Objective: Currently, limited information is available on the milk transfer properties of drugs when consumed by lactating women. Therefore, we aim to construct a prediction model of milk transfer of drugs using machine learning methods.

Methods: We obtained data from Hale’s Medications & Mothers’ Milk (MMM) and SciFinder^®, and then constructed the datasets. The physicochemical and pharmacokinetic data were used as feature variables with M/P ratio ≥ 1 and M/P ratio < 1 as the objective variables, classified into two groups as the classification of milk transferability. In this study, analyses were conducted using machine learning methods: logistic regression, linear support vector machine (linear SVM), kernel method support vector machine (kernel SVM), random forest, and k-nearest neighbor classification. The results were compared to those obtained with the linear regression equation of Yamauchi et al. from a previous study. The analysis was performed using scikit-learn (version 0.24.2) with python (version 3.8.10).

Results: Model construction and validation were performed on the training data comprising 159 drugs. The results revealed that the random forest had the highest accuracy, area under the receiver operating characteristic curve (AUC), and F value. Additionally, the results with test data A and B (n = 36, 31), which were not used for training, showed that both F value and accuracy for the random forest and the kernel method SVM exceeded those with the linear regression equation of Yamauchi et al.

Conclusion: We were able to construct a predictive model of milk transferability with relatively high performance using a machine learning method capable of nonlinear separation. The predictive model in this study can be applied to drugs with unknown M/P ratios for providing a new source of information on milk transfer.

Objective: Tocilizumab and infliximab are biologic drugs that are widely used for the treatment of rheumatoid arthritis (RA). The dosage of these injectable RA drugs is calculated based on body weight. However, injectable RA drugs are used only once due to stability and sterility concerns. For expensive biologic drugs, drug disposal wastage needs to be reduced. Tocilizumab is approved in three vial sizes: 80, 200, and 400 mg. In this study, we evaluated the validity of these tocilizumab vial sizes to help resolve the issue of excess residual drug.

Methods: A log-normal distribution was assumed for body weight, and 10,000 hypothetical cases were created using the programming language R. We analyzed the average wasted dose rate per vial (%) by gender after considering different vial size combinations.

Results: The average wasted dose rate per vial of tocilizumab was estimated to be 3.7% for males and 4.7% for females.

Conclusion: The three vial sizes of 80, 200, and 400 mg are reasonable for tocilizumab. The average wasted dose rate per vial of infliximab was estimated to be 17.7% for males and 22.6% for females. The average wasted dose rate per vial was lower for tocilizumab than infliximab. Tocilizumab is administered in a dose range of 200 to 1,100 mg with three different vial sizes in multiples of 40 mg. However, infliximab is administered in a dose range of 50 to 400 mg with a single vial size of 100 mg. Multiple vial sizes should be prepared to ensure the efficient use of limited medical resources. It is also expected that the method employed for this hypothetical case model will be applied to other drugs for which disposal wastage is a problem and used to set appropriate vial size combinations.

Objective: Management of low-density lipoprotein cholesterol (LDL-C) and triglyceride (TG) is important for patients with type 2 diabetes merger hyperlipidemia. Pemafibrate (PF) has different characteristics from conventional fibrates. In this study, we retrospectively compared the efficacy and safety of PF and bezafibrate (BF) in patients with type 2 diabetes merger hypertriglyceridemia.

Methods: Patients who were administered PF (0.2 mg/day) or BF (400 mg/day) for 24 weeks or longer were included. Twenty patients in each group were extracted using propensity score matching (PS). PS was calculated using the patient background (before the start of administration) of PF or BF. We investigated lipid-related parameters (TG, high density lipoprotein cholesterol [HDL-C], and LDL-C) and other laboratory test parameters pre administration and 24 weeks post administration.

Results: TG decreased significantly in both groups (p<0.05). However, there were no significant differences between the two groups in the TG treatment target (<150 mg/dL) achievement rate (p =1.00), TG change rate (p=0.84), and TG change amount (p=0.77). In addition, there were no significant changes in HDL-C and LDL-C in both groups. In the PF group, alanine transaminase (ALT) (p< 0.05), alkaline phosphatase (p<0.05) decreased. In the BF group, ALT (p<0.05) and γ-GTP (p<0.05) decreased. Both groups showed improvement in liver function after 24 weeks. eGFR (p<0.05) significantly decreased only BF group. There were no significant changes in renal function, creatine kinase (CK), or hemoglobin A1c (HbA1c) in either group.

Conclusion: Our study suggests that there is no difference in the TG lowering effect and safety of PF and BF in type 2 diabetic patients.

Objective: Two types of symbols have been established as industry standards in terms of two-dimensional (2D) symbols with prescription information: one for objects to be printed on prescriptions and the other for electronic versions of medication diaries. However, no studies have investigated the system for using 2D symbols in pharmacies and hospitals/clinics as well as the quality of the information actually stored in these 2D symbols. Therefore, we conducted a survey to clarify the current status and problems pertaining to prescription information sharing via 2D symbols.

Methods: We distributed questionnaires to community pharmacies through the Fukui Pharmaceutical Association and asked them to cooperate with us during the survey. The list of items in the survey　included the installation status of devices necessary for reading 2D symbols at each pharmacy, receipt computer in use, and status of the support issued by hospitals/clinics for reading 2D symbols. At the same time, we received 2D symbols created by community pharmacies and conducted reading tests to examine issues related to the collection of prescription information via 2D symbols at medical institutions.

Results: The response rate for the survey was 21.8%. Among the 57 stores that responded to the survey, 26 (45.6%) answered that they could read prescription symbols, and 22 of them had actually used the system till date. In addition, 38 community pharmacies were able to provide the 2D symbols for medication diaries. Of the 30 provided symbols for medication diaries, 16 (53.3%) could be read as Japanese data by the barcode reader used.

Conclusions: It has become clear that the 2D symbols with stored prescription information are not being completely utilized at present, as both community pharmacies and hospitals/clinics face several issues such as hardware maintenance, software updates, and time and effort required for the usage.