Chinese Rheumatism Biobank(CRB)

Systemic lupus erythematosus (SLE) is a complex autoimmune disease characterized by the presence of multiple autoantibodies in the serum and multisystemic damage.The clinical manifestations of SLE are highly heterogeneous, and damage to different organs appears gradually over time. Lupus nephritis (LN) and thrombocytopenia (TP) represent the most common major organ damage in SLE, with an incidence of about 30-40%; pulmonary arterial hypertension (PAH) is a rare but difficult complication of SLE. Pulmonary arterial hypertension (PAH) is a rare but intractable complication of SLE, which clinically manifests itself as a progressively worsening increase in pulmonary arterial pressure, and ultimately causes right heart failure or even death. Once combined with LN, TP, PAH and other important organ damage, the long-term prognosis and quality of life of SLE patients are severely impaired.

Some predictive models for the occurrence of organ involvement in SLE have been reported in previous studies. Among them, several incidence prediction models for lupus nephritis (LN) included demographic, clinical and laboratory indicators such as male, anti-dsDNA antibody positivity, age and duration of SLE onset, erythrocyte sedimentation rate, mucosal ulceration, proteinuria, hematuria, etc., with the area under the curve (AUC) of the prediction model of 0.707. A prediction model for incidence prediction of neuropsychiatric lupus (NPSLE) incorporated both patient disease and environmental indicators, including anti-dsDNA antibody positivity, anti-SSA antibody positivity, lymphocyte count, erythrocyte pressure volume, erythrocyte sedimentation rate, prealbumin, retinol-binding protein, creatine kinase isoenzyme MB, brain natriuretic peptide precursor, creatinine, indirect bilirubin, fibrinogen, ultrasensitive C-reactive protein, CO, and mild contamination, which had an AUC in the discovery and validation cohorts of up to 0.849-0.895. Our center team has developed 2 predictive models for the onset of SLE organ involvement: the SLE-associated pulmonary arterial hypertension (PAH) onset prediction model contains 5 clinical variables (acute/subacute cutaneous lupus, arthritis, renal involvement, thrombocytopenia, and interstitial lung disease) and 3 autoantibodies (anti-RNP antibody positivity, anti-Ro/SSA antibody positivity, and anti-La/SSB antibody positivity). La/SSB antibody positive). The above predictive models for the occurrence of SLE organ involvement basically used easily accessible clinical data, such as clinical manifestations and laboratory indicators, and the models were practical; however, most of the models were not constructed based on prospective cohorts, and some of the models were not validated in large-scale patient populations, so the predictive efficacy of the models is still unsatisfactory, and their clinical application is still difficult.

There is still a lack of effective biomarkers and prediction models for important organ damage in SLE. The fundamental problem is that there is no high-quality cohort of SLE patients before and after the onset of organ damage. Therefore, early prediction of vital organ damage in SLE can only be achieved by dynamically tracking the evolution of SLE patients before and after the onset of vital organ damage, and analyzing the microscopic molecular evolution patterns synchronized with the macroscopic pathophysiological changes.