Therapeutic Drug Monitoring最新文献

{"title":"Biomonitoring of Oxycodone and Its Metabolites in Conventional and Nonconventional Biological Matrices: A Systematic Critical Review.","authors":"Maria Sofia Fede, Alida Likey, Adele Minutillo, Giulia Bambagiotti, Giuseppe Basile, Francesco Paolo Busardò, Anastasio Tini, Manuela Pellegrini","doi":"10.1097/FTD.0000000000001541","DOIUrl":"https://doi.org/10.1097/FTD.0000000000001541","url":null,"abstract":"<p><strong>Background/purpose: </strong>Oxycodone is widely prescribed for moderate-to-severe pain, but is associated with a high risk of misuse and overdose. Global data indicate an increase in opioid-related hospitalizations and fatalities, highlighting the need for effective therapeutic drug monitoring (TDM). This review systematically evaluates the analytical approaches and biomonitoring of oxycodone in conventional and nonconventional matrices across clinical and medicolegal contexts.</p><p><strong>Methods: </strong>A systematic search of PubMed and Scopus (2015-2026) identified studies reporting oxycodone detection, quantification, and pharmacokinetics in human biological matrices. Blood, plasma, urine, hair, oral fluid, meconium, and cerebrospinal fluid were collected. Data on sample preparation, analytical techniques, detection limits, and applications in therapeutic, forensic, and emergency settings were extracted.</p><p><strong>Results: </strong>Plasma, whole blood, and urine were the primary matrices for oxycodone monitoring, with liquid chromatography-tandem mass spectrometry (LC-MS/MS) as the gold standard. Microsampling strategies including dried blood spots and volumetric absorptive microsampling, enable minimally invasive collection with reliable plasma correlations. Urine profiling supports the simultaneous assessment of parent drugs and metabolites, enabling the evaluation of CYP3A4 and CYP2D6 activity. Nonconventional matrices (hair, oral fluid, meconium, and cerebrospinal fluid) offer complementary information for chronic, recent, prenatal, and central nervous system exposure. Several LC-MS-based methods with validated analytical workflows and certified reference materials are currently available, providing comprehensive quantification of oxycodone and its metabolites across all matrices. These workflows often incorporate advanced extraction approaches, such as solid-phase extraction, liquid-liquid extraction, and electromembrane-based methods, which, combined with high-throughput LC-MS/MS, enhance sensitivity, specificity, and analytical sustainability. Limits of quantification across matrices range from 0.1 to 1 ng/mL, confirming the robustness of current TDM and forensic applications.</p><p><strong>Conclusions: </strong>Current evidence supports a shift toward faster analytical methodologies and the complementary use of alternative biological matrices for oxycodone biomonitoring. As no single specimen is suitable for every clinical or forensic application, matrix selection should be tailored to a specific interpretative context. Harmonization of interpretative criteria and integration with clinical, pharmacogenomic, and artificial intelligence-based approaches may further enhance oxycodone TDM.</p>","PeriodicalId":23052,"journal":{"name":"Therapeutic Drug Monitoring","volume":" ","pages":""},"PeriodicalIF":2.3,"publicationDate":"2026-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149324863","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparative Evaluation of 11 Machine Learning-Based Risk Prediction Models for Vancomycin-Associated Acute Kidney Injury During Initial Dosing.","authors":"Moeko Iida, Yasuhiro Horita, Masato Noda, Yoshinori Hisada, Masaya Nagamizu, Kaori Tsuzuki, Sakurako Muramatsu, Kazuki Ohashi, Yuki Nomura, Minami Asaoka, Tomoyo Mizuno, Hiroki Asakura, Tomoaki Hayakawa, Chiharu Wachino, Masaharu Kudo, Yoshihisa Mimura, Taketo Miyamoto, Masami Kawahara, Nobuyuki Morishita, Masahiro Kondo, Yuji Hotta, Atsushi Nakamura, Yoko Furukawa-Hibi","doi":"10.1097/FTD.0000000000001537","DOIUrl":"https://doi.org/10.1097/FTD.0000000000001537","url":null,"abstract":"<p><strong>Background: </strong>Vancomycin-associated acute kidney injury (AKI) can develop early during therapy and remains an important clinical concern. Although area under concentration-time curve (AUC)-guided dosing is recommended, the clinical relevance of early non-steady-state exposure for predicting AKI remains unclear. In this study, the authors aimed to develop and compare machine-learning models for predicting vancomycin-associated AKI using early exposure metrics.