Journal of pharmacological and toxicological methods最新文献

{"title":"Optogenetic modulation of excitability using ChR2-NaV1.5 and automated patch clamp platforms for translational safety screening","authors":"Kefan Yang,&nbsp;Irene Lu,&nbsp;Tim Strassmaier,&nbsp;Patrick Mumm,&nbsp;Nadine Becker,&nbsp;Tom A. Goetze,&nbsp;Elena Dragicevic,&nbsp;Alison Obergrussberger,&nbsp;Michael George,&nbsp;Niels Fertig","doi":"10.1016/j.vascn.2026.108511","DOIUrl":"10.1016/j.vascn.2026.108511","url":null,"abstract":"<div><div>Channelrhodopsins (ChRs) are light-gated proteins that control phototaxis in protists and are commonly used in optogenetics to manipulate neuron activity. ChR2-Na_V1.5, a fusion of ChR2 with cardiac-specific Na⁺ channel, modulates sodium channel kinetics and produces photosensitive inward currents.</div><div>Using automated low and high throughput patch clamp (APC), we found that blue-light intensity directly correlates with ChR2 channel activity, with higher intensities resulting in larger peak amplitudes under constant light exposure durations. We also found that light could be used to activate ChR2, depolarizing the membrane sufficiently in current clamp mode to elicit action potentials.</div><div>This is demonstrated on both the Port-a-Patch, a semi-automated patch clamp device recording from a single cell at a time, and the high throughput fully automated device, the SyncroPatch 384. The SyncroPatch 384 could also be used to investigate different ChR2 variants. Several different constructs could be activated using blue light on a single NPC-384 chip. In this study we used HEK cells stably expressed ChR2-Na_V1.5 but the technology could be extended to hiPSC-neurons or other stem cells or primary cells which express light activated ion channels.</div><div>Our findings underscore the utility of APC combined with optogenetics in enhancing assay throughput, consistency, and mechanistic resolution. These advanced platforms could support early-stage safety screening and contribute to more predictive, human-relevant models for drug-induced excitability disorders and off-target toxicities.</div></div>","PeriodicalId":16767,"journal":{"name":"Journal of pharmacological and toxicological methods","volume":"140 ","pages":"Article 108511"},"PeriodicalIF":2.1,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148730091","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":"Extracellular field potentials and contractility in human induced pluripotent stem cell-derived cardiomyocytes in-vitro under optical stimulation","authors":"Annika Rexa,&nbsp;Pia Wanders,&nbsp;Larissa de Biasi,&nbsp;Sarah Rimpel,&nbsp;Herbert M. Himmel","doi":"10.1016/j.vascn.2026.108492","DOIUrl":"10.1016/j.vascn.2026.108492","url":null,"abstract":"<div><div>Utilizing human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CM) in the CardioExcyte/FlexCyte system (96CE/FLX), we report on establishing an optical pacing assay and present first pharmacological validation data.</div><div>Extracellular field potentials (EFP) and contractile force (Fc) were measured in spontaneously beating and optically stimulated hiPSC-CM (iCell^2, Fujifilm-CDI). For optical stimulation, iCell^2 were transduced with channelrhodopsin-2 (ChR2) via adeno-associated virus 9 and 6 (AAV9, AAV6) over wide ranging multiplicities of infection (MoI) and followed for up to 4 weeks. Selected compounds known for their high (E4031) and low/no (mexiletine, nifedipine) arrhythmia risk and for their inotropic/chronotropic properties (e.g. isoprenaline) were investigated. At 7 days post-infection with AAV9 (MoI 5000/cell), 50% of wells responded to optical stimulation. AAV6 transduction required 10-fold higher viral load and longer duration. Under controlled conditions, unstimulated hiPSC-CM exhibit a frequency of approximately 0.65 Hz with a field potential duration (FPD) of 859 ± 30 ms (mean ± SD). Upon stimulation, FPD decreases to 786 ± 12 ms at 1.0 Hz, 720 ± 9 ms at 1.25 Hz, and 700 ± 4 ms at 1.5 Hz, indicating a frequency-dependent shortening of FPD. Under the influence of 10 nM E4031, FPD increases in unstimulated