkhead family transcription factor 3)



    • TWEAK (tumor necrosis factor–like weak inducer of apoptosis)



    • Osteoprotegerin




  • Neural




    • Antihistone



    • Anti-N



    • AECA (antiendothelial cell antibodies)



    • MMP-9 (matrix metalloproteinase 9)



    • Anti-NMDA (anti–N-methyl-D-asapartate)



    • Anti-NR2 (anti–N-methyl-D-asapartate receptor subunit NR2)



    • Anti-P ribosome




  • Skin




    • Anticyclic citrullinated peptide Ab




Biomarkers in lupus nephritis


Biomarkers


Nonspecific Biomarkers


Nonspecific biomarkers occur in many inflammatory states and are not disease specific. For example, prolactin and ferritin serum levels may be elevated in autoimmune diseases like SLE. Similarly, erythrocyte sedimentation rate, C-reactive protein, and resistin are not lupus specific.


Genetic Markers


Genetic markers typically identify patients at risk for disease (eg, lupus or specific disease manifestations). They are useful for early diagnosis and monitoring patients at risk (eg, family members) ( Box 1 ).


Protein tyrosine phosphatase 22


Protein tyrosine phosphatase 22 (PTPN22) is a tyrosine-specific phosphotase in T cell signaling, expressed preferentially in memory and effector T cells. Expression of this lymphoid specific protein was predictive in Caucasians and Europeans for increased risk of developing autoimmune diseases, including lupus.


Interferon regulatory factor-5


Interferon regulatory factor-5 (IRF-5), which encodes a transcription factor regulating interferon α (IFN-α), may be a useful diagnostic marker. A single-nucleotide polymorphism was associated with disease susceptibility and SLE activity. However, it was also elevated in other autoimmune diseases.


STAT 4


STAT 4, a transcription factor that regulates cytokine production, is helpful in early detection of SLE, although increased protein levels (rs7574865) were present with other autoimmune diseases (eg, rheumatoid arthritis, inflammatory bowel disease, type 1 diabetes mellitus).


Type I interferons


Type I interferons have also been studied as potential biomarkers of disease activity (eg, IFIT 1 , OAS1, LY6E, ISG15, and MX ). IFIT 1 correlated with active lupus but lagged behind clinical remission.


Other candidates


Other candidates under evaluation include poly–adenosine diphosphate-ribose polymerase (PARP), B lymphoid tyrosine kinase (BLK), ITGAM , LYN, and intergrin alpha X (ITGAX), and they await validation.


Serologic Diagnosis


Anti–double-stranded DNA (anti-dsDNA), anti-Ro, anti-La, anti-Sm and antiribonucleoprotein antibodies remain useful for diagnosis, although limited for monitoring activity. By flow cytometry, erythrocyte-bound complement-activation product C4d (E-C4d) was higher, and complement receptor C1 (E-CR1) was lower when compared with healthy patients. The sensitivity and specificity of the ratio was 81% and 94 %, respectively, compared with healthy controls, and 72% and 79%, respectively, compared with inactive patients. In other studies, platelet-bound C4d (P-C4d) was 98% specific in diagnosing lupus, although its utility requires further validation. Antiguanosine antibodies distinguished active lupus from inactive disease. However, they were not predictive of organ involvement.


Overall Disease Activity


Reticulocyte-C4d has promise. Reticulocytes have a short half-life (120 days), and during active autoimmunity and complement activation, C4d binds to them. Manzi and colleagues correlated reticulocyte-C4d levels with disease activity, and also correlated the combination of decreased E-CR-1 and decreased E-C4d, with enhanced sensitivity and specificity (83% and 89%, respectively). During active disease with C3 and C4 levels decreasing, E-C4d increased. CD27 levels on B cells also correlate with disease activity.


Type I interferons (ie, IFN-α) may help distinguish the severe form of such disease as cerebritis, nephritis, and hematological pathologies from other forms. Type I interferons were shown to be associated with anti-Ro, U1-RNP, and Smith antibodies, although their utility has been disputed.


B-lymphocyte stimulator (BLyS), a transmembrane protein located on monocytes, macrophages, and dendritic cells, is crucial for B cell growth and survival. In animal models, overexpression of this protein and lupus activity has been linked. However, in humans, BLyS levels have been variable and not useful in monitoring either disease activity or specific organ involvement. Nevertheless, in a comparison of rheumatoid arthritis and SLE patients, the lupus patients had higher levels, and they correlated better with overall disease activity.


