Predicting\nthe Long-Term Stability of Biologics with\nShort-Term Data

Understanding the long-term stability of biologics is\ncrucial to\nensure safe, effective, and cost-efficient life-saving therapeutics.\nCurrent industry and regulatory practices require arduous real-time\ndata collection over three years; thus, reducing this bottleneck while\nstill ensuring product quality would enhance the speed of medicine\nto patients. We developed a parallel-pathway kinetic model, combined\nwith Monte Carlo simulations for prediction intervals, to predict\nthe long-term (2+ years) stability of biotherapeutic critical quality\nattributes (aggregates, fragments, charge variants, purity, and potency)\nwith short-term (3–6 months) data from intended, accelerated,\nand stressed temperatures. We rigorously validated the model with\n18 biotherapeutic drug products, composed of IgG1 and IgG4 monoclonal\nantibodies, antibody–drug conjugates, dual protein coformulations,\nand a fusion protein, including high concentration (≥100 mg/mL)\nformulations, in liquid and lyophilized presentations. For each drug\nproduct, we accurately predicted the long-term trends of multiple\nquality attributes using just 6 months of data. Further, we demonstrated\nsuperior stability prediction via our methods compared with industry-standard\nlinear regression methods. The robust and repeatable results of this\nwork across an unprecedented suite of 18 biotherapeutic compounds\nsuggest that kinetic models with Monte Carlo simulation can predict\nthe long-term stability of biologics with short-term data.