Prediction Of Clearance Research Articles

Mechanistic Modeling of In Vitro Biopharmaceutic Data for a Weak Acid Drug: A Pathway Towards Deriving Fundamental Parameters for Physiologically Based Biopharmaceutic Modeling.

Mechanistic modeling of in vitro experiments using metabolic enzyme systems enables the extrapolation of metabolic clearance for in vitro-in vivo predictions. This is particularly important for successful clearance predictions using physiologically based pharmacokinetic (PBPK) modeling. The concept of mechanistic modeling can also be extended to biopharmaceutics, where in vitro data is used to predict the in vivo pharmacokinetic profile of the drug. This approach further allows for the identification of parameters that are critical for oral drug absorption in vivo. However, the routine use of this analysis approach has been hindered by the lack of an integrated analysis workflow. The objective of this tutorial is to (1) review processes and parameters contributing to oral drug absorption in increasing levels of complexity, (2) outline a general physiologically based biopharmaceutic modeling workflow for weak acids, and (3) illustrate the outlined concepts via an ibuprofen (i.e., a weak, poorly soluble acid) case example in order to provide practical guidance on how to integrate biopharmaceutic and physiological data to better understand oral drug absorption. In the future, we plan to explore the usefulness of this tutorial/roadmap to inform the development of PBPK models for BCS 2 weak bases, by expanding the stepwise modeling approach to accommodate more intricate scenarios, including the presence of diprotic basic compounds and acidifying agents within the formulation.

Prediction of human clearance using human liver microsomes and hepatocytes with 455 drugs

Prediction of human clearance using human liver microsomes and hepatocytes with 455 drugs

Prediction and validation of human hepatobiliary clearances and hepatic concentrations of transported drugs using the relative expression factor (REF) and PET imaging approaches

Prediction and validation of human hepatobiliary clearances and hepatic concentrations of transported drugs using the relative expression factor (REF) and PET imaging approaches

Dynamic free fraction: concept, methodology and impact on hepatic clearance prediction using the well-stirred model

Dynamic free fraction: concept, methodology and impact on hepatic clearance prediction using the well-stirred model

Application of machine learning combined with population pharmacokinetics to improve individual prediction of vancomycin clearance in simulated adult patients.

Vancomycin, a glycopeptide antimicrobial drug. PPK has problems such as difficulty in accurately reflecting inter-individual differences, and the PPK model may not be accurate enough to predict individual pharmacokinetic parameters. Therefore, the aim of this study is to investigate whether the application of machine learning combined with the PPK method can improve the prediction of vancomycin CL in adult Chinese patients. In the first step, a vancomycin CL prediction model for Chinese adult patients is given by PPK and Hamilton Monte Carlo sampling is used to obtain the reference CL of 1,000 patients; the second step is to obtain the final prediction model by machine learning using an appropriate model for the predictive factor and the reference CL; and the third step is to randomly select, in the simulated data, a total of 250 patients for prediction effect evaluation. XGBoost model is selected as final machine learning model. More than four-fifths of the subjects' predictive values regarding vancomycin CL are improved by machine learning combined with PPK. Machine learning combined with PPK models is more stable in performance than the PPK method alone for predicting models. The first combination of PPK and machine learning for predictive modeling of vancomycin clearance in adult patients. It provides a reference for clinical pharmacists or clinicians to optimize the initial dosage given to ensure the effectiveness and safety of drug therapy for each patient.

Sialic Acid Mediated Endothelial and Hepatic Uptake: A Mechanism based Mathematic Model Elucidating the Complex Pharmacokinetics and Pharmacodynamics of Efmarodocokin Alfa, a Variably Glycosylated Fusion Protein

