Advances in antisense oligonucleotide (ASO) chemistry and screening have enabled the design and selection of molecules that are optimized for a particular therapeutic application in terms of both potency and tolerability. The most-well studied of the chemically modified ASOs are single-stranded antisense inhibitors with phosphorothioate backbones and 2'-O-methoxyethyl modifications (2'-MOE ASO). The 2'-MOE chemical modification in the design of the ASO has conferred increased hybridization affinity, increased stability, and/or enhanced tissue residence time, resulting in better potency and pharmacokinetics. Compound screening and selection are also important in optimizing the tolerability of intended therapeutic antisense inhibitors. In this study, we report the chronic toxicity of multiple 2'-MOE ASOs in mice for several representative compounds that have progressed to later phases of clinical development. The results show that these 2'-MOE ASOs selected for development have consistent behavior between sequences, have tolerability profiles suitable for chronic administration, and exhibit a relative lack of progression of findings observed in subchronic studies in mice.

No abstract

The 6-month Tg.rasH2 mouse carcinogenicity model provides an acceptable alternative to the 2-year carcinogenicity study in CD-1 mice. However, key questions related to the use of this model for testing antisense oligonucleotides (ASOs) include the similarity in the biologic response between mouse strains and the feasibility of using data from the CD-1 mouse to set doses and dose schedules for a Tg.rasH2 carcinogenicity study. To evaluate the potential strain differences, four distinct 2 0-O-(2-methoxyethyl) ASOs were administered to CByB6F1 (wild type), Tg.rasH2 (hemizygous), and CD-1 mice. There were no meaningful differences in clinical signs, body weight, food consumption, or serum chemistry and hematology parameters. Histopathology evaluation indicated little to no difference in the spectrum or magnitude of changes present. The cytokine/chemokine response was also not appreciably different between the strains. This was consistent with the similarity in ASO concentration in the liver between the mouse strains tested. As the class effects of the ASOs were not meaningfully different between CD-1, CByB6F1, or Tg.rasH2 mice, data from nonclinical studies in CD-1 mice can be used for dose selection and expectation of effect in the Tg.rasH2 mouse.

No abstract

Safety assessment of drug impurities is a routine part of the drug development process. For oligonucleotide-based drugs, impurities can arise from impurities in starting materials, as by-products of the manufacturing process or from degradation, and are generally structurally similar to the parent oligonucleotide. To study the potential impact of impurities, a representative batch of a 2¢-O-methoxyethyl (MOE) antisense oligonucleotide (ASO) was compared to batches of drug that were enriched with nine of the common impurities encountered with the chemical class. Mice were treated for 3 months with weekly subcutaneous injection of 10 or 30 mg/kg. The impurity content of the parent batch was 0.25%-2.5% of total drug substance. The enriched impurity mixtures contained from 3% to 10% of the various impurities. The expected common class effects were observed at the 30 mg/kg/week dose level in hematology, serum chemistry, and histopathology. However, there were no differences between the representative batch of material and those enriched with impurities. Based on these data, common oligonucleotide impurity studies do not appear to contribute to the overall toxicology profile.