Method for predicting pharmacokinetic behavior of perfluoro and polyfluoroalkyl substances in living organisms

Abstract

The application discloses a method for predicting the pharmacokinetic behavior of perfluoro and polyfluoro alkyl substances in vivo, and relates to the technical field of environmental health. The method comprises the following steps: establishing a PBPK model comprising multiple physiological compartments, explicitly introducing parameters representing the binding process of PFAS and key proteins in vivo into the kinetic equation, integrating the metabolic and biliary excretion double elimination mechanisms in the liver compartment equation, explicitly introducing the contribution of enterohepatic circulation in the intestinal compartment equation, and integrating the renal tubular active transport mechanism based on Michaelis-Menten kinetics in the kidney compartment equation; obtaining data through animal experiments and in vitro metabolism experiments, fitting the PFAS specific parameter set by using a genetic optimization algorithm; inputting the parameter set into the model for simulation and prediction; and finally verifying the model using independent dose experimental data, and making predictions based on the verified model. The application improves the accuracy, mechanism and cross-dose extrapolation reliability of the prediction of PFAS in vivo behavior, and is suitable for risk assessment and alternative product screening of PFAS.

Description

Technical Field

[0001] This invention relates to the fields of environmental health and pharmacokinetics, and more specifically, to a method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in organisms. Background Technology

[0002] Per- and polyfluoroalkyl substances (PFAS) are a class of synthetic persistent organic pollutants widely present in environmental media and organisms. Their potential bioaccumulation and toxic effects have become a global environmental and health concern. Accurately predicting the absorption, distribution, metabolism, and excretion of PFAS in organisms is crucial for scientifically assessing their health risks, establishing environmental safety limits, and guiding the development of alternatives. Among existing predictive tools, physiologically based pharmacokinetic (PBPK) models are considered one of the most promising quantitative prediction methods due to their ability to simulate mechanisms based on the physiological structure and biochemical processes of biological systems. PBPK models, by constructing multiple physiological compartments connected by blood flow (such as the liver, kidneys, and adipose tissue) and integrating compound-specific parameters, theoretically enable extrapolation predictions from high-dose experimental data to low-dose environmental exposures and even across species.

[0003] However, applying PBPK models to perfluorinated and polyfluoroalkyl substances still faces the following major challenges: First, in terms of model mechanism, existing PBPK models fail to fully integrate the core process of PFAS binding to key proteins. Most models still use empirically fixed partition coefficients instead of treating protein binding as a modifiable explicit mechanism parameter, thus making it difficult to accurately explain the significant differences in tissue distribution (especially hepatic blood partitioning) of PFAS with different chain lengths, resulting in limited predictive stability. Second, there are deficiencies in the description of key clearance processes. Renal excretion of PFAS is highly dependent on transporter-mediated active transport, but existing models often oversimplify this process, failing to accurately quantify the active transport kinetics in the renal tubules, leading to inaccurate predictions of clearance rates and tissue accumulation. Furthermore, there are shortcomings in modeling key disposal and circulation processes. PFAS clearance in the liver involves multiple pathways, including metabolism and bile excretion, and a portion excreted in bile may be reabsorbed through enterohepatic circulation, significantly affecting its in vivo kinetics. Existing models often fail to adequately characterize this process, failing to fully quantify the coupling mechanism between hepatic multi-pathway elimination and enterohepatic reabsorption, leading to biases in predicting the persistence of PFAS, especially long-chain homologues. Finally, there are weaknesses in model support. On the one hand, model parameters are mostly derived from scattered literature data, lacking a unified parameter set covering multiple PFAS obtained through systematic experiments; on the other hand, most models are only validated at a single dose, and their reliability in cross-dose prediction has not been fully confirmed, limiting their application in practical risk assessment. Therefore, an improved method with clearer mechanisms, more reliable parameters, and more thorough validation is needed to enhance the accuracy and practicality of predicting PFAS behavior in vivo. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances (PFAS) in vivo. This method improves the accuracy, mechanistic basis, and reliability of cross-dose extrapolation in predicting PFAS behavior in vivo, and is applicable to risk assessment of PFAS and screening of alternatives.

[0005] This invention provides a method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo, comprising the following steps: A multi-compartment PBPK model was established based on physiological structure. In the kinetic equation of the PBPK model, parameters characterizing the binding process of PFAS with key proteins in vivo were explicitly introduced to quantify tissue distribution. In the kinetic equation of the liver compartment, two elimination mechanisms, metabolic clearance and bile excretion, were integrated. In the kinetic equation of the intestinal compartment, the contribution of enterohepatic circulation was explicitly introduced. In the kinetic equation of the kidney compartment, the active tubular transport mechanism based on Michaelis-Menten kinetics was integrated. Multi-tissue concentration data of the target PFAS at multiple time points were obtained through animal experiments, and in vitro metabolic data were obtained through in vitro liver microsome experiments. A genetic optimization algorithm was used to fit and obtain the PFAS-specific parameter set for the PBPK model. The PFAS-specific parameter set is input into the PBPK model, the model is run to simulate the dynamic process of the PFAS in vivo, and the pharmacokinetic characteristics are output. The accuracy of the PBPK model prediction results was verified using independent exposure experimental data at doses different from those used for parameter fitting. Based on the verified PBPK model, the pharmacokinetic behavior of PFAS in vivo was predicted.

[0006] In this scheme, the multi-compartment PBPK model includes a blood compartment, a liver compartment, a kidney compartment, a stomach compartment, an intestinal compartment, an adipose tissue compartment, a muscle tissue compartment, a spleen compartment, a lung compartment, a heart compartment, a brain compartment, and a remaining body tissue compartment, all of which are connected by blood flow.

[0007] In this protocol, parameters characterizing the binding process of PFAS to key proteins in vivo are explicitly introduced to quantify tissue allocation, including: The binding characteristics of PFAS to plasma proteins were determined by in vitro protein binding assays, and the plasma free fraction was calculated. ; Introducing plasma free fraction As a preconditioning factor, among all blood flow-based tissue drug transport rate terms, the concentration gradient will be driven by the total blood concentration. Corrected to blood free concentration ,in, The plasma free fraction, which characterizes the degree of binding of PFAS to plasma proteins, represents the drug concentration in the blood. Based on blood and multiple tissue samples collected during the elimination period in animal experiments, the steady-state concentration of PFAS in each tissue was measured, and the tissue-plasma partition coefficient of each tissue was calculated. The tissue-plasma partition coefficient of each tissue Introduced as an independent model parameter to describe the concentration balance between tissue and plasma; The plasma free fraction Used for in vitro-in vivo extrapolation calculations, converting in vitro metabolic data into in vivo metabolic clearance rates to determine the metabolic rate constant in the model; The plasma free fraction With tissue-plasma partition coefficient This constitutes a set of model parameters that explicitly characterize the protein binding process.