Cytokines


Serum levels of IL-17 and IL-23 were increased in both active and inactive SLE patients, whereas IL-22 levels were decreased in patients with active lupus. Chun and colleagues found that IL-6, IL-10, IL-12, and IFN-γ levels were higher in the SLE patients, while IL-2 levels were lower. IL-6 and IL-10 levels correlated with some aspects of disease activity but not with lupus nephritis. In separate studies, IL-6 elevations and tumor necrosis factor α levels were found to correlate with active disease. By contrast, although IL-12 and IL-18 were elevated in lupus patients, the levels did not correlate with activity. IL-2Rα levels were found to be increased, especially in those with severe lupus nephritis, and this biomarker may prove to be useful if confirmed.


Chemokines


Fu and colleagues measured IFN-inducible chemokines (RANTES [regulated on activation, normal T expressed and secreted], CXCL-11 [C-X-C chemokine ligand 11]), IP-10 (IFN inducible protein 10), MIG (monokine induced by IFN-γ), CCL-19 (C-C chemokine ligand 19), monocyte chemoattractant protein-1 (MCP-1), and IL-8 as biomarkers for disease activity. An IFN score was created from measuring the gene expression, a chemokine score was created by measuring chemokine serum levels, and the results were compared with those of patients with rheumatoid arthritis and those of healthy controls. The chemokine score correlated with disease activity compared with C3 levels and SLEDAI-2K. As the disease activity improved with treatment, the score decreased.


CXCL-13 (B cell attractant to areas of inflammation) levels were increased in lupus sera. Active nephritic patients had higher levels. However, CXCL-13 levels were not useful in determining histologic class.


Type I interferon-inducible protein-10 (IP-10, a chemokine produced by peripheral mononuclear cells) was elevated in chronic inflammatory diseases and SLE. However, it was associated more closely with mucocutaneous disease and hematological manifestations but not nephritis.




Biomarkers for renal involvement


Most lupus nephritis patients have antichromatin/nucleosome antibodies (specificity ∼98%; sensitivity 69%), and they may be positive when the anti-dsDNA antibodies are negative. Similar findings were observed with anti-C1q antibodies, especially with nephritic flares (negative positive predictive value of 97%–100%,) although their precise role has been debated.


Anti-α actinin antibodies are prevalent in patients with active lupus nephritis, and they may be more predictive of nephritis than anti-dsDNA antibodies, although larger studies are needed for confirmation. Anti-Sc-70 (topoisomerase) antibodies have also had mixed results. Adrenomedullin released from macrophages and smooth muscle cells is elevated in SLE, pregnancy, hypertension with left ventricular hypertrophy, diabetes, and other chronic diseases, and it appears to be elevated in active lupus nephritis, although its specificity needs to be determined.




Biomarkers for renal involvement


Most lupus nephritis patients have antichromatin/nucleosome antibodies (specificity ∼98%; sensitivity 69%), and they may be positive when the anti-dsDNA antibodies are negative. Similar findings were observed with anti-C1q antibodies, especially with nephritic flares (negative positive predictive value of 97%–100%,) although their precise role has been debated.


Anti-α actinin antibodies are prevalent in patients with active lupus nephritis, and they may be more predictive of nephritis than anti-dsDNA antibodies, although larger studies are needed for confirmation. Anti-Sc-70 (topoisomerase) antibodies have also had mixed results. Adrenomedullin released from macrophages and smooth muscle cells is elevated in SLE, pregnancy, hypertension with left ventricular hypertrophy, diabetes, and other chronic diseases, and it appears to be elevated in active lupus nephritis, although its specificity needs to be determined.




Urine biomarkers


Urine biomarkers are influenced by urine concentration, breakdown in the urine, protein binding, interpretation, and other factors. Nevertheless, they have the advantage of being noninvasive and easily obtainable. Their utility is discussed below.


