Published papers

Our previous studies demonstrated that the ethyl acetate extract of the juice of jabara (JEx), a Japanese citrus fruit, inhibited cytochrome P450 (CYP) 3A4 activity in a time-dependent manner, but does not inhibit CYP2C19 activity. Its components, 3,3′,4′,5,6,7,8-heptamethoxyflavone (HpMF) and 3,3′,4′,5,6,7-hexamethoxyflavone (HxMF), have been shown to be responsible for this CYP3A4 inhibition. This study aimed to investigate the nature of CYP2C9 inhibition by JEx, HpMF, and HxMF. The time-dependent inhibition kinetics of JEx against human recombinant CYP2C9 were investigated in vitro using warfarin as a substrate. The time-independent inhibition kinetics of JEx, HpMF, and HxMF against CYP2C9 were also determined. JEx did not cause time-dependent inhibition of CYP2C9, while JEx, HpMF, and HxMF inhibited CYP2C9 activity in a time-independent manner with IC50 values of 2.65 % equivalent, 16.2 µM, and 57.8 µM, respectively. HpMF and HxMF were considered to be primarily responsible for CYP2C9 inhibition by JEx. The ingestion of 250 mL of jabara juice was estimated to reduce intestinal CYP2C9 activity to 3 % of the control value assuming of dilution with 210 mL of intestinal fluid, suggesting that jabara juice may cause food-drug interactions via intestinal CYP2C9 inhibition.

Background: Patients’ oral expressions serve as valuable sources of clinical information to improve pharmacotherapy. Natural language processing (NLP) is a useful approach for analyzing unstructured text data, such as patient narratives. However, few studies have focused on using NLP for narratives in the Japanese language. Objective: We aimed to develop a high-performance NLP system for extracting clinical information from patient narratives by examining the performance progression with a gradual increase in the amount of training data. Methods: We used subjective texts from the pharmaceutical care records of Keio University Hospital from April 1, 2018, to March 31, 2019, comprising 12,004 records from 6559 cases. After preprocessing, we annotated diseases and symptoms within the texts. We then trained and evaluated a deep learning model (bidirectional encoder representations from transformers combined with a conditional random field [BERT-CRF]) through 10-fold cross-validation. The annotated data were divided into 10 subsets, and the amount of training data was progressively increased over 10 steps. We also analyzed the causes of errors. Finally, we applied the developed system to the analysis of case report texts to evaluate its usability for texts from other sources. Results: The F1-score of the system improved from 0.67 to 0.82 as the amount of training data increased from 1200 to 12,004 records. The F1-score reached 0.78 with 3600 records and was largely similar thereafter. As performance improved, errors from incorrect extractions decreased significantly, which resulted in an increase in precision. For case reports, the F1-score also increased from 0.34 to 0.41 as the training dataset expanded from 1200 to 12,004 records. Performance was lower for extracting symptoms from case report texts compared with pharmaceutical care records, suggesting that this system is more specialized for analyzing subjective data from pharmaceutical care records. Conclusions: We successfully developed a high-performance system specialized in analyzing subjective data from pharmaceutical care records by training a large dataset, with near-complete saturation of system performance with about 3600 training records. This system will be useful for monitoring symptoms, offering benefits for both clinical practice and research.

Background: Although transferring drugs to a third party, even in a personal capacity, is illegal in Japan, many drug transactions are carried out via internet auction sites. Pharmacy consumers might not be aware of the illegality of transferring drugs to others due to various factors. To clarify pharmacy consumers' awareness of this issue and its covariates, we carried out a questionnaire survey.Methods: A self-completed questionnaire, consisting of yes/no and multiple-choice questions, was used. The adult patients who visited Keio University Pharmacy were asked to fill out the questionnaire.Results and Discussion: One hundred and fifty-one response sheets were effectively collected from 173 responders. The subjects'awareness of the illegality of transferring drugs was significantly affected by the formulation (oral vs. patches) and pharmacological category of the drug, as well as the source of the drug (prescribed vs. over the counter). Patient education about the illegality of personally transferring drugs, especially patches, might be required to avoid illegal drug transfers, including via the internet.