hiPSC-CM at a frequency of 0.6 Hz by 21.0 ∆∆%. When stimulated, FPD is prolonged by 14.9 ∆∆% (1.0 Hz), 7.6 ∆∆% (1.25 Hz), and 5.8 ∆∆% (1.5 Hz), thus demonstrating the reverse frequency-dependent effect of E4031. The Fc in unstimulated hiPSC-CM is 1.79 ± 0.02 mN/mm² at 40 ± 0.14 beats/min. Following 100 nM isoprenaline, the frequency increases by 95.5 ∆∆% while the Fc increases by 13.6 ∆∆% showing positive chronotropy and inotropy. Drug effects align with known outcomes from other assay systems. In conclusion, we successfully established and validated optically stimulated hiPSC-CM for EFP and Fc measurement. The results obtained with the reference compounds investigated are consistent with publications in hiPSC-CM [1–2] and primary human CM [3–4] and may improve translation to clinical trials.<ul><li><span>1.</span><span><div>Patel, et al. Toxicol. Sci. 2019; 170: 167.</div></span></li><li><span>2.</span><span><div>Wie, et al. Toxicol. Sci. 2020; 174: 254.</div></span></li><li><span>3.</span><span><div>Hasenfuss, et al. Circ. Res. 1994; 75: 434.</div></span></li><li><span>4.</span><span><div>Pieske, et al. Circ. Res. 1999; 85: 38.</div></span></li></ul></div></div>","PeriodicalId":16767,"journal":{"name":"Journal of pharmacological and toxicological methods","volume":"140 ","pages":"Article 108492"},"PeriodicalIF":2.1,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148730257","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":"Heart rate variability-based machine learning model for predicting convulsions in non-human primates: Validation with a human convulsant compound","authors":"Motohiro Shiotani,&nbsp;Hiroshi Mizuno,&nbsp;Yusaku Matsushita,&nbsp;Takashi Yoshinaga,&nbsp;Shoji Asakura","doi":"10.1016/j.vascn.2026.108435","DOIUrl":"10.1016/j.vascn.2026.108435","url":null,"abstract":"<div><div>[Background] Drug-induced convulsions represent a significant toxicological risk in drug development, requiring reliable biomarkers that can detect early signs of convulsive activity. We have experienced that our compound induced a convulsion in a clinical trial (”Compound X\"; sodium channel blocker). For Compound X, we conducted an electroencephalogram (EEG) study in non-human primates (NHPs) but found no EEG abnormalities at dose levels lower than convulsive dose. Then, we co-developed a machine learning-based convulsion prediction model using heart rate variability (HRV) data from NHPs, trained on both convulsant and non-convulsant compounds (Kuga et al., 2024). This model calculates a “convulsive index\" by identifying deviations in autonomic nervous system activity. Recently, we refined the model to reduce false positives caused by non-convulsant compounds that influence HRV. [Objectives] Here, we introduced an evaluation result of Compound X with the refined model and discussed the usefulness of our model for the preclinical convulsion assessment. [Methods] Telemetry-implanted male NHPs (<em>n</em> = 4) received oral doses of Compound X (0 [vehicle], 3, 10, and 30 mg/kg) in a Latin square design. Heart rate data were collected for approximately 6 h post-dose and analyzed using our refined model to calculate the convulsive index. For the 10 and 30 mg/kg doses, approaching convulsion-inducing levels in NHPs, retrospective toxicokinetic analysis was conducted to compare the C_max in NHPs with the blood concentration at which convulsions occurred in humans. [Results] No convulsions were observed in any animal. However, the mean convulsive index increased in a dose-dependent manner. Apparently higher index values were observed in 1 animal at 10 mg/kg and 2 animals at 30 mg/kg (approximately 8-fold higher than the individual vehicle-treated values). The C_max values in these 3 animals were close to the blood concentration at convulsion occurrence in humans (approximately 0.5 to 2 times that in humans). [Conclusions] These findings suggest that our HRV-based model can sensitively detect autonomic changes associated with convulsion risk prior to convulsion occurrence. The convulsive index may serve as a valuable preclinical biomarker, and future applications in human studies could support proactive safety assessments and the prevention of unexpected convulsions.