[0008] In this scheme, the hepatic kinetic equations simultaneously integrate two elimination mechanisms: metabolic clearance and bile excretion, including: By introducing independent metabolic clearance rate terms and bile excretion rate terms, the two elimination mechanisms of metabolic clearance and bile excretion are integrated simultaneously. The metabolic clearance mechanism is described by first-order kinetics and is used to quantitatively characterize the rate at which PFAS are metabolized and converted by the enzyme system in the liver, expressed in the equation as follows: ,in This is the metabolic rate constant. The amount of drug in the liver compartment; The bile excretion mechanism is described by first-order kinetics, which quantitatively characterizes the rate at which PFAS enter the intestine with bile secretion, and is expressed in the equation as follows: ,in This is the bile excretion rate constant; The amount of drug excreted by the bile excretion mechanism As input to the kinetic equations coupled to the intestinal chamber, the enterohepatic circulation or fecal excretion process is quantified.

[0009] In this scheme, the enterohepatic circulation contribution is explicitly incorporated into the intestinal compartment dynamics equations, including: The intestinal reabsorption of bile-excreted drugs is quantitatively described by explicitly introducing an enterohepatic circulation contribution term, which is expressed in the kinetic equation as follows: ,in This represents the bile excretion rate constant of the liver compartment. The amount of drug in the liver compartment. This indicates the amount of drug excreted into bile from the liver per unit of time. The enterohepatic circulation fraction, with a value between 0 and 1, represents the proportion of drugs excreted into the intestine via bile that are reabsorbed and returned to the systemic circulation; The enterohepatic circulation contribution item indicates the amount of [something] per unit time. The dose of PFAS re-enters the gastric and intestinal chambers via enterohepatic circulation, participating in subsequent intestinal absorption or fecal excretion.

[0010] In this scheme, the active tubular transport mechanism is specifically characterized by the following equation: The kidney chamber is divided into a renal tissue subcompartment, a proximal tubular cell subcompartment, and a renal tubular filtrate subcompartment; In the kinetic equations for the renal subcompartments, the Michaelis-Menten equations are adopted, and a rate term characterizing the active uptake from the renal tissue to the proximal tubular cell subcompartments is defined, expressed as follows: ;in, This represents the maximum transport rate of basolateral active transport proteins. For the corresponding Michaelis constant, This refers to the drug concentration in the kidney tissue. The dosage of the drug in the kidney tissue. This refers to the weight of the kidney tissue. In the kinetic equation for the proximal tubular cell subcompartment, the Michaelis-Menten equation is adopted, and a rate term characterizing the active reabsorption from the renal tubular filtrate subcompartment to the proximal tubular cell subcompartment is defined as follows: ;in, This represents the maximum transport rate of active transport proteins on the apical membrane side. For the corresponding Michaelis constant, This refers to the drug concentration in the renal tubular filtrate. The dosage of the drug in the renal tubular filtrate. This represents the weight of the renal tubular filtrate. The basal-side active uptake rate and the apical-side active reabsorption rate, through mass balance, serve as the input term for the proximal tubular cell subcompartment and the output term for the renal tubular filtrate subcompartment, respectively, forming a bidirectional active transport kinetic closed loop.

[0011] In this protocol, multi-tissue concentration data of the target PFAS at multiple time points were obtained through animal experiments, and in vitro metabolic data were obtained through in vitro liver microsomal experiments, including: The animal experiment involved using healthy female Sprague-Dawley rats as experimental animals, administering the target PFAS via repeated oral gavage, and setting at least two dose levels including a high-dose exposure group and a low-dose exposure group. During the exposure period of continuous administration and the subsequent elimination period after discontinuation of administration, blood, liver, kidney, spleen, fat, lung, muscle, heart and brain tissue samples were collected from each animal at multiple pre-set time points. All tissue samples were pretreated and analyzed by high performance liquid chromatography-tandem mass spectrometry to obtain the concentration data of each PFAS at each time point and in each tissue, and a time-tissue-concentration dataset was generated. The in vitro liver microsomal experiment involved establishing an in vitro incubation system using liver microsomals from female Sprague-Dawley rats, an NADPH regeneration system, and phosphate buffer, and adding the target PFAS as a substrate to initiate the reaction. Samples were taken at multiple preset time points and the reaction was quenched. The residual concentration of parent PFAS in the incubation solution at each time point was determined by high performance liquid chromatography-tandem mass spectrometry. Based on the curve of the remaining concentration decreasing over time, the in vitro first-order metabolic rate constant of the PFAS was obtained by fitting a single exponential degradation model.

[0012] In this scheme, based on the concentration data and the in vitro metabolic data, PFAS-specific parameters for the PBPK model are obtained, including: Based on the in vitro metabolic data obtained from the in vitro liver microsomal experiments, the initial value of the in vivo metabolic rate constant of the PFAS was determined by the in vitro-in vivo extrapolation method. The parameters other than the metabolic rate constant in the PFAS-specific parameters are defined as the set of parameters to be fitted. A genetic algorithm is used to perform global optimization on the set of parameters to be fitted. The objective function is constructed based on minimizing the overall error between the simulated PFAS concentration values ​​of each tissue at the sampling time point output by the PBPK model and the measured concentration values ​​obtained by the high-dose exposure group in the animal experiment. During the calculation process, the in vivo metabolic rate constant is fixed at the initial value to set the search range constraint, and the global approximate optimal parameter solution is obtained through the iteration of the genetic algorithm; Using the global approximate optimal parameter solution as the initial point, the L-BFGS-B algorithm is used to perform a local fine search, and the set of parameters to be optimized after local fine optimization is output. The determined in vivo metabolic rate constant is combined with the locally refined set of parameters to be optimized to form a PFAS-specific parameter set for the PBPK model.

