Background/Purpose:

Traditional biomarkers for systemic lupus erythematosus (SLE) and lupus nephritis (LN) include serum creatinine (Cr), complement levels (C3/C4), double-stranded DNA antibody (dsDNA), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), urine red blood cells (uRBC), and urine protein/Cr ratio (uPCR). However, these clinical parameters have limited specificity for active LN and are inconsistent predictors of renal pathology. Whether these biomarkers are associated with LN class has important implications in therapeutic decision-making. The objectives of this study are: 1) to determine whether these biomarkers are associated with LN on renal biopsy; 2) to determine whether they can distinguish between LN classes; and 3) to assess correlation between these biomarkers and activity and chronicity indices on renal biopsy.

Methods:

Using the University of Pittsburgh Health Sciences Tissue Bank (HSTB) and Renal Pathology Department stored biopsy specimens, we identified 37 cases of LN diagnosed on renal biopsy from 2010-2016. Using the electronic database, we obtained LN classes and biomarkers checked within a few days to a month of biopsy date for each sample. Descriptive summaries, correlations, Fisher’s exact test, and rank-sum tests were used to evaluate the relationship between the biomarkers and LN.

Results:

Of the 37 LN samples, 1 had class I, 3 had class II, 10 had class III, 15 had class IV, and 13 had class V (5 with isolated class V) LN. Excluding samples with missing data, 14 (47%) had Cr >1.3 mg/dL. There were 17 (56.7%) with low C3, 18 (60.0%) with low C4, and 16 (59.3%) with dsDNA positivity. Of the 11 samples negative for dsDNA, 8 had class III or IV, and 3 had class I, II, or V LN. Overall mean ± SD ESR was 53.3 ± 35.6 mm/h and CRP was 1.2 ± 2.5 mg/dL. Hematuria was variably reported, but only 11 (37.9%) had >5 RBC/hpf on urinalysis. Mean ± SD uPCR was 2.6 ± 2.9 mg/g Cr. Comparison of biomarkers between classes III and IV vs. other classes (I, II, V, and II+V) showed significantly greater proportion with low C3 (66.7% vs. 16.7%; p=0.027) and significantly higher activity index (7.2 ± 3.8 vs. 1.0 ± 0.7; p=0.002) in class III-IV vs. other classes. Comparison of biomarkers for class V only vs. classes III and IV with or without class V (III, IV, III+V, and IV+V) showed significantly lower proportion with low C3 (0.0% vs. 66.7%; p=0.008) and significantly lower activity index (0.5 ± 0.7 vs. 7.2 ± 3.8; p=0.034) in class V vs. class III-IV. There was no significant correlation between any biomarker and activity index. There was borderline significant correlation between CRP (rho -0.4, p=0.058) and uPCR (rho 0.4, p=0.057) and chronicity index.

Conclusion:

Traditional biomarkers of SLE and LN are imperfect predictors of LN class on renal biopsy. Surprisingly, only C3 was significantly different between LN classes, and none were significantly correlated with activity and chronicity indices. Because clinical laboratory and urine biomarkers do not distinguish LN classes well, renal biopsy is necessary to reliably diagnose LN class. Given the lack of associations between traditional biomarkers and renal pathology, more specific LN biomarkers are urgently needed.

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/are-traditional-biomarkers-of-lupus-associated-with-renal-pathology-in-lupus-nephritis/