Artificial intelligence (AI) is transforming how scientific hypotheses are formulated, tested, and applied. Nucleic acid therapeutics can benefit from emerging opportunities but will also face new challenges. AI lowers the barrier to hypothesis generation, increasing the emphasis on critical evaluation, rigorous experimental testing, and expert judgement. The goal of this perspective is to begin that discussion, with an emphasis on basic and applied research at the interface of nucleic acid science and medicine.
Some, but not all, gapmer antisense oligonucleotides (ASOs) delivered into cerebrospinal fluid exhibit fatal subacute toxicity in late-stage prion-infected mice but not in uninfected mice. Here, we performed a battery of in vivo studies demonstrating that the toxicity is compound- and disease- stage-dependent, and not predicted by behavioral or molecular measurements in uninfected animals. A panel of toxic ASOs all elicited a consistent transcriptomic response signature in prion-infected mice, which emerged over the course of 7 days post dose. This signature, driven chiefly by decreased expression of a specific set of astrocyte-expressed genes, was absent in uninfected mice receiving the same ASOs, was absent, less intense, and/or more transient in infected mice receiving tolerated ASOs, and was distinct from the transcriptomic effect of prion disease itself. Our data argue that the toxicity is an interaction between ASO and prion disease at high doses for specific compounds, and does not represent an exacerbation of ASO toxicity nor of prion disease.
Oligonucleotide-based therapeutics (ONTs) present unique nonclinical development challenges that are not fully addressed by paradigms established for small molecule pharmaceuticals or biologic therapeutics. To better understand current industry practice, the IQ DruSafe ONT Working Group conducted an industry survey examining species selection and high-dose selection in GLP repeat-dose toxicology studies, parameters used to determine safety margins, and human equivalent dose (HED) scaling used to support initial dose selection for first-in-human studies. Twenty companies provided responses, largely reflecting experience with systemically administered siRNA and RNase H antisense ONTs. Common themes identified in the survey included frequent use of both rodent and non-rodent species, with nonhuman primates used extensively, and high-dose selection based on an integrated consideration of maximum tolerated dose, pharmacodynamic response, and exposure-based criteria. Safety margins were most often based on steady-state exposure, generally using plasma rather than tissue exposure. For HED scaling, body surface area was the predominant approach, although some respondents reported use of body weight- or platform-informed alternatives. Overall, the survey identifies areas of convergence in current ONT nonclinical practice while highlighting unresolved areas where nonclinical practices diverge. The survey also identified areas where future ONT-specific regulatory guidance could improve consistency across ONT development programs.
Extravascular administration of mRNA lipid nanoparticles (mRNA-LNPs) offers practical and clinical advantages but is limited by tissue barriers that restrict dispersion, lymphatic access, systemic exposure, and ultimately, target tissue expression. Recombinant human hyaluronidase PH20 (rHuPH20) transiently modifies the hyaluronan-rich extracellular matrix in the subcutaneous (SC) space, enhancing SC delivery of protein therapeutics; however, its effect on mRNA-LNP delivery has not been systematically evaluated. Here, the impact of rHuPH20 on extravascular mRNA-LNP delivery was assessed across routes of administration and formulations. In a coordinated series of mouse and minipig studies, mRNA-LNP expression, biodistribution, lymphatic trafficking, systemic exposure, tissue deposition, and inflammatory responses were evaluated following SC, intramuscular (IM), and intravenous (IV) administration. In mice, co-administration with rHuPH20 increased local expression after IM dosing of luciferase mRNA-LNPs up to 7-fold and whole-body expression after SC dosing up to 38-fold, relative to the corresponding route without rHuPH20. In minipigs, rHuPH20-enabled SC administration of huEPO mRNA-LNPs produced serum huEPO exposure that approached IV benchmarks. rHuPH20 also increased hepatic expression following extravascular delivery, demonstrated consistent effects across ionizable lipid chemistries, and was associated with reduced pro-inflammatory cytokine responses following repeat-dosing. Collectively, these findings support rHuPH20 as an effective enabler of extravascular mRNA-LNP delivery.
Rondaptivon pegol (BT200) is a PEGylated RNA aptamer that prolongs von Willebrand factor (VWF) and factor VIII half-lives in patients with hemophilia A and von Willebrand disease (VWD) type 2B. Embryo-fetal safety is particularly relevant given the disproportionate disease burden in women with VWD. Data on placental transfer and developmental safety of PEGylated aptamers remain limited. An embryo-fetal development study was conducted in pregnant rabbits. Animals received rondaptivon pegol subcutaneously (0, 1.5, 5, or 15 mg/kg/day) from gestational days (GDs) 7-19. Maternal toxicokinetic samples and pooled fetal plasma concentrations were collected during gestation to quantify transplacental exposure. Rondaptivon pegol was well tolerated, with no maternal toxicity, no effects on implantation, and no treatment-related fetal malformations. Dose-related reductions in mean fetal body weight (<11%) were considered adverse only at the highest dose level. Maternal systemic exposure was supratherapeutic, with mean Cmax ranging from 133 to 807 µg/mL on GD19, far exceeding the human target concentration (∼1.2 µg/mL). Fetal plasma concentrations were <0.25% of maternal levels across all dose groups, confirming minimal placental transfer. The no-observed-adverse-effect level was 5 mg/kg/day. At maternal systemic exposures ranging from 110- to 670-fold above anticipated clinical levels, rondaptivon pegol did not adversely affect implantation, embryonic viability, or fetal development. These findings provide the first quantitative maternal-fetal pharmacokinetic and developmental safety data for a PEGylated aptamer, supporting further clinical evaluation of rondaptivon pegol in women of reproductive age.