Subcutaneous daratumumab DARA-SC is superior to intravenous daratumumab with a significantly shorter treatment time for multiple myeloma. To prevent infusion reaction (IR) during DARA-SC, oral diphenhydramine, a first-generation antihistamine, has been used in the phase Ⅲ clinical trial and following clinical settings. We recently substituted diphenhydramine with levocetirizine, a second-generation antihistamine, to diminish antihistamine-induced sedation in our regimens for daratumumab.In this study, we aimed to retrospectively compare the prevention of IR and rate adverse events between diphenhydramine and levocetirizine. We retrospectively analyzed the clinical records of patients in whom DARA-SC treatment was introduced for the first time from August 1st, 2021 to July 31, 2022, at Keio University Hospital.For the first period, 7 and 24 patients were treated with diphenhydramine and levocetirizine, respectively. Neither patient experienced IR. Two out of 7 patients experienced drowsiness by diphenhydramine while only one of 24 patients experienced it by levocetirizine (P = 0.125). Five and one patients initially treated with diphenhydramine were switched to levocetirizine upon the second and third periods of DARA-SC treatment, respectively. After the switching, their drowsiness was reduced.In conclusion, levocetirizine was considered more beneficial than diphenhydramine to prevent IR with a lower rate of drowsiness.

The subjective and qualitative survey reported that ward-based clinical pharmacy practices were reported to be rated highly by physicians and nurses. However, no reports have quantitatively evaluated the reduction of their workload from the viewpoint of task-shifting to pharmacists from other medical professionals. In this study, we conducted a questionnaire survey of physicians and nurses working in wards receiving the pharmacy practices service in Keio University Hospital. Survey items include the reduction in the workload of physicians and nurses and subjective ratings regarding the performance of pharmacy practices. The ward-based pharmacy practices (5.9 hours/day, average for a ward) were estimated to be equivalent to the workloads of 3.1 and 2.6 hours/day for a physician and nurse, respectively, adding up to 36.1 hours of their workload per ward. More than 80％ of the responders positively rated the performance of pharmacy practices, except for sterile admixture. This is the first study to evaluate the quantitative contribution of ward-based pharmacy practices on task-shifting from physicians and nurses.

New adverse events not detected in pre-marketing clinical trials sometimes emerge after medicines have entered the market. Therefore, systems that can extract the terminology of diseases and symptoms from unstructured data such as medical records are needed to enable automatic monitoring of adverse events. In this study, we analyzed nursing records of patients treated with vancomycin hydrochloride using a disease extraction system and a terminology list of allergic skin symptoms to identify cases with these symptoms and their occurrence frequency in a period with many occurrences and other period. The accuracy of the proposed method in detecting cases was assessed using the F-measure, which was found to be 0.71-0.74, and changes in the occurrence frequency between two periods were almost equal for the proposed method and visual confirmation by the researcher. These results suggest that the proposed method can be used for monitoring allergic skin symptoms due to vancomycin hydrochloride.

Summary: Objective: Among various adverse reactions of atypical antipsychotics, weight gain and impaired glucose tolerance are clinically significant. The aim of this study is to analyze quantitatively the contributions of various receptors to these antipsychotics-induced adverse reactions based on the receptor occupancy theory. Methods: Two indices of antipsychotics-induced weight gain (the values estimated by a metaanalysis and the observed values in clinical trials) and the morbidity rate of type 2 diabetes mellitus during treatment with antipsychotics were taken from the literature. We calculated the estimated mean receptor occupancies of α1adrenergic, α2adrenergic, dopamine D2, histamine H1, muscarinic acetylcholine (mACh), serotonin 5-HT1A, 5-HT2Aand 5-HT2Creceptors by antipsychotics by using the pharmacokinetic parameters and receptor dissociation constants, and analyzed the correlation between the occupancies and the extent of adverse reactions as assessed using the aforementioned indices. Results: There were statistically significant correlations between the estimated occupancies of H1and mACh receptors and antipsychotics-induced weight gain estimated by metaanalysis (rs= 0.81 and rs= 0.83, respectively, p<0.01). There were also statistically significant correlations between these receptor occupancies and observed weight gain in clinical trials (rs= 0.66 in each case, p<0.01). The morbidity rate of type 2 diabetes mellitus was highly correlated with H1, mACh, and 5-HT2Creceptor occupancies (rs= 0.90 in each case, p<0.05). However, H1receptor occupancy was also highly correlated with mACh receptor occupancy among antipsychotics, so that only one of them may be critically associated with the adverse reactions. Considering that these adverse reactions have not been reported for drugs with mACh receptor antagonistic action, other than antipsychotics, the H1receptor may contribute predominantly to the antipsychotics-induced weight gain and diabetes mellitus. Discussion/Conclusion: Model analysis based on receptor occupancy indicates that H1receptor blockade is the primary cause of antipsychotics-induced weight gain and diabetes mellitus. © 2005, The Japanese Society for the Study of Xenobiotics. All rights reserved.