Sialic acid (SA) is crucial for protecting glycoproteins from clearance. Efmarodocokin alfa (IL-22Fc), a fusion protein agonist that links IL-22 to the crystallizable fragment (Fc) of human IgG4, contains 8 N-glycosylation sites and exhibits heterogeneous and variable terminal sialylation biodistribution. This presents a unique challenge for Pharmacokinetic (PK) and Pharmacodynamic (PD) analysis and cross-species translation. In this study, we sought to understand how varying SA levels and heterogeneous distribution contribute to IL-22Fc's complex PKPD properties. We initially used homogenous drug material with varying SA levels to examine PKPD in mice. Population PKPD analysis based on mouse data revealed that SA was a critical covariate simultaneously accounting for the substantial between subject variability (BSV) in clearance (CL), distribution clearance (CLd), and volume of distribution (Vd). In addition to the well-established mechanism by which SA inhibits ASGPR activity, we hypothesized a novel mechanism by which decrease in SA increases the drug uptake by endothelial cells. This decrease in SA, leading to more endothelial uptake, was supported by the neonatal Fc receptor (FcRn) dependent cell-based transcytosis assay. The population analysis also suggested in vivo EC50 (IL-22Fc stimulating Reg3β) was independent on SA, while the in-vitro assay indicated a contradictory finding of SA-in vitro potency relationship. We created a mechanism based mathematical (MBM) PKPD model incorporating the decrease in SA mediated endothelial and hepatic uptake, and successfully characterized the SA influence on IL-22Fc PK, as well as the increased PK exposure being responsible for increased PD. Thereby, the MBM model supported that SA has no direct impact on EC50, aligning with the population PKPD analysis. Subsequently, using the MBM PKPD model, we employed 5 subpopulation simulations to reconstitute the heterogeneity of drug material. The simulation accurately predicted the PKPD of heterogeneously and variably sialylated drug in mouse, monkey and human. The successful prospective validation confirmed the MBM's ability to predict IL-22Fc PK across variable SA levels, homogenous to heterogeneous material, and across species (R2=0.964 for clearance prediction). Our model prediction suggests an average of 1 mol/mol SA increase leads to a 50% increase in drug exposure. This underlines the significance of controlling sialic acid levels during lot-to-lot manufacturing.

In Vitro to In Vivo Scalars for Drug Clearance in Nonalcoholic Fatty Liver and Steatohepatitis.

In vitro-in vivo extrapolation (IVIVE) allows prediction of clinical outcomes across populations from in vitro data using specific scalars tailored to the biologic characteristics of each population. This study experimentally determined scalars for patients with varying degrees of nonalcoholic fatty liver disease (NAFLD), ranging from fatty liver to nonalcoholic steatohepatitis (NASH) and cirrhosis. Microsomal, S9, and cytosol fractions were extracted from 36 histologically normal and 66 NAFLD livers (27 nonalcoholic fatty liver [NAFL], 13 NASH, and 26 NASH with cirrhosis). Corrected microsomal protein per gram liver (MPPGL) progressively decreased with disease severity (26.8, 27.4, and 24.3 mg/g in NAFL, NASH, and NASH/cirrhosis, respectively, compared with 35.6 mg/g in normal livers; ANOVA, P < 0.001). Homogenate, S9, and cytosolic protein showed a consistent trend of decline in NASH/cirrhosis relative to normal control (post-hoc t test, P < 0.05). No differences across the groups were observed in homogenate, S9, cytosolic, and microsomal protein content in matched kidney samples. MPPGL-based scalars that combine protein content with liver size revealed that the reduction in MPPGL in NAFL and NASH was compensated by the reported increase in liver size (relative scalar ratios of 0.96 and 0.99, respectively), which was not the case with NASH/cirrhosis (ratio of 0.63), compared with healthy control. Physiologically based pharmacokinetics-informed global sensitivity analysis of the relative contribution of IVIVE scalars (hepatic CYP3A4 abundance, MPPGL, and liver size) to variability in exposure (area under the curve) to three CYP3A substrates (alprazolam, midazolam, and ibrutinib) revealed enzyme abundance as the most significant parameter, followed by MPPGL, whereas liver volume was the least impactful factor. SIGNIFICANCE STATEMENT: Nonalcoholic fatty liver disease-specific scalars necessary for extrapolation from in vitro systems to liver tissue are lacking. These are required in clearance prediction and dose selection in nonalcoholic fatty liver and steatohepatitis populations. Previously reported disease-driven changes have focused on cirrhosis, with no data on the initial stages of liver disease. The authors obtained experimental values for microsomal, cytosolic, and S9 fractions and assessed the relative impact of microsomal scalars on predicted exposure to substrate drugs using physiologically based pharmacokinetics.