</p><p><strong>Methods: </strong>Early AUC values, trough concentrations, concomitant medications, and comorbidities were incorporated into machine learning models. Eleven algorithms were evaluated using nested cross validation to minimize optimistic bias and prevent data leakage.</p><p><strong>Results: </strong>Among 442 patients (median age, 72 years), 5.7% developed AKI. Across the models, day 2 AUC and trough concentration consistently emerged as the most influential predictors, followed by concomitant use of piperacillin/tazobactam, diuretics, and vasopressors. The random forest model demonstrated the best overall discriminative performance, with an area under the receiver-operating characteristic curve of 0.89 and area under the precision-recall curve of 0.47. SHapley Additive exPlanations indicated nonlinear increases in predicted AKI risk at a day 2 AUC of approximately ≥550 mg·h/L and trough concentration of approximately ≥20 mcg/mL. These findings suggest that early vancomycin exposure, particularly cumulative exposure during the first 48 hours, is strongly associated with AKI risk.</p><p><strong>Conclusions: </strong>Machine learning models integrating early AUC values and trough concentrations may provide a clinically useful framework for individualized prediction of AKI risk during the early phase of vancomycin therapy.</p>","PeriodicalId":23052,"journal":{"name":"Therapeutic Drug Monitoring","volume":" ","pages":""},"PeriodicalIF":2.3,"publicationDate":"2026-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149324864","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Isoform-specific Modulation of Drug Metabolism in a Fatality Involving Olanzapine Overexposure: Inflammation-Induced Phenoconversion as a Mechanistic Framework.","authors":"Maria Pena Martin, Chadi Abbara, Nicolas Picard, Florent Voillet, Stéphane Malbranque, Nathalie Jousset, Marie Briet, Guillaume Drevin","doi":"10.1097/FTD.0000000000001548","DOIUrl":"https://doi.org/10.1097/FTD.0000000000001548","url":null,"abstract":"<p><strong>Background: </strong>Interindividual variability influences adverse drug reactions and unexpected drug-related deaths. Despite increasing pharmacogenetics incorporation in clinical practice, acquired metabolic activity modifiers remain insufficiently integrated into toxicological interpretation. Here, the authors report an olanzapine overexposure fatality occurring despite therapeutic dosing, highlighting isoform-specific inflammation-induced phenoconversion.</p><p><strong>Methods: </strong>A medico-legal investigation (including review of the medical file, full autopsy, toxicological [femoral blood and urine], and pharmacogenetic analyses [CYP1A2/2D6/2C8/2C9 and UGT1A4 genes]) was conducted in a 57-year-old woman hospitalized for bipolar disorder with psychotic features. She was administered titrated therapeutic does of olanzapine.</p><p><strong>Results: </strong>The investigation indicated an acute systemic inflammatory state, characterized by neutrophilic leukocytosis (16 × 109/L) and elevated C-reactive protein (62 mg/L). A toxic femoral blood olanzapine concentration (172 mcg/L) and CYP1A2*30/*30, CYP2C8*1/*3, CYP2D*62/*5, and UGT1A4*1B/*1 genotypes were revealed by the toxicological and pharmacogenetics analyses, respectively. None of which alone accounted for the observed olanzapine accumulation.</p><p><strong>Conclusions: </strong>The selective accumulation of olanzapine, contrasted with preserved exposure to drugs predominantly metabolized by CYP3A4/2D6/alternative UGT isoforms, supports isoform-specific metabolic inhibition. CYP1A2 was the most plausible target of suppression. Although the impact of inflammation on UGT activity is less well characterized and appears isoform- and substrate-dependent, the selective modulation of UGT1A4 cannot be excluded. Overall, this case emphasizes that inflammation may differentially affect metabolic isoenzymes, and that both cytochrome P450- and UGT-mediated pathways should be considered when interpreting unexpected psychotropic drug overexposure.