The drug discovery process is demanding and time-consuming, and machine learning-based research is increasingly proposed to enhance efficiency. A significant challenge in this field is predicting whether a drug molecule’s structure will interact with a target protein. A recent study attempted to address this challenge by utilizing an encoder that leverages prior knowledge of molecular and protein structures, resulting in notable improvements in the prediction performance of the drug-target interactions task. Nonetheless, the target encoders employed in previous studies exhibit computational complexity that increases quadratically with the input length, thereby limiting their practical utility. To overcome this challenge, we adopt a hint-based learning strategy to develop a compact and efficient target encoder. With the adaptation parameter, our model can blend general knowledge and target-oriented knowledge to build features of the protein sequences. This approach yielded considerable performance enhancements and improved learning efficiency on three benchmark datasets: BIOSNAP, DAVIS, and Binding DB. Furthermore, our methodology boasts the merit of necessitating only a minimal Video RAM (VRAM) allocation, specifically 7.7GB, during the training phase (16.24% of the previous state-of-the-art model). This ensures the feasibility of training and inference even with constrained computational resources.

A lack of information regarding whether genetic polymorphisms of SLCO1B1 and ABCG2 affect the pharmacokinetics (PKs)/pharmacodynamics (PDs) of rosuvastatin in elderly subjects prevents optimal individualized pharmacotherapy of rosuvastatin in clinical settings. This study aimed to investigate the effect of age and genetic polymorphisms and possible differences in genetic effects on the PKs/PDs of rosuvastatin between elderly and young subjects. Two separate clinical studies designed as open-label, one-sequence studies with multiple-dose administration for elderly (n = 20) and young (n = 32) subjects were conducted. All subjects received 20 mg of rosuvastatin once daily for 21 days. The exposure to rosuvastatin, characterized by the area under the time curve (AUC), increased by 23% in the elderly subjects compared with that of young subjects, which was not significant. When compared to the subjects with breast cancer resistance protein (BCRP) normal function, the exposure to rosuvastatin increased by 44% in young subjects (p = 0.0021) with BCRP intermediate function (IF) and by 35% and 59% (p > 0.05 for both) in elderly subjects with BCRP IF and low function, respectively. SLCO1B1 521T > C was also partially associated with a higher AUC of rosuvastatin in young subjects and a less pronounced increasing trend in elderly subjects (p > 0.05 for both). The lipid-lowering effect of rosuvastatin was less pronounced in the elderly subjects than in the young subjects, and genetic polymorphisms of neither SLCO1B1 nor ABCG2 significantly affected the PDs of rosuvastatin. The ABCG2 421C > A polymorphism was associated with the PKs of rosuvastatin and was identified as a more important determinant than the SLCO1B1 521T > C polymorphism in both elderly and young subjects.

Sarcopenia is characterized by a skeletal muscle disorder with progressive and generalized loss of muscle mass and function, and it increases the risk of adverse outcomes with considerable prevalence in patients with chronic liver disease. Sarcopenia in chronic liver disease underlies complicated and multifactorial mechanisms for pathogenesis, including alterations in protein turnover, hyperammonemia, energy disposal, hormonal changes, and chronic inflammation. The key contribution to sarcopenia in patients with chronic liver diseases can be the hyperammonemia-induced upregulation of myostatin, which causes muscle atrophy via the expression of atrophy-related genes. Several clinical studies on emerging treatment options for sarcopenia have been reported, but only a few have focused on patients with chronic liver diseases, with mostly nutritional and behavioral interventions being carried out. The inhibition of the myostatin-activin receptor signaling pathway and hormonal therapy might be the most promising therapeutic options in combination with an ammonia-lowering approach in sarcopenic patients with chronic liver diseases. This review focuses on current and emerging treatment options for sarcopenia in chronic liver diseases with underlying mechanisms to counteract this condition.

Highly variable and non-linear pharmacokinetics of voriconazole are mainly caused by CYP2C19 polymorphisms. This study aimed to develop a mechanistic population pharmacokinetic model including the CYP2C19 phenotype, and to assess the appropriateness of various dosing regimens based on the therapeutic target. A total of 1,828 concentrations from 193 subjects were included in the population pharmacokinetic analysis. A three-compartment model with an inhibition compartment appropriately described the voriconazole pharmacokinetics reflecting auto-inhibition. Voriconazole clearance in the CYP2C19 intermediate metabolizers (IMs) and poor metabolizers (PMs) decreased by 17% and 53% compared to that in the extensive metabolizers (EMs). There was a time-dependent inhibition of clearance to 16.2% of its original value in the CYP2C19 EMs, and the extent of inhibition differed according to the CYP2C19 phenotypes. The proposed CYP2C19 phenotype-guided initial dosing regimens are 400 mg twice daily (bid) for EMs, 200 mg bid for IMs, and 100 mg bid for PMs. This CYP2C19 phenotype-guided initial dosing regimen will provide a rationale for individualizing the optimal voriconazole therapy.

Baloxavir marboxil, a novel influenza therapeutic agent, is a prodrug rapidly metabolized into its active form, baloxavir acid, which inhibits cap‐dependent endonuclease. This study evaluated the pharmacokinetics (PKs) and safety of baloxavir acid in healthy Korean subjects and compared them with published data in Japanese subjects. This open‐label and single‐ascending dose study was conducted in 30 Korean male subjects, with a single oral dose of baloxavir marboxil (20, 40, or 80 mg) administered to eight subjects each; additionally, 80 mg was administered to six subjects (body weight >80 kg). Noncompartmental and population PK analyses were performed, and results were compared with those of Japanese subjects. Appropriateness of the body weight‐based dosing regimen was evaluated by simulation. PK profiles of baloxavir acid revealed multicompartment behavior with a long half‐life (80.8–98.3 h), demonstrating a dose‐proportional increase. Baloxavir acid reached peak plasma concentration from 3.5 to 4.0 h postdosing. Body weight was identified as a significant covariate of apparent oral clearance and apparent volume of distribution, which was similar to that observed in Japanese subjects. Body weight‐adjusted analysis revealed that exposure to baloxavir acid did not significantly differ between Korean and Japanese subjects. Simulated exposures to baloxavir acid demonstrated that the body weight‐based dosing regimen for baloxavir marboxil was appropriate. Based on a PK study, clinical data including dosing regimen developed in Japan were adequately extrapolated to Korea, supporting the approval of baloxavir marboxil in Korean as a new treatment option for influenza.