The aims of this study were to evaluate the transplacental transfer properties of diclofenac and to determine the effect of L-lactic acid on the transplacental transfer of diclofenac. The maternal and fetal vessels of human placenta were perfused in a single-pass mode with a solution containing diclofenac and antipyrine. The transplacental pharmacokinetic model was fitted to the time profiles of the drug concentrations in the effluent and placenta to obtain transplacental pharmacokinetic parameters. In addition, chloride ion in the perfusate was partially replaced with L-lactic acid to see the change in the transplacental transfer properties of diclofenac. The TPT(ss) value (ratio of the rate of amount transferred across the placenta to that infused in the steady state) of diclofenac was 2.22%, which was approximately one-third that of antipyrine and was significantly reduced in the presence of L-lactic acid. The transplacental pharmacokinetic model could adequately explain the transplacental transfer of diclofenac with influx clearances from maternal and fetal perfusates to placental tissue of 0.276 and 0.0345 ml/min/g cotyledon and efflux rate constants from placental tissue to maternal and fetal perfusates of 0.406 and 0.0337 min(-1), respectively. By taking into account protein binding, the placental tissue/plasma concentration ratio in humans for diclofenac was estimated to be 0.108 ml/g of cotyledon and was smaller than that of antipyrine. In conclusion, human placental perfusion and transplacental pharmacokinetic modeling allowed us to determine the transplacental transfer properties of diclofenac quantitatively. Diclofenac may share transplacental transfer system(s) with L-lactic acid.

Cytochrome P450 2D6 (CYP2D6) is a highly polymorphic enzyme that is involved in the metabolism of many drugs. Terbinafine (TER) is a CYP2D6 inhibitor and causes persistent drug interactions in the clinical setting; however, its inhibitory mechanism and the differences in its inhibitory potency among genetic variants of CYP2D6 remain to be investigated. This study aimed to investigate the inhibitory mechanism of TER and the differences in its inhibitory potency among three CYP2D6 variants, CYP2D6.1, CYP2D6.2, and CYP2D6.10. In a competitive inhibition study, the metabolic activity of the CYP2D6 was assessed based on their demethylation of dextromethorphan in the presence or absence of TER, and the time-dependency of the inhibitory effects were examined by preincubating the enzymes with TER. TER had weaker inhibitory effects on CYP2D6.2 and CYP2D6.10 than on CYP2D6.1; i.e., TER exhibited Ki values (the concentration of inhibitor that results in half-maximal inhibition) of 0.0525, 0.355, and 1.85 μM for CYP2D6.1, CYP2D6.2, and CYP2D6.10, respectively. The inhibitory effects of TER were not time-dependent. Since TER's Ki value for CYP2D6.10 was 35.2-fold higher than its Ki value for CYP2D6.1, the CYP2D6 genotype of subjects should be taken into account when estimating the severity of drug interactions involving TER.

Objective: The contribution ratios (CR) of metabolic enzymes to the systemic clearance of a drug can be estimated from in vitro studies. Another feasible approach is to calculate them based on the increase in the area under the time-concentration curve (AUC) caused by the co-administration of a potent and selective inhibitor in a clinical drug-drug interaction (DDI) study. However, some factors, such as the inhibitory potency of the inhibitor and the inhibition of first-pass metabolism, might affect the estimation of CR based on clinical DDI studies. We aimed to validate the accuracy of the DDI-based estimation of CR using an in silico approach. Materials and methods: An in silico DDI study was conducted using a population-based physiological pharmacokinetic simulator to estimate the CR of cytochrome P450 (CYP)3A4 for zolpidem, sildenafil, omeprazole, triazolam, and repaglinide. The ratio of the AUC value seen in the presence of an inhibitor (ketoconazole or itraconazole) to that observed in the absence of the inhibitor (AUC ratio) was also calculated. The CR for CYP3A4 obtained using the simulator (CRdef) were compared with those calculated from the AUC ratio (CRest). Results: When ketoconazole was used, good correlations between the CRest and CRdef were obtained for all examined substrates (inconsistencies were seen in < 10% of subjects). CR estimates derived from the AUC ratio were found to be accurate. Some underestimation was observed, possibly due to incomplete inhibition, and some overestimation caused by extensive first-pass metabolism was noted. Conclusion: This study verified that CR obtained from AUC ratios in DDI studies are quite reliable.

