Background Nirmatrelvir, in combination with ritonavir, is one of the first orally available antiviral treatment for coronavirus disease 2019 (COVID‐19). Symptomatic pregnant women are at increased risk for severe illness and complications that can affect the developing baby. No malformations or lower embryo‐fetal survival have been observed when nirmatrelvir were administered to pregnant rats and rabbits. Safety evaluation of drugs used for treating COVID‐19 also in pregnancy is urgent for public health, then in this study we further investigated nirmatrelvir developmental toxicity using zebrafish as in vivo model. Material and Methods Using the standardized Fish Embryo Toxicity (FET) test, we first determined the lethal concentration 50 (LC50), exposing embryos from gastrula stage up to 120 hr post fertilization (hpf) and daily recording lethality. Then, we exposed embryos to five doses comprising the human therapeutic one and up to the LC50 (25 μM). Morphology was evaluated at 72 and 120 hpf. Results Nirmatrelvir did not affect survival rate and did not induce morphological defects up to the human therapeutic dose. Exposure at higher doses (2.4× and 3× the human Cmax) however resulted in decreased hatching rate, reduced growth, slower heartbeat with pericardial edema, reduction of eye dimension, absence of the swim bladder and disruption of the anterior–posterior axis, with lack of tail detachment, spinal curvature and straight and smaller head. Conclusions Our findings in zebrafish embryos add further information about developmental nirmatrelvir safety. Further studies are needed for pharmacological safety assessment of nirmatrelvir exposure during pregnancy.

In the past, one of the most widely used non-nucleoside reverse transcriptase inhibitors (NNRTI) in first-line antiretroviral therapy (ART) of HIV infection was efavirenz (EFV), which is already used as a cost-effective treatment in developing countries due to its efficacy, tolerability, and availability. However, EFV also demonstrates several adverse effects, like hepatotoxicity, altered lipid profile, neuropsychological symptoms, and behavioral effects in children after in utero exposure. In 2018, another NNRTI, doravirine (DOR), was approved due to its similar efficacy but better safety profile. Preclinical safety studies demonstrated that DOR is not genotoxic and exhibits no developmental toxicity or effects on fertility in rats. Zebrafish (Danio rerio) embryos have been widely accepted as a vertebrate model for pharmacological and developmental studies. We used zebrafish embryos as an in vivo model to investigate the developmental toxicity of DOR compared to EFV. After exposure of the embryos to the drugs from the gastrula stage up to different developmental stages (30 embryos for each arm, in three independent experiments), we assessed their survival, morphology, hatching rate, apoptosis in the developing head, locomotion behavior, vasculature development, and neutral lipid distribution. Overall, DOR showed a better safety profile than EFV in our model. Therapeutic and supra-therapeutic doses of DOR induced very low mortality [survival rates: 92, 90, 88, 88, and 81% at 1, 5, 10, 25, and 50 μM, respectively, at 24 h post fecundation (hpf), and 88, 85, 88, 89, and 75% at the same doses, respectively, at 48 hpf] and mild morphological alterations compared to EFV exposure also in the sub-therapeutic ranges (survival rates: 80, 77, 69, 63, and 44% at 1, 5, 10, 25, and 50 μM, respectively, at 24 hpf and 72, 70, 63, 52, and 0% at the same doses, respectively, at 48 hpf). Further, DOR only slightly affected the hatching rate at supra-therapeutic doses (97, 98, 96, 87, and 83% at 1, 5, 10, 25, and 50 μM, respectively, at 72 hpf), while EFV already strongly reduced hatching at sub-therapeutic doses (83, 49, 11, 0, and 0% at 1, 5, 10, 25, and 50 μM, respectively, at the same time endpoint). Both DOR at therapeutic doses and most severely EFV at sub-therapeutic doses enhanced apoptosis in the developing head during crucial phases of embryo neurodevelopment and perturbed the locomotor behavior. Furthermore, EFV strongly affected angiogenesis and disturbed neutral lipid homeostasis even at sub-therapeutic doses compared to DOR at therapeutic concentrations. Our findings in zebrafish embryos add further data confirming the higher safety of DOR with respect to EFV regarding embryo development, neurogenesis, angiogenesis, and lipid metabolism. Further studies are needed to explore the molecular mechanisms underlying the better pharmacological safety profile of DOR, and further human studies are required to confirm these results in the zebrafish animal model.