Breastfeeding is important in childhood development, and medications are often necessary for lactating individuals, yet information on the potential risk of infant drug exposure through human milk is limited. Establishing a lactation modeling framework can advance our understanding of this topic and potentiate clinical decision making. We expanded the modeling framework previously developed for sotalol using pregabalin as a second prototypical probe compound with similar absorption, distribution, metabolism, and elimination (ADME) properties. Adult oral models were developed in PK-Sim® and used to build a lactation model in MoBi® to simulate drug transfer into human milk. The adult model was applied to breastfeeding pediatrics (ages 1 to 23 months) and subsequently integrated with the lactation model to simulate infant drug exposure according to age, size, and breastfeeding frequency. Physiologically based pharmacokinetic (PBPK) model simulations captured the data used for verification both in adults and pediatrics. Lactation simulations captured observed milk and plasma data corresponding to doses of 150 mg administered twice daily to lactating individuals, and estimated a relative infant dose (RID) of approximately 7% of the maternal dose. The infant drug exposure simulations showed peak plasma concentrations of 0.44 μg/mL occurring within the first 2 weeks of life, followed by gradual decline with age after week four. The modeling framework performs well for this second prototypical drug and warrants expansion to other drugs for further validation. PBPK modeling and simulation approaches together with clinical lactation data could ultimately help inform infant drug exposure risk assessments to guide clinical decision making.

Organizations such as the American Academy of Pediatrics,WorldHealth Organization, and Centers for Disease Control and Prevention recommend breastfeeding because of multiple health benefits and early motherinfant bonding. However, it is reported that lactating individuals take an average of 4 medications during the lactation period.1 Information on drug presence in human milk and resulting exposure to the breastfed infant is necessary to assess the potential risk to the breastfed infant and critical to the lactating individual’s benefit-risk decision whether to use the medication. Unfortunately, data collected from clinical lactation studies may be limited, and evaluating all possible drug combinations would be challenging, if not impossible, to study experimentally. New opportunities in clinical pharmacology are arising in the translation of available concentration (exposure) data to further inform risk to breastfeeding infants.

Levonorgestrel (LNG) is a progestin used in many contraceptive formulations, including subcutaneous implants. There is an unmet need for developing long-acting formulations for LNG. To develop long-acting formulations, release functions need to be investigated for LNG implant. Therefore, a release model was developed and integrated into an LNG physiologically-based pharmacokinetic (PBPK) model. Utilizing a previously developed LNG PBPK model, subcutaneous administration of 150 mg LNG was implemented into the modeling framework. To mimic LNG release, ten functions incorporating formulation-specific mechanisms were explored. Release kinetic parameters and bioavailability were optimized using Jadelle® clinical trial data (n = 321) and verified using two additional clinical trials (n = 216). The First-order release and Biexponential release models showed the best fit with observed data, the adjusted R-squared (R2) value is 0.9170. The maximum released amount is approximately 50% of the loaded dose and the release rate is 0.0009 per day. The Biexponential model also showed good agreement with the data (adjusted R2 = 0.9113). Both models could recapitulate observed plasma concentrations after integration into the PBPK simulations. First-order and Biexponential release functionality may be useful in modeling subcutaneous LNG implants. The developed model captures central tendency of the observed data as well as variability of release kinetics. Future work focuses on incorporating various clinical scenarios into model simulations, including drug-drug interactions and a range of BMIs.

