Background/Purpose: The interferon alpha (IFN) gene signature is frequent in SLE (~50% of patients) and important in pathogenesis.  However, multiple studies have found that the IFN signature does not predict flares.  In SLE peripheral blood, the level of the B-cell activating factor (BAFF) transcript is highly associated with the IFN signature.  We asked whether an association of the interferon gene signature with disease activity depended on the level of the BAFF transcript. 

Methods:

292 patients (59% Caucasian, 34% African-American, 92% female, mean age 46 ±12 years) were enrolled in a prospective observational study. At baseline, gene expression levels were assessed in peripheral blood RNA samples using microarray (Affymetrix) and select transcripts validated with qPCR. The IFN signature “score” was calculated based on the geometric mean of the expression levels (chip signal intensity) of 16 IFN-induced genes and divided into low, medium and high groups. Patients were divided in subgroups based on BAFF and IFN signatures.  Subgroups were compared with respect to frequency of clinical events in the next year.  P-values were calculated using generalized estimating equations (SAS 9.2).

Results:

Table 1 shows the strong correlation of the IFN signature with BAFF gene expression. Of those individuals with a high level of BAFF mRNA, 97% also had a high IFN signature. 

Table 1:  Number (%) of SLE Patients with Each Level of IFN Signature, by BAFF Levels

		IFN signature
		Low	Medium	High
BAFF mRNA	Low (n=91)	77 (85%)	7 (8%)	7 (8%)
	Medium (n=88)	38 (43%)	8 (9%)	42 (48%)
	High (n=93)	1 (1%)	2 (2%)	90 (97%)

Because there were only 17 patients with medium IFN, they were pooled together with high IFN in Table 2 below. 

Table 2: Proportion of Visits with Various Disease Manifestations, by IFN within BAFF Subgroups over 1 Year

 

Disease activity	Low BAFF		Medium BAFF		High BAFF		P-value for association between IFN and item, controlling for BAFF	P-value for association between BAFF and item, controlling for IFN
	Low IFN (n=77) (304 total visits)	High IFN (n=14) (67  total visits)	Low IFN (n=38) (159 total visits)	High IFN (n=50) (210 total visits)	Low IFN (n=1) (4 total visits)	High IFN (n=92) (381 total visits)
Physician global assessment >1	8%	3%	24%	20%	0%	26%	0.37	0.0008
SLEDAI ≥2	34%	27%	55%	65%	0%	69%	0.55	0.0002
Urine protein/cr (≥0.5)	3%	0%	13%	14%	0%	14%	0.83	0.012
Anti-dsDNA ≥ 10	7%	7%	7%	37%	0%	36%	0.0070	0.074
C3 <79 mg/dl	3%	0%	4%	20%	20%	21%	0.018	0.028
C4 <12 mg/dl	3%	0%	7%	18%	18%	20%	0.14	0.023
ESR >20	32%	45%	50%	62%	0%	61%	0.053	0.068
SLEDAI Arthritis	1%	3%	6%	3%	0%	5%	0.84	0.29
SLEDAI Skin	19%	23%	29%	36%	0%	36%	0.28	0.19
SLEDAI Heme	3%	5%	2%	6%	0%	6%	0.27	0.88

In those patients with low or medium BAFF gene expression, there was no association between the IFN signature and global disease activity measures (such as PGA, SLEDAI).  Thus, controlling for BAFF, the IFN signature was not related to global measures of disease activity (p=0.37 for PGA, p=0.55 for SLEDAI).

Conclusion: In this study, we found that the association of the IFN signature with SLE global disease parameters is dependent on the BAFF gene expression (with which it is closely correlated). The IFN gene signature, however, is associated with serologic activity (high anti-dsDNA and low complement), controlling for BAFF.  These results may shed light on confusing data from anti-IFN clinical trials.

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/correlation-of-the-interferon-gene-signature-with-systemic-lupus-erythematosus-disease-activity-is-dependent-on-the-associated-level-of-the-baff-gene-transcript/