In-Silico Aided Screening and Characterization Results in Stability Enhanced Novel Roxadustat Co-Crystal.

• ΔpK a , Molecular Complementarity (MC), Molecular Electrostatic Potential (MEP) Maps and Machine Learning approaches were used for the coformer's selection. • MEPS model and Machine learning blend model using MACCs descriptors appears to be the top-performing models for the coformer selection. • Experimental co-crystal screening resulted in Roxadustat-Nicotinamide (RXD-NA) co-crystal, confirmed by single crystal structure, solid-state FTIR and NMR analysis. • Physical and chemical stability of RXD-NA co-crystal was established. • The phenomenon of photo isomer impurity generation was evaluated by computational analysis. Roxadustat (RXD) is an approved drug substances for the treatment of renal anemia. It has poor aqueous solubility and photochemical stability. This study employs a comprehensive approach to enhance the stability and physicochemical properties RXD through coformer selection and characterization. The investigation integrates delta pKa analysis, molecular complementary assessment, molecular electrostatic potential surface analysis, and machine learning techniques to predict potential co-crystal formation and binding interactions between drug molecules and coformers. The co-crystal screening which lead to in a novel RXD-nicotinamide co-crystal (RXD-NA). Experimental characterization underscores the physical and chemical stability of the co-crystals. To elucidate the supramolecular synthons and understand the intermolecular interactions in the RXD-NA co-crystal, Hirshfeld surfaces analysis, quantum theory of atoms in molecules (QTAIM) analysis and non-covalent interaction (NCI) analysis were performed. Computational analysis of photo-isomer formation aligns with experimental observations, further enhancing our understanding of RXD-coformer interactions. RXD-NA co-crystal was found photo-chemically stable as compared to free base API drug substance. This integrated methodology provides a systematic framework for informed co-crystal design, holding promise for optimizing RXD formulations based on molecular interactions and stability considerations. Consequently, this study contributes valuable insights to the field of rational drug design and formulation optimization. [ABSTRACT FROM AUTHOR]