Characterization of infant drug exposure through human milk is important and underexplored. Because infant plasma concentrations are not frequently collected in clinical lactation studies, modeling and simulation approaches can integrate physiology, available milk concentrations, and pediatric data to inform exposure in breastfeeding infants. A physiologically based pharmacokinetic model was built for sotalol, a renally eliminated drug, to simulate infant drug exposure from human milk. Intravenous and oral adult models were built, optimized, and scaled to an oral pediatric model for a breastfeeding‐relevant age group (<2 years). Model simulations captured the data that were put aside for verification. The resulting pediatric model was applied to predict the impacts of sex, infant body size, breastfeeding frequency, age, and maternal dose (240 and 433 mg) on drug exposure during breastfeeding. Simulations suggest a minimal effect of sex or frequency on total sotalol exposure. Infants in the 90th percentile in height and weight have predicted exposures ≈20% higher than infants of the same age in the 10th percentile due to increased milk intake. The simulated infant exposures increase throughout the first 2 weeks of life and are maintained at the highest concentrations in weeks 2‐4, with a consistent decrease observed as infants age. Simulations suggest that breastfeeding infants will have plasma concentrations in the lower range observed in infants administered sotalol. With further validation on additional drugs, physiologically based pharmacokinetic modeling approaches could use lactation data to a greater extent and provide comprehensive information to support decisions regarding medication use during breastfeeding.

Background: The treatment of cancer during pregnancy remains challenging with knowledge gaps in drug dosage, safety, and efficacy due to the under-representation of this population in clinical trials. Our aim was to investigate physiological changes reported in both pregnancy and cancer populations into a PBPK modeling framework that allows for a more accurate estimation of PK changes in pregnant patients with cancer. Methods: Paclitaxel and docetaxel were selected to validate a population model using clinical data from pregnant patients with cancer. The validated population model was subsequently used to predict the PK of acalabrutinib in pregnant patients with cancer. Results: The Simcyp pregnancy population model reasonably predicted the PK of docetaxel in pregnant patients with cancer, while a modified model that included a 2.5-fold increase in CYP2C8 abundance, consistent with the increased expression during pregnancy, was needed to reasonably predict the PK of paclitaxel in pregnant patients with cancer. Changes in protein binding levels of patients with cancer had a minimal impact on the predicted clearance of paclitaxel and docetaxel. PBPK modeling predicted approximately 60% lower AUC and Cmax for acalabrutinib in pregnant versus non-pregnant patients with cancer. Conclusions: Our results suggest that PBPK modeling is a promising approach to investigate the effects of pregnancy and cancer on the PK of oncology drugs and potentially inform dosing for pregnant patients with cancer. Further evaluation and refinement of the population model are needed for pregnant patients with cancer with additional compounds and clinical PK data.

The main objective of this tutorial is to provide the readers with a roadmap of how to establish increasingly complex target-mediated drug disposition (TMDD) models for monoclonal antibodies. To this end, we built mathematical models, each with a detailed visualization, starting from the basic TMDD model by Mager and Jusko to the well-established, physiologically based model by Li et al. in a step-wise fashion to highlight the relative importance of key physiological processes that impact mAb kinetics and system dynamics. As the models become more complex, the question of structural and parameter identifiability arises. To address this question, we work through a trastuzumab case example to guide the modeler's choice for model and parameter optimization in light of the context of use. We leave the readers of this tutorial with a brief summary of the advantages and limitations of each model expansion, as well as the model source codes for further self-guided exploration and hands-on analysis.

Worldwide, 922 million women of reproductive age (or their partners) use some sort of contraception to prevent pregnancy. Oral combined hormonal contraceptives (CHCs) typically utilize a combination of a progestin and an estrogen. CHCs are potentially at risk to metabolic drug–drug interaction (DDI) via CYP3A4, the main enzyme involved in the oxidative metabolism of ethinyl estradiol and most progestins (e.g., levonorgestrel (LNG) and drospirenone (DRSP)). Recently, the US Food and Drug Administration (FDA) issued a guidance addressing metabolic DDIs in the realm of CHC, establishing an overall class‐based recommendation with respect to avoidance of CYP3A4 induction interactions. Given that different progestins have varying magnitudes of fraction metabolized by CYP3A4 (fmCYP3A4), it would be of clinical benefit to determine if all progestins are at the same risk to CYP3A4‐mediated metabolic DDIs. LNG and DRSP are commonly used progestins that are at the margins of the rifampicin induction effect observed in vivo because they have the relatively lowest and highest fmCYP3A4 among commonly used CHC formulations containing norgestimate, desogestrel, norgestrel, and norethindrone. Therefore, we applied a multi‐pronged strategy (i.e., (i) development of the physiologically‐based pharmacokinetic models; (ii) comparison of the effect of CYP3A inducers and inhibitors on DRSP vs. LNG; and (iii) providing the clinical‐practice context based on real‐world data, to explore the difference in DDI risk for oral CHCs.

