Background: New precision medicine guided therapies targeting disease drivers are required to improve clinical outcomes in rheumatoid arthritis (RA). Anti-citrullinated protein autoantibodies (ACPA) promote inflammation and joint tissue injury and define a poor prognostic group of patients with RA. Citrullinated autoantigens that drive this autoimmune response are generated by Peptidyl Arginine Deiminase (PAD) enzymes which are predominately expressed and released by neutrophils. Accordingly, blocking the enzymatic activity of PADs and thereby the generation of autoantigens that drive autoimmunity and tissue injury may provide clinical benefit.

Objectives: To develop a first-in-class precision medicine guided therapy targeting PADs for the treatment of RA.

Methods: PAD2 and PAD4 protein levels were measured by ELISA. Anti-PAD2 and anti-PAD4 antibodies were generated by phage display technology. Optimized high affinity inhibitory antibody sequences were cloned into bi-specific expression vectors and mutations were integrated into the Fc region to promote FcRn recycling and to suppress C1q and FcγR-mediated effector functions. The resultant anti-PAD2/4 Ab AZD1163 was evaluated for binding to recombinant PADs using surface plasmon resonance. Inhibition of recombinant and endogenous PAD activity was measured using a Histone H3 citrullination assay. In vitro assays were used to evaluate the impact of AZD1163 on phagocytosis with labelled S. aureus bioparticles, ROS production, NETosis and cytokine release. Single and multi-dose studies in non-human primates were used to evaluate the pharmacokinetics, pharmacodynamics and safety profile of this novel antibody.

Results: PAD2 and PAD4 levels and PAD enzymic activity are increased in RA patient serum compared to healthy controls. Using phage display, we generated the novel anti-PAD2/4 bi-specific antibody AZD1163 and showed that it binds with high affinity to recombinant human PAD2 (6.76 ± 4 pM) and PAD4 (34 ± 7 pM). AZD1163 inhibited all endogenous PAD activity in the serum and synovial fluid of RA patients. PAD2 and PAD4 were detected on the surface of neutrophils and monocytes from healthy controls. Binding of AZD1163 to PADs on neutrophils did not induce proinflammatory cytokine induction or otherwise impact normal neutrophil phagocytosis, ROS production or NETosis. In a non-human primate PK/PD study, a single 5 mg/kg dose of AZD1163 was sufficient for a rapid and durable suppression of endogenous PAD activity for eight weeks. In plasma samples spiked with 400 ng of PAD4 to mimic the higher levels detected in some patients’ serum, AZD1163 completely inhibited PAD activity for 4 weeks.

Conclusion: We describe generation of a novel potent bi-specific anti-PAD2/4 antibody (AZD1163) that inhibits neutrophil PADs, the enzymes responsible for the generation of citrullinated autoantigens in RA. Moreover, we found no evidence for neutrophil activation by this bi-specific antibody. In non-human primates, AZD1163 exhibited favourable PK/PD properties that merit progression to clinical trials.

REFERENCES: NIL.

Acknowledgements: NIL.

Disclosure of Interests: Dorothy A Sims AstraZeneca, AstraZeneca, John Andrews AstraZeneca, AstraZeneca, Teneema Kuriakose AstraZeneca, AstraZeneca, Lacie Scaletta AstraZeneca, AstraZeneca, Lichchavi Rajasinghe AstraZeneca, AstraZeneca, Anna Sigurdardottir AstraZeneca, AstraZeneca, Martin Strain AstraZeneca, AstraZeneca, Fanyi Jiang AstraZeneca, AstraZeneca, Elin Boger AstraZeneca, AstraZeneca, Liselotte Björsson AstraZeneca, AstraZeneca, Garry Douglas AstraZeneca, AstraZeneca, Nicholas White AstraZeneca, AstraZeneca, Katie Day AstraZeneca, AstraZeneca, Iain B. Mc Innes AstraZeneca, AstraZeneca, Jessica Neisen AstraZeneca, AstraZeneca, David Close AstraZeneca, AstraZeneca, Catherine Huntington AstraZeneca, AstraZeneca, Gary P Sims AstraZeneca, AstraZeneca.