</p>","PeriodicalId":23052,"journal":{"name":"Therapeutic Drug Monitoring","volume":" ","pages":""},"PeriodicalIF":2.3,"publicationDate":"2026-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149324865","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessment of Interpatient and Intrapatient Variability of Dasatinib and Imatinib Concentrations in Patients With Chronic Myeloid Leukemia via Volumetric Absorptive Microsampling at Home.","authors":"Nick Verougstraete, Dominiek Mazure, Christophe P Stove","doi":"10.1097/FTD.0000000000001549","DOIUrl":"https://doi.org/10.1097/FTD.0000000000001549","url":null,"abstract":"<p><strong>Background: </strong>Therapeutic drug monitoring (TDM) of tyrosine kinase inhibitors (TKIs) may optimize treatment outcomes in chronic myeloid leukemia (CML). Conventional blood sampling is invasive and logistically demanding, whereas finger-prick microsampling, such as volumetric absorptive microsampling (VAMS), offers a less invasive, patient-centric alternative that can be performed at home, which is particularly relevant for trough-level monitoring and frequent sampling. This study evaluated the feasibility of VAMS-based home-sampling for dasatinib and imatinib in patients with CML and assessed interindividual and intraindividual variability in trough levels.</p><p><strong>Methods: </strong>Patients were trained to collect replicate capillary VAMS samples at their home immediately before TKI intake, every 3 days, on 4 occasions. Collected samples were returned to the laboratory via prepaid mail. In addition, participants completed a questionnaire to assess the user-friendliness and feasibility of home-sampling.</p><p><strong>Results: </strong>Forty-nine patients were included (dasatinib, 32; imatinib, 17), and 47 sets of VAMS samples were received. Ninety percent of the home-collected VAMS samples were of good quality, with an acceptable analytical variability between replicates. Most participants reported a positive experience with home-sampling. The results confirmed large interpatient variations in dasatinib and imatinib trough levels and demonstrated that within a patient, trough levels may vary substantially, particularly for dasatinib.</p><p><strong>Conclusions: </strong>Home-based VAMS sampling for dasatinib and imatinib TDM in patients with CML is feasible, well received, and reliable. The observed interpatient and intrapatient variability in trough levels underscores the potential clinical value of TDM in guiding individualized therapy and in the longitudinal monitoring of patients.</p>","PeriodicalId":23052,"journal":{"name":"Therapeutic Drug Monitoring","volume":" ","pages":""},"PeriodicalIF":2.3,"publicationDate":"2026-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149300422","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Serum Profiles of Pembrolizumab, Cardiac Troponin T, and Interleukin-6 in a Plasma Exchange-Treated Patient With Immune Checkpoint Inhibitor-Related Myocarditis.","authors":"Kazumasa Narita, Sho Suzuki, Koichiro Kuwahara, Takafumi Naito","doi":"10.1097/FTD.0000000000001546","DOIUrl":"https://doi.org/10.1097/FTD.0000000000001546","url":null,"abstract":"<p><strong>Background: </strong>A 77-year-old female patient with renal cell carcinoma, treated with intravenous pembrolizumab, developed pembrolizumab-related myocarditis accompanied by myasthenia gravis. For myasthenia gravis-related symptoms, intravenous methylprednisolone was administered on days 24-26 and 29-31 after the administration of 200 mg of pembrolizumab, and plasma exchange (PE) therapy was conducted on days 24, 26, 28, and 30.</p><p><strong>Methods: </strong>Serum pembrolizumab levels were monitored in conjunction with serum markers of myocarditis and interleukin-6 (IL-6) in the present patient treated with PE.</p><p><strong>Results: </strong>An elevated serum cardiac troponin T (cTnT) level of 0.542 ng/mL was observed 24 days after the first administration, and the serum pembrolizumab level was 15.7 µg/mL. After the first PE session, the serum levels of cTnT and IL-6 on day 25 were 0.406 ng/mL and 27.9 pg/mL, respectively, in conjunction with a decline of the serum pembrolizumab level to 4.83 µg/mL. Although the patient had a decreased serum pembrolizumab level of 3.86 µg/mL on day 29, the serum cTnT and IL-6 levels had increased to 0.843 ng/mL and 40.6 pg/mL, respectively. After the fourth PE session, serum levels of pembrolizumab, cTnT, and IL-6 on day 31 decreased to 3.62 µg/mL, 0.381 ng/mL, and 10.1 pg/mL, respectively. The overall removal rate after 4 sessions of PE was 76.9%, after which myasthenia gravis-related symptoms gradually disappeared.