Background/Purpose: Despite the heterogeneity of SLE, specific pathways of immune dysregulation characterize significant subsets of patients. We examined cytokines reflecting IL6-TH17 pathology in lupus as an objective standard to test performance of clinical outcome measures.

Methods: 91 patients, all with active disease at baseline (SLEDAI >/= 6) participated in two visits at which BILAG and SLEDAI were prospectively scored. A clinician retrospectively assessed the second visit as overall worse, the same, or improved. Cytokines were measured by xmap multiplex bead-based assay.

Results: Using physician-rated improvement (PRI), 68 patients improved and 23 were same or worse at follow up. Cytokine changes were evaluated in patients who did or did not meet response criteria of PRI, SRI (SLE Responder Index) and BICLA (BILAG-based Composite Lupus Assessment). Of patients with detectable IL6 at baseline and/or follow up (IL6+) (n=51), those who improved by PRI, SRI or BICLA had a significant drop in IL6, not observed in patients who did not improve (Table 1). IL21 and IL23 decreased in those who improved by PRI, but did not track with SRI or BICLA. IL17 was only detected in 3 patients and could not be evaluated.

IL21 and IL23 were detectable when IL6 was, both at baseline (p<0.001 and p=0.002 respectively) and at follow up (p=0.032 and p=0.006). In IL6+/IL21+ (double positive) patients (n=35), IL21 tracked well with IL6 (p=0.0028), driven by the patients who improved by PRI (n=26) (p=0085).

In IL6+ patients the PRI was tested as a standard to compare performance of SRI and BICLA (sensitivities 94.9% and 74.4%, specificities 58.3% and 75%). Using decrease in IL6 as an objective standard of improvement, sensitivities of SRI vs BICLA were 85.2% vs 66.7% and specificities were 35.9% vs 39.1%.

Conclusion: Decreases in IL6, IL21 and IL23 reflected improvements by SRI, BICLA and clinical judgement. IL21 and IL23 were detectable when IL6 was, consistent with expected relatedness in an IL6-TH17 pathway. The SRI was more sensitive than and as specific as the BICLA in detecting objective biologic change in this subset of IL6 producing patients. Low specificities underscore the multiple pathways of lupus pathology, but we speculate that a range of predictable cytokine changes in definable patient subsets could be formulated into objective standards for testing clinical outcome instruments.