[0013] In this scheme, the PFAS-specific parameter set is input into the PBPK model, the model is run to simulate the dynamic process of PFAS in vivo, and the pharmacokinetic characteristics are output, including: The PBPK model, which includes the chamber structure and dynamic equations, is integrated with the PFAS-specific parameter set. Initial simulation conditions are set, and exposure scenarios are defined according to the prediction purpose. The exposure scenarios include drug dosage, route of administration, frequency of administration, and total simulation duration. The integrated model is solved to calculate the dynamic changes in drug content in each tissue compartment and drug concentration in the blood from the initial time to the end of the simulation. Based on the dynamic changes, pharmacokinetic parameters for characterizing the in vivo behavior of PFAS are calculated and output, including: systemic clearance, apparent volume of distribution, and half-life. The dynamic change process is output in the form of time-concentration curves, and the pharmacokinetic characteristic parameters are output in the form of structured data as prediction results of the in vivo behavior of the PFAS under the set exposure scenario. The whole body clearance rate It is calculated based on the ratio of the average infusion rate to the steady-state blood drug concentration, and is expressed as: ; The apparent volume of distribution is calculated as the ratio of the total amount of drug in the body to the blood drug concentration at steady state, and is expressed as: ; The half-life The total drug exposure under steady-state conditions is estimated and expressed as follows: ; in Here, represents the steady-state blood drug concentration, and represents the administered dose. This refers to the dosing interval. For weight, This represents the sum of drug concentrations in each compartment of the tissue at steady state. This represents the total amount of drug in the body at steady state.

[0014] In this approach, the PBPK model is validated using independent exposure experimental data at doses different from those used for parameter fitting, and predictions are made based on the validated model, including: The experimental data of the low-dose exposure group in the animal experiment were obtained to form an independent observation dataset for model validation. The drug dose of the low-dose exposure group was 1 / 10 of the drug dose of the high-dose exposure group. Keeping the PBPK model structure and PFAS-specific parameter set unchanged, the exposure scenario is set to the same dosage, regimen and duration as the low-dose exposure group experiment, and the model is run to output the predicted PFAS concentration values ​​of each tissue under this condition. The predicted PFAS concentration values ​​of each tissue were quantitatively compared with the independent observed data; the accuracy of the model's prediction was evaluated by calculating the ratio of the predicted value to the observed value for each data point. If the model's prediction accuracy meets the preset validation criteria, the validated PBPK model will be used to predict the pharmacokinetic behavior of the PFAS under unvalidated new exposure scenarios. The new exposure scenarios include: environmentally relevant dose exposure, exposure at different dosing frequencies, and exposure to specific populations or species simulated by adjusting the model's physiological parameters.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention achieves mechanistic and quantitative simulation of PFAS behavior in vivo by explicitly introducing and integrating four core mechanisms into a multi-compartment PBPK model. The explicitly introduced protein binding parameters (plasma free fraction and tissue-plasma partition coefficient) replace empirical constants, enabling the explanation and prediction of the distribution differences of PFAS with different chain lengths and structures among key tissues such as the liver and blood from a physicochemical perspective, significantly improving the accuracy and stability of the model's prediction of the tissue accumulation tendency of different PFAS. Simultaneously, the integration of metabolic and bile excretion dual elimination pathways in the liver compartment, the integration of active transport and passive processes based on Michaelis-Menten kinetics in the kidney compartment, and the explicit closed-loop enterohepatic circulation in the intestinal compartment allow the model to more realistically characterize the complex clearance kinetics and circulation processes of PFAS in vivo, especially accurately simulating the long half-life characteristics of long-chain PFAS and their special accumulation behavior in excretory organs such as the kidneys, overcoming the predictive bias caused by process simplification in traditional models. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments or examples of the present invention, the drawings used in the embodiments or examples will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained according to these drawings without creative effort.

[0017] Figure 1 A flowchart is shown for a method to predict the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo. Figure 2 A schematic diagram of the PBPK model structure is shown; Figure 3 The figure shows a comparison between predicted and observed PFAS concentrations in blood, liver, and kidney in an independent validation of the PBPK model under low-dose exposure in female rats. Detailed Implementation

[0018] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0019] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0020] like Figure 1As shown, this embodiment provides a method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo, including: A multi-compartment PBPK model was established based on physiological structure. In the kinetic equations of the PBPK model: Explicitly introduce parameters characterizing the binding process of PFAS to key proteins in vivo to quantify tissue distribution; The hepatic kinetic equations simultaneously integrate both metabolic clearance and bile excretion as elimination mechanisms. In the dynamic equations of the intestinal compartments, the contribution of enterohepatic circulation is explicitly introduced; The active tubular transport mechanism based on Michaelis-Menten dynamics is integrated into the kinetic equations of the renal compartment. Multi-tissue concentration data of the target PFAS at multiple time points were obtained through animal experiments, and in vitro metabolic data were obtained through in vitro liver microsome experiments. A genetic optimization algorithm was used to fit and obtain the PFAS-specific parameter set for the PBPK model. The PFAS-specific parameter set is input into the PBPK model, the model is run to simulate the dynamic process of the PFAS in vivo, and the pharmacokinetic characteristics are output. The accuracy of the PBPK model prediction results was verified using independent exposure experimental data at doses different from those used for parameter fitting. Based on the verified PBPK model, the pharmacokinetic behavior of PFAS in vivo was predicted.

[0021] It should be noted that the multi-compartment PBPK model was established based on physiological parameters of female Sprague-Dawley rats. For example... Figure 2 As shown, the model includes a blood chamber, a liver chamber, a kidney chamber, a stomach chamber, an intestinal chamber, an adipose tissue chamber, a muscle tissue chamber, a spleen chamber, a lung chamber, a heart chamber, a brain chamber, and the remaining body tissue chambers. Each chamber is connected by blood flow, and the transport of compounds follows the principles of blood flow rate limiting and mass balance.

[0022] The rate of change in drug dosage in each room ( () equals the sum of all rates flowing into the chamber minus the sum of all rates flowing out of the chamber.