Firing Accuracy in Rotor Dynamics Prediction of Unbalanced Mass and Bearing Clearance

The XYZ center of gravity of the barrel part of the M134 Gatling gun and the components composed of bullets was calculated from the five types of clearances between the bearings without clearance to the bearings with clearance. According to the calculation of five clearances, the angle deviation is about 11.50, and based on the target of 3000m, the sideflight is about 600 meters. From the perspective of Gatling weapons, it has been officially determined that transfer weapons must primarily rely on surface killing.

Improving In Vitro-In Vivo Extrapolation of Clearance Using Rat Liver Microsomes for Highly Plasma Protein-Bound Molecules.

It is common practice in drug discovery and development to predict in vivo hepatic clearance from in vitro incubations with liver microsomes or hepatocytes using the well-stirred model (WSM). When applying the WSM to a set of approximately 3000 Novartis research compounds, 73% of neutral and basic compounds (extended clearance classification system [ECCS] class 2) were well-predicted within 3-fold. In contrast, only 44% (ECCS class 1A) or 34% (ECCS class 1B) of acids were predicted within 3-fold. To explore the hypothesis whether the higher degree of plasma protein binding for acids contributes to the in vitro-in vivo correlation (IVIVC) disconnect, 68 proprietary compounds were incubated with rat liver microsomes in the presence and absence of 5% plasma. A minor impact of plasma on clearance IVIVC was found for moderately bound compounds (fraction unbound in plasma [fup] ≥1%). However, addition of plasma significantly improved the IVIVC for highly bound compounds (fup <1%) as indicated by an increase of the average fold error from 0.10 to 0.36. Correlating fup with the scaled unbound intrinsic clearance ratio in the presence or absence of plasma allowed the establishment of an empirical, nonlinear correction equation that depends on fup Taken together, estimation of the metabolic clearance of highly bound compounds was enhanced by the addition of plasma to microsomal incubations. For standard incubations in buffer only, application of an empirical correction provided improved clearance predictions. SIGNIFICANCE STATEMENT: Application of the well-stirred liver model for clearance in vitro-in vivo extrapolation (IVIVE) in rat generally underpredicts the clearance of acids and the strong protein binding of acids is suspected to be one responsible factor. Unbound intrinsic in vitro clearance (CLint,u) determinations using rat liver microsomes supplemented with 5% plasma resulted in an improved IVIVE. An empirical equation was derived that can be applied to correct CLint,u-values in dependance of fraction unbound in plasma (fup) and measured CLint in buffer.

Prediction of Human Clearance Using In Silico Models with Reduced Bias.

Predicting human clearance with high accuracy from in silico-derived parameters alone is highly desirable, as it is fast, saves in vitro resources, and is animal-sparing. We derived random forest (RF) models from 1340 compounds with human intravenous pharmacokinetic (PK) data, the largest data set publicly available today. To assess the general applicability of the RF models, we systematically removed structural-therapeutic class analogues and other compounds with structural similarity from the training sets. For a quasi-prospective test set of 343 compounds, we show that RF models devoid of structurally similar compounds in the training set predict human clearance with a geometric mean fold error (GMFE) of 3.3. While the observed GMFE illustrates how difficult it is to generate a useful model that is broadly applicable, we posit that our RF models yield a more realistic assessment of how well human clearance can be predicted prospectively. We deployed the conformal prediction formalism to assess the model applicability and to determine the prediction confidence intervals for each prediction. We observed that clearance can be predicted better for renally cleared compounds than for other clearance mechanisms. We show that applying a classification model for predicting renal clearance identifies a subset of compounds for which clearance can be predicted with higher accuracy, yielding a GMFE of 2.3. In addition, our in silico RF human clearance models compared well to models derived from scaling human hepatocytes or preclinical in vivo data.