Recently, clinical cases have been reported of QT prolongation and torsades de pointes associated with the use of tacrolimus (FK506). We examined the relationship between QTc prolongation and the pharmacokinetics of FK506 in guinea pigs in order to evaluate the arrhythmogenicity of FK506 in comparison with quinidine (QND). FK506 (0.1 or 0.01 mg/hr/kg) or QND (30 mg/hr/kg) was intravenously infused to guinea pigs and time profiles of drug concentration in blood and QTc interval were examined during and after infusion. Both FK506 and QND evoked a significant QTc prolongation, and the dose-response relationship showed an anti-clockwise hysteresis, FK506-induced QTc prolongation persisted throughout the duration of the experiment despite a decline in the plasma FK506 concentration, whilst QND-induced QTc prolongation disappeared as plasma concentrations decreased. FK506 induced a sustained QTc prolongation in guinea pigs at drug concentrations in blood that correspond to its therapeutic range in human, suggesting that it might be of clinical significance to monitor the electrocardiogram, especially when patients have congenital or acquired QT-prolonging risk factors.

1 Nateglinide, a novel oral hypoglycemic agent, rapidly reaches the maximum serum concentration after oral administration, suggesting that it is rapidly absorbed in the gastrointestinal tract. The aim of this work is to clarify the intestinal absorption mechanism of nateglinide by means of in vitro studies. 2 We examined the transcellular transport and the apical uptake of [ 14C]nateglinide in a human colon carcinoma cell line (Caco-2). We also examined whether nateglinide is transported via monocarboxylate transport-1 (MCT1) by means of an uptake study using MCT1-expressing Xenopus laevis oocytes. 3 In Caco-2 cells, the transcellular transport of [ 14C]nateglinide from the apical to basolateral side was greater than that in the opposite direction. The uptake of [ 14C]nateglinide from the apical side was concentration-dependent, H +-dependent, and Na +-independent. Kinetic analysis revealed that the Kt and Jmax values of the initial uptake rate of [ 14C]nateglinide were 448 μM and 43.2 nmol mg protein -1 5 min -1, respectively. Various monocarboxylates, including salicylic acid and valproic acid, and glibenclamide significantly inhibited the uptake of [ 14C]nateglinide. 4 The uptake study using MCT1-expressing oocytes showed that nateglinide inhibits the MCT1-mediated uptake of [ 14C]L-lactic acid, though nateglinide itself is not transported by MCT1. 5 Taken together, these results suggest that the uptake of nateglinide from the apical membranes of Caco-2 cells is, at least in part, mediated by a proton-dependent transport system(s) distinct from MCT1.

Gastroesophageal reflux (GER) is a common episode in pediatric patients with severe motor and intellectual disabilities (SMID) and occasionally leads to a severe clinical state accompanied with nausea, hematemesis, melena, wheezing, pneumonia, anemia and/or failure to thrive. We report here a case of a 14-year-old male with Lennox syndrome who had been treated with a histamine H2 blocker intravenously or via a nasogastric tube for repeated gastric hemorrhage due to severe GER. Since his gastric hemorrhage became resistant to the H2 blocker, we decided to replace it with a proton pump inhibitor (PPI). Although lansoprazole can be decapsulated for administration via a nasogastric tube, it tends to block fine tubes. The acid-sensitive drug omeprazole, another oral PPI, is commercially available as enteric-coated tablets. Therefore, we pulverized the tablets and administered omeprazole, mixed with a small amount of antacid, via a nasogastric tube. The patient's gastric hemorrhage was dramatically improved. Thus, administration of pulverized omeprazole concomitantly with antacid via a fine nasogastric tube may provide a novel approach for the treatment of chronic GER in pediatric patients with SMID. © 2004 Dustri-Verlag Dr. K. Feistle.