[0023] As the central hub for drug transport between tissues, the changes in drug levels in the blood chamber are influenced by the net transport between tissues and blood, as well as the drug reflux in the proximal tubules of the kidneys. The kinetic equation for the blood chamber is expressed as: , in Indicates blood flow to the tissues. Tissue-plasma partition coefficient, For the organization weight fraction, and For the organization Dosage of the medicine It refers to one of the following: kidney, spleen, fat, lungs, muscles, brain, heart, or other components. Hepatic blood flow fraction; The liver-plasma partition coefficient; Blood weight fraction This represents the liver weight fraction; The dosage of medicine for the liver. The amount of drug in the proximal tubule cells; The amount of drug in the blood; This is the external discharge transfer rate constant; Proximal tubule cell volume factor In traditional models, the net transport of compounds between tissues and blood is usually directly determined by the empirical allocation coefficient P. t Determined by the concentration difference, this protocol explicitly introduces parameters characterizing the binding process of PFAS to key proteins in vivo to quantify tissue distribution. The binding characteristics of PFAS to plasma proteins are determined through in vitro protein binding experiments, and the plasma free fraction is calculated. ; Introducing plasma free fraction As a preconditioning factor, among all blood flow-based tissue drug transport rate terms, the concentration gradient will be driven by the total blood concentration. Corrected to blood free concentration ,in, The plasma free fraction is used to characterize the degree of binding of PFAS to plasma proteins. The concentration of the drug in the blood was used; based on blood and multiple tissue samples collected during the elimination period in animal experiments, the steady-state concentration of PFAS in each tissue was measured, and the tissue-plasma partition coefficient of each tissue was calculated. The tissue-plasma partition coefficient of each tissue Introduced as an independent model parameter to describe the concentration balance between tissue and plasma; the plasma free fraction Used for in vitro-in vivo extrapolation calculations, it converts in vitro metabolic data into in vivo metabolic clearance rates to determine the metabolic rate constant in the model. In the tissue compartment dynamics equations, it represents the amount of PFAS in the tissue. Through its corresponding and tissue weight Converted to equivalent plasma free drug concentration The plasma free fraction With tissue-plasma partition coefficient A set of model parameters was constructed to explicitly characterize the protein binding process, quantifying the regulatory role of PFAS binding to plasma proteins and tissue components on their in vivo distribution and clearance behavior.

[0024] In the hepatic kinetic equation, two elimination mechanisms, metabolic clearance and bile excretion, are integrated. The change in drug dosage is determined by three net inflows and two net outflows. The blood perfusion infusion delivers PFAS (mainly in free state) to the liver through the hepatic artery and portal vein. The intestinal absorption directly delivers PFAS absorbed orally. When PFAS enters the systemic circulation through the intestinal wall, it first enters the liver entirely through the portal vein. The mass outflow includes metabolic clearance and bile excretion.

[0025] The hepatic ventricular kinetic equations were designed to simultaneously characterize metabolism and bile excretion: , in, This is the gastrointestinal transport rate constant. This is the intestinal transport rate constant. This refers to the amount of drug in the intestines.

[0026] Metabolic clearance refers to the process by which PFAS undergoes oxidation, reduction, and hydrolysis under the catalysis of liver cells, transforming it into metabolites. This metabolic clearance mechanism is described by first-order kinetics and used to quantitatively characterize the rate at which PFAS is metabolized and converted by the enzyme system in the liver. It is expressed in the equation: , where is the metabolic rate constant. This represents the amount of drug in the liver compartment. This linear term assumes that the metabolic enzymes are not saturated, and the metabolic rate is proportional to the amount of drug in the liver. It is applicable to most low-dose exposure scenarios in the environment. The amount of drug metabolized is ultimately included in the metabolites excreted in feces.

[0027] Bile excretion refers to the process by which PFAS or their conjugates (such as glucuronide conjugates) are actively or passively transported from hepatocytes into bile and subsequently released into the intestine. This bile excretion mechanism is described by first-order kinetics, which quantitatively characterizes the rate at which PFAS enter the intestine with bile secretion, and is expressed in the equation: ,in The bile excretion rate constant; the amount of drug excreted by the bile excretion mechanism. As input to the kinetic equations coupled to the intestinal compartments, the enterohepatic circulation or fecal excretion process is quantified. A portion of the drug entering the intestine is reabsorbed and re-enters the liver via the portal vein; the remaining portion is not reabsorbed and is ultimately excreted in the feces, which is included in the cumulative fecal excretion. : , In the kinetic equations of the intestinal compartments, the contribution of enterohepatic circulation is explicitly introduced. The basic pathway is: liver → bile → duodenum → (possibly intestinal reabsorption → portal vein → liver). The kinetic equation is expressed as: , in, This refers to the amount of medication in the stomach. The amount of drug in the intestines. This is the gastrointestinal transport rate constant. This is the intestinal transport rate constant. This refers to the intestinal absorption rate of the drug. This represents the reabsorption rate of drugs excreted by the liver.

[0028] The material source for enterohepatic circulation is bile excretion from the liver, and the rate of bile excretion is quantified as follows: This represents the amount of drug excreted from the liver into the bile duct per unit time, which is a prerequisite and input quantity for enterohepatic circulation. The intestinal reabsorption of bile-excreted drugs is quantitatively described by explicitly introducing an enterohepatic circulation contribution term, which is expressed in the kinetic equation as follows: ,in This represents the bile excretion rate constant of the liver compartment. The amount of drug in the liver compartment. This indicates the amount of drug excreted into bile from the liver per unit of time. The enterohepatic circulation fraction, with a value between 0 and 1, represents the proportion of drugs excreted into the intestine via bile that are reabsorbed and returned to the systemic circulation.

[0029] The enterohepatic circulation contribution item indicates the amount of [something] per unit time. The dosage of PFAS, and the original drug expelled from the stomach. It mixes in the intestinal compartment and participates in subsequent intestinal absorption or fecal excretion. The amount of drug in the mixed intestinal fluid... will at a rate The absorbed portion enters the portal vein bloodstream and returns to the liver; the unabsorbed portion will be absorbed at a rate... It enters the feces and is eventually excreted from the body. The enterohepatic circulation contribution is related to the liver chambers. The direct coupling of these components forms a closed loop of liver excretion and intestinal reabsorption, which is key to accurately predicting the long half-life of long-chain PFAS.

[0030] The renal compartment is divided into renal tissue subcompartments, proximal tubular cell subcompartments, and renal tubular filtrate subcompartments. The renal tissue subcompartments represent the parenchymal tissue portion of the kidney and are the initial distribution space for PFAS after reaching the kidney via the bloodstream, corresponding to the amount of drug in the renal tissue. The proximal tubular cell subcompartment represents the renal tubular epithelial cells that perform active transport functions, corresponding to the intracellular drug concentration. The renal tubular filtrate subcompartment represents the filtrate within the renal tubule lumen and the final urine formed, corresponding to the amount of drug in the filtrate. And the amount of medication accumulated in the urine.

[0031] The kinetic equations for each subcompartment were established, and the Michaelis transport terms were explicitly integrated. The kinetic equations for the renal subcompartments were expressed as follows: , in This represents the fractional blood flow to the kidney tissue.