</div></div>","PeriodicalId":16767,"journal":{"name":"Journal of pharmacological and toxicological methods","volume":"140 ","pages":"Article 108435"},"PeriodicalIF":2.1,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148730680","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":"All roads lead to Rome with a full heart: Comparison of in silico clinical trials for the assessment of QT-interval prolongation using two electrophysiology cell models","authors":"Jazmin Aguado-Sierra","doi":"10.1016/j.vascn.2026.108522","DOIUrl":"10.1016/j.vascn.2026.108522","url":null,"abstract":"<div><div>Cardiac safety testing is critical to drug development to ensure patient safety and regulatory approval. Currently, computational, or in silico modelling has become a key component as an approach endorsed by regulators. Recently, 3-dimensional full heart simulations have shown excellent quantitative correlations to human clinical trials of QT exposure-response. Full heart simulations provide as output the QT-interval prolongation as the main marker and can be directly compared to clinical trial data.</div><div>The aim of this study was to assess the performance of two different cell models: Optimized O'Hara-Rudy and Tomek-Rodriguez when employed to produce in silico trials of human heart populations. The context of use is the estimation of the concentration-QT interval prolongation after the administration of various drugs: moxifloxacin, mexiletine, nilotinib and procainamide.</div><div>A finite element model of the human heart was created from high resolution ex-vivo MRI images. The biventricular mesh with regular maximum side length elements of 380 μm was used to solve the electrophysiology problem using either the optimized O'Hara-Rudy (OHR) or the Tomek-Rodriguez (ToR) human cell models. Eight virtual subjects were defined by variating of conductances of 5 different ion channels, following the single cell population approach. Each virtual patient was simulated using either the OHR or the ToR models and administered three concentrations of the drugs tested: Cmax, ½ Cmax and ¼ Cmax. Results were compared to human clinical trial data when available.</div><div>A total of 288 simulations were performed. The clinical trials employed with both cell models present results within the intervals of clinical data measurements for moxifloxacin, nilotinib and procainamide. The biggest difference between the solutions was observed with the compound mexiletine, where the ToR model showed an equal but opposite dose-dependent QT-interval reduction.</div><div>Both cell models provide quantitative results that exist within the error bounds of human clinical data. The methodology employed allows for the exploitation of state-of-the-art computational models and can provide clinically relevant markers for the assessment of QT-interval prolongation regardless of the cell model of choice.</div></div>","PeriodicalId":16767,"journal":{"name":"Journal of pharmacological and toxicological methods","volume":"140 ","pages":"Article 108522"},"PeriodicalIF":2.1,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148730629","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":"A twist in the translational tale using JASPER: Delayed effects causing Torsade de Pointes","authors":"Maria Ghita,&nbsp;Henk Van Der Linde,&nbsp;David J. Gallacher","doi":"10.1016/j.vascn.2026.108485","DOIUrl":"10.1016/j.vascn.2026.108485","url":null,"abstract":"<div><div>A translational analysis was performed to investigate a compound's effects across multiple nonclinical assays conducted in <em>in vitro</em> and <em><em>in vivo</em></em> models for early cardiovascular safety risk assessment using Janssen Application tool for Safety Pharmacology and Exposure Responses (JASPER). This data science initiative serves as a pilot program showcasing the significance of integrating supplementary data from toxicology studies alongside Safety Pharmacology (SP) studies into a comprehensive end-to-end (E2E) data warehouse through