ABSTRACT
Population pharmacokinetics/pharmacodynamics (pop-PK/PD) consolidates pharmacokinetic and pharmacodynamic data from many subjects to understand inter- and intra-individual variability due to patient backgrounds, including disease state and genetics. The typical workflow in pop-PK/PD analysis involves the determination of the structure model, selection of the error model, analysis based on the base model, covariate modeling, and validation of the final model. Machine learning is gaining considerable attention in the medical and various fields because, in contrast to traditional modeling, which often assumes linear or predefined relationships, machine learning modeling learns directly from data and accommodates complex patterns. Machine learning has demonstrated excellent capabilities for prescreening covariates and developing predictive models. This review introduces various applications of machine learning techniques in pop-PK/PD research.
Subject(s)
Machine Learning , Models, Biological , Pharmacokinetics , HumansABSTRACT
This study aimed to investigate whether renin-angiotensin system inhibitors (RAS-I) have an advantage over calcium channel blockers (CCB) for suppression of proteinuria in hypertensive patients with gastric cancer receiving ramucirumab (RAM) treatment. Adult Japanese patients with gastric cancer who were outpatients at Asahikawa Medical University Hospital, National Hospital Organization Hokkaido Cancer Center, and Iwate Medical University Hospital between July 1, 2015, and March 31, 2021, were included in this study. Of these patients, those who had received first-time RAM treatment, and those treated with antihypertensive agents including RAS-I or a CCB at initial RAM administration were included. A total of 36 patients were analyzed in this study. Of these patients, 17 patients were classified into the RAS-I group and the remaining 19 into the CCB group. After 12 weeks of RAM administration, the prevalence of proteinuria in the RAS-I group was significantly lower than that in the CCB group. Additionally, Kaplan-Meier analysis showed that the cumulative occurrence of proteinuria in the RAS-I group over 12 weeks following RAM administration was significantly lower than that in the CCB group. Furthermore, simulation of the time course of RAM blood concentrations based on the O'Brien model showed that there may not be differences in the RAM blood concentration profiles over 12 weeks between the two groups. RAS-I may have an advantage over CCB for suppressing proteinuria in hypertensive patients with gastric cancer treated with blood pressure antihypertensive agents. Our results provide useful information to healthcare professionals involved in the administration of RAM treatment.
ABSTRACT
BACKGROUND: Plasma teicoplanin concentrations do not reach the therapeutic range in several patients with hematological malignancies. Nevertheless, the characteristics of the population pharmacokinetic (PPK) models have not been clarified for malignancy. The decrease in the teicoplanin concentration in patients with cancer has been attributed to augmented renal clearance (ARC). It is essential to identify the causative factors of ARC to construct a PPK model to optimize the administration method. The authors aimed to establish a PPK model and develop an appropriate dosing regimen for teicoplanin in patients with hematological malignancies. METHODS: PPK analysis was performed using therapeutic drug monitoring (TDM) data from 119 patients with hematological malignancies. The developed model was verified by predictive performance. RESULTS: The covariates affecting systemic clearance were serum creatinine, presence or absence of neutropenia (<500/µL), and body size descriptor. Patients with hematologic malignancies and neutropenia showed a 25% increase in clearance compared with those with a normal neutrophil count. The PPK model was constructed based on the presence or absence of neutropenia. This model allowed the selection of the most appropriate dosage regimen out of those recommended by the TDM guidelines for patients with eGFR of >60 mL/min/1.73 m2. The PPK model predicted a dosing regimen for achieving a 10% improvement in the coverage probability of the target concentration range during the loading and maintenance phases. CONCLUSIONS: The PPK model may help optimize dose regimens and evaluate dosing methods, using comparative simulations, in patients with hematological malignancies.