BACKGROUND:Modeling approaches offer a novel way to detect and predict coagulopathy in trauma patients. A dynamic model, built and tested on thromboelastogram (TEG) data, was used to generate a virtual library of over 160,000 simulated RapidTEGs. The patient-specific parameters are the initial platelet count, platelet activation rate, thrombus growth rate, and lysis rate (P(0), k1, k2, and k3, respectively).METHODS:Patient data from both STAAMP (n = 182 patients) and PAMPer (n = 111 patients) clinical trials were collected. A total of 873 RapidTEGs were analyzed. One hundred sixteen TEGs indicated maximum amplitude (MA) below normal and 466 TEGs indicated lysis percent above normal. Each patient's TEG response was compared against the virtual library of TEGs to determine library trajectories having the least sum-of-squared error versus the patient TEG up to each specified evaluation time ∈ (3, 4, 5, 7.5, 10, 15, 20 minutes). Using 10 nearest-neighbor trajectories, a logistic regression was performed to predict if the patient TEG indicated MA below normal (<50 mm), lysis percent 30 minutes after MA (LY30) greater than 3%, and/or blood transfusion need using the parameters from the dynamic model.RESULTS:The algorithm predicts abnormal MA values using the initial 3 minutes of RapidTEG data with a median area under the curve of 0.95, and improves with more data to 0.98 by 10 minutes. Prediction of future platelet and packed red blood cell transfusion based on parameters at 4 and 5 minutes, respectively, provides equivalent predictions to the traditional TEG parameters in significantly less time. Dynamic model parameters could not predict abnormal LY30 or future fresh-frozen plasma transfusion.CONCLUSION:This analysis could be incorporated into TEG software and workflow to quickly estimate if the MA would be below or above threshold value within the initial minutes following a TEG, along with an estimate of what blood products to have on hand.LEVEL OF EVIDENCE:Therapeutic/Care Management: Level IV.

Background Maternal hemorrhage protocols involve risk screening. These protocols prepare clinicians for potential hemorrhage and transfusion in individual patients. Patient-specific estimation and stratification of risk may improve maternal outcomes. Study Design and Methods Prediction models for hemorrhage and transfusion were trained and tested in a data set of 74 variables from 63 973 deliveries (97.6% of the source population of 65 560 deliveries included in a perinatal database from an academic urban delivery center) with sufficient data at pertinent time points: antepartum, peripartum, and postpartum. Hemorrhage and transfusion were present in 6% and 1.6% of deliveries, respectively. Model performance was evaluated with the receiver operating characteristic (ROC), precision-recall curves, and the Hosmer-Lemeshow calibration statistic. Results For hemorrhage risk prediction, logistic regression model discrimination showed ROCs of 0.633, 0.643, and 0.661 for the antepartum, peripartum, and postpartum models, respectively. These improve upon the California Maternal Quality Care Collaborative (CMQCC) accuracy of 0.613 for hemorrhage. Predictions of transfusion resulted in ROCs of 0.806, 0.822, and 0.854 for the antepartum, peripartum, and postpartum models, respectively. Previously described and new risk factors were identified. Models were not well calibrated with Hosmer-Lemeshow statistic P values between .001 and .6. Conclusions Our models improve on existing risk assessment; however, further enhancement might require the inclusion of more granular, dynamic data. With the goal of increasing translatability, this work was distilled to an online open-source repository, including a form allowing risk factor inputs and outputs of CMQCC risk, alongside our numerical risk estimation and stratification of hemorrhage and transfusion.