Hydraulic resistance of three-dimensional pial perivascular spaces in the brain

BackgroundPerivascular spaces (PVSs) carry cerebrospinal fluid (CSF) around the brain, facilitating healthy waste clearance. Measuring those flows in vivo is difficult, and often impossible, because PVSs are small, so accurate modeling is essential for understanding brain clearance. The most important parameter for modeling flow in a PVS is its hydraulic resistance, defined as the ratio of pressure drop to volume flow rate, which depends on its size and shape. In particular, the local resistance per unit length varies along a PVS and depends on variations in the local cross section.MethodsUsing segmented, three-dimensional images of pial PVSs in mice, we performed fluid dynamical simulations to calculate the resistance per unit length. We applied extended lubrication theory to elucidate the difference between the calculated resistance and the expected resistance assuming a uniform flow. We tested four different approximation methods, and a novel correction factor to determine how to accurately estimate resistance per unit length with low computational cost. To assess the impact of assuming unidirectional flow, we also considered a circular duct whose cross-sectional area varied sinusoidally along its length.ResultsWe found that modeling a PVS as a series of short ducts with uniform flow, and numerically solving for the flow in each, yields good resistance estimates at low cost. If the second derivative of area with respect to axial location is less than 2, error is typically less than 15%, and can be reduced further with our correction factor. To make estimates with even lower cost, we found that instead of solving for the resistance numerically, the well-known resistance of a circular duct could be scaled by a shape factor. As long as the aspect ratio of the cross section was less than 0.7, the additional error was less than 10%.ConclusionsNeglecting off-axis velocity components underestimates the average resistance, but the error can be reduced with a simple correction factor. These results could increase the accuracy of future models of brain-wide and local CSF flow, enabling better prediction of clearance, for example, as it varies with age, brain state, and pathological conditions.

Prediction Model for the Clearance of Hepatitis B Surface Antigen in Patients with Chronic Hepatitis B before Interferon Therapy: A Prospective Case-Control Study.

To evaluate the prediction model comprised of patients' laboratory results and single-nucleotide polymorphism (SNP) markers of host gene for the clearance of hepatitis B surface antigen (HBsAg) in patients with chronic hepatitis B (CHB) who underwent interferon (IFN)-α therapy, this prospective case-control study enrolled 131 patients with CHB who underwent IFN-α-based regimens in our hospital between January 2015 and September 2019. Among them, 56 cases were without HBsAg clearance, while the other 75 cases had HBsAg clearance. Multivariable logistic regression analysis showed that CYP27B1 rs4646536 (odd ratio [OR] = 0.155, 95% CI: 0.030-0.807, p = 0.027), PAK4 rs9676717 (OR = 11.237, 95% CI: 1.768-71.409, p = 0.010), IL28B rs12979860 (OR = 0.059, 95% CI: 0.006-0.604, p = 0.017), baseline HBsAg (OR = 0.170, 95% CI: 0.040-0.716, p = 0.016), and HBeAg status (OR = 3.971, 95% CI: 1.138-13.859, p = 0.031) were independently associated with HBsAg clearance. The model that included rs3077, rs4646536, rs9676717, rs2850015, rs12979860, baseline HBsAg, HBeAg status, and HBV DNA had the best prediction performance for HBsAg clearance prediction, with AUC = 0.877, 80% sensitivity, and 81% specificity. In conclusion, laboratory results and gene polymorphisms before treatment might have a good predictive value for HbsAg clearance after IFN-α treatment in CHB.