a retroSEND initiative. This study aimed to evaluate the delayed time/exposure-effects of a test compound on I_Kr and I_Ks currents in preclinical studies, emphasizing the translation of QT-profile parameters. The study investigates net effects from preclinical models to differentiate between acute and delayed effects using JASPER. This tool enables the visualization of effects for translational analysis of cardiovascular markers relevant to assessing QT prolongation risk and Torsade de Pointes (TdPs). Several approaches were implemented to adapt the study protocols for assessing both acute and chronic effects, comparable to 1-month toxicology studies. The translational analysis examined compound-induced changes, focusing on endpoints in acute SP studies versus modified delayed exposure protocols. Following the traditional acute SP protocols, no effects were observed, highlighting the need to integrate additional data for timely and efficient translation. <em>In vitro</em> I_Kr and I_Ks assays, a synchronously beating human stem-cell-derived cardiomyocytes (CTCM) assay, a Multi-Electrode Array human cardiomyocyte (MEA-CM) assay and <em>in vivo</em> QT assays (telemetry and anesthetized) results revealed no significant acute effects. In contrast, delayed protocols showed impacts on QT-related parameters, particularly at higher exposure levels. Comparing a 9-day repeat dose SP dog study and a 30-day chronic toxicology dog study indicated a median QTc VDW (Van de Water's QT corrected formula) effect of +45% on the last day of the SP study, resembling the high-dose group in the toxicology study (63% - final day). The effect is significantly different from the vehicle group starting from the 4th day. The translation of effects observed in nonclinical repeat-dose SP and 1-month toxicology studies reveal statistically significant delayed effects predicting QTc prolongation, associated with a potential risk of TDPs.</div></div>","PeriodicalId":16767,"journal":{"name":"Journal of pharmacological and toxicological methods","volume":"140 ","pages":"Article 108485"},"PeriodicalIF":2.1,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148730633","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":"Enhancing cardiovascular safety pharmacology via blood pressure and ECG waveform analysis using symmetric projection attractor reconstruction","authors":"Miquel Serna Pascual,&nbsp;Marieke Van Daele,&nbsp;Matt Skinner,&nbsp;Jeanette Woolard,&nbsp;James Rickard,&nbsp;Dilshat Djumanov,&nbsp;Manasi Nandi","doi":"10.1016/j.vascn.2026.108467","DOIUrl":"10.1016/j.vascn.2026.108467","url":null,"abstract":"<div><div>In preclinical and clinical studies, the study of vital organ systems is critical for evaluating drug safety and informing decisions related to compound progression. Enhanced characterization of physiological changes reflected in vital signs could help decision-making, potentially reducing time-to-market of safe new drugs. While vital-sign monitors record cyclical physiological signals with high fidelity both clinically and pre-clinically, conventional analysis is typically reduced to single point markers which can miss important information. A novel technology, Symmetric Projection Attractor Reconstruction (SPAR), transforms vital signs into time-contained images (attractors) which amplify whole-waveform morphology and variability features. Change in SPAR quantifications could provide additional information on the cardiovascular effects of drugs and SPAR images may be more easily interpreted by non-experts. We tested the ability of SPAR to identify drug effects of new drug candidates in cardiovascular systems of animal models and human subjects. Two existing databases were retrospectively analyzed in this study. First, we analyzed 48-h blood pressure traces from adult male Sprague-Dawley rats (<em>n</em> = 5) administered either vehicle or one of 5 different drug compounds: two Syk inhibitors (fostamatinib, 20 mg/kg and entospletinib, 6 mg/kg), two vasodilators (vardenafil, 10 mg/kg, and molsidomine,10 mg/kg) and one vascular endothelial growth factor receptor (VEGFR) inhibitor (sunitinib, 7 mg/kg). Second, we analyzed triplicate 10-s ECG recordings from a placebo-controlled, 8-h phase-1 study of a known cardioactive drug (Moxifloxacin, 400 mg) in 30 patients (15 female, 17 male subjects).