[0032] In the kinetic equations for the renal subcompartments, the Michaelis-Menten equations are used, with a rate term representing the active uptake from the renal tissue to the proximal tubular cell subcompartment, denoted as basal-side active uptake, expressed as: ;in, This represents the maximum transport rate of basolateral active transport proteins. For the corresponding Michaelis constant, This refers to the drug concentration in the kidney tissue. The dosage of the drug in the kidney tissue. This represents the weight of kidney tissue; this indicates that the rate of active uptake varies with kidney tissue concentration. Increases with increasing, but exhibits saturation characteristics, when At that time, the rate approaches its maximum value. .

[0033] The kinetic equation for the subcompartments of proximal tubule cells is expressed as: , in The external transport rate constant is denoted as . This refers to the intracellular drug concentration in the proximal tubules. The proximal tubule is a cell volume factor. This represents the drug concentration within the proximal tubule cells; In the kinetic equation of the proximal tubular cell subcompartment, the Michaelis-Menten equation is adopted, and a rate term characterizing the active reabsorption from the renal tubular filtrate subcompartment to the proximal tubular cell subcompartment is set, denoted as apical active reabsorption, to describe the process of PFAS being reabsorbed from the renal tubular filtrate back into the cell, expressed as: ;in, This represents the maximum transport rate of active transport proteins on the apical membrane side. For the corresponding Michaelis constant, This refers to the drug concentration in the renal tubular filtrate. The dosage of the drug in the renal tubular filtrate. This represents the weight of the renal tubular filtrate. The kinetic equation for the renal tubular filtrate subcompartment is expressed as: , Cumulative drug amount in urine Represented as: , in, GFR is the diffusion rate constant between kidney tissue and proximal tubular cells, BW is the animal's body weight. The proportion of filtered drugs; This is the drug urinary excretion rate constant.

[0034] The basal-side active uptake rate and the apical-side active reabsorption rate, through mass balance, serve as inputs to the proximal tubular cell subcompartment and outputs to the renal tubular filtrate subcompartment, respectively, forming a bidirectional active transport kinetic closed loop. The Michaelis-Menten equation introduces a transport saturation effect, accurately reflecting the biological characteristics resulting from the limited number of transport proteins. This allows the model to predict nonlinear behavior that may occur in renal clearance at high exposure doses. In all the above equations, the glomerular filtration rate term is multiplied by the plasma free fraction, reflecting the classic pharmacokinetic principle that only free PFAS can be filtered by the glomerulus. This unifies the protein binding (blood side) and active transport (renal cell side) mechanisms, jointly determining the renal disposition of PFAS.

[0035] Taking the spleen as an example, other tissues (fat, muscle, lungs, heart, brain, and other tissues) are similar in form, controlled only by the perfusion process limited by blood flow rate: , in This is the dosage of Chinese medicine for the spleen. Splenic blood flow fraction The spleen-blood partition coefficient. This refers to the quality of the spleen.

[0036] It should be noted that multi-tissue concentration data of the target PFAS at multiple time points were obtained through animal experiments, and in vitro metabolic data were obtained through in vitro liver microsome experiments. The animal experiments involved using healthy female Sprague-Dawley rats as experimental animals, administering the target PFAS via repeated oral gavage, with at least two dose levels including a high-dose exposure group and a low-dose exposure group. The preferred high-dose exposure group consisted of short-chain PFAS (PFHxA, PFHpA) at a dose of 2 mg / kg / day, and medium- and long-chain PFAS (PFOA, PFNA, etc.) and emerging alternatives (GenX, F-53B) at a dose of 1 mg / kg / day. The low-dose exposure group received one-tenth the dose of the high-dose group, i.e., 0.2 mg / kg / day of short-chain PFAS and 0.1 mg / kg / day of the other PFAS. The solvent control group received an equal volume of 5% (v / v) dimethyl sulfoxide aqueous solution for background monitoring.

[0037] During the continuous dosing exposure period and the subsequent elimination period, multiple tissue samples were collected from each animal at several pre-set time points (9 time points in total: 0 (baseline), 24, 96, 168, 240, 336, 408, 534, and 702 hours). These samples included: brain, heart, liver, bilateral kidneys, spleen, lungs, stomach, small intestine, thyroid gland, uterus, ovary, mammary gland, inguinal white adipose tissue, quadriceps femoris muscle, and femur. Immediately after collection, the tissues were rinsed with pre-cooled saline, blotted dry, weighed, and aliquoted. Urine and fecal samples were collected simultaneously over 24 hours during the experiment. All biological samples (serum, tissue, urine, and feces) were immediately flash-frozen in liquid nitrogen after collection and then transferred to a -80°C cryogenic freezer for later analysis. Sample pretreatment and high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS / MS) analysis were performed to obtain the concentration data of each PFAS in each time point and tissue, generating a time-tissue-concentration dataset. Based on the time-tissue-concentration dataset obtained from the aforementioned animal experiments, for each PFAS, concentration data of PFAS in each tissue and plasma were extracted during the elimination period (after drug administration cessation). Since the concentrations of PFAS in each tissue and plasma decrease synchronously during the elimination period, and the concentration ratio between tissue and plasma tends to be constant, this constant ratio is the tissue-plasma partition coefficient under steady-state conditions. The tissue-plasma partition coefficient, as an important component of the PFAS-specific parameter set, together with the plasma free fraction, constitutes a set of model parameters explicitly characterizing the protein binding process, used for subsequent PBPK model construction and prediction.

[0038] The in vitro liver microsomal assay involved the following steps: An in vitro incubation system was established using liver microsomes from female Sprague-Dawley rats, an NADPH regeneration system, and phosphate buffer. Female SD rat liver microsomal suspension at a concentration of 2 mg protein / mL was used to prepare a mixed solution containing all target PFAS, with low-concentration groups (200 μM each) and high-concentration groups (2000 μM each), dissolved in DMSO. In a 2 mL polypropylene centrifuge tube, the following were added sequentially: 1 μL PFAS substrate solution, 100 μL liver microsomal suspension, 39 μL PBS buffer, 50 μL NADPH regeneration system solution A, and 10 μL solution B. The mixture was immediately vortexed, the tube was sealed, and incubated at 37°C in a shaker (120 rpm).