Scaling approaches for the prediction of human clearance of LNA-i-mir-221: A retrospective validation

LNA-i-miR-221 is a novel microRNA(miRNA)-221 inhibitor designed for the treatment of human malignancies. It has recently undergone phase 1 clinical trial (P1CT) and early pharmacokinetics (PKs) data in cancer patients are now available. We previously used multiple allometric interspecies scaling methods to draw inferences about LNA-i-miR-221 PKs in humans and estimated the patient dose based on the safe and pharmacodynamic (PD) active dose observed in mice, therefore providing a framework for the definition of safe starting and escalation doses for the P1CT. The preliminary data collected during the P1CT showed that the LNA-i-miR-221 anticipated doses, according to our human PK estimation approach, were indeed well tolerated and effective. PD data demonstrated concentration-dependent downregulation of miR-221 and upregulation of its CDKN1B/p27 and PTEN canonical targets as well as stable disease in 8 (50.0%) patients and partial response in 1 (6.3%) colorectal cancer case. Here, we detail the experimentally evaluated PK parameters of LNA-i-miR-221 in human, using both a non-compartmental and a population PKs approach. The population approach was adequately described by a three-compartments model with first-order elimination. The recorded age, sex and body weight of patients were evaluated as potential covariates. The estimated typical population parameter values were clearance (CL = 200 mL/h/kg), central volume of distribution (V1 = 45 mL/kg), peripheral volume of distribution (V2 = 200 mL/kg, volume of the second peripheral compartment V3 = 930 mL/h/kg) and inter-compartmental clearance (Q2 = 480 mL/h/kg and Q3 = 68 mL/h/kg). Age was found to be a predictor of Q3, with a statistically significant correlation. This work aimed also at retrospectively comparing the measured plasmatic clearance values with those predicted by different allometric scaling approaches. Our comparative analysis showed that the most accurate prediction was achieved by applying the single species allometric scaling approach and that the use of more than one species in allometric scaling to predict therapeutic oligonucleotides PKs would not necessarily generate the best prediction. Finally, our predictive approach was found accurate not only in predicting the main PK parameters in human but suggesting the range of effective and safe dose to be applied in the next clinic phase 2.

Transient Tip Clearance Prediction Model Considering Transient Radial Temperature Distribution of Discs in a Gas Turbine Engine

Abstract The tip clearance of turbomachinery components, such as compressor and turbine, significantly impacts the gas turbine engine performance, including the thrust and fuel consumption. Therefore, an appropriate prediction of the turbomachinery tip clearance is indispensable in improving the accuracy of engine performance evaluations. The traditional 1D transient tip clearance prediction model does not consider the radial temperature distribution of the disk (also called disc). However, as the engine performance keeps improving, the radial temperature gradient of the disk also increases. In order to maintain the prediction accuracy within an acceptable range, a generic transient tip clearance prediction model is developed in this paper. Based on the special treatment of the axial and radial convective heat transfer boundary regions of the disk, a 2D transient heat transfer model is established, and the radial temperature of the disk is predicted. At the same time, the interaction effect of the thermal expansion between the two different layers of the shroud is studied to improve the prediction accuracy further. In addition, the centrifugal deformation of the blade and disk with variable-thickness characteristics is considered in the new transient model. Compared with the traditional 1D transient prediction model, the model developed in this paper further improves the predicting accuracy of the transient tip clearance, which is verified by the high-fidelity fluid–solid–thermal coupling method. The prediction model provides a useful tool for evaluating the transient tip clearance. It may also benefit the future design strategy of active clearance control.

Real-world application of PBPK in drug discovery.

The utility of PBPK models in support of drug development has been well documented. During the discovery stage, PBPK has increasingly been applied for early risk assessment, prediction of human dose, toxicokinetic dose projection and early formulation assessment. Previous review articles have proposed model building and application strategies for PBPK-based first in human predictions with comprehensive descriptions of the individual components of PBPK models. This includes the generation of decision trees, based on comprehensive literature reviews, to guide the application of PBPK in the discovery setting. The goal of this mini review is to provide additional guidance on the real-world application of PBPK, in support of the discovery stage of drug development. In this mini review, our goal is to provide guidance on the typical steps involved in the development and application of a PBPK model during drug discovery to assist in decision making. We have illustrated our recommended approach through description of case examples, where PBPK has been successfully applied to aid in human PK projection, candidate selection and prediction of drug interaction liability for parent and metabolite. Through these case studies, we have highlighted fundamental issues, including pre-verification in preclinical species, the application of empirical scalars in the prediction of in vivo clearance from in vitro systems, in silico prediction of permeability and the exploration of aqueous and biorelevant solubility data to predict dissolution. In addition, current knowledge gaps have been highlighted and future directions proposed. Significance Statement Through description of three case studies, we have highlighted the fundamental principles of PBPK application during drug discovery. These include pre-verification of the model in preclinical species, application of empirical scalars where necessary in the prediction of clearance, in silico prediction of permeability, and the exploration of aqueous and biorelevant solubility data to predict dissolution. In addition, current knowledge gaps have been highlighted and future directions proposed.

