Background Cutaneous lupus eyrthematosus (CLE) is frequently refractory to immunosuppressive therapies including B-cell depletion, but this varies by morphology with the chronic discoid (DLE) subtype being particularly resistant. Local production and response to type-I interferon (IFN-I) is implicated in all subtypes of CLE. Therapeutic blockade of the IFN-I receptor with anifrolumab has direct effects on IFN-I signaling, and subsequent more widespread effects on other immune functions regulated by IFN-I.

Response to anifrolumab by lesion subtype have not been described, and it is unclear which effects of IFN-blockade are responsible for cutaneous response. We hypothesise that the efficacy of anifrolumab will differ dependent on the relative contribution of direct IFN-I effects vs. the downstream immunostimulatory effects of IFN-I on other immune functions.

Objectives To evaluate the effect of anifrolumab on (i) rituximab-refractory CLE; (ii) on DLE; (iii) to compare clinical responses with IFN-specific biomarkers and transcriptomic evaluation of broader immune responses; (iv) to compare early and late immunophenotypic and clinical responses.

Methods SLE patients with active recalcitrant CLE received anifrolumab 300 mg IV every 4 weeks and evaluated using the Cutaneous lupus erythematosus disease area and severity index (CLASI) and dermatology life quality index (DLQI). Fluorescence intensity of tetherin (CD317), a cell surface interferon biomarker, was evaluated by multiparameter flow cytometry of peripheral blood mononuclear cells (PBMCs). Previously validated IFN-Scores-A and B, in addition to gene expression scores annotated to Inflammation, Myeloid lineage and Plasmablasts modules [3], were measured in PBMCs using customised Taqman array at serial time points.

Results 7 patients (DLE n=5, chillblain/nodular vasculitis n= 1, subacute CLE n=1) have commenced therapy. Median number of previously failed standard therapies is 6, including rituximab in 6/7 patients, belimumab in 2/7 and thalidomide in 4/7. Three patients required long-term oral prednisolone >10 mg daily. Median baseline CLASI activity score was 17 and DLQI was 17/30.

Rapid clinical responses were evident at 1 month, with more rapid effects observed in patients with SCLE and DLE compared with chillblain lesions. Median fall in CLASI activity score at 1 month was 6 points with a median percentage change from baseline of 31%. In all patients, a rapid and marked suppression of IFN-Score-A (mean difference 2.92, p<0.01) and plasmablast tetherin (p=0.01), was evident by 1 month. Small and variable downward trends were observed in Inflammation- and IFN-Score-B (p=0.06), Myeloid (p=0.27) and Plasmablast (p=0.15) -annotated gene expression scores. Major cell population numbers were proportionally unaltered in flow cytometry.

Conclusions These preliminary results suggest that anifrolumab: (i) may be effective in rituximab-resistant CLE, (ii) is effective in DLE; (iii) rapidly suppresses IFN-I response, but with lesser effects on non-IFN immune biomarkers and (iv) early direct effects on IFN-I are associated with rapid clinical response.