</p><p><strong>Conclusions: </strong>This patient with a serum pembrolizumab level within the previously reported range developed myocarditis accompanied by myasthenia gravis. PE therapy efficiently extracted circulating pembrolizumab and IL-6 in this patient with immune checkpoint inhibitor-related myocarditis.</p>","PeriodicalId":23052,"journal":{"name":"Therapeutic Drug Monitoring","volume":" ","pages":""},"PeriodicalIF":2.3,"publicationDate":"2026-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149300486","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantification of Upadacitinib, Dasatinib, Ruxolitinib, Imatinib, and Cabozantinib in Plasma With Liquid Chromatography-Tandem Mass Spectrometry.","authors":"Mohsin El Amrani, Pim M A Keur, Kim C M van der Elst, Bognàr Tim","doi":"10.1097/FTD.0000000000001535","DOIUrl":"https://doi.org/10.1097/FTD.0000000000001535","url":null,"abstract":"<p><strong>Background: </strong>Tyrosine kinase inhibitors (TKIs) are increasingly used to treat hematological malignancies and solid tumors. As pharmacokinetic variability is substantial among patients, therapeutic drug monitoring may help optimize treatment efficacy and minimize toxicity. However, simultaneous quantification of multiple TKIs with markedly different therapeutic concentration ranges is challenging. A sensitive and selective liquid chromatography-tandem mass spectrometry (LC-MS/MS) method was developed and validated for simultaneous quantification of upadacitinib, dasatinib, ruxolitinib, imatinib, and cabozantinib in human plasma.</p><p><strong>Methods: </strong>Proteins were precipitated in methanol containing stable isotope-labeled internal standards. Rapid chromatographic separation (5 minutes) was achieved on a BEH C18 column using a gradient of 0.025% ammonium hydroxide in water and methanol. TKIs were detected using a triple-quadrupole mass spectrometry in positive electrospray ionization mode. EMA ICH M10 guidelines were used for method validation.</p><p><strong>Results: </strong>All analytes showed excellent linearity (R2 ≥ 0.9992) over calibration ranges of 0.1-10,000 mcg/L depending on the analyte. Lower limits of quantification were 0.1 mcg/L for upadacitinib and dasatinib, 0.5 mcg/L for ruxolitinib, 2.25 mcg/L for cabozantinib, and 10 mcg/L for imatinib. Within-run and between-run precision were ≤4.7% and ≤3.1%, respectively, and accuracy ranged from 0.1% to 13.2%. No significant matrix effects, carryover, or interference were observed. All analytes were stable during short-term storage (3 days at 5°C and 20°C), freeze-thaw cycles, and long-term storage (12 months at -20°C and -80°C).</p><p><strong>Conclusions: </strong>This simple, rapid, sensitive, and robust LC-MS/MS method enables simultaneous quantification of 5 TKIs at different concentrations with minimal sample volume and excellent analytical performance, supporting its use in therapeutic drug monitoring and pharmacokinetic studies of kinase inhibitors in clinical practice.</p>","PeriodicalId":23052,"journal":{"name":"Therapeutic Drug Monitoring","volume":" ","pages":""},"PeriodicalIF":2.3,"publicationDate":"2026-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149288854","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantification of Antibiotic Concentration at the Site of Infection: Analytical Challenges and Current Practices.","authors":"Soma Bahmany, Amanda J Holst, Wichor M Bramer, Rawi Ramautar, Robert B Flint, Birgit C P Koch","doi":"10.1097/FTD.0000000000001501","DOIUrl":"https://doi.org/10.1097/FTD.0000000000001501","url":null,"abstract":"<p><strong>Background: </strong>Quantifying antibiotic concentrations at infection sites is challenging because of invasive sampling and complex matrices. Despite limited knowledge, target-site data are crucial for optimizing dosing, minimizing resistance, ensuring safer strategies, and reducing toxicity risks through adequate target attainment.</p><p><strong>Methods: </strong>This review aims to provide an overview of the most commonly reported challenges associated with quantifying target-site concentrations of antibiotics. Antibiotics were selected as the focus because of their use in hospital and outpatient settings. An extensive literature search was conducted using Embase.com. After title and abstract screening of 828 articles, 398 articles were included for full-text screening, and 72 were included based on relevance to analytical challenges in target-site quantification. Reported challenges were categorized in subgroups: sampling difficulties, sample heterogeneity, unavailability of free drug concentrations, drug complexities, and others.