PRediction Of Flares In Lupus With autoantibodiEs and Chemokines

Systemic lupus erythematosus (SLE) is a chronic relapsing-remitting autoimmune disease with a wide range of clinical manifestations affecting several organs. Although the management of lupus patients has improved in the last years, accurate models for predicting disease progression are lacking.

In clinical practice, SLE patients can be categorized into three groups:

1. A large group of patients has 'quiescent' disease; after diagnosis of SLE and possibly a short induction treatment with corticosteroids, patients remain in a state of remission for years with hydroxychloroquine treatment only.
2. A substantial group of patients has relapsing-and-remitting disease, with mostly cutaneous inflammation but without further organ involvement.
3. A smaller group of patients has severe inflammation with extensive organ involvement, including lupus nephritis and neuropsychiatric SLE (npSLE) as the most threatening complications.

This stratification of patients is reflected by the frequency of hospital visits. Patients with more severe disease activity have a higher frequency of visits to the outpatient clinic and more frequently receive biologicals.(1) In the UMC Utrecht, patients in group 1. normally visit the outpatient clinic on a yearly basis, patients in group 2. visit 2-4 times a year and patients in group 3. are seen more frequently. However, some patients with quiescent disease for years can still present with an SLE exacerbation, and patients with frequent bouts of inflammation can eventually reach a remission state.

It would be of great value to be able to distinguish these patient categories early after diagnosis. This would aid the clinician in identifying those patients who can safely visit the outpatient clinic only once a year, and conversely, it would be possible to point out which patients should be monitored more closely. Ideally, this early distinction would also correspond with treatment decisions, that is, intensifying treatment when a flare is suspected, or tapering medication when there is a low risk profile. The ultimate goal would be to enable clinicians to identify high risk patients and to treat them with immunomodulating therapy before any inflammatory damage has occurred, for instance with the combination of rituximab and belimumab as is being studied right now.(2,3) The pathogenesis of SLE is highly complex. Genetic predispositions, proinflammatory and anti-inflammatory cytokines, autoantibodies, lymphocyte subset abnormalities as well as defects in the complement systems all have putative roles in the development of SLE. At present, tools that enable early patient stratification are lacking. Furthermore - except for a possible association of rise in anti-dsDNA-antibodies and the development of lupus nephritis - factors that can predict SLE flares have not been identified.

Active disease has a large impact on the life of SLE patients, since it is known to result in a lower health-related quality of life.(4) Fatigue is an important factor in patient reported quality of life, but interestingly, disease activity is not associated with fatigue.(5) This can lead to a discrepancy, where the clinician assesses the patient's disease activity to be under control, where the patient is still experiencing a large burden of disease. Due to lack of objective parameters correlating with patient reported outcomes, this discrepancy is difficult to interpret for physicians.

When considering factors that can be relevant for patient risk stratification, the role of (novel) autoantibodies in the pathogenesis of SLE should not be overlooked. Several autoantibodies have been shown to be of diagnostic value for SLE (anti-dsDNA, anti-SmD, anti-Rib-P, anti-PCNA, anti-Chromatin, anti-complement (C1q)) and changes in the level of autoantibodies can reflect disease activity.(6) However, dynamics of antibody levels on their own are insufficient to be used as predictors of lupus activity in individual patients. Moreover, some types of treatment, such as B-cell targeted therapy, are known to influence the dynamics of antibody levels, making them harder to interpret.

Some chemokines and matrix metalloproteases (MMPs) and their inhibitors (TIMPs) have been shown to provide an indication of subclinical injury or inflammation in other inflammatory diseases and in renal inflammation.(7-9) For example, urinary CXCL9 levels were found to be associated with risk of acute rejection of renal transplants and a decline in renal function.(10) In addition, serum levels of CXCL10 have been shown to correlate with lupus activity in patients with SLE.(11) Furthermore, urinary levels of CXCL10 have shown very promising results in the early detection of allograft rejection in kidney transplants. Currently, an international randomized controlled trial is investigation if early treatment of rejection, detected by urinary CXCL10, will improve outcomes. (12) As of yet, these urinary markers have not been studied in the context of SLE and lupus nephritis. Dynamic levels of these urine markers and the relation to their corresponding serum levels could serve as early indicators of nephritis.

The primary objective of this study will be to prospectively evaluate the predictive value of a combination of chemokines, MMPs/TIMPs, and autoantibody levels for predicting flares in patients with SLE. This approach allows identification of the markers with the best predictive value.

As a secondary objective, this study will investigate whether dynamic changes in autoantibody- and chemokine levels differ between patients treated with or without anti-B-cell therapy (for instance, with rituximab or belimumab). In patients treated with anti-B-cell therapy lower autoantibody levels are observed, which could influence the combination of markers being studied. Until now, it is unclear whether the degree of lowering autoantibody levels after treatment is correlated with treatment efficacy and whether changes in autoantibody levels in SLE patients treated with anti-B-cell therapy can be indicative of lupus flares. This study will give insight into this relationship.

The relationship of these biomarkers with patient reported quality of life and fatigue will be assessed. These markers might provide the physicians extra tools on how to interpret the discrepancy between objective measurements by the physician and patient reported outcomes.

This study will be conducted in the department of Rheumatology and Clinical Immunology in the UMC Utrecht, a tertiary care center with approximately 400 patients with SLE in care. This department is very experienced and successful in conducting large clinical and translational trials in the field of autoimmunity.(13-18) Together with two other Dutch academic hospitals, the Dutch SLE registry was founded, including not only all patients who are treated with belimumab, but at present almost 1000 Dutch SLE patients who are being followed clinically at least once a year. The department actively takes part in the current multi-center international BLISS-BELIVE study.(2) The proposed number of patients to be included in this study is 100, making it a large prospective cohort study in SLE related research.

Main hypothesis: A combination of levels of autoantibodies and chemokines can be used to predict flares in SLE patients.