Defects in blood clotting (coagulopathies) are linked to severe outcomes in mothers suffering from obstetrical hemorrhage. Identifying patients with a coagulopathy poses a challenge for clinicians, who are required to make quick treatment decisions in fast-paced environments with a high degree of uncertainty. Integrating data from point-of-care coagulation tests with mathematical models of coagulation presents an exciting opportunity to improve patient outcomes by reducing this uncertainty. A model with parameters estimated from individual patient data can provide clinicians with a way to compare patients and group them into categories of probable coagulopathy based on biologically-interpretable parameters. With this in mind, we developed a mechanism-inspired model of blood coagulation calibrated against thromboelastogram (TEG) data. Markov Chain Monte Carlo and sensitivity analysis were used to assess the identifiability and distribution of model parameters for 25 obstetric patients. The ability of our model to separate patients in parameter space based on differences in observed TEG response lends credence to the feasibility of using dynamic models as tools for identifying coagulopathy subtypes within the obstetric population.

mitted: July 23, 2019, Revised: Published online: February 6, 2 m the Department of Chemical Swanson School of Engineerin M.D.N., J.L.S., G.C.), Departm of Bioengineering (R.S.P., G.C Pittsburgh, Pittsburgh, PA. plemental digital content is avai the printed text, and links to the article on the journal’s Web sit dress for reprints: Gilles Clermon University of Pittsburgh, 602A email: cler@pitt.edu.

Obstetric patients show an increased risk of developing venous thromboembolism (VTE). Modeling the changes that occur during and after delivery may help determine which patients will develop VTE and when they might be likely to develop this dangerous complication after delivery. Combining a model of blood volume changes with a model of activated clotting factor concentrations, which are both affected during delivery, can identify scenarios that may lead to an increased risk of developing clots in the venous vascular space. This Ordinary Differential Equation (ODE) model recapitulates known phenomena including an elevated coagulation response during delivery and hemorrhage leading to increased clotting factor concentration in the vascular space. The simulation from normal activation without hemorrhage results in a spike in clotting factors in the vascular space to reestablish hemostasis after delivery. With twice the activation rate, simulations show elevated and extended duration of activated clotting factor presence in the vascular space. With response to a hemorrhage with normal activation, the resulting elevation and duration is further increased. This model, when tailored to individual patients, could lead to the development of a VTE risk assessment tool for clinicians to help mitigate and reduce an individual's risk of developing this deadly complication.

Modeling clotting dynamics that occur during a thromboelastogram (TEG), a functional ex-vivo assay of coagulation, has the potential to provide useful clinical insights. Taking a systems view, the ex vivo system (TEG) dynamics are modeled without explicitly representing the complex in vivo clotting cascade. The model utilizes reaction engineering principles represented as ordinary differential equations (ODEs), which are solved in MATLAB ( c ©2016, The Mathworks Inc.) or Python. The model structure uses linear kinetics, except for the thrombus growth rate, which has a quadratic dependence on activated platelets. Estimated parameters include the initial platelet count, platelet activation rate, thrombus growth rate, and lysis rate (P0, k1, k2, and k3, respectively). Parameter estimation via nonlinear least-squares optimization of model predictions against clinical data used PYOMO 4.3 with the IPOPT solver. For 60 sample TEG tracings, the model yielded an average relative error of 6.42%, which demonstrates a simple mathematical description of coagulation is able to capture TEG data. Multi-parameter distributions demonstrates patient variability, suggesting it may be possible to identify patient endotypes from these distributions. A model-informed endotyping of patient-specific coagulopathic state could guide clinicians in selecting individualized treatment decisions leading to improved patient outcomes.

Copyright © 2016 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.