[0039] Samples were taken and the reaction quenched at multiple preset time points (minutes 0, 5, 15, 30, 60, 120, 240, and 360). The residual concentration of parent PFAS in the incubation solution at each time point was determined by high-performance liquid chromatography-tandem mass spectrometry. Based on the curve of the residual concentration decreasing over time, a single-exponential degradation model was used to fit and obtain the in vitro first-order metabolic rate constant of the PFAS. The first-order metabolic rate constant... The parent PFAS concentration measured at each time point was fitted. Compared with the initial concentration natural logarithm ratio Plotting the graph against time t, we fit the slope using linear regression. The absolute value of the slope is the in vitro first-order metabolic rate constant. The single-exponential degradation model is expressed as follows: .

[0040] In vitro inherent clearance rate according to The calculation shows that, among which The concentration of liver microsomal protein in the incubation system was 1 mg / mL in this experiment; To convert in vitro data into in vivo metabolic parameters usable in the PBPK model, an in vitro-in vivo extrapolation method was used in the in vitro protein binding experiment. This method, combined with liver physiological parameters (liver blood flow, liver microsomal protein content, and blood free protein fraction), was used to calculate the predicted in vivo intrinsic liver clearance rate. This provides a crucial estimation basis for the metabolic rate constant in the model, expressed as: , in For liver blood flow, this experiment used 16.1 mg / h / g liver. The liver microsomal protein content (unit: mg MSP / g liver) refers to the amount of liver microsomal protein contained in each gram of liver tissue. This needs to be determined experimentally or by referring to species-specific data. In this experiment, 45 mg MSP / g liver was selected. This represents the unbound fraction of the drug in liver microsomes. Since the protein content during in vitro incubation (1 mg / mL) is lower than the protein content in blood (45 mg albumin / mL), it can be assumed to be 1 (ignoring the effect of microsome binding). The fraction of plasma free plasma. , For the associative constant, For maximum load capacity, the total protein concentration used in this compartment was 486 µmol / L.

[0041] Based on the in vitro metabolic data obtained from the in vitro liver microsome experiment, the initial value of the in vivo metabolic rate constant of the PFAS is determined by the in vitro-in vivo extrapolation method. The parameters in the PFAS-specific parameters other than the metabolic rate constant and plasma free fraction are defined as the set of parameters to be fitted, such as: tissue-plasma partition coefficients, plasma free fraction, absorption coefficients, excretion correlation coefficients, kinetic constants, fractional parameters, etc.

[0042] A genetic algorithm is used to globally optimize the parameter set to be fitted. The objective function is constructed based on minimizing the overall error between the simulated PFAS concentrations of each tissue at the sampling time points output by the PBPK model and the measured concentrations obtained from the high-dose exposure group in the animal experiment. A commonly used error metric is the weighted sum of squared residuals. During the calculation, the in vivo metabolic rate constant is fixed at the initial value to set the search range constraint. The process involves initializing the population, evaluating individual fitness, performing selection, crossover, and mutation operations, and iteratively updating the population until a preset iteration termination condition is met, thereby obtaining a globally approximate optimal parameter solution.

[0043] Using the global approximate optimal parameter solution as the initial point, the L-BFGS-B algorithm is used to perform a local fine-tuning search under the same parameter boundary constraints. The gradient information of the objective function is used for iteration until the convergence tolerance condition is met, and the set of parameters to be optimized after local fine-tuning is output. The determined in vivo metabolic rate constant is merged with the set of parameters to be optimized after local fine-tuning to form the PFAS-specific parameter set for the PBPK model.

[0044] It should be noted that the PBPK model, which includes the chamber structure and dynamics equations, is integrated with the PFAS-specific parameter set, and initial simulation conditions are set. Exposure scenarios are defined according to the prediction objective, including drug dosage, route of administration, frequency of administration, and total simulation duration. The PFAS-specific parameters in the PFAS-specific parameter set are shown in Table 1 below.

[0045] At the simulation start time (usually t=0), the drug concentration in all compartments is zero. A robust numerical integration algorithm applicable to rigid or non-rigid ordinary differential equation systems is used to solve the integrated model, calculating the dynamic changes in drug concentration in each tissue compartment and in the blood from the initial time to the end of the simulation. Based on these dynamic changes, pharmacokinetic parameters characterizing the in vivo behavior of PFAS are calculated and output. These pharmacokinetic parameters include: systemic clearance, apparent volume of distribution, and half-life. The dynamic changes are output as time-concentration curves, preferably extracting blood concentration-time curves, concentration-time curves of major target tissues (e.g., liver, kidney, fat), and cumulative excretion (urine, feces)-time curves. The pharmacokinetic parameters are output as structured data as a prediction of the in vivo behavior of the PFAS under the specified exposure scenario.

[0046] The whole body clearance rate It is calculated based on the ratio of the average infusion rate to the steady-state blood drug concentration, and is expressed as: ; The apparent distribution volume It is calculated by the ratio of the total amount of drug in the body to the blood drug concentration at steady state, and is expressed as: ; The half-life The total drug exposure under steady-state conditions is estimated and expressed as follows: ; in To achieve steady-state blood drug concentration, This is the dosage. The dosing interval is [value], and body weight is [value]. This represents the sum of drug concentrations in each compartment of the tissue at steady state. This represents the total amount of drug in the body at steady state.

[0047] It should be noted that experimental data from the low-dose exposure group in the animal experiments, such as 0.2 mg / kg / day for short-chain PFAS and 0.1 mg / kg / day for others, were obtained to form an independent observation dataset for model validation. The dosage of the low-dose exposure group was 1 / 10 of that of the high-dose exposure group. Keeping the PBPK model structure and PFAS-specific parameter set unchanged, the exposure scenario was set to the same dosage, regimen, and duration as the low-dose exposure group experiment. The model was run to output the predicted PFAS concentration values ​​for each tissue under these conditions. The predicted PFAS concentration values ​​for each tissue were quantitatively compared using independent observation data. By calculating the ratio of the predicted value to the observed value for each data point, the percentage of data points whose ratio fell within the 2x error interval (0.5 to 2) and the 5x error interval (0.2 to 5) was statistically analyzed to evaluate the model's predictive accuracy. Figure 3 As shown, for most PFAS and tissues, more than 80% of the predicted values ​​fall within the 5-fold error range, with particularly accurate predictions for key tissues such as blood and liver, fully demonstrating the model's reliable cross-dose prediction capability.

