Short oligonucleotide therapeutics, including antisense oligonucleotides and siRNAs, are an emerging class of biopharmaceuticals to treat a wide variety of diseases, many of which are inaccessible to small molecule- and protein-based approaches. An increase in FDA approvals over the past few years has fueled interest in this molecular class, yet there remains uncertainty on how to define them structurally, since they fall into a regulatory gray area in between small molecules and biologics. Although high-resolution NMR spectroscopy is increasingly being used to characterize the structure of protein-based biologics, to date this technology is only sparingly applied to short oligonucleotide therapeutics. Here, we demonstrate NMR structural fingerprinting methods using a model antisense oligonucleotide, a model siRNA, and a simulated version of the oligonucleotide drug product givosiran. Unlike current analytical methods that involve a combination of techniques with experimental conditions deviating significantly from the formulated state, we show that the NMR fingerprinting methods can be performed on fully formulated oligonucleotides. In addition to 1D and 2D NMR experiments that correlate 1H, 13C, and 15N, the non-native chemistries of this molecular class allowed for more diverse fingerprinting strategies using the NMR active handles of 19F and 31P. We further examine the sensitivity of the NMR fingerprinting methods through forced degradation studies, and applied chemometric strategies to tease out important spectral changes. This benchmarking study provides a pathway for implementation of 2D NMR fingerprinting as a powerful method for the characterization of oligonucleotide therapeutic in the biopharmaceutical laboratory to support assessment of quality attributes at the atomic level, providing greater confidence in the overall chemical and structural integrity of drug substance.
