Systems, methods, and devices for predicting pharmacokinetic effects on ketogenic effects after administration of glycerol tricaprylate

Abstract

Systems, methods, and devices predict pharmacokinetic parameters after administration of a formulation, such as a medium chain triglyceride (MCT). The method includes generating training data indicative of a human metabolic outcome of the formulation. The training data includes a gastric emptying parameter, a digestion rate, an absorption constant, or a conversion rate, wherein the training target variable is a ketone concentration or a plasma free fatty acid concentration. The method further includes training a physiology-based biopharmaceutical (PBB) model with the training data, and predicting at least one of a target plasma free fatty acid concentration or a target ketone concentration using the patient data using the PBB model. Further, as part of a parametric sensitivity analysis of the PBB model, the method varies input parameters of the training data according to a simulated average plasma free fatty acid curve or a simulated average ketone plasma curve.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. Provisional Patent Application No. 63 / 323,305, filed on March 24, 2022, entitled “METHODS FOR PREDICTING PHARMACOKINETIC INFLUENCES ON KETOGENESIS FOLLOWING ADMINISTRATION OF TRICAPRILIN TO HUMANS,” and U.S. Provisional Patent Application No. 63 / 423,349, filed on November 7, 2022, entitled “METHODS FOR PREDICTING PHARMACOKINETIC INFLUENCES ON KETOGENESIS FOLLOWING ADMINISTRATION OF TRICAPRILIN TO HUMANS,” the entire contents of each of which are incorporated herein by reference.

[0003] field

[0004] Aspects of the presently disclosed technology relate generally to systems and methods for predicting pharmacokinetics, and more particularly, to predicting ketosis parameters in Alzheimer's disease patients.

[0005] background

[0006] Alzheimer's disease (AD) is a progressive neurodegenerative disorder and the most common cause of dementia, accounting for approximately 60-80% of all cases. In the United States alone, an estimated 6.2 million patients aged 65 and over are ill. Currently, there are multiple theories about the cause and potential mechanisms of AD. They include the formation of beta-amyloid plaques, neurofibrillary tangles in neurons, and chronic inflammation of brain tissue. Many pathophysiological pathways that contribute to the disease remain to be explored.

[0007] Studies have demonstrated reduced expression of glucose transporters in the central nervous system (CNS) of patients with Alzheimer's disease. In addition, enhancing their glucose metabolism can contribute to an extended lifespan. In this context, ketones acetoacetate and f3-hydroxybutyrate serve as the main energy sources and provide up to 80% of brain energy when glucose is not available. This has led to the development of ketogenic diets to improve energy supply and achieve functional improvements in patients with AD.

[0008] It is with these observations in particular in mind that various aspects of the present disclosure have been conceived and developed.

[0009] Overview

[0010] The embodiments described and claimed herein can solve the aforementioned problems by providing a method for predicting pharmacokinetic parameters after administration of a formulation. The method may include: generating training data indicating the human metabolic results of the formulation; training a physiologically based biopharmaceutics (PBB) model with the training data through one or more computer devices executing computer-readable instructions; and / or as part of a parameter sensitivity analysis of the PBB model, changing the input parameters of the training data according to a simulated mean plasma free fatty acid curve (e.g., octanoic acid) or a simulated mean ketone plasma curve. In addition, the method may include: receiving patient data corresponding to a patient who orally receives the formulation through the one or more computer devices; and using the patient data to predict at least one of a target plasma free fatty acid concentration or a target ketone concentration using the PBB model.

[0011] In some instances, the formulation includes medium chain triglycerides (MCT). The MCT may be tricaprylin. The training data may also include a data set based on a gastric emptying model. For example, the gastric emptying model may include a sinusoidal algorithm corresponding to a rhythmic emptying pattern. In addition, the gastric emptying model may include a first approach for solid food and a second approach for liquid food; and / or a gastric amplitude parameter, a complete emptying parameter, or a gastric cycle parameter.

[0012] In some examples, predicting at least one of a target plasma free fatty acid concentration or a target ketone concentration comprises determining a digestion rate constant obtained by fitting a release curve of the formulation in the presence of a lipase while assuming a first order kinetic process. The training data may include in vitro parameters, which are digestion rate and / or bioavailability digestion fraction. The training data may also include in vivo parameters, which are one or more of metabolic fraction, absorption constant, conversion rate, elimination rate, volume distribution of free fatty acids, or volume distribution of ketone bodies. In addition, the patient data may include one or more of patient physiological parameters or formulation dosage information.

[0013] In some examples, the method may further include modeling a digestion rate using a release profile of the formulation, and may further include obtaining a digestion rate constant based on modeling the digestion rate, wherein predicting at least one of a target plasma free fatty acid concentration or a target ketone concentration may include using the digestion rate with the PBB model. Predicting at least one of a target plasma free fatty acid concentration or a target ketone concentration may include predicting the bioavailability of the free fatty acids of the patient based on the digestion rate with the PBB model. In addition, training the PBB model may include instructing the PBB model to treat the enzymatic conversion of free fatty acids to plasma ketones as a non-saturated primary process.

[0014] In some examples, a method for predicting a pharmacokinetic parameter after administration of a formulation includes: generating training data indicating one or more gastric emptying parameters, digestion rate, absorption constant, or conversion rate, wherein the training target variable is ketone concentration; training a PBB model with the training data by one or more computer devices executing computer-readable instructions; receiving patient data corresponding to a patient receiving the formulation by the one or more computer devices; and / or predicting a target ketone concentration using the PBB model using the patient data.

[0015] In some examples, predicting the target ketone concentration includes predicting the patient's plasma free fatty acid concentration using the PBB model and the patient data. The method may also include validating the PBB model using clinical trial data that quantifies free fatty acids and ketone bodies in plasma from volunteers.

[0016] In some examples, a method for predicting a pharmacokinetic parameter after administration of a formulation includes: generating training data indicating a gastric emptying parameter, a digestion rate, an absorption constant, or a conversion rate, wherein the training target variable is a plasma free fatty acid concentration; training a PBB model with the training data by one or more computer devices executing computer-readable instructions; receiving patient data corresponding to a patient receiving the formulation by the one or more computer devices; and / or predicting a target plasma free fatty acid concentration using the PBB model using the patient data. In some examples, treatment measures may be taken in response to predicting a target plasma free fatty acid concentration and / or a target ketone concentration, such as adjusting the dosage, concentration, or substance of the formulation.

[0017] In some examples, the method further comprises predicting ketone concentration using the PBB model using the patient data, the plasma free fatty acid concentration, and the conversion rate estimate. Training the PBB model may include providing a gastric absorption emptying model with a sinusoidal algorithm, the model having an amplitude variable, a period variable, and a complete emptying time variable. In addition, the training data may include: a first cluster including physiologically based parameters, a second cluster including population parameters, and a third cluster including in vitro parameters.

[0018] Other embodiments are also described and described herein. In addition, although multiple embodiments are disclosed, other embodiments of the currently disclosed technology will become apparent to those skilled in the art based on the following detailed description, which shows and describes illustrative embodiments of the currently disclosed technology. As will be appreciated, the currently disclosed technology can be modified in various aspects, all without departing from the spirit and scope of the currently disclosed technology. Therefore, the drawings and detailed description should be regarded as illustrative and non-restrictive in nature. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 An example biopharmaceutics modeling system including a physiologically based biopharmaceutics (PBB) model is shown.