[0048] If the model's prediction accuracy meets the preset validation criteria, the validated PBPK model will be used to predict the pharmacokinetic behavior of the PFAS under unvalidated new exposure scenarios. The new exposure scenarios include: environmentally relevant dose exposure, exposure at different dosing frequencies, and exposure to special populations or species simulated by adjusting the model's physiological parameters. After validation, the model can safely, economically, and quickly simulate various real or hypothetical exposure scenarios on a computer, providing a powerful decision support tool for the safety assessment and risk management of chemicals. Preferably, the validated model can be used for a variety of prediction scenarios: (1) Screening for new PFAS alternatives: Basic parameters of new compounds (such as protein binding and in vitro metabolism) are obtained through in vitro experiments, and their in vivo half-life and accumulation potential can be preliminarily predicted by substituting them into the model. (2) Risk assessment for special populations: By adjusting the physiological parameters in the model (such as glomerular filtration rate GFR and liver volume), the exposure risk of people with renal insufficiency or different age groups can be simulated. (3) Environmental dose exposure assessment: Environmentally relevant exposure doses (usually very low) are input into the model to predict the steady-state in vivo load under long-term low-dose exposure, providing a basis for setting environmental benchmarks.

[0049] A second embodiment of the present invention provides a computer-readable storage medium comprising a program for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo. When the program for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo is executed by a processor, it implements the steps of a method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo.

[0050] In the several embodiments provided in this application, it should be understood that the disclosed methods and systems can be implemented in other ways. The system embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, and can be electrical, mechanical, or other forms. Furthermore, in the various embodiments of the present invention, all functional units can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0051] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0052] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo, characterized in that, Includes the following steps: A multi-compartment PBPK model was established based on physiological structure. In the kinetic equation of the PBPK model, parameters characterizing the binding process of PFAS with key proteins in vivo were explicitly introduced to quantify tissue distribution. In the kinetic equation of the liver compartment, two elimination mechanisms, metabolic clearance and bile excretion, were integrated. In the kinetic equation of the intestinal compartment, the contribution of enterohepatic circulation was explicitly introduced. In the kinetic equation of the kidney compartment, the active tubular transport mechanism based on Michaelis-Menten kinetics was integrated. Multi-tissue concentration data of the target PFAS at multiple time points were obtained through animal experiments, and in vitro metabolic data were obtained through in vitro liver microsomal experiments. A genetic optimization algorithm was used to fit and obtain the PFAS-specific parameter set for the PBPK model. The PFAS-specific parameter set is input into the PBPK model, the model is run to simulate the dynamic process of the PFAS in vivo, and the pharmacokinetic characteristics are output. The accuracy of the PBPK model prediction results was verified using independent exposure experimental data at doses different from those used for parameter fitting. Based on the verified PBPK model, the pharmacokinetic behavior of PFAS in vivo was predicted.

2. The method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo according to claim 1, characterized in that, The multi-compartment PBPK model includes a blood compartment, a liver compartment, a kidney compartment, a stomach compartment, an intestinal compartment, an adipose tissue compartment, a muscle tissue compartment, a spleen compartment, a lung compartment, a heart compartment, a brain compartment, and remaining body tissue compartments, all of which are connected by blood flow.

3. The method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo according to claim 1, characterized in that, Explicitly introduced parameters characterizing the binding process of PFAS to key proteins in vivo quantify tissue distribution, including: The binding characteristics of PFAS to plasma proteins were determined by in vitro protein binding assays, and the plasma free fraction was calculated. ; Introducing plasma free fraction As a preconditioning factor, among all blood flow-based tissue drug transport rate terms, the concentration gradient will be driven by the total blood concentration. Corrected to blood free concentration ,in, The plasma free fraction is used to characterize the degree of binding of PFAS to plasma proteins. This refers to the concentration of the drug in the blood. Based on blood and multiple tissue samples collected during the elimination period in animal experiments, the steady-state concentration of PFAS in each tissue was measured, and the tissue-plasma partition coefficient of each tissue was calculated. The tissue-plasma partition coefficient of each tissue Introduced as an independent model parameter to describe the concentration balance between tissue and plasma; The plasma free fraction Used for in vitro-in vivo extrapolation calculations, converting in vitro metabolic data into in vivo metabolic clearance rates to determine the metabolic rate constant in the model; The plasma free fraction With tissue-plasma partition coefficient This constitutes a set of model parameters that explicitly characterize the protein binding process.

4. The method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo according to claim 1, characterized in that, The hepatic kinetic equations simultaneously integrate two elimination mechanisms: metabolic clearance and bile excretion, including: By introducing independent metabolic clearance rate terms and bile excretion rate terms, the two elimination mechanisms of metabolic clearance and bile excretion are integrated simultaneously. The metabolic clearance mechanism is described by first-order kinetics and is used to quantitatively characterize the rate at which PFAS are metabolized and converted by the enzyme system in the liver, expressed in the equation as follows: ,in This is the metabolic rate constant. The amount of drug in the liver compartment; The bile excretion mechanism is described by first-order kinetics, which quantitatively characterizes the rate at which PFAS enter the intestine with bile secretion, and is expressed in the equation as follows: ,in This is the bile excretion rate constant; The amount of drug excreted by the bile excretion mechanism As input to the kinetic equations coupled to the intestinal chamber, the enterohepatic circulation or fecal excretion process is quantified.

5. The method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo according to claim 1 or 4, characterized in that, In the kinetic equations of the intestinal compartments, the contribution of the enterohepatic circulation is explicitly introduced, including: The intestinal reabsorption of bile-excreted drugs is quantitatively described by explicitly introducing an enterohepatic circulation contribution term, which is expressed in the kinetic equation as follows: ,in This represents the bile excretion rate constant of the liver compartment. The amount of drug in the liver compartment. This indicates the amount of drug excreted into bile from the liver per unit of time. The enterohepatic circulation fraction, with values ​​between 0 and 1, represents the proportion of drugs excreted into the intestine via bile that are reabsorbed and returned to the systemic circulation; the enterohepatic circulation contribution indicates the percentage of drugs excreted into the intestine via bile that are reabsorbed and returned to the systemic circulation per unit time. The dose of PFAS re-enters the gastric and intestinal chambers via enterohepatic circulation, participating in subsequent intestinal absorption or fecal excretion.

