Background/Purpose: Serum levels of chemokines have been previously used as downstream markers of the interferon (IFN) pathway in SLE. CXCL2 (GROB) and CXCL10 (IP-10) have been found to be elevated in SLE (over controls) and in SLE with high type I IFN gene signature over controls (Bauer et al, 2006).  CXCL2, however, may be an IFN-independent marker, in contrast to previous reports. We compared CXCL2 with CXCL10 (which is IFN-induced) over time in SLE, and determined the relationship with disease activity. 

Methods: 25 SLE patients (84% female, 72% Caucasian, 28% African-American, mean age at baseline 48 + 10 yrs) had CXCL10 and CXCL2 levels measured in plasma by ELISA at baseline and at follow-up visit 0.72 years (mean) later.  Chemokine levels were defined as low (<150 ng/ml for CXCL10 and <100 ng/ml for CXCL2) or high (>150 ng/ml for CXCL10 and >100 ng/ml for CXCL2).

Results: Baseline high CXCL2 levels were associated with a history of proteinuria.  High CXCL10 was associated with low C3 (p=0.007), low C4 (p<0.0001) and anti-dsDNA (p=0.0094).  The high CXCL10 group had a significant increase in SLEDAI at the second visit (2.4 to 4.5, p=0.021).  The high CXCL2 group had an increase in activity by both PGA (p=0.02) and SLEDAI (p=0.011), as well as lower C3 (p=0.023) and lower C4 (p=0.096) at the second visit.  Urine pr/cr went up (0.3 to 0.7), but was not statistically significant.

Changes in Measures of Disease Activity by Chemokine High/Low

 

CXCL10	Mean at baseline	p-value for baseline between groups	Mean at followup	Mean Change between baseline and followup	p-value for change within group	p-value for change between groups
PGA
Low CXCL10 (n=12)	0.6	0.47	0.7	0.1	0.57	0.73
High CXCL10 (n=13)	0.8		0.8	0.0	0.98
SLEDAI
Low CXCL10 (n=12)	0.8	0.061	1.3	0.5	0.54	0.16
High CXCL10 (n=13)	2.4		4.5	2.2	0.021
C3
Low CXCL10 (n=12)	118.5	0.0007	119.8	1.3	0.85	0.97
High CXCL10 (n=13)	79.0		80.5	1.5	0.74
C4
Low CXCL10 (n=12)	25.7	<0.0001	25.9	0.3	0.88	0.63
High CXCL10 (n=13)	12.2		13.4	1.2	0.31

 

CXCL2	Mean at baseline	p-value for baseline between groups	Mean at followup	Mean Change between baseline and followup	p-value for change within group	p-value for change between groups
PGA
Low CXCL2 (n=18)	0.8	0.48	0.6	-0.2	0.49	0.052
High CXCL2 (n=7)	0.5		1.1	0.6	0.022
SLEDAI
Low CXCL2 (n=18)	1.8	0.47	2.5	0.7	0.33	0.058
High CXCL2 (n=7)	1.1		4.3	3.1	0.011
C3
Low CXCL2 (n=18)	96.9	0.81	103.4	6.4	0.19	0.033
High CXCL2 (n=7)	100.6		89.0	-11.6	0.023
C4
Low CXCL2 (n=18)	18.1	0.66	20.2	2.1	0.080	0.026
High CXCL2 (n=7)	20.0		17.3	-2.7	0.096

 Conclusion: The lack of correlation between levels of CXCL2 and CXCL10 strongly supports that these chemokines are driven by distinct mechanistic pathways in SLE. Although the IFN pathway is likely to be responsible of changes in CXCL10 levels in SLE, the mechanism(s) that regulate CXCL2 in SLE is unclear. High CXCL2 is predictive of increasing disease activity over a mean of 0.7 marks, by both PGA and SLEDAI, and reduction in C3 and C4.  Neither chemokine, however, significantly changed from baseline to second visit.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/elevated-plasma-levels-of-cxcl2-and-cxcl10-have-distinct-predictive-value-in-systemic-lupus-erythematosus/