[0021] Figure 2 Schematic diagram showing an example metabolic configuration for designing a PBB model, which can form Figure 1 At least a portion of the system depicted in .

[0022] Figure 3 An example dispersibility and lipolysis study of an oral formulation for a PBB model is shown, which can form Figure 1 At least a portion of the system depicted in .

[0023] Figure 4 Example Phase I clinical trial data used for model validation are shown.

[0024] Figure 5 An example diagram showing a gastric emptying model that can be formed Figure 1 At least a portion of the system depicted in .

[0025] Figure 6 shows some parameter sensitivity analysis of the PPB model, which can be formed Figure 1 At least a portion of the system depicted in .

[0026] Figure 7 Figure 2 shows simulated in vivo data from a PBB model including gastric emptying, which can be formed Figure 1 At least a portion of the system depicted in .

[0027] Figure 8 Figure 2 shows simulated in vivo data for the PBB model omitting gastric emptying, which can be formed Figure 1 At least a portion of the system depicted in .

[0028] Fig. 9 Figure 2 shows simulated in vivo data of the PBB model, which shows the effect of digestion rate on the PBB model, which can form Figure 1 At least a portion of the system depicted in .

[0029] Fig.10 Figure 2 shows simulated in vivo data of the PBB model, which shows the effect of bioavailability on the PBB model, which can form Figure 1 At least a portion of the system depicted in .

[0030] Fig.11 A system including a computing device is shown that can be used with Figure 1The system is the same as or forms at least a part of the system depicted in.

[0031] Fig.12 An example method for predicting pharmacokinetic parameters following administration of a formulation is shown.

[0032] Detailed Description

[0033] Aspects of the present disclosure include systems, methods and devices for predicting pharmacokinetic parameters (e.g., plasma free fatty acid concentrations, such as caprylic acid concentrations or ketone concentrations) after administration of a formulation comprising a medium chain triglyceride (MCT) such as tricaprylin. "Medium chain triglycerides" as used herein or "MCT" as used herein refers to triglycerides in which all three fatty acid moieties are medium chain fatty acid moieties. Medium chain fatty acids (MCFA) are fatty acids having 6 to 12 carbon atoms. In some aspects, MCT refers to triglycerides containing almost exclusively caproic acid (C6:0), caprylic acid (caprylic acid; C8:0), capric acid (capric acid; C10:0), or any combination thereof. In some aspects, the MCT is caprylic acid (caprylic acid; C8:0). In some aspects of the method, the MCT is tricaprylin. Tricaprylin is a semi-synthetic medium chain triglyceride (MCT) consisting of 8 carbon fatty acids, which has a water solubility of 0.4 mg / L. Other MCTs can also be used for the dietary management of AD, and are within the scope of the present disclosure. The MCT preparations comprising tricaprylin are used as examples throughout the present disclosure, and those skilled in the art will appreciate that preparations comprising other MCTs can be utilized in the methods disclosed herein. After oral administration, MCTs (such as tricaprylin) are digested into free fatty acids (e.g., caprylic acid, also known as caprylic acid) by lipases in the gastrointestinal (GI) tract, and then fatty acids are absorbed from the duodenum. Free fatty acids (such as caprylic acid) are then converted into ketone bodies by oxidation in the liver. Medium-chain fatty acids (such as caprylic acid) show preferential absorption and oxidation over long-chain fatty acids. These ketones serve as an important energy source for the central nervous system (CNS). Therefore, due to the considerable ketogenic potential of free fatty acids (such as caprylic acid), MCTs such as tricaprylin can provide advantages over previous technologies. Glucose enters the brain through an active transport mechanism that can be saturated. The nonlinear relationship between brain and plasma glucose levels can be determined. Compared to the transporter-mediated uptake of glucose into the CNS, there is a linear relationship between the concentrations of free fatty acids (such as caprylic acid) in the blood and brain. Therefore, elevated plasma ketone levels following administration of MCTs (such as tricaprylin) can lead to a corresponding increase in brain ketone concentrations.

[0034] Some pharmacokinetic studies mainly involve direct quantification of plasma ketone levels after MCT (e.g., tricaprylin) administration. However, the contribution of each continuous process related to the digestion of MCT (e.g., tricaprylin) and the release, absorption, and conversion of free fatty acids to ketone bodies is determined and used by the technology disclosed herein. Machine learning modeling such as physiologically based biopharmaceutics (PBB) modeling can utilize the knowledge of the physiological processes involved, and together with selected in vitro and in vivo data sets, generate predictions of the biopharmaceutical behavior of the drug. In order to cover the complexity behind the pharmacokinetics of MCT (such as tricaprylin), a PBB model can be designed based on in vitro digestion studies and human clinical data. The model can be validated using the results of a Phase I clinical trial. For example, a Phase I clinical trial evaluating the administration of a single dose of tricaprylin formulation AC-SD-03 to healthy volunteers can be used.

[0035] Additional advantages of the technology will become apparent from the disclosure herein.

[0036] Figure 1 An example biopharmaceutical modeling system 100 including a PBB model 102 is shown. The PBB model 102 can be developed using MlxEditor2020R1 (e.g., from Lixoft, Antony, France) to describe the absorption, metabolism, and / or elimination of MCTs (such as tricaprylin) after oral administration. Probability values ​​indicating the effect of each physiological process on ketogenesis and the generation of ketone bodies (e.g., β-hydroxybutyrate and acetoacetate) can be generated by the PBB model 102 as model outputs 104. Once the PBB model 102 is trained and / or validated, the PBB model 102 can receive patient data 105 associated with a target patient. The PBB model 102 can analyze the patient data 105 using the techniques discussed herein and output a predicted plasma free fatty acid concentration 107 (such as caprylic acid concentration 107) and / or a predicted ketone concentration 109 based at least in part on the patient data 105. The design and operation of the PBB model 102 are discussed in more detail below.

[0037] The PBB model 102 can receive various model inputs 106, including MCT (such as tricaprylin), experimental data 108 and / or human metabolic data 110. For example, the characteristics of individual patient curves can be bimodal phenomena. The data can indicate that the MCT preparation, such as tricaprylin preparation (AC-SD-03), can be orally administered at high doses. Therefore, the effect of gastric emptying on in vivo performance can be assumed based on the gastric emptying model. For example, gastric emptying can be regulated by various factors such as neural circuits and hormones, and occurs in different ways in fasting and feeding states. As indicated by the data, in the fed state, gastric emptying of solid food may take several hours. Another approach, i.e., gastric path (gastric diameter), can allow caloric free liquid to be emptied from the stomach in a much shorter time. Through the PBB model 102, such an emptying pattern can be observed after the intake of a high-fat homogeneous meal. In order to mathematically explain this complex physiological process, the rhythmic emptying pattern can be incorporated into the PBB model 102. For example, a rhythmic emptying pattern may include a sinusoidal algorithm similar to a regular transport pattern, followed by a hysteresis period:

[0038]