</div><div>SPAR revealed blood pressure waveform morphology changes in vardenafil (<em>p</em> &lt; 0.0001) and molsidomine (p &lt; 0.0001). Vardenafil also produced a decrease in SPAR blood pressure waveform variability (<em>p</em> = 0.0148). None of the other tested compounds in the preclinical analysis showed significant SPAR metric changes. In the clinical ECG analysis, SPAR described a moxifloxacin-induced whole-waveform morphology change (p &lt; 0.0001) in the first 4 h of the study. SPAR trends observed in this study described distinct waveform changes from the ones quantified by conventional metrics (i.e. HR, BP, QTc) whilst simultaneously providing an at-a-glance visual representation of complex waveform features. Further studies are required to validate the ability of SPAR to assist in safety assessment workflows and impact the decision making process.</div></div>","PeriodicalId":16767,"journal":{"name":"Journal of pharmacological and toxicological methods","volume":"140 ","pages":"Article 108467"},"PeriodicalIF":2.1,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148730759","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":"Addressing the confounding influence of activity on heart rate measurements in conscious rats","authors":"Kamila J. Sadko,&nbsp;Derek J. Leishman,&nbsp;Avery VandenBussche,&nbsp;Hannah Garver,&nbsp;Marc B. Bailie,&nbsp;Greg D. Fink,&nbsp;D. Adam Lauver","doi":"10.1016/j.vascn.2026.108461","DOIUrl":"10.1016/j.vascn.2026.108461","url":null,"abstract":"<div><div>Telemetered models are essential tools in cardiovascular safety studies, but spontaneous activity can confound heart rate (HR) measurements. To isolate direct drug effects, we applied a “one-step” correction method (previously used for QT interval adjustment) to HR. This approach improves sensitivity and may reduce the number of animals needed to detect treatment-related changes. We hypothesize that it will increase statistical power and better detect small HR differences.</div><div>We analyzed a historical study of 9 Sprague Dawley (SD) rats implanted with PA-C4 telemeters with 80 days of recordings. Data were collected for 10 s every 10 min. Least significant differences (LSDs) were determined by subtracting the lower confidence interval (CI) from the upper CI and dividing by 2. Group sizes of 3, 4, 6, and 8 were analyzed. A separate 48-day study of 17 Dahl S rats implanted with HD-S10 telemeters, treated with semaglutide at 0.1 mg/kg SC (<em>N</em> = 9) or vehicle (saline, <em>N</em> = 8) daily, was analyzed to evaluate treatment effect.</div><div>The one-step method produced smaller LSD values compared to the conventional method. For group sizes of 8, average LSDs were 3 ± 0.001 BPM versus 15 ± 1 BPM for the one-step and conventional methods, respectively. Across all two-day and animal combinations, one-step LSDs averaged 3 ± 0.1 BPM compared to 14 ± 1 BPM with the conventional method. Weekly averages supported this advantage (1 ± 0.001 BPM vs. 15 ± 3 BPM). In the semaglutide-treated group, detected HR differences were similar (65 ± 0.5 BPM with one-step vs. 64 ± 2 BPM with conventional method), but LSDs were lower with one-step (1 BPM vs. 19 BPM). Super-interval (2–8 h) analyses at a group size of 8 demonstrated decreasing LSDs (median ± SD) with longer intervals: one-step yielded 11 ± 2, 8 ± 1, 6 ± 0.6, and 6 ± 0.5 BPM, while the conventional method returned 20 ± 4, 17 ± 3, 16 ± 3, and 16 ± 2 BPM, respectively.</div><div>The one-step correction method enables more precise detection of drug-induced HR changes by accounting for activity-related variability, offering greater sensitivity. This supports the use of smaller group sizes while maintaining statistical power, reducing unnecessary animal use in preclinical studies.