Valproic acid is an anticonvulsant widely used for the treatment of epilepsy. However, valproic acid is known to show fetal toxicity, including teratogenicity. In the present study, to elucidate the mechanisms of valproic acid transport across the blood-placental barrier, we carried out transcellular transport and uptake experiments with human placental choriocarcinoma epithelial cells (BeWo cells) in culture. The permeability coefficient of [3H]valproic acid in BeWo cells for the apical-to-basolateral flux was greater than that for the opposite flux, suggesting a higher unidirectional transport in the fetal direction. The uptake of [3H]valproic acid from the apical side was temperature-dependent and enhanced under acidic pH. In the presence of 50 μM carbonyl cyanide p-trifluoromethoxylhydrazone, the uptake of [3H]valproic acid was significantly reduced. A metabolic inhibitor, 10 mM sodium azide, also significantly reduced the uptake of [3H]valproic acid. Therefore, valproic acid is actively transported in a pH-dependent manner on the brush-border membrane of BeWo cells. Kinetic analysis of valproic acid uptake revealed the involvement of a non-saturable component and a saturable component. The Michaelis constant for the saturable transport (Kt) was smaller under acidic pH, suggesting a proton-linked active transport mechanism for valproic acid in BeWo cells. In the inhibitory experiments, some short-chain fatty acids, such as acetic acid, lactic acid, propanoic acid and butyric acid, and medium-chain fatty acids, such as hexanoic acid and octanoic acid, inhibited the uptake of [3H]valproic acid. The uptake of [3H]valproic acid was also significantly decreased in the presence of 4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid, salicylic acid and furosemide, which are well-known inhibitors of the anion exchange system. Moreover, p-aminohippuric acid significantly reduced the uptake of [3H]valproic acid. These results suggest that an active transport mechanism for valproic acid exists on the brush-border membrane of placental trophoblast cells and operates in a proton-linked manner. © 2001 Published by Elsevier Science B.V.

Clinical cases have been reported of tacrolimus (FK506)-induced QT prolongation. We have previously demonstrated sustained QT prolongation by FK506 in guinea pigs. Herein, we aimed to conduct a pharmacokinetic/pharmacodynamic (PK/PD) analysis of FK506, using a model involving the myocardial compartment. The pharmacokinetics of FK506 and its effects on QTc intervals were investigated in guinea pigs. In the pharmacokinetic study, whole blood and ventricular FK506 concentrations were analyzed, using a 4-compartment model during and after intravenous infusion of FK506 (0.01 or 0.1 mg/hr/kg). Subsequently, the concentration-response relationship between ventricular FK506 concentration and change in QTc interval was analyzed, using the maximal effect (Emax) model. Pharmacokinetic profiles of FK506 showed a delayed distribution of FK506 into the ventricle. Furthermore, the observed QT prolongation paralleled the ventricular FK506 concentrations, with no lag-time between the two. The Emax model successfully described the relationship between changes in QTc interval and ventricular FK506 concentrations. In conclusion, the PK/PD model where the myocardial drug concentration of FK506 was linked with its adverse effect could describe, for the first time, the anti-clockwise hysteresis observed in the relationship between blood FK506 concentration and QT prolongation. Such a hysteresis pattern for QT prolongation might be caused, therefore, mainly by the delayed disposition of FK506 to ventricular myocytes.

Background and aims: Histamine H1 receptor (H1R) antagonists, or antihistamines, mainly act on the peripheral tissues but can induce sedation. This undesirable central side effect is caused by blockade of nerve transmission in the histaminergic neuron system. Variation in sedative properties of antihistamines is due to difference in their blood-brain barrier (BBB) permeability, partly defined by activity of P-glycoprotein (P-gp), an efflux pump located in the membrane of endothelial cells in brain capillaries. The aim of the present study is to evaluate the effects of single nucleotide polymorphisms (SNP) of the P-gp encoding "multi-drug resistance gene (MDR)-1" on BBB permeability of a non-sedative antihistamine, fexofenadine, that is a substrate of P-gp, using positron emission tomography (PET) with [11C]doxepin. Methods: Healthy young male volunteers (n=39, ranging 19-36 y.o.) were recruited and examined for genotyping of the common SNP in exon 26 (C3435T). Twenty-two out of the 39 subjects were studied by PET with [11C]doxepin following oral administration of placebo or fexofenadine 120 mg, in a single blind cross-over design. PET scan started 90 min following administration of each drug and the dynamic data, obtained taking 90 min, were analyzed based on Logan reference method to yield binding potential in various cortical structures. Based on the biding potential data of each subject in two drug conditions, histamine H1 receptor occupancy (H1RO) due to fexofenadine was calculated for each subject and the individual H1RO values were compared between wild-type (CC: n=6), heterozygous (CT: n=13) and homozygous (TT: n=3) subgroups. Results: Overall mean H1RO due to fexofenadine in the 22 subjects was calculated as -5%. The mean H1ROs of the 3 subgroups, CC, CT and TT, were -13.4, -3.7 and -1.4%, respectively, although the difference was not significant. Conclusions: The reason that H1RO of fexofenadine was calculated as lower than zero is not known. The PET results might suggest the presence of slight decrease in efflux function of P-glycoprotein in BBB among holders of C3435T mutation in comparison to wild-type in human. Further investigation is needed to draw a definitive conclusion with larger sample size, including studies on other mutations and other influx and efflux transporters. Functional and molecular imaging will be a strong tool for establishment of "individualized medicine".