[0039] The amplitude (a) can represent the maximum transport rate, while the duration of the lag phase can be defined by the period (P). After a certain time (h), the gastric compartment can be completely emptied. For example, Architect (v1.5.1, iseesystems, Lebanon, USA) was used to determine appropriate simulation parameters. Studies measuring the volume of gastric contents after ingestion of a high-fat homogenous meal can be used to define one or more gastric emptying parameter ranges. The absolute average multiple error (AAFE) can be calculated using the following equation to compare model simulations with clinical data:

[0040]

[0041] To estimate the digestion of MCT (e.g., tricaprylin) after entering the small intestine, in vitro lipid digestion kinetics 114 can be used as model input 106. To model the digestion rate 116, a first-order kinetic process can be assumed to fit the release curve of fatty acids in the presence of lipase 118. The digestion rate constant (k d ). Under conditions that simulate gastrointestinal (GI) physiology, the digestion curve can reach a plateau of about 83%. Phase I clinical trials may also involve quantification of MCTs (such as tricaprylin). Negligible uptake of undigested triglycerides (C max=1.05μ,M). It can be determined that the total digestibility of MCT (such as tricaprylin) in the GI tract mainly determines the overall bioavailability (F) of free fatty acids (e.g., caprylic acid). The total digestibility of MCT (e.g., tricaprylin) in the GI tract mainly determines the overall bioavailability (F) of free fatty acids (e.g., caprylic acid).

[0042] Additionally, input parameters are provided in Table 1. Table 1 depicts computer parameters 130 and parameter ranges that may be used as initial estimates for computer analysis of free fatty acids and ketones:

[0043]

[0044] (Table 1)

[0045] To estimate parameters related to the distribution, metabolism, and elimination of free fatty acids (such as octanoic acid), reliable data sources can be determined for generating training data sets to describe the plasma pharmacokinetics following intravenous administration of free fatty acids (such as octanoic acid) or ketones. Pharmacokinetic data 120 related to MCTs (such as tricaprylin) and their metabolites in humans are summarized in Table 2, which can be included in the training data 122 and / or used by the validation system 124 of the biopharmaceutical modeling system:

[0046]

[0047] (Table 2)

[0048] In Table 2, DBHB is D-β-hydroxybutyrate; ET is essential tremor; and SD is standard deviation.

[0049] In some examples, ketone plasma concentrations can be quantified after an intravenous push of sodium acetoacetate (n=12). Acetoacetate and D-β-hydroxybutyrate (n=6), two main components of plasma ketones, can also be quantified after intravenous infusion of three different doses of D-β-hydroxybutyrate. After five different doses of free fatty acids (e.g., octanoic acid) are orally administered to a patient population (n=3 per dose), blood free fatty acid (such as octanoic acid) concentrations can be measured. The pharmacokinetic data 120 can be analyzed using PKanalix2020R1 (e.g., Lixoft, Antony, France). These injections, infusions, and / or administrations of the doses of the target patient can be provided to the PBB model 102 (e.g., as patient data 105).

[0050] In some instances, a one-compartment model can fully describe the pharmacokinetics of free fatty acids (e.g., octanoic acid) and ketones, and can be applied to a mixed model later. When the plasma concentration-time curve obtained from the Phase I clinical trial is modeled during model validation, the parameter range obtained (minimum, maximum, and mean values ​​of all studies) can be used to define the initial estimate within a fixed range. For the development of pharmacokinetic models, several key assumptions can be made. They involve model structures and differential equations for simulating a single process. The pharmacokinetic studies provided to the PBB model 102 can involve a wide range of dosages. Here, the scalability of the rate between different dose levels and patient populations can be determined or assumed. This can include the enzymatic conversion of free fatty acids (e.g., octanoic acid) to plasma ketones, which is considered to be an unsaturated first-order process.

[0051] In some cases, the PBB model 102 can be written in the Mlxtran programming language and executed using Monolix 2020R1 (e.g., Antony, France). Initial parameters and fixed parameter ranges (e.g., such as those shown in Table 1 and / or model inputs 106) can be obtained (e.g., from experimental data and / or literature data).

[0052] Next, the population parameters of the target population treated with MCT preparations (e.g., AC-SD-03) can be obtained using the stochastic approximation expectation maximization (SAEM) algorithm. The population parameters can be considered to be log-normally distributed. In addition, the estimation tasks of empirical Bayesian estimation (EBE) and conditional distribution can be performed to calculate individual predictions and give the uncertainty of each individual parameter and data respectively. The individual clinical data 124 of free fatty acids (e.g., caprylic acid) and ketones obtained from pharmacokinetic studies can be used for model validation. For example, the accuracy of the prediction can be evaluated by calculating the relative mean absolute prediction error (MAPE, %) using the following equation:

[0053]

[0054] In Equation 3, V O is the observed concentration, and V P is the corresponding predicted concentration. Predictions with a MAPE of <10% or 10%-20% may be interpreted as excellent or good performance. In addition, the model may be evaluated by visual inspection of goodness-of-fit plots, observed values ​​versus individual predicted values, individual weighted residuals (IWRES) versus time, and normalized prediction distribution error (NPDE) versus time plots, which are performed by one or more validation / simulation systems 126.

[0055] In some examples, a partial parameter sensitivity analysis (PPSA) can be performed by studying the effects of model inputs 106 on the results of simulations performed using the PBB model 102. Average PK profiles for both free fatty acids (e.g., octanoic acid) and ketones in humans can be used for simulations. Three different combinations of input variables can be varied. Specifically, gastric emptying parameters a, h, and P and the in vivo parameter p metabo , k a , k conv , k e 、V OA and V 酮 can be varied within a range of ±20%. To visualize the in vitro digestion rate k d The effect on the pharmacokinetic curve is that the release rate can be changed by 50% of the absolute value. metabo , k a , k conv ) Additional PPSA was performed to determine its impact.

[0056] In some examples, simulations with (multiple) validation / simulation systems 126 can be performed using Simulx 2020R1 (Lixoft, Antony, France) from the PK model using individual estimates generated by Monolix software. One hundred virtual individuals can be simulated, including plasma concentration-time curves of free fatty acids (e.g., caprylic acid) and ketones, considering a single dose administration of 42,500 μmol of tricaprylin (formulated as AC-SD-03) at 0-16 hours. In some cases, statistical analysis can be performed using t-tests on GraphPad Prism 8.

[0057] A variety of neurodegenerative disorders lead to hypometabolism of the brain due to a significant reduction in glucose utilization. Ketogenesis can provide an alternative energy source and can have a beneficial effect on cognitive function in patients with Alzheimer's disease. Medium chain triglycerides (e.g., tricaprylin) are metabolized into ketone bodies, which can provide 80% of the energy required for CNS. Other systems have been working to determine the impact of the bioavailability of these ketone bodies. For this purpose, the PBB model 102 can be a suitable computer model that is developed to quantitatively describe the various effects on ketogenesis and the effects of physiological parameters on pharmacokinetics. In vitro digestion studies of MCT (such as tricaprylin) in gastric and intestinal media can provide an understanding of its expected degradation rate in the human intestine. Based on information obtained from the literature, other in vivo parameters that have a potential impact on ketone body concentrations can be estimated. Phase I clinical trials evaluating MCT preparations (such as AC-SD-03) in healthy volunteers can provide pharmacokinetic data required for model validation.