</p><p><strong>Results: </strong>Direct measurement of antibiotic concentrations at infection sites is constrained by invasive procedures and complex analytical methods. Key challenges include sampling heterogeneity, difficulty in obtaining samples, and the lack of free drug concentration data. Several opportunities have been proposed to address these challenges.</p><p><strong>Conclusions: </strong>We recommend updating international validation guidelines to include standardized tissue sampling protocols, encouraging the commercial sectors to develop reference standards and blank materials for alternative matrices, and promoting the use of minimally invasive techniques, such as microneedle biosensors, to improve the accuracy and safety of target-site quantification. These advances offer promising avenues to enhance antimicrobial therapy optimization despite current.</p>","PeriodicalId":23052,"journal":{"name":"Therapeutic Drug Monitoring","volume":" ","pages":""},"PeriodicalIF":2.3,"publicationDate":"2026-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149288814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Clinical Interactions Between Antiseizure Medications and Direct Oral Anticoagulants: A Scoping Review on Therapeutic Failures and Knowledge Gaps.","authors":"Chiara Cancellerini, Roberta Esposto, Alice Caravelli, Luca Vignatelli, Elena Pasini, Andrea Zini, Barbara Mostacci, Laura Licchetta, Francesca Bisulli","doi":"10.1097/FTD.0000000000001528","DOIUrl":"https://doi.org/10.1097/FTD.0000000000001528","url":null,"abstract":"<p><strong>Purpose: </strong>Antiseizure medications (ASMs) are often coprescribed with direct oral anticoagulants (DOACs). The pharmacokinetic (PK) interactions between ASMs and DOACs may result in altered exposure to anticoagulants, potentially increasing the risk of thromboembolic events or bleeding. This review summarizes the clinical evidence on drug-drug interactions between ASMs and DOACs.</p><p><strong>Methods: </strong>In accordance with PRISMA-ScR guidelines, a systematic literature search was conducted in PubMed (up to November 2025). All clinical reports and observational datasets reporting the concomitant use of at least 1 ASM and DOAC were included. ASM-DOAC combinations, PK findings, and reported clinical outcomes are presented.</p><p><strong>Results: </strong>Thirty studies were included. Most reports involved enzyme-inducing ASMs (carbamazepine, phenobarbital, primidone, and oxcarbazepine), which were consistently associated with subtherapeutic DOAC plasma concentrations and severe clinical failures. Doubling the DOAC dose did not prevent these events in several cases. Rivaroxaban and apixaban were the most frequently described DOACs, whereas clinical data for edoxaban were limited. Evidence of non-enzyme-inducing ASMs, such as levetiracetam and valproic acid, was limited.</p><p><strong>Conclusions: </strong>Although enzyme-inducing ASMs compromised DOAC efficacy, the available clinical evidence was fragmented and prone to publication bias. The observed variability in interactions involving noninducing ASMs such as levetiracetam underlines the clinical complexity that may exceed theoretical PK predictions. Given the unpredictability of these interactions and the failure of empirical dose adjustments, a standardized approach for therapeutic drug monitoring and prospective, comparative trials are essential to guide personalized therapy in this high-risk population.</p>","PeriodicalId":23052,"journal":{"name":"Therapeutic Drug Monitoring","volume":" ","pages":""},"PeriodicalIF":2.3,"publicationDate":"2026-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149239432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Toward Reinforcement Learning-Enabled Model-Informed Precision Dosing: Concepts, Applications, and Implementation.","authors":"Kei Irie, Wen Rui Tan, Tomoyuki Mizuno","doi":"10.1097/FTD.0000000000001540","DOIUrl":"https://doi.org/10.1097/FTD.0000000000001540","url":null,"abstract":"<p><strong>Purpose: </strong>Reinforcement learning is suited to sequential decision making in clinical settings, where treatment strategies are adjusted based on patient response. This aligns with therapeutic drug monitoring and model-informed precision dosing, in which dosage is iteratively refined as conditions evolve. Thus, reinforcement learning is a potential dosing optimization framework. This review provides a structured and clinically interpretable overview of these applications and their implementation.