Background/Purpose: Lupus nephritis (LN) is a serious complication of SLE. Reliable biomarkers to assess and/or predict renal involvement in SLE patients are needed. The aim of this study was to better assess the link between blood transcriptional signatures and LN.

Methods: Consecutive SLE patients (ACR criteria) were followed-up prospectively. Blood samples were split in: group 1, samples collected at the time of a biopsy-proven LN with active lesions, either proliferative (class III or IV) or not (class II or V); group 2, patients sampled at the time of an extra-renal flare; group 3, patients sampled at their first clinically quiescent visit (no flare or treatment modification in the past 60 days and SLEDAI ≤4). Microarray data were generated using Illumina beadchips and analyzed using modular repertoire analyses. Modules with ³ 20% transcripts differentially expressed compared to matched healthy controls were considered active.

Results: In addition to the IFN-related modules (M1.2, M3.4 and M5.12), modular repertoire analysis in SLE patients revealed a strong upregulation of M5.15, a module of 24 transcripts annotated “neutrophil”. There was no significant correlation between IFN modules and M5.15 activity. At the individual level, spearman correlations between modules and SLEDAI were significant for M5.12 (r=0.25, p=0.03), but not for M5.15 (r=0.21, p=0.09). M5.15, however, was the only module strongly associated with LN (Wilcox t-test p=0.009), although there was a trend for M5.12 (p=0.099). M5.15 was not associated with cutaneous, articular or hematological flares.

Group 1 comprised 24 patients, with proliferative (n=14) or non-proliferative (n=10) LN. Group 2 comprised 11 patients with an extra-renal flare. Group 3 comprised 34 patients, among whom 22 had a past history of LN. A neutrophil modular signature (M5.15 ≥ 20%) was observed in 16/24 (67%), 2/11 (18%) and 16/34 (47%) of patients respectively from group 1, 2 and 3 (Fisher’s exact test p=0.027). In group 2, 1 patient with a neutrophil signature had a history of LN, the other subsequently had a LN flare (class III at M24). In group 3, 9/16 had a previous LN and 4/16 subsequently developed a LN flare (class IV at M18, M21 and M36; class V at M13).

M5.15 activity was neither correlated to age, gender or ethnicity, nor with the titer of anti-dsDNA (p=0.7). There was a strong correlation between daily corticosteroid dose and M5.15 (r=0.45, p<0.0001).  M5.15 was not significantly correlated with 24h proteinuria or serum creatinine but was correlated with acute renal failure (p=0.03) and serum albumin (r=-0.30, p=0.01). In group 1, the median value of M5.15 in patients was significantly higher in patients with proliferative than non proliferative LN (66.7 vs 18.8 %, p=0.04). After initiation of treatment in patients with proliferative LN and ≥ 3 consecutive evaluation (n=10), a trend was observed between the decrease of M5.15 activity and remission at M6 (p=0.13).

Conclusion: Modular repertoire analysis demonstrates that neutrophil signature in SLE patients is correlated with the occurrence and severity of lupus nephritis and may help in the design of disease prognostic and/or activity biomarkers.

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/modular-transcriptional-neutrophil-signature-as-predictive-of-nephritis-and-of-its-severity-in-sle-patients/