[0058] Go to Figure 2 , depicts a schematic diagram 202 of a metabolic design configuration 204 for designing a PBB model 102 .

[0059] like Figure 2 As shown, PBB model 102 can be used for computer analysis and data simulation.The preparation can enter the stomach, release MCT (such as tricaprylin) in the stomach, and then transfer to the duodenum.MCT (such as tricaprylin) can be digested into free fatty acids (such as caprylic acid from tricaprylin), and then fatty acids are absorbed in the intestinal tract, and then metabolized / converted.A part of the absorbed free fatty acids (for example, caprylic acid) can be converted into ketone bodies, which are then eliminated from the body.A certain part can enter the metabolic cycle and disappear from blood plasma.

[0060] As previously noted, the preparation can undergo gastric transport before MCT (e.g., tricaprylin) reaches the intestinal tract. Dissolution and lipolysis of MCT preparations (e.g., AC-SD-03) can lead to the release of free fatty acids (e.g., caprylic acid), which are then absorbed into the blood circulation. Subsequently, conversion of free fatty acids (e.g., caprylate) to ketone bodies occurs. This may involve certain losses to other metabolic pathways including the Krebs cycle. The ketone bodies may appear in plasma, where they serve as an energy source for the brain and other tissues. Clinical trials for model validation may involve quantification of free fatty acids (such as caprylic acid) and ketone bodies in plasma from healthy volunteers. The in vitro and in vivo data sources for model design and validation will be discussed in more detail below.

[0061] In some examples, a preliminary dispersibility study can be performed to evaluate the physical stability of MCT formulations (such as AC-SD-03) in gastric media (pH 1.2, 37°C). It can provide more information about the physical state of the intragastric dispersion that may affect gastric emptying. Considering that multiple peaks can be observed during the first 4 hours after administration, this period can be studied. Other time periods can also be studied based on the observed peaks of the tested MCT formulations.

[0062] Figure 3 Dispersibility and lipolysis studies of AC-SD-03 formulations at 37° C. in 50 mM phosphate buffer (pH 1.2) are shown. Image 302 is a photograph showing the sedimentation of AC-SD-03 formulations over 4 hours. Any of the information from these studies can be used as model input 106 and / or for validating the PBB model 102.

[0063] like Figure 3As depicted, Figure 304 shows the percentage of the amount of tricaprylin in the supernatant of an AC-SD-03 suspension over 4 hours (n=3). Figure 306 depicts the digestion curve of AC-SD-03, expressed as the percentage of digestion per unit time. The solid line represents the first order kinetic fit. The data are provided as mean ± standard deviation (SD) (n=3).

[0064] In some instances, sedimentation of MCT formulations such as AC-SD-03 can occur rapidly, such as Figure 3 As shown in images 302 and 304. Although no creaming or phase separation of oil and water phases was observed, a considerable portion may settle with the insoluble excipients. The experiment can be performed without stirring. Therefore, sedimentation may be overestimated. Nevertheless, the adsorption of MCT (e.g., tricaprylin) to solid silica particles may lead to a considerable degree of gastric retention. The first dose can be given in a total volume of 300 mL, resulting in low viscosity. Therefore, most of the MCT (e.g., tricaprylin) may travel along the gastric route from the fundus through the center of the gastric antrum to the duodenum. After 1 hour, repeated drinking may lead to additional "clearance" of the drug. Therefore, in this example, a more stable dispersion system can increase the amount of tricaprylin delivered by this route.

[0065] As a next step, the in vitro digestion of MCT preparations (such as AC-SD-03) can be measured. The preparation can be exposed to gastric media and then lipolysis is performed in a medium simulating intestinal conditions. The latter can cause the release of fatty acids, expressed as the percentage of digested free fatty acids (e.g., tricaprylin), as shown in Figure 306. For each mole of tricaprylin, three moles of titratable fatty acids can be released. In the first 5 minutes, 40 ± 3% of tricaprylin can be converted into caprylic acid, and then the digestion rate gradually decreases. After 15 minutes, the digestion curve reaches a plateau at 83 ± 4%. Assuming rapid absorption from the intestine, digestion rate (kd) and lipolysis degree (F) mainly determine the bioavailability of caprylic acid. In the simulation performed by (multiple) verification / simulation system 126, this score (83%) can be used as a baseline value when simulating different bioavailabilities. By the way, there is the possibility that undigested tricaprylin can also be absorbed into the blood circulation. However, the estimated rate of this process may be 30-52 gmol / h. This corresponds to less than 0.5% of the total dose during the first 4 hours. Furthermore, negligible amounts of tricaprylin could be found in the blood circulation during Phase I clinical trials, also indicating the minor relevance of this pathway. Metabolic processes such as the hepatic conversion of tricaprylin generally result in the formation of caprylic acid, which is subsequently converted into ketone bodies.

[0066] As mentioned above, the digestive parameter k that has a strong influence on the absorption of free fatty acids (e.g., octanoic acid) d F and F can be determined and used by the digestion parameter model of the PBB model 102. The first-order kinetic model is used to fit the in vitro fat decomposition curve (eg Figure 3 306) produces 4.1h -1 However, some evidence suggests that the in vitro system does not resemble physiological processes and that even higher digestion rates can be expected in vivo. In addition, prior hydrolysis of medium-chain triglycerides by gastric lipase may increase digestion. Considering the potential differences between in vitro and in vivo lipolysis, it can be determined that the digestion rate of 1-5 h -1 range of digestion parameters. Thus, the PBB model 102 can fit an individual curve with a digestion rate that is approximately 20% higher. Furthermore, based on the observations from the in vitro experiments, a lipolysis of 83 ± 4% of the initial dose can be estimated. This is accompanied by the uncertainty of the in vitro experiments. Therefore, higher and lower bioavailabilities in the virtual clinical trial can be defined, as discussed below.

[0067] In some instances, the PBB model 102 can generate and use a gastric emptying model. As a non-limiting example, the MCT formulation AC-SD-03 is a powder formulation with a high tricaprylin content. The elevated dose, together with the low density of the drug, promotes the effect of gastric emptying on pharmacokinetics. The uneven distribution of tricaprylin partially combined with gastric contents is very likely. Therefore, an algorithm can be integrated to simulate these emptying patterns 128. Training data 122 of measured values ​​indicating the change in the volume of gastric contents after ingestion of a high-fat homogeneous meal in humans can be used. The PBB model 102 can assume that this is similar to the intake of a standard breakfast with tricaprylin. The sinusoidal algorithm can be used to simulate both the slow emptying pattern without additional drinking water and the repeated "clearing" of tricaprylin. The latter is the result of a clinical protocol that allows free drinking of water 1 hour after administration. The appropriate parameter range can be determined by fitting literature data. An AAFE of 1.11 can indicate the best representation of the curve. The parameters used by the algorithm can include the amplitude (a) and period (P) of the sinusoid and the time to complete emptying (h). They can be 1.65h -1 , 2.03h and 0.62h. Parameter variations within the range of ±50% of the initial values ​​can be estimated (Table 1).