</div></div>","PeriodicalId":16767,"journal":{"name":"Journal of pharmacological and toxicological methods","volume":"140 ","pages":"Article 108461"},"PeriodicalIF":2.1,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148731190","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":"In silico study: ADMET-driven pharmacokinetic profiling of antiviral compounds for pre-clinical screening framework of intranasal drug development","authors":"Jeerakit Kerdsiri,&nbsp;Tasana Pitaksuteepong,&nbsp;Neal M. Davies,&nbsp;Nitra Nuengchamnong,&nbsp;Kowit Hengphasatporn,&nbsp;Shengnan Zhao,&nbsp;Siwaporn Boonyasuppayakorn,&nbsp;Parvapan Bhattarakosol,&nbsp;Yasuteru Shigeta,&nbsp;Raimar Loebenberg,&nbsp;Neti Waranuch","doi":"10.1016/j.vascn.2026.108565","DOIUrl":"10.1016/j.vascn.2026.108565","url":null,"abstract":"<div><div>The ongoing emergence of respiratory viral pathogens, including novel variants and established threats such as influenza, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and Respiratory Syncytial Virus (RSV), continues to challenge global health systems due to limited specific therapeutic options. Intranasal antiviral delivery represents a promising approach for targeting respiratory tract infections; however, drug development frequently fails due to inappropriate pharmacokinetic properties. This study employed computational ADMET Predictor tools to systematically evaluate the physicochemical and potential pharmacokinetic properties of ten reference compounds, including nine approved antiviral drugs and ivermectin, a repurposed antiparasitic agent with reported antiviral activity and prior intranasal investigation: Zanamivir, Ribavirin, Oseltamivir, Favipiravir, Remdesivir, Nitazoxanide, Acyclovir, Lopinavir, Molnupiravir, and Ivermectin. Computational predictions were validated against experimental literature data to establish evidence-based criteria ranges. Key parameters related to intranasal drug delivery, covering physicochemical, absorption, distribution, metabolism, elimination, and toxicity properties, were systematically analyzed. The analysis resulted in revised criteria including molecular weight (<em>M</em>_wt), pH, lipophilicity (Log <em>P</em>), water solubility (<em>S</em>_w and <em>S</em>_pH), Volume of distribution (<em>V</em>_d), P-glycoproteins substrate (Pgp substrate), and skin sensitization (Sens_Skin) to demonstrate 80% concordance with approved drug properties. This framework provides the preliminary evidence-based screening criteria as a tool to identify promising drug candidates suitable for preclinical evaluation in intranasal formulation development.</div></div>","PeriodicalId":16767,"journal":{"name":"Journal of pharmacological and toxicological methods","volume":"140 ","pages":"Article 108565"},"PeriodicalIF":2.1,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148537879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Detecting drug-induced EEG alterations with machine learning: A methodological proof-of-concept study in rats treated with MK-801","authors":"Matthieu Basset,&nbsp;Paul-Emeric Saunier,&nbsp;Geoffrey Viardot,&nbsp;Philippe L'Hostis,&nbsp;Eric Delpy,&nbsp;Kevin Carvalho,&nbsp;Florian W. Adraoui","doi":"10.1016/j.vascn.2026.108566","DOIUrl":"10.1016/j.vascn.2026.108566","url":null,"abstract":"<div><h3>Introduction</h3><div>Although deep learning methods for electroencephalogram (EEG) analysis are rapidly advancing, architectures developed for human multichannel recordings may not be suited to animal EEG, where recordings typically involve substantially fewer channels. This study investigated whether deep learning could detect EEG alterations induced by the NMDAr antagonist MK-801 (dizocilpine), recorded in rats subjected or not to auditory stimulations, to determine the potential of this method for brain circuit-related efficacy and safety assessment.</div></div><div><h3>Methods</h3><div>EEG recordings were obtained from male Sprague-Dawley rats implanted with telemetry implants studied in a randomized crossover treatment design. Signals from frontal and temporal electrodes were segmented into 0.5-s epochs and processed using a minimal preprocessing pipeline. Several candidate deep-learning architectures were evaluated on classification tasks.