Human Dose and Pharmacokinetic Predictions for Biologics at Boehringer Ingelheim: A Retrospective Analysis.

Accurate predictions of pharmacokinetics and efficacious doses for biologics in humans are critical for selecting appropriate first-in-human starting doses and dose ranges and for estimating clinical material needs and cost of goods. This also impacts clinical feasibility, particularly for subcutaneously administered biologics. We performed a comprehensive comparison between predicted and observed clearances and doses in humans for a set of 22 biologic drugs developed at Boehringer Ingelheim (BI) over the last 2decades. The analysis included biologics across three therapeutic areas comprising a wide variety of modalities: mono- and bispecific monoclonal antibodies (mAbs) and nanobodies and a Fab fragment. Our analysis showed that observed clearances in humans were within twofold of predicted clearances for 17 out of 20 biologics (85%). Six biologics had uncharacteristically high observed human clearances (range 32-280mL/h) for their respective molecular classes, impacting their clinical developability. For three molecules, molecular characteristics contributed to the high clearance. Clinically selected doses were within twofold of predicted for 58% of projects. With 42% and 25% of projects selecting clinical doses higher than two- or threefold the predicted value, respectively, the importance of better understanding not only the pharmacokinetic (PK) but also the predictivity of pharmacodynamic models is highlighted. We provide a clinical pharmacology perspective on the commonly accepted twofold range of human clearance predictions as well as the implications of higher than predicted targeted efficacious plasma concentration on clinical development. Finally, an analysis of key success factors for biologics at BI was conducted, which may be relevant for the entire pharmaceutical industry. This is one of the largest retrospective analyses for biologics and provides further evidence that successful predictions of human PK and efficacious dose will be further facilitated by gathering key translational data early in research.

Semi-empirical and machine learning-based prediction of site of metabolisms mediated by aldehyde oxidase

Prediction of drug clearance by aldehyde oxidase (AO) catalyzed metabolism has been of great interest in drug discovery. We report a fast, accurate, and fully-automated tool that combines semi-empirical quantum mechanical (SQM) calculations and machine-learning (ML) technique to predict AO-mediated site of metabolism (SOM). We infer that calculated probability scores together with those based on approximate C(sp2)−H bond dissociation energies accomplish accurate and immediate prediction of the number and rank ordering of SOMs. Tests on a series of 150 AO substrates demonstrate that our computational strategy can be applicable to high-throughput screening of metabolically stable compounds.

Population pharmacokinetic model of cefepime for critically ill adults: a comparative assessment of eGFR equations.

Cefepime exhibits highly variable pharmacokinetics in critically ill patients. The purpose of this study was to develop and qualify a population pharmacokinetic model for use in the critically ill and investigate the impact of various estimated glomerular filtration rate (eGFR) equations using creatinine, cystatin C, or both on model parameters. This was a prospective study of critically ill adults hospitalized at an academic medical center treated with intravenous cefepime. Individuals with acute kidney injury or on kidney replacement therapy or extracorporeal membrane oxygenation were excluded. A nonlinear mixed-effects population pharmacokinetic model was developed using data collected from 2018 to 2022. The 120 included individuals contributed 379 serum samples for analysis. A two-compartment pharmacokinetic model with first-order elimination best described the data. The population mean parameters (standard error) in the final model were 7.84 (0.24) L/h for CL1 and 15.6 (1.45) L for V1. Q was fixed at 7.09 L/h and V2 was fixed at 10.6 L, due to low observed interindividual variation in these parameters. The final model included weight as a covariate for volume of distribution and the eGFRcr-cysC (mL/min) as a predictor of drug clearance. In summary, a population pharmacokinetic model for cefepime was created for critically ill adults. The study demonstrated the importance of cystatin C to prediction of cefepime clearance. Cefepime dosing models which use an eGFR equation inclusive of cystatin C are likely to exhibit improved accuracy and precision compared to dosing models which incorporate an eGFR equation with only creatinine.