[0068] In some examples, the PBB model 102 can generate and / or use a free fatty acid (such as octanoic acid) absorption and metabolism model. The training data 122 can include measured values ​​of the pharmacokinetics of free fatty acids (such as octanoic acid) after oral administration at various dose levels, with absorption rate constants ranging from 0.23-3.16h -1Poor absorption may reduce the effect of gastric emptying on pharmacokinetics and bioavailability. However, for free fatty acids (such as caprylic acid), the absorption rate may be high enough to ensure complete absorption within the intestinal transit time. Nevertheless, the absorption rate can be considered as one of the influencing parameters and further evaluated in PPSA.

[0069] In some examples, the PBB model 102 may include a ketone body model for determining the ketogenic effect of free fatty acids (such as octanoic acid) and the pharmacokinetics of ketone bodies. The conversion rate constant (k) of free fatty acids (such as octanoic acid) to ketone bodies can be measured in rats and / or humans. conv ). To estimate this parameter, human clinical trials can be used as training data 122. The plasma ketone levels observed after sodium caprylate infusion in male Sprague-Dawley rats can be considered as confirmation of first order kinetics without using this rate for further analysis (e.g., as used by validation / simulation system 126). The conversion rate k of MCT (e.g., tricaprylin) can be estimated based on the elimination rate of free fatty acids (such as caprylic acid in the case of tricaprylin). conv Here, ketogenesis can be considered as the main metabolic pathway. Certain losses to other biochemical pathways can be determined. The Krebs cycle uses the available chemical energy from the conversion of acetyl-CoA (intermediate of octanoic acid) to nicotinamide adenine dinucleotide (NADH) as an energy source. In addition, β-hydroxyβ-methylglutaryl-CoA can be used in the synthesis of cholesterol. This model imposes a certain amount of energy on the conversion rate k conv or metabolic fraction p metabo The high sensitivity of the PBB models 102 may reduce their predictive power. Therefore, the PBB models 102 were systematically analyzed for their sensitivity in a partial parameter sensitivity analysis (PPSA).

[0070] In some cases, the PBB model 102 can perform simulations of human pharmacokinetics, for example, using the validation / simulation system 126. The PBB model 102 can fill the knowledge gap between the administration of MCTs (such as tricaprylin) and the appearance of ketone bodies in plasma. Breaking this process down into individual steps and then performing partial validation can be one of the goals of the PBB model 102. The final structure can include a simulation of gastric emptying in the fed state with and without additional drinking water. Dispersibility studies can confirm the formation of a stable low-viscosity emulsion, and during in vitro lipolysis experiments, for example, more than 80% of tricaprylin can be rapidly digested into caprylic acid. By performing pharmacokinetic analysis of clinical data, the absorption and conversion rate constants and distribution volume (VOA) of free fatty acids (such as caprylic acid) can be obtained by the PBB model 102.

[0071] In some instances, such as Figure 4As depicted, training data 122 may include Phase I clinical trial data 400, which provides the plasma concentrations of free fatty acids (in this example, octanoic acid) and ketone bodies over time in 20 healthy volunteers. For model validation, pharmacokinetic curves of up to 20 subjects may be used. Compared with the in vivo pharmacokinetic data of octanoic acid (blue dots) and ketones (red dots), the computer individual simulation of octanoic acid is displayed as a purple line, and ketones are displayed as black lines. Although the model structure still provides a variety of possible situations for the simulation, the fixed parameter range can improve the estimation by weighing each influence. As expected, some subjects still cause high prediction errors. However, the overall performance of the PBB model 102 for Phase I clinical trials may be acceptable. A single parameter may be narrowly distributed, wherein the distribution volume of free fatty acids (e.g., octanoic acid) varies the most. The prediction errors are summarized in Table 3. Table 3 shows the mean absolute prediction error (MAPE) of the PBB model 102 for simulating human clinical data of all subjects.

[0072]

[0073] (Table 3)

[0074] The absence of systematic bias can be confirmed by visualizing the relationship between the individual weighted residual (IWRES) and the normalized prediction error (NPDE) after administration and time. The difference between the mean and zero can be insignificant (p=0.06 for IWRES and p=0.09 for NPDE, respectively). Each analysis can indicate accurate predictions.

[0075] In some instances, gastric emptying represents a strong influence on pharmacokinetics and can be difficult to simulate. For example, initially, administration of AC-SD-03 in 300 mL of water can potentially result in rapid emptying along the gastric tract. Figure 5 The emptying patterns corresponding to 20 healthy volunteers are provided in.

[0076] Figure 5 Validation techniques for the PBB model 102 are depicted. Graph 500 depicts the predicted amount of tricaprylin in the stomach of 20 subjects within 3 hours after administration of AC-SD-03. The thick curves (green, blue, and red) represent typical conditions for subjects with different gastric emptying patterns and different water intake during the trial. Schematic 502 depicts the emptying of AC-SD-03 suspension from a fed stomach along the gastric tract (created with biorender.com).

[0077] In some instances, the first emptying may affect 50-100% of the total dose. Since water intake was only allowed 1 hour after administration of tricaprylin, additional water intake may be assumed for some patients and at later time points. Thus, a model algorithm with the flexibility to simulate various gastric emptying patterns may be used. Emptying of the enthalpy may occur much slower, resulting in a plateau (e.g., Figure 5 500). The reason for the observed emptying pattern 1 hour after ingestion of tricaprylin may be due to uneven distribution of tricaprylin in the stomach or additional fluid intake.

[0078] In some cases, the initial values ​​(Table 1) can provide estimates of physiological and formulation-related effects, given that all population parameters can be within the initial range and prediction errors can be kept low. This leads to the following conclusions. -1 The rate at which fatty acids reach the small intestine is too low, and absorption of fatty acids can become the rate-limiting step. In addition, compared with the much faster lipolysis (4.6 h -1 ) and ketosis (2.6h -1 ) compared to the absorption rate of different individuals. The conversion of caprylic acid in the liver may play a minor role in the availability of the active form of tricaprylin. One of the major effects may be related to the drug formulation, which in most patients results in some retention of tricaprylin in the stomach. These observations can be systematically evaluated in PPSA and / or can be provided to the PBB model 102 as model input 106 or as part of the validation process.

[0079] In some instances, a partial parameter sensitivity analysis (PPSA) can be performed. Model parameters can be accompanied by considerable errors. However, simulations based on in vitro data are expected to have the highest uncertainty. PPSA clarifies parameters that have a strong impact on the simulation. The agreement between the high uncertainty of one or more parameters and the model sensitivity can reduce the predictive power.

[0080] For example, the input parameters can be initially divided into three clusters. To identify combinations with significant effects, physiologically based parameters (e.g., gastric emptying) and population parameters (e.g., elimination rate) can be varied within a range of 20%. In vitro parameters (e.g., digestion rate) can be varied within a wider range.

[0081] Figure 6 Depicted are (A and D) gastric emptying parameters (a, h, P), (B and E) in vitro parameters kd and F, and (C and F) in vivo parameters (p metabo , k a , k conv , k e , VOA and V酮 ) A sensitivity analysis of some parameters of the computer model using incremental changes 600. These parameters can be varied based on simulated mean (A, B, and C) octanoate and (D, E, and F) ketone plasma profiles. The observed plasma concentrations can be plotted as symbols, with error bars representing standard deviations (n=20).

