Patients with membranous lupus nephritis form two clusters with different prognoses

Area of Science:

• Nephrology
• Immunology
• Rheumatology

Background:

• Membranous lupus nephritis (MLN) classification, including class V alone or with classes III/IV, has controversial clinical, therapeutic, and prognostic relevance.
• Traditional classification may not accurately differentiate patient outcomes.

Purpose of the Study:

• To identify novel phenotypic clusters in MLN patients.
• To enhance the accuracy of high-risk profile identification and prognostic prediction in MLN.
• To compare the predictive power of novel clusters against traditional classifications.

Main Methods:

• Retrospective cohort study of 412 MLN patients.
• Unsupervised clustering analysis (K-means), principal component analysis, and decision tree analysis were used to identify phenotypes.
• Primary outcomes included adverse renal events, all-cause death, and end-stage renal disease (ESRD).

Main Results:

• Distinct clinical and pathological differences were observed between traditional class IV + V and classes V + III/V.
• K-means clustering identified high-risk (n=180) and low-risk (n=232) groups with significantly different adverse renal outcomes (9.2% vs 4.1%, P < 0.001).
• A decision tree model incorporating SLEDAI score, hemoglobin, serum creatinine, traditional classification, and activity index achieved 95.8% accuracy in the development cohort and 87.1% in the validation cohort for high-risk MLN patient identification.

Conclusions:

• Two novel phenotypic clusters were identified in MLN patients.
• These novel clusters demonstrate superior predictive accuracy for high-risk profiles and prognosis compared to traditional classifications.
• The findings support the use of these novel clusters for improved patient management and risk stratification in MLN.