6. The method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo according to claim 1, characterized in that, The active transport mechanism of renal tubules is specifically characterized by the following equation: The kidney chamber is divided into a renal tissue subcompartment, a proximal tubular cell subcompartment, and a renal tubular filtrate subcompartment; In the kinetic equations for the renal subcompartments, the Michaelis-Menten equations are adopted, and a rate term characterizing the active uptake from the renal tissue to the proximal tubular cell subcompartments is defined, expressed as follows: ;in, This represents the maximum transport rate of basolateral active transport proteins. For the corresponding Michaelis constant, This refers to the drug concentration in the kidney tissue. The dosage of the drug in the kidney tissue. This refers to the weight of the kidney tissue. In the kinetic equation for the proximal tubular cell subcompartment, the Michaelis-Menten equation is adopted, and a rate term characterizing the active reabsorption from the renal tubular filtrate subcompartment to the proximal tubular cell subcompartment is defined as follows: ;in, This represents the maximum transport rate of active transport proteins on the apical membrane side. For the corresponding Michaelis constant, This refers to the drug concentration in the renal tubular filtrate. The dosage of the drug in the renal tubular filtrate. This represents the weight of the renal tubular filtrate. The basal-side active uptake rate and the apical-side active reabsorption rate, through mass balance, serve as the input term for the proximal tubular cell subcompartment and the output term for the renal tubular filtrate subcompartment, respectively, forming a closed loop of bidirectional active transport kinetics.

7. The method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo according to claim 1, characterized in that, Multi-tissue concentration data of the target PFAS at multiple time points were obtained through animal experiments, and in vitro metabolic data were obtained through in vitro liver microsomal experiments, including: The animal experiment involved using healthy female Sprague-Dawley rats as experimental animals, administering the target PFAS via repeated oral gavage, and setting at least two dose levels including a high-dose exposure group and a low-dose exposure group. During the exposure period of continuous administration and the subsequent elimination period after discontinuation of administration, blood, liver, kidney, spleen, fat, lung, muscle, heart and brain tissue samples were collected from each animal at multiple pre-set time points. All tissue samples were pretreated and analyzed by high performance liquid chromatography-tandem mass spectrometry to obtain the concentration data of each PFAS at each time point and in each tissue, and a time-tissue-concentration dataset was generated. The in vitro liver microsomal experiment involved establishing an in vitro incubation system using liver microsomals from female Sprague-Dawley rats, an NADPH regeneration system, and phosphate buffer, and adding the target PFAS as a substrate to initiate the reaction. Samples were taken at multiple preset time points and the reaction was quenched. The residual concentration of parent PFAS in the incubation solution at each time point was determined by high performance liquid chromatography-tandem mass spectrometry. Based on the curve of the remaining concentration decreasing over time, the in vitro first-order metabolic rate constant of the PFAS was obtained by fitting a single exponential degradation model.

8. The method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo according to claim 1 or 7, characterized in that, Based on the concentration data and the in vitro metabolic data, PFAS-specific parameters for the PBPK model are obtained, including: Based on the in vitro metabolic data obtained from the in vitro liver microsomal experiments, the initial value of the in vivo metabolic rate constant of the PFAS was determined by the in vitro-in vivo extrapolation method. The parameters other than the metabolic rate constant in the PFAS-specific parameters are defined as the set of parameters to be fitted. A genetic algorithm is used to perform global optimization on the set of parameters to be fitted. The objective function is constructed based on minimizing the overall error between the simulated PFAS concentration values ​​of each tissue at the sampling time point output by the PBPK model and the measured concentration values ​​obtained by the high-dose exposure group in the animal experiment. During the calculation process, the in vivo metabolic rate constant is fixed at the initial value to set the search range constraint, and the global approximate optimal parameter solution is obtained through the iteration of the genetic algorithm; Using the global approximate optimal parameter solution as the initial point, the L-BFGS-B algorithm is used to perform a local fine search, and the set of parameters to be optimized after local fine optimization is output. The determined in vivo metabolic rate constant is combined with the locally refined set of parameters to be optimized to form a PFAS-specific parameter set for the PBPK model.

9. The method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo according to claim 1, characterized in that, The PFAS-specific parameter set is input into the PBPK model, the model is run to simulate the dynamic process of PFAS in vivo, and the pharmacokinetic characteristics are output, including: The PBPK model, which includes the chamber structure and dynamic equations, is integrated with the PFAS-specific parameter set. Initial simulation conditions are set, and exposure scenarios are defined according to the prediction purpose. The exposure scenarios include drug dosage, route of administration, frequency of administration, and total simulation duration. The integrated model is solved to calculate the dynamic changes in drug content in each tissue compartment and drug concentration in the blood from the initial time to the end of the simulation. Based on the dynamic changes, pharmacokinetic parameters for characterizing the in vivo behavior of PFAS are calculated and output, including: systemic clearance, apparent volume of distribution, and half-life. The dynamic change process is output in the form of time-concentration curves, and the pharmacokinetic characteristic parameters are output in the form of structured data as prediction results of the in vivo behavior of the PFAS under the set exposure scenario. The whole body clearance rate It is calculated based on the ratio of the average infusion rate to the steady-state blood drug concentration, and is expressed as: ; The apparent distribution volume It is calculated by the ratio of the total amount of drug in the body to the blood drug concentration at steady state, and is expressed as: ; The half-life The total drug exposure under steady-state conditions is estimated and expressed as follows: ; in To achieve steady-state blood drug concentration, This is the dosage. This refers to the dosing interval. For weight, This represents the sum of drug concentrations in each compartment of the tissue at steady state. This represents the total amount of drug in the body at steady state.

10. The method for predicting the pharmacokinetic behavior of perfluorinated and polyfluoroalkyl substances in vivo according to claim 1, characterized in that, The PBPK model was validated using independent exposure experimental data at doses different from those used for parameter fitting, and predictions were made based on the validated model, including: The experimental data of the low-dose exposure group in the animal experiment were obtained to form an independent observation dataset for model validation. The drug dose of the low-dose exposure group was 1 / 10 of the drug dose of the high-dose exposure group. Keeping the PBPK model structure and PFAS-specific parameter set unchanged, the exposure scenario is set to the same dosage, regimen and duration as the low-dose exposure group experiment, and the model is run to output the predicted PFAS concentration values ​​of each tissue. The predicted PFAS concentration values ​​of each tissue were quantitatively compared with the independent observed data; the accuracy of the model's prediction was evaluated by calculating the ratio of the predicted value to the observed value for each data point. If the model's prediction accuracy meets the preset validation criteria, the validated PBPK model will be used to predict the pharmacokinetic behavior of the PFAS under unvalidated new exposure scenarios. The new exposure scenarios include: environmentally relevant dose exposure, exposure at different dosing frequencies, and exposure to specific populations or species simulated by adjusting the model's physiological parameters.

Images