[0082] In some examples, the effects of system parameter changes on simulated pharmacokinetic profiles of octanoate and ketones are provided in Figure 6 The corresponding pharmacokinetic parameters (C ma x, T max AU Call ), and when gastric emptying parameters (a, h and P), in vitro parameters (k d and F) and in vivo parameters (p metabo , k a , k conv , k e 、V OA and V 酮 ) change together, the model can be sensitive to them.

[0083] In some instances, gastric emptying can affect C max had the strongest effect, while in vitro and in vivo parameters particularly affected AU Call The variation of a single parameter can further resolve the absorption rate (k a ). In the absence of changes, the high metabolic capacity of the liver (measured in k conv ) will not be the rate-limiting factor for ketogenesis and can be overwhelmed by formulation-related parameters. conv ) or metabolic rate (p metabo ) may have little or no effect on pharmacokinetics. Nevertheless, they may be more important for some patient populations, for example, due to reduced liver function. In cases where the absorption rate is lower than the estimated digestion and conversion rate, it may act as a gatekeeper for ketone concentrations.

[0084] In some instances, as a next step, virtual clinical trials can be performed that simulate the expected pharmacokinetics in a larger population. Here, the digestion rate (k d ), intestinal digestion extent (F) and gastric emptying.

[0085] In some examples, pharmacokinetics can be predicted in silico by the PBB model 102. For example, a Phase I clinical trial can be used for model validation, which includes only a small number of subjects (n=20). Therefore, additional simulations can be performed on the expected performance of AC-SD-03 in a larger population (n=100).

[0086] Figure 7 A first graph 700 depicts simulated in vivo data for octanoic acid; and a second graph 702 depicts simulated in vivo data for ketones. The blue band shows the percentiles calculated for the simulated data for all individuals and all replicates. The solid black curve represents the median.

[0087] like Figure 7 As depicted, the biopharmaceutical modeling system 100 can provide predictions of the expected plasma concentration-time profiles of octanoate and ketones in 100 virtual subjects. The C max , T max and AU Call The medians of the concentrations of octanoate and ketones were 124.7 μM, 1 h, and 388.2 μM·h, respectively. While the powder formulations can lead to higher variability in plasma octanoate concentrations, plasma ketone levels can remain within a narrower range, indicating a minor effect of the conversion rate.

[0088] Figure 8 Depicted are a first graph 800 of simulated in vivo caprylic acid and a second graph 802 of simulated ketone concentrations using a physiologically based biopharmaceutical model (without gastric emptying). The blue band shows the percentiles calculated for all individuals and all replicates of the simulated data. The solid black curve represents the median.

[0089] In some instances, several formulation parameters can be expected to have considerable effects on plasma ketone levels. These can include gastric emptying and the rate and extent of lipid digestion. By completely removing / omitting the gastric compartment from the PBB model 102, the effect of gastric emptying on the in vivo fate of AC-SD-03 can be studied. Results are provided in Figure 8 Simulations showed that gastric emptying significantly reduced plasma concentrations, and especially C max (70% and 26% increase for caprylic acid and ketones respectively). Figure 8 A comparison of predictions using the PBB model 102 with and without gastric emptying is provided in FIG. max From 2 hours to 1 hour, while the AU of caprylic acid and ketones Call Remain almost the same. C max It may be a strong indicator of adequate brain ketone body supply. Therefore, maximizing peak concentrations is most likely to contribute to beneficial effects in the treatment of AD.

[0090] Fig. 9 Depicted are a first graph 900 and a second graph 902 showing the digestion rate (k d ) on the effects of octanoic acid in the body; and a third figure 904 and a fourth figure 906, which show their effects on ketone concentrations. The first figure 900 and the third figure 904 depict k d50% increase; and the second graph 902 and the fourth graph 906 depict k d The blue band shows the percentiles calculated for the simulated data for all individuals and all replicates. The solid black curve represents the median.

[0091] In some instances, further simulations can determine the digestion rate (k d ) on plasma ketone concentration. The digestion rate can be increased by 50% and 100%. d When the increase is 50%, Fig. 9 As shown, a significant increase in plasma ketone levels can be observed. Median C ma x and AU Call These can be calculated to be 129.5 μM and 311.5 μM·h (octanoic acid) and 488.7 μM and 1126.5 μM·h (ketone), respectively.

[0092] In some cases, taking into account the results of dispersibility studies, an increase in digestion rate can be achieved, for example, by developing a more stable liquid emulsion system with optimized dispersibility and physical stability. Higher surface area can provide a larger portion of tricaprylin to be digested at a higher rate.

[0093] Fig.10 Depicted are a first graph 1000 and a second graph 1002 showing the effect of bioavailability (e.g., tricaprylin digestion (F)) on caprylic acid in vivo, and in a third graph 1004 and a fourth graph 1006 showing the effect of bioavailability on ketone concentration. The F for the first graph 1000 and the third graph 1004 is set to 50%; and the F for the second graph 1002 and the third graph 1004 is set to 100%. The blue band shows the percentiles calculated for the simulated data for all individuals and all replicates. The solid black curve represents the median.

[0094] In some examples, the extent to which lipolysis affects bioavailability can be evaluated in silico. Degradation of MCTs (e.g., tricaprylin) can be the primary cause of a portion of the initial dose becoming available in plasma. For AC-SD-03, the digestion fraction found in vitro (83%) can be used as a base value to simulate the increase in bioavailability, and a maximum increase in bioavailability of 17% (corresponding to 100% bioavailability) can be observed. It is further assumed that estimates of absolute bioavailability may not be accurate and even higher values ​​are possible, such as Fig.10 However, all aspects considered, the limited dispersibility and incomplete lipolysis of AC-SD-03 may still be the most critical formulation parameters limiting bioavailability.

[0095] Therefore, additional approaches may focus on formulation strategies that increase the rate and extent of digestion and reduce the effects of gastric emptying.

[0096] In some examples, studies / analyses performed with the biopharmaceutical modeling system 100 describe and explain the ketogenic effect after ingestion of tricaprylin using the PBB model 102. The simulation can accurately predict the in vivo results of the spray-dried powder formulation in humans. Estimating the physiological effects and formulation-related effects on pharmacokinetics, the extent of tricaprylin digestion can have a strong impact on bioavailability. Not only can incomplete in vitro lipolysis be observed, but dispersibility studies can provide additional evidence that a certain portion of tricaprylin is adsorbed to the insoluble excipients of AC-SD-03. In addition, the uneven distribution and adsorption of tricaprylin to gastric contents may be the main reason for the strong effect of gastric emptying on pharmacokinetics. A significant delay in absorption and a decrease in the peak plasma concentration of ketone bodies can be observed. Once caprylic acid reaches the small intestine, the absorption of fatty acids (rather than enzymatic conversion) can have a significant impact on the C of ketones. max Has a certain gatekeeping effect. Even at high doses, the enzyme capacity can avoid limiting plasma concentrations. When formulating the medium chain triglyceride tricaprylin, the formulation strategy can target the reduction of gastric transit time as well as improved dispersibility to enhance the digestive process. The spray-dried powder formulation can be at least partially processed into a low-density solid, thereby promoting the effect of gastric emptying and reducing its digestion extent.