</div></div><div><h3>Results</h3><div>EEG Conformer provided the best balance between performance and robustness. On the recording-level test set, the final model reached an F-score close to 0.90 on the recording-level test set. In a complementary leave-one-subject-out (LOSO) cross-validation, it reached a mean weighted F1-score of 0.646 ± 0.034 (<em>n</em> = 5 folds), above the reference chance level and consistent with transfer of a learned MK-801-related EEG signature to unseen animals. In MK-801-treated rats, abnormal-class outputs increased with dose, from 0.17 at 0.025 mg/kg to 0.62 at 0.2 mg/kg. Auditory stimulation increased abnormal classification relative to non-stimulated epochs, indicating improved sensitivity under stimulation conditions.</div></div><div><h3>Conclusion</h3><div>These findings support the feasibility of using deep learning to detect compound-related EEG alterations in low-channel rat recordings with limited preprocessing. Two complementary forms of generalization were assessed: transfer to unseen treatment conditions within known subjects in the crossover setting, and generalization to entirely unseen animals in the LOSO setting. The lower LOSO performance is consistent with the more stringent validation conditions. Together, these results suggest that the approach may complement conventional analytical pipelines and support more automated, standardized preclinical EEG workflows for both safety and efficacy pharmacology studies.</div></div>","PeriodicalId":16767,"journal":{"name":"Journal of pharmacological and toxicological methods","volume":"140 ","pages":"Article 108566"},"PeriodicalIF":2.1,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148551801","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":"Predicting thorough QT safety profile using hERG IC10 versus IC50","authors":"Andrew G. Edwards,&nbsp;William J. Crumb Jr.","doi":"10.1016/j.vascn.2026.108564","DOIUrl":"10.1016/j.vascn.2026.108564","url":null,"abstract":"<div><div>The human ether-à-go-go–related gene (hERG) K⁺ channel is the principal off-target mediator of drug-induced QT prolongation, and hERG <em>IC</em>_<em>50</em> remains the only required in vitro cardiac safety assay. However, clinically relevant QT prolongation (ΔΔQTc ≥10 ms) typically occurs at much lower hERG inhibition (5–20%), and requires that <em>IC</em>_<em>50</em> be transformed into a safety margin to assess risk at therapeutic exposures. Thus, we examined the relationship between hERG inhibition and clinical ΔΔQTc to determine whether inhibition metrics closer to clinical concentrations improve prediction. Across 44 drug/dose pairs, 13.1% hERG inhibition corresponded to mean 10-ms ΔΔQTc, while 10.1% inhibition elicited 10-ms ΔΔQTc at the upper confidence interval (UCI). As a predictor, raw <em>IC</em>_<em>10</em> demonstrated comparable sensitivity (mean ΔΔQTc: 95.5% vs. 100%; UCI: 85.7% vs. 92.9%) and specificity (mean: 77.3% vs. 76.2%; UCI: 87.5% vs. 53.3%) to FDA safety margins calculated from <em>IC</em>_<em>50</em>. Logistic regression using <em>IC</em>_<em>10</em> improved discrimination of ΔΔQTc UCI ≥ 10 ms over <em>IC</em>_<em>50</em> (AUC 0.95 vs. 0.90, <em>p</em> = 0.076). In contrast, performance for mean ΔΔQTc ≥10 ms was similar (AUC ∼0.96 for both). The improvement in ΔΔQTc UCI prediction by <em>IC</em>_<em>10</em> was explained by differences in Hill slope between drugs. Those with shallower Hill slope were more readily predicted unsafe by <em>IC</em>_<em>10</em> and those with steeper Hill slope were more readily predicted safe. These findings suggest that hERG <em>IC</em>_<em>10</em> provides a direct index of QT liability, which takes advantage of differences in Hill slope to improve prediction of ΔΔQTc liabilities near critical hERG inhibition of ∼10%.</div></div>","PeriodicalId":16767,"journal":{"name":"Journal of pharmacological and toxicological methods","volume":"140 ","pages":"Article 108564"},"PeriodicalIF":2.1,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148537915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}