INTRODUCTION

Systemic lupus erythematosus (SLE) is a heterogeneous and multifactorial autoimmune disease, revealed by autoantibody production and dysregulated high expression of Type I interferon (IFN)-stimulated genes, causing a wide variety of clinical manifestations¹. About 50-80% of SLE patients will develop lupus nephritis (LN), which represents a significant cause of morbidity and mortality in these patients²,³. In the last decade, therapeutic approaches for LN have remained essentially unchanged, with a probability of achieving complete or partial remission not beyond 60-70%; unfortunately, the rest of the patients usually progresses to end-stage renal disease within 5 years after the onset⁴,⁵. Moreover, LN diagnosis and/or monitoring remain suboptimal in most clinical contexts. Besides the low possibility of causing kidney damage and being an invasive procedure, renal biopsy represents the gold standard for LN diagnosis and to determine the degree of tissue involvement5; furthermore, its usefulness in directing the choice of one therapeutic approach over another is still limited²,⁴,⁶,⁷. Consequently, there are different areas for improvement in the field of LN, such as the optimization of long-standing treatment strategies and the search for novel therapeutic agents to improve outcomes such as mortality or development of end-stage renal disease. Among these issues, one of the most relevant is the search for new clinical biomarkers with a good correlation between clinical manifestations and actual histological damage that allow the follow-up of patients with LN to guide personalized treatments.

In recent years, the investigation of LN biomarkers has produced many publications; however, the same serum and urinary measurements have continued to be used in clinical practice for over three decades. These conventional diagnostic and prognostic LN biomarkers include proteinuria, hematuria, urinary casts, serum creatinine levels, anti-double-stranded DNA autoantibodies, and complement levels⁵,⁷. Given the heterogenetic nature of LN patients and the not-yet reached needs in precision diagnosis and classification of these individuals in terms of personalized therapy, identification of novel biomarkers would be helpful to provide auxiliary information about the disease activity, the early recognition of renal flares, treatment selection, and responsiveness to a particular medication scheme. Although urine component-related studies are promising, the more robust blood/serum biomarkers may still be useful in developing point-of-care systems that can be adapted to most clinical settings or even home testing of LN.

NOVEL BLOOD/SERUM BIOMARKERS FOR LUPUS NEPHRITIS

A relatively large number of non-classical blood/serum candidate biomarkers have been evaluated in LN cohorts. Most of these candidates have only been assessed in cross-sectional analyses, with few reports of markers prospectively evaluated. Limitations of most-recent studies also include the relatively small number of patients tested and the lack of confirmation by independent research groups. Although further validation must be performed in most cases, the following candidates, not yet incorporated into routine clinical practice, still represent promising alternatives to enhance the diagnostic accuracy, prognostic stratification, and monitoring of treatment response in LN patients. Detailed information will be next presented for these potential LN biomarkers, while these and some of the most recently reported candidates are enumerated in table 1.

MONOCYTE CHEMOATTRACTANT PROTEIN-1 (MCP-1 OR CCL₂)

Monocyte chemoattractant protein 1 (MCP-1 or CCL2) belongs to the C-C chemokine family⁸. Increased expression of this chemokine on epithelial or endothelial cells and infiltrating mononuclear leukocytes in the tubulointerstitial regions can be demonstrated in kidney biopsies of LN patients⁸,⁹. In addition, different studies have found a potential diagnostic or prognostic value for this molecule in the context of LN; some reported an association between increased levels of MCP-1/CCL2 and LN, while others looked at its prognostic role through renal disease activity as an outcome, showing significant correlations with the chemokine circulating levels⁷,⁸.

SURROGATE MARKERS OF TYPE I INTERFERON SIGNATURE

SLE patients are characterized by an increased expression of type I IFN-regulated genes¹⁰, indicated as an IFN gene signature that comprises some LN biomarker candidates, including SIGLEC-1, CXCL10, and Galectin-9. The sialic acid-binding Ig-like lectin 1 (SIGLEC-1) is a cell-adhesion molecule involved in the initial contact of phagocytic cells with sialylated pathogens¹¹. It could also be found in its soluble form in serum, where elevated concentrations have been associated with increased frequency of renal complications but not with SLE disease activity index⁷,¹².

CXCL10 is a chemokine that has shown a high correlation with SLE activity and the best predictive ability for flares among chemokines regulated by IFN-I⁸,¹³. Remarkably, renal flares are accurately predicted by CXCL10 serum levels but not by classical biomarkers such as complement levels or anti-double-stranded DNA autoantibodies⁷,⁸. Furthermore, the β-galactoside-binding lectin Galectin-9 has shown a higher correlation with SLE activity than CXCL10 levels. Interestingly, in the context of LN, serum levels of Galectin-9 also reflect kidney damage and not only SLE activity⁷,¹⁴.

LYMPHOCYTE POPULATIONS

Although several circulating lymphocyte subsets have been reported to act as SLE markers, including plasmablasts (CD19^lo CD27^hi CD38^hi)¹⁹, transitional B cells (CD24^hi CD38^hi)²⁰, and follicular helper T cells (CD4⁺ CXCR5⁺)²¹, none of them are currently linked to nephritis complication⁶.

Interestingly, our group recently confirmed the potential utility of the CD11c⁺ T-bet⁺ CD21^lo/– B lymphocytes (known as age-associated B cells or ABCs) to assess lupus nephritis; however, we also reported the finding of a related cell subset CD11c⁺ T-bet⁺ CD21^hi B cells that are almost absent when renal manifestations arise²². Importantly, this CD^21hi subset could be used as a prognostic factor as their numbers strongly correlate with a lower SLE activity. In addition, we found that either ABCs or CD^21hi B cell subsets could be considered to evaluate the response to induction therapy in LN²².

CONCLUSIONS

Despite the large quantity of recently published data that have validated new serum markers for lupus nephritis (Table 1), several research groups keep working to identify novel molecules or cells that account for the diagnosis, prognosis, and management of these patients. We believe that exploratory efforts regarding serum biomarker searching should now be focused on generating prospective randomized high-scale clinical trials including patients with different ethnic backgrounds. These approaches will probably deliver schemes that include combining these novel LN markers with conventional clinical parameters to enhance sensitivity and specificity regarding the prediction of renal flares, staging, and therapeutical management of LN patients.