[0097] In some examples, the biopharmaceutical modeling system 100 can use various materials to perform the techniques disclosed herein. For example, the spray-dried tricaprylin powder formulation (AC-SD-03) can be composed of 40% of the active pharmaceutical ingredient tricaprylin and 60% of inactive ingredients, including Kolliphor RH40 (BASF, Germany), monoolein (Abitec Corp., USA), soy phosphatidylcholine (Lipoid LLC, USA), polyvinyl pyrrolidone (JRS Pharma GmbH & Co. KG, Germany), Fumed silica (Koninklijke DSM NV, Netherlands), sucralose, acesulfame potassium and flavoring. It can be produced by Lonza (Oregon, USA) under license from Cerecin Inc. (Singapore). Caprylic acid (>99% purity) can be purchased from Nu-Chek Prep Inc. (Minnesota, USA). Anhydrous sodium dihydrogen orthophosphate (98% purity) can be purchased from Chem Supply (SA, Australia). Methanol (LiChrosolv) can be purchased from Merck Millipore (Bayswater, Australia). Sodium hydroxide granules (>97% purity) can be purchased from Ajax Finechem (NSW, Australia). Trifluoroacetic acid (HPLC,>99%) and 4-bromophenylboronic acid (4-BPBA,>95% purity) can be purchased from Sigma-Aldrich (St. Louis, USA). FaSSIF (fasted state simulated intestinal fluid) can be purchased from Biorelevant.com Ltd (London, UK). USP grade pancreatic enzyme extract from porcine pancreas can be purchased from Southern Biologicals (VIC, Australia). Triglyceride quantitative colorimetric kit can be purchased from Sigma-Aldrich Pte. Ltd. (MO, USA). Distilled water and deionized water (18.2 Mil / cm at 25°C) can be obtained from Millipore water purification system (Billerica, USA). It should be understood by those of ordinary skill in the art that other MCT preparations can be used in the biopharmaceutical modeling system and method disclosed herein, and AC-SD-03 is only an example of such a preparation.

[0098] In some examples, MCT formulations (eg, AC-SD-03) can be characterized based on in vitro dispersibility and digestion behavior.

[0099] In some instances, the dispersibility study may include a method for evaluating the dispersibility and lipolysis of medium chain triglycerides (MCT). The evaluation uses simulated gastric media to determine the physical stability and gastric emptying of the preparation containing MCT in the stomach. The method may include dispersing the preparation in the gastric medium, incubating the preparation at 36°C to 38°C without stirring for 3.5 to 4.5 hours, taking a sample of the supernatant from the preparation at intervals of 0 minutes, 5 minutes, 10 minutes, 15 minutes, 30 minutes, 45 minutes, 1 hour, 1.25 hours, 1.5 hours, 1.75 hours, 2 hours, 2.25 hours, 2.5 hours, 2.75 hours, 3 hours, 3.25 hours, 3.5 hours, 3.75 hours, 4 hours, 4.25 hours and 4.5 hours, quantifying the amount of MCT in each sample, analyzing the amount of quantified MCT to determine the time when the MCT in the preparation settles, wherein the sedimentation time indicates gastric retention, the bioavailability of MCT and / or the degree of digestion.

[0100] In some instances, any gastric medium known in the art can be used in the method for determining the physical stability of a formulation. The gastric medium can include a phosphate buffer having a pH of 1.2. The phosphate buffer can include a 50 mM phosphate buffer. The gastric medium can be prepared from a mixture of sodium chloride (2.0 g), dilute hydrochloric acid (24 ml) and water (approximately 1000 ml). In some aspects, the gastric medium is prepared by dissolving pepsin (1 g), gastric mucin (1.5 g) and NaCl (8.775 g) in 1 L of distilled water and adjusting to a pH of 1.3 using 6N HCl. The gastric medium can be commercially available artificial gastric fluid or simulated gastric fluid. The gastric medium can include a solution of the composition and pH of simulated gastric fluid. Non-limiting examples of components in the gastric medium include sodium chloride, dilute acid, pepsin, mucin, etc.

[0101] The method may include dispersing the preparation comprising MCT in a gastric medium. Dispersing the preparation comprises mixing a powder of the preparation comprising MCT in a phosphate buffer or in an artificial gastric medium. Dispersing the preparation may produce a suspension of the preparation comprising MCT in a phosphate buffer or in an artificial gastric medium. Incubation may be performed with further stirring. The dispersion may then be incubated at 35° C. to 39° C. without stirring for 3 to 5 hours.

[0102] Incubation can be performed for about 3 hours, about 3.5 hours, about 4 hours, about 4.5 hours, or about 5 hours. In some examples, incubation can be performed for about 4 hours.

[0103] The dispersion can be incubated at a temperature of about 35° C. to about 39° C. In some aspects, the dispersion is incubated at a temperature of about 35° C., about 36° C., about 37° C., about 38° C., or about 39° C. In some aspects, the dispersion is incubated at a temperature of about 37° C.

[0104] The method can also include obtaining a sample of the supernatant from the preparation after incubation. Samples can be obtained at several intervals. Samples can be obtained at the beginning of incubation, at 0 minute and at one or more intervals. The one or more intervals can include obtaining samples at 0 minute, 5 minutes, 10 minutes, 15 minutes, 30 minutes, 45 minutes, 1 hour, 1.25 hours, 1.5 hours, 1.75 hours, 2 hours, 2.25 hours, 2.5 hours, 2.75 hours, 3 hours, 3.25 hours, 3.5 hours, 3.75 hours, 4 hours, 4.25 hours and / or 4.5 hours.

[0105] The method may also include quantifying the amount of MCT in the sample obtained at several intervals. Quantifying the MCT in the sample may include quantifying using a colorimetric assay. In the disclosed method, any known commercially available quantitative colorimetric or fluorescence assay kit may be used to quantify MCT (e.g., Sigma-Aldrich MAK266-1KT). During the assay, triglycerides may be converted into free fatty acids and glycerol by chemical hydrolysis, enzymatic hydrolysis, or any combination thereof. In some instances, chemical hydrolysis may be performed by adding an alcohol solution of an alkali metal hydroxide. In some instances, enzymes may be used for enzymatic hydrolysis, and non-limiting examples of enzymes include lipase, cholesterol esterase, carboxylesterase, protease, or any combination thereof. Glycerol is then oxidized to generate a product that reacts with a probe to produce color and fluorescence, which is detected using a spectrophotometer. Glycerol may be measured as follows: using ATP and Mg ++ Glycerol is enzymatically converted to glycerol-1-phosphate, which is enzymatically converted to dihydroxyacetone phosphate using oxidized nicotinamide adenine dinucleotide (NAD), thereby reducing NAD to the reduced form NADH, and determining the amount of NADH formed. NADH can be determined by reacting NADH with a colorless tetrazolium dye suitable for forming a colored product in such a reaction in the presence of an electron carrier, and determining the amount of the resulting colored tetrazolium product by photometry.

