Model-Based Design of Sustained-Release Formulations of Anti-TNF-α Monoclonal Antibodies for Intravitreal...

Background/Objectives: While intravitreal administration allows for increased ocular exposure to anti-TNF-α monoclonal antibodies, there is still a need for developing delivery systems able to prolong ocular drug exposure and alleviate patient compliance and safety concerns because of repeated injections. Therefore, the objective of this work was to guide the design of sustained-release formulations of anti-TNF-α monoclonal antibodies for intravitreal administration through a model-based strategy in non-infectious uveitis in the preclinical setting. Methods: Using an in-house-developed anterior uveitis disease model in rats, an intravenous reference dose reducing free TNF-α by 90% at the biophase was established. Intravitreal administrations of sustained-release formulations every 24 weeks were then simulated for adalimumab, golimumab and infliximab to evaluate TNF-α kinetics in the anterior chamber of the eye at different release rates. The selected sustained-release formulation was further evaluated for possible formulation issues causing device emptying before the next administration. Results: Intravitreal administration of sustained-release formulations releasing adalimumab, golimumab or infliximab at 1.802, 0.979 and 1.442 μg/week, respectively, met the predefined criteria of ≥90% reduction in free TNF-α at the biophase. TNF-α levels in aqueous humour were anticipated to be the most sensitive to detect possible formulation issues. Formulation emptying 10, 4 or 8 weeks for adalimumab, golimumab and infliximab, respectively, before next administration triggered TNF-α reaching pathological levels at week 24 post-dose. Conclusions: This work underscores the potential of new approach methodologies in the preclinical drug development of sustained-release formulations for intravitreal administration in ocular inflammatory disorders with less animal testing and without compromising the accuracy of model-informed predictions for human translation.