</p><p><strong>Methods: </strong>A narrative literature review of studies that applied reinforcement learning to dosing optimization was conducted, using database searches and manual screening. Studies were qualitatively analyzed for drug type, algorithm, training environment, and their state, action, and reward definitions. A simplified implementation example using a deep Q-network illustrates the core workflow.</p><p><strong>Results: </strong>Reinforcement learning has been applied across diverse clinical settings, including anesthesia, oncology, and critical care. Most studies used simulation-based environments (pharmacokinetic, pharmacodynamic, or mathematical models) to enable safe, scalable training. Off-policy methods (eg, Q-learning and deep Q-networks) were commonly used for discrete dosing decisions and retrospective data; on-policy methods (eg, actor-critic approaches) were applied when adaptive dose adjustment was required. Substantial variability was observed in algorithm selection and in state, action, and reward definitions. The simplified example illustrates how deep Q-network-based reinforcement learning can be integrated with pharmacokinetic models to support dose optimization.</p><p><strong>Conclusions: </strong>Reinforcement learning provides a flexible framework that complements pharmacometric models to support dosing optimization. Further research is required to develop best-practice recommendations for study design, algorithm selection, reward design, and model evaluation to facilitate translation into clinical practice.</p>","PeriodicalId":23052,"journal":{"name":"Therapeutic Drug Monitoring","volume":" ","pages":""},"PeriodicalIF":2.3,"publicationDate":"2026-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149239492","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Therapeutic Drug Monitoring of Golimumab to Predict Disease Activity and Antidrug Antibodies Status in Patients with Ankylosing Spondylitis.","authors":"Sadeel A Shanshal, Shamil H Othman, Ali A Younis","doi":"10.1097/FTD.0000000000001527","DOIUrl":"https://doi.org/10.1097/FTD.0000000000001527","url":null,"abstract":"<p><strong>Background and aim: </strong>Golimumab is a fully human tumor necrosis factor-alpha inhibitor that is highly effective in the management of ankylosing spondylitis (AS). A major limitation of golimumab is nonresponse, resulting from low drug concentrations, which is primarily due to antidrug antibodies (ADAs). The aim of this study was to assess the trough concentration of golimumab and the prevalence of ADA formation in patients with AS.</p><p><strong>Methods: </strong>An enzyme-linked immunosorbent assay was used to quantify golimumab trough concentrations and ADAs in 59 Iraqi patients with AS; AS disease activity was also evaluated using the Ankylosing Spondylitis Disease Activity Score (ASDAS). Regression analysis was performed to determine predictors of disease activity and ADA formation.</p><p><strong>Results: </strong>The mean age of the participants was 37.86 years, most of them (52 patients) were men, and the median number of golimumab doses administered was 7. The median ASDAS score was 3.1, and approximately 35% of patients had low disease activity (ASDAS score <2.1). The range of golimumab trough concentration was 0.21-1.38 mg/L with a median of 0.41 mg/L, and ADAs were detected in 8 patients (13.6%). Trough concentrations were higher in ADA-negative patients and were significantly negatively correlated with ASDAS scores. Regression analysis identified the trough drug concentration as the only predictor of both disease activity and ADA formation, with lower concentrations responsible for higher disease activity and a higher chance of developing ADAs.</p><p><strong>Conclusions: </strong>Drug concentration is an important determinant of disease activity, and routine monitoring of trough concentration is recommended for therapeutic drug monitoring of golimumab.</p>","PeriodicalId":23052,"journal":{"name":"Therapeutic Drug Monitoring","volume":" ","pages":""},"PeriodicalIF":2.3,"publicationDate":"2026-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149220735","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}