The Comparison of Machine Learning and Mechanistic In Vitro-In Vivo Extrapolation Models for the Prediction of Human Intrinsic Clearance.

Accurate prediction of human pharmacokinetics (PK) remains one of the key objectives of drug metabolism and PK (DMPK) scientists in drug discovery projects. This is typically performed by using in vitro-in vivo extrapolation (IVIVE) based on mechanistic PK models. In recent years, machine learning (ML), with its ability to harness patterns from previous outcomes to predict future events, has gained increased popularity in application to absorption, distribution, metabolism, and excretion (ADME) sciences. This study compares the performance of various ML and mechanistic models for the prediction of human IV clearance for a large (645) set of diverse compounds with literature human IV PK data, as well as measured relevant in vitro end points. ML models were built using multiple approaches for the descriptors: (1) calculated physical properties and structural descriptors based on chemical structure alone (classical QSAR/QSPR); (2) in vitro measured inputs only with no structure-based descriptors (ML IVIVE); and (3) in silico ML IVIVE using in silico model predictions for the in vitro inputs. For the mechanistic models, well-stirred and parallel-tube liver models were considered with and without the use of empirical scaling factors and with and without renal clearance. The best ML model for the prediction of in vivo human intrinsic clearance (CLint) was an in vitro ML IVIVE model using only six in vitro inputs with an average absolute fold error (AAFE) of 2.5. The best mechanistic model used the parallel-tube liver model, with empirical scaling factors resulting in an AAFE of 2.8. The corresponding mechanistic model with full in silico inputs achieved an AAFE of 3.3. These relative performances of the models were confirmed with the prediction of 16 Pfizer drug candidates that were not part of the original data set. Results show that ML IVIVE models are comparable to or superior to their best mechanistic counterparts. We also show that ML IVIVE models can be used to derive insights into factors for the improvement of mechanistic PK prediction.

Prediction of negative axillary node clearance by sentinel node-positive to total node ratio: a retrospective cohort study.

Increasing evidence suggests that de-escalation of axillary surgery is safe, without significantly impacting patient outcome. Obtaining positive lymph nodes at a sentinel lymph node biopsy (SNB) can guide decisions toward the requirement of axillary nodal clearance (ANC). However, methods to predict how many further nodes will be positive are not available. This study investigates the feasibility of predicting the likelihood of a negative ANC based on the ratio between positive nodes and the total number of lymph nodes excised at SNB. Retrospective data from January 2017 to March 2022 was collected from electronic medical records. Patients with oestrogen receptor (ER) positive and HER2 negative receptor disease were included in the study. ER-negative and HER2-positive disease was excluded, alongside patients who had chemotherapy before ANC. Of 102 patients, 58.8% (n=60) had no macrometastasis at ANC. On average, 2.76 lymph nodes were removed at SNB. A higher SNB ratio of positive to total nodes [OR 11.09 (CI 95% 2.33-52.72), P=0.002] had a significant association with positive nodes during ANC. SNB ratio less than or equal to 0.33 (1/3) had a specificity of 79.2% in identifying cases that later had a negative completion ANC, with a 95.8% specificity of no further upgrade of nodal staging. A low SNB ratio of less than 0.33 (1/3) has a high specificity in excluding the upgradation of nodal staging on completion of ANC, with a false-negative rate of less than 5%. This may be used to identify patients with a low risk of axillary metastasis, who can avoid ANC.