Summary: The purpose of this study was to investigate the transport mechanism of tolbutamide across the blood-brain barrier (BBB) using MBEC4 cells as an in vitro BBB model. Methods: The BBB transport of tolbutamide was studied by using a mouse brain capillary endothelial cell line, MBEC4, cultured on dishes with their luminal membrane facing the culture medium. Results: The uptake of [14Cltolbutamide by MBEC4 cells was dependent on temperature and energy. The uptake coefficient of [14C]tolbutamide increased markedly with decreasing pH of the external medium from neutral to acidic. Valinomycin and replacement of chloride with sulfate or gluconate significantly increased the initial uptake of [14C]tolbutamide, while replacement with nitrate significantly decreased it. The uptake was significantly reduced by a proton ionophore, FCCP, and an anion-exchange inhibitor, DIDS. The initial uptake of [14C]tolbutamide was saturable with Kt of 0.61 ± 0.03 mM (pH 7.4) and 1.76±0.19 mM (pH 6.5). At pH 6.5, the initial uptake of [14C]tolbutamide was significantly reduced by several sulfa drugs, salicylic acid, valproic acid and probenecid, and was competitively inhibited by sulfaphenazole (Ki = 3.47 ±0.50 mM) and valproic acid (Ki = 2.29 ±0.43 mM). Conclusion: These observations indicate the existence of a pH-and membrane-potential-dependent anion exchange and/or proton-cotransport system(s) for concentrative uptake of tolbutamide and sulfa drugs in MBEC4 cells. © 2004, The Japanese Society for the Study of Xenobiotics. All rights reserved.

The cardiac effect of edrophonium (2-20 μmol kg-1), pyridostigmine (0.5-5 μmol kg-1), neostigmine (0.05-0.5 μmol kg-1) and ambenonium (0.02-0.3 μmol kg-1) was investigated after intravenous administration to rats. For pyridostigmine and neostigmine, the heart rate decreased in a dose-dependent manner, and then gradually recovered to the basal level at about 10 min. Rapid decrease of heart rate was observed after edrophonium and ambenonium administration, and rapid recovery to the basal level within 1 min. For ambenonium, a dose-dependent tachycardiac response was observed. The time-course of heart rate change was analysed by the effect-compartment model. Significant correlation was observed between bradycardiac EC50 values obtained by effect-compartment model analysis and inhibitory constant (K(i)) to acetylcholinesterase in-vitro, suggesting that the bradycardiac response was induced by inhibition of this enzyme and following elevation of acetylcholine concentration in the synaptic cleft. On the other hand, the tachycardiac EC50 values of edrophonium and ambenonium based on the effect-compartment model analysis were similar to dissociation constants (K(d)) of these drugs to muscarinic receptors in-vitro, suggesting that the tachycardiac activity of these drugs may be associated with antagonistic activity to postsynaptic muscarinic receptors. We conclude that, clinically, edrophonium and ambenonium are safer drugs than pyridostigmine and neostigmine, at least as regards muscarinic side-effects, including bradycardia.

The aim of this study was to develop a method for predicting the extent of grapefruit juice (GFJ)-drug interactions and their interindividual variations from the pharmacokinetic profile in the absence of GFJ. The pharmacokinetic profiles of 13 drugs after intravenous and oral administration were used to develop and validate the method. For each drug, the proportion absorbed into the intestine and the intestinal availability (Fg) were calculated from clinical data taken from the literature. Then, the AUC ratio (the ratio of the AUC with GFJ to that without GFJ) was predicted by assuming that Fg was 1.0 when GFJ was concomitantly ingested. According to the developed method, the AUC ratio of felodipine was 2.50 and its coefficient of variation (CV) was 45%, which agreed well with the observed AUC ratio of 2.48 and CV of 51%. Although the developed method overestimated the AUC ratios of some drugs such as nisoldipine, no underestimation occurred. The predicted CV values were consistent with those observed. The developed method might be useful to predict the AUC ratio, along with its interindividual variation, from the pharmacokinetic profile in the absence of grapefruit juice.