Endothelial-1 (ET-1) is a 21–amino acid peptide produced in the vasculature, and it participates in cell proliferation, inflammation, vasoconstriction, and fibrosis. Urinary ET-1 reflects both renal and extrarenal production. In one study, serum ET-1, urinary ET-1, and the fractional excretion of ET-1 (feEt-1) were evaluated as biomarkers for chronic kidney disease, rheumatoid arthritis, and SLE. FeET-1 increased during progression of both chronic kidney disease and lupus nephritis. FeET-1 decreased after therapy in lupus nephritis, so it may have utility.


Lipocalin-2, secreted by leukocytes and epithelial cells, is important for iron transport. Urinary levels were evaluated in 70 patients with lupus (with or without lupus nephritis) and controls, with normalization to urinary creatinine. Urinary levels were found to be predictive of active nephritis.


Urinary MCP-1 has also been demonstrated to be predictive of disease activity. Increased levels were found to precede lupus flare by as much as 4 months, and urinary MCP-1 fell with successful treatment. Also, polymorphism of the MCP-1 gene vary in predicting renal involvement in lupus (ie, A/A polymorphism indicates a lower likelihood of involvement compated with a A/G or G/G polymorphism). In a study of 123 SLE patients and 53 controls, urinary MCP-1 was elevated in active lupus nephritis, whereas other markers (eg, IL-6, IL-10, and IL-8 [CXCL-8]) were not useful. In separate studies, urinary message of MCP-1 and transforming growth factor β (TGF-β) correlated with active disease severity but not with the level of fibrosis. In a small group of pediatric patients, migration inhibition factor/creatinine ratio was elevated in the five active lupus nephritis patients versus the five inactive patients.


Zhang and colleagues confirmed that infiltrating leukocytes in the kidney generated hepcidin, and they found that urinary hepcidin increased during active nephritis and decreased with disease resolution. Urinary hepcidin 20 was especially useful as a flare predictor, whereas urinary hepcidin 25 decreased during flares and increased during treatment. Urinary adiponectin ( a cytokine produced by the adipodcytes with an anti-inflammatory effect) increased 2 months before flares, remained elevated during nephritic activity, and decreased 4 months postflare.


Various adhesion molecules have also been studied, including vascular cell adhesion molecule-1(VCAM-1), P-selectin, CXCL-16, and tumor necrosis factor receptor-1 (TNFR-1). VCAM-1, which is found mostly in the kidney, recruits monocytes, dendritic cells, and endothelial cells to inflamed areas. Urinary excretion of VCAM-1, P-selectin, CXCL-16, and TNFR-1 increased in lupus patients compared with the other groups. In separate studies, FOXP3 (forkhead family transcription factor 3, which is important for regulation of T cells) urinary messenger RNA levels correlated with proliferative lupus nephritis in a small cohort.


Chan and colleagues analyzed T cells, B cells, and natural killer cells in the urine of 12 patients with active lupus nephritis, 17 patients in remission, and 12 lupus patients without renal involvement. CD3+ (T cells) and CD20+ (B cells) correlated best with lupus disease activity while the CD56+ cells did not recognize disease activity with any accuracy.


Chemokines


Avihingsanon and colleagues studied urinary messenger RNA of various chemokines to differentiate the different classes of lupus nephritis. For diagnosing class IV nephritis, the messenger RNA of interferon-producing protein 10 (IP-10) was most useful, followed by vascular endothelial growth factor, and then CXCR3 (chemokine [C-X-C-motif] receptor 3).


Tumor necrosis factor–like weak inducer of apoptosis (TWEAK ) regulates other chemokines, such as IP-10, RANTES, and MCP-1. Binding of TWEAK to the Fn14 receptor leads to activation of these chemokines. Schwartz and colleagues evaluated this regulator in lupus nephritis in patients from the Ohio State SLE Study and the Albert Einstein College Lupus Cohort. TWEAK increased in active lupus nephritis patients compared with patients with only extrarenal lupus and with patients with inactive disease, and TWEAK levels appeared to increase with nephritic flares, but not extrarenal flares.


Osteoprotegerin


Osteoprotegerin (tumor necrosis factor family) causes bone reabsorption and is found in many other organs. Urinary levels of osteoprotegerin correlate well with the presence of renal lupus but not with the severity of disease.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Oct 1, 2017 | Posted by in RHEUMATOLOGY | Comments Off on Biomarkers in Lupus Nephritis

Full access? Get Clinical Tree

Get Clinical Tree app for offline access