Background/Purpose: The efficacy
and safety of anifrolumab (ANIFR), a type I IFN receptor antagonist, were
assessed in a Phase II, randomized, double-blind, placebo-controlled study in
SLE (NCT01438489).

Methods: Three hundred and
five adults with seropositive moderate to severe SLE despite standard of care
medication were randomized and received intravenous (iv)
ANIFR (300 mg, 1000 mg) or placebo (PBO) every 4 weeks for 48 weeks. Patients were
stratified by SLEDAI score, oral corticosteroid (OCS) dose, and IFN gene
signature (IFN high vs. IFN low) based on a 4-gene expression assay. Disease
activity was assessed by SLEDAI-2K, BILAG 2004, Physician’s Global Assessment, CLASI,
BICLA, and 28-joint count. The primary endpoint was a composite of SRI(4) response at Day 169 with sustained reduction of OCS (<10
mg/day and ≤Day 1 dose maintained between Days 85 and 169). Two secondary
efficacy endpoints were assessed at Day 365: A) Composite of SRI(4)
response at Day 365 with sustained reduction of OCS maintained between Days 281
and 365; B) Reduction of OCS to ≤7.5 mg/day by Day 365 in those taking ≥10
mg/day at baseline. Other efficacy measures of systemic and organ-specific
disease activity and safety were also assessed at Day 365.

Results: The primary
endpoint at Day 169 was met by a greater proportion of ANIFR-treated patients vs.
PBO (PBO: 17.6%; 300 mg: 34.3%, p=0.014;
1000 mg: 28.8%, p=0.063). At Day 365,
the secondary SRI endpoint was met by 25.5% of PBO, 51.5% of 300 mg (p<0.001) and 38.5% of 1000 mg (p=0.048) patients. OCS reduction to
≤7.5 mg/day at Day 365 was achieved by 26.6% of PBO, 56.4% of 300 mg (p=0.001) and 31.7% of 1000 mg (p=0.595) patients. ANIFR demonstrated consistent
benefit across multiple global and organ-specific measures at Day 365 (Table),
as well as lower rates of BILAG moderate/severe flares/patient-year (PBO: 0.611;
300 mg: 0.266; 1000 mg: 0.391). 75% of patients were IFN high at baseline. The ANIFR
efficacy observed in the entire cohort was similar or more pronounced in IFN
high patients. Median suppression of 21 IFN-regulated genes was ~90% for both
doses of ANIFR. Serious adverse events were reported in 18.8% of patients on PBO
and 16.7% of patients in the pooled ANIFR groups. A higher frequency of
influenza (most unconfirmed) (PBO: 1.0%; 300 mg: 6.1%; 1000 mg: 7.6%) and a dosage-dependent
increase in Herpes zoster (PBO: 2.0%;
300 mg: 5.1%; 1000 mg: 9.5%) occurred in the ANIFR arms.

Conclusion: Anifrolumab
significantly reduced disease activity compared with PBO across multiple clinical
endpoints. Enhanced effects in IFN high patients support the pathobiology of
this treatment strategy. The lack of dose response can be explained by the
nearly similar degrees of IFN gene signature inhibition achieved with the two anifrolumab
doses. These data strongly support further development of anifrolumab.

Acknowledgements: Funded by MedImmune. Editorial Assistance:
S Cotterill, QXV Comms, an Ashfield business, UK.