Background/Purpose: The interaction of inducible T cell costimulatory ligand (ICOSL) and the ICOS receptor is key in the pathogenesis of primary Sjögren’s syndrome (pSS). MEDI5872 interferes with this pathway by binding to ICOSL. We evaluated the safety and efficacy of MEDI5872 in patients (pts) with pSS.

Methods: We conducted a Phase 2a, randomized, placebo (PBO)-controlled study (NCT02334306) in pts with active pSS (according to American-European Consensus Group criteria for pSS) and active disease defined by EULAR Sjögren’s syndrome disease activity index (ESSDAI) ≥6 and markers of B-cell hyperactivity (target enrollment N=42). Pts were randomized (1:1) to MEDI5872 210 mg or PBO subcutaneously (SC) once weekly for 3 weeks, then every 2 weeks for 9 weeks. Beginning on Day 99, all pts received MEDI5872 210 mg for an additional 12 weeks “open-label period” (OLP). The primary endpoint was mean change in ESSDAI score from baseline to Day 99. Secondary endpoints at Day 99 included: proportion of ESSDAI responders (≥3 points decrease in ESSDAI, no discontinuation of investigational product and no use of prohibited medications); changes in ESS patient reported index (ESSPRI) score from baseline; and biomarker endpoints (including changes in plasma cells [PC] and T-follicular helper [Tfh] cells in the peripheral blood [PB] and minor salivary glands [MSG], and changes in MSG biopsy focus score). Duration of response was assessed during the OLP. P-values were not adjusted for multiplicity; P< 0.1 was considered statistically significant.

Results: Baseline characteristics were similar between the MEDI5872 (n=16) and PBO (n=16) arms (enrollment stopped early due to slow recruitment). Mean (SD) ESSDAI score was 11.8 (5.4) and 11.6 (4.5); and mean (SD) ESSPRI score was 6.5 (2.2) and 6.4 (1.8) for MEDI5872 and PBO, respectively. At Day 99 mean (SE) ESSDAI score decreased by 3.8 (0.9) and 2.3 (0.8) with MEDI5872 and PBO, respectively (adjusted mean difference (SE) of –1.4 (1.3), P=0.262). ESSDAI responder criteria were met by 7/16 and 4/16 patients in the MEDI5872 vs. PBO groups, respectively (P=0.458). Two patients in the PBO group had ≥3 points decrease in ESSDAI but were considered non-responders due to drop-out or use of prohibited medication. No other clinical efficacy measurement or PRO showed significant improvement at Day 99. No further improvement was noticed past Day 99 during the OLP.

On MSG biopsies at Day 99, CD4+/ICOS Tfh-like cells decreased from baseline with MEDI5872 and increased from baseline in the PBO group (geometric mean ratio to baseline (SE), MEDI5872 vs. PBO: 0.75 (0.18) vs. 1.76 (0.21), P=0.008). There was no difference in change from baseline in total PC or PD1/ICOS+ Tfh-like cells. No statistically significant differences were observed in PBPC and Tfh cell levels in the blood. IgA, IgG and IgM rheumatoid factor (RF) levels decreased with MEDI5872 but not with PBO; the difference compared to PBO was statistically significant at Day 99. Adverse event rates were balanced: 11 pts (68.8%) with MEDI5872 vs. 14 pts (87.5%) with PBO experienced ≥1.

Conclusion: In patients with active pSS, despite decreasing the level of RF, MEDI5872 210 mg did not achieve consistent improvement of clinical or other biomarker measures of disease activity.

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/a-phase-2a-study-of-medi5872-amg557-a-fully-human-anti-icos-ligand-monoclonal-antibody-in-patients-with-primary-sjogrens-syndrome/