[0106] The method may also include measuring the time for the MCT in the formulation to settle. The time for the MCT in the formulation to settle may be qualitatively measured by visual methods.

[0107] The amount of MCT quantified and the time it takes for the MCT in the formulation to settle can be used to determine the gastric retention, bioavailability, and / or extent of digestion of the MCT.

[0108] In some examples, the dispersibility study of AC-SD-03 can be carried out in 50mM phosphate buffer at pH 1.2. The powder (0.5g) can be dispersed in 8mL of phosphate buffer and incubated at 37°C for 4 hours without further stirring. To evaluate stability, 0.1mL aliquots of the supernatant can be collected at 0, 5, 10, 30min, 1, 2, 3 and 4h, respectively. The amount of tricaprylin can be measured using a triglyceride quantitative colorimetric kit (n=3).

[0109] In some examples, the present disclosure provides methods for determining digestion of a formulation comprising MCT. The disclosed methods may include incubating a formulation comprising MCT in a gastric medium at 35°C to 39°C for about 10 minutes to about 60 minutes under constant agitation, taking samples of the formulation in the gastric medium at constant intervals over 30 minutes, and adding a lipase inhibitor to each of the collected samples at the time of collection, adjusting the pH of the formulation in the gastric medium to 6.80 after the 30-minute incubation, adding a FaSSIF solution to the pH-adjusted formulation, initiating fat decomposition in the formulation by adding a lipase to the formulation, incubating the formulation for 60 minutes while maintaining a pH of 6.8, taking a sample of the formulation before incubation and then taking a sample of the formulation at constant intervals during the 60-minute incubation, and adding a lipase inhibitor to each of the collected samples at the time of collection, optionally quantifying the amount of digested MCT in each sample using HPLC, and analyzing the quantified information to determine the digestion rate, extent, and / or profile of the formulation comprising MCT.

[0110] Any gastric medium known in the art can be used in the method for determining the digestion of the preparation. The gastric medium can include a phosphate buffer at a pH of 1.2. The phosphate buffer can include a 50mM phosphate buffer. The gastric medium can be prepared from a mixture of sodium chloride (2.0g), dilute hydrochloric acid (24ml) and water (about 1000ml). The gastric medium can be prepared by dissolving pepsin (1g), gastric mucin (1.5g) and NaCl (8.775g) in 1L distilled water and adjusting the pH to 1.3 using 6N HCl. The gastric medium can be commercially available artificial gastric fluid or simulated gastric fluid. The gastric medium can include a solution of the composition and pH of simulated gastric fluid. Non-limiting examples of components in the gastric medium include sodium chloride, dilute acid, pepsin, mucin, etc.

[0111] The formulation can be incubated at 36°C to 38°C for about 10 minutes to about 60 minutes with constant stirring. The formulation can be incubated for about 10 minutes, about 15 minutes, about 20 minutes, about 25 minutes, about 30 minutes, about 35 minutes, about 40 minutes, about 45 minutes, about 50 minutes, about 55 minutes, or about 60 minutes. The formulation can be incubated for about 30 minutes. The formulation can be incubated at a temperature of about 35°C to about 39°C. The formulation can be incubated at a temperature of about 35°C, about 36°C, about 37°C, about 38°C, or about 39°C. In some instances, the formulation can be incubated at a temperature of about 37°C.

[0112] As disclosed herein, the method can also include obtaining a sample of the preparation in the gastric medium at a constant interval within 30 minutes and adding a lipase inhibitor. Samples can be obtained at about 0 minutes, about 5 minutes, about 10 minutes, about 15 minutes, about 20 minutes, about 25 minutes and / or about 30 minutes. Samples can be obtained at about 0 minutes, about 15 minutes and about 30 minutes. The lipase inhibitor can include orlistat, 4-BPBA, lipstatin, panclicins, hesperidin, ebelactones, esterase inhibitor, valilactone, benzoxazinone or their derivatives. The lipase inhibitor can include 4-BPBA. In some instances, a lipase inhibitor of 0.5M of about 1 μL to about 5 μL can be added. In some examples, about 1 μL, about 1.5 μL, about 2 μL, about 2.5 μL, about 3 μL, about 3.5 μL, about 4 μL, about 4.5 μL, or about 5 μL of lipase inhibitor is added. In some examples, 2.5 μL of 0.5 M 4-BPBA in methanol can be added.

[0113] The disclosed method may further comprise adjusting the pH of the formulation in the gastric medium to about 6.80. For example, the pH may be adjusted to about 6.80 using a 5M NaOH solution.

[0114] According to some examples of the method, fasting state simulated intestinal fluid (FaSSIF) solution is used to simulate intestinal conditions. FaSSIF solution can be prepared by mixing 3mM sodium taurocholate, 0.75mM lecithin, 0.174g NaOH (granules), 1.977g NaH2PO4.H2O or 1.719g anhydrous NaH2PO4 and 3.093g NaCl in 500mL of purified water. FaSSIF solution can have a pH of about 6.5 and an osmotic pressure of about 270mOsmol / kg. Commercially available FaSSIF (e.g., Biorelevant V2FAS) can be used for the disclosed method. A 10-fold concentrated FaSSIF solution of about 1-4mL can be added. A 10-fold concentrated FaSSIF solution of about 1ml, about 2ml, about 3ml or about 4ml can be added. A 10-fold concentrated FaSSIF solution of about 2ml is added.

[0115] In some instances, the method can include the following lipolysis in the startup preparation: add lipase to the preparation, and incubate the preparation for 60 minutes while maintaining a pH of 6.8. Examples of lipases suitable for the present invention include, but are not limited to, animal lipases (for example, porcine lipase, pancreatic enzyme lipases), bacterial lipases (for example, Pseudomonas lipases and / or Burkholderia lipases), fungal lipases, plant lipases, recombinant lipases (for example, produced by a suitable host cell by recombinant DNA technology, the host cell being selected from any one of the bacteria, yeast, fungi, plants, insects or mammalian host cells cultivated, or such recombinant lipases: it includes amino acid sequences homologous to or substantially identical to naturally occurring sequences, lipases encoded by nucleic acid homologous to or substantially identical to naturally occurring lipase encoding nucleic acids, etc.), synthetic lipases, chemically modified lipases and mixtures thereof. In some instances, the lipase is a pancreatic enzyme lipase suspension. Any method known in the art can be used to prepare a lipase suspension. For example, lipases can be dispersed in an aqueous solvent (for example, water, acetone, ethanol). In some instances, about 1-4 ml of lipase can be added to the preparation. In some instances, about 1 ml, about 2 ml, about 3 ml, or about 4 ml of lipase can be added to the preparation. In some instances, about 2 ml of lipase can be added to the preparation.

[0116] During or after the start of lipolysis, the method can also include obtaining a sample. Samples can be obtained at 0 minutes, 5 minutes, 10 minutes, 15 minutes, 20 minutes, 25 minutes, 30 minutes, 35 minutes, 40 minutes, 45 minutes, 50 minutes, 55 minutes and 60 minutes. The sample may be further processed. In some instances, the sample is centrifuged and the supernatant can be used for further analysis.

[0117] The method may also include analysis of free fatty acids released by digestion. Any known method in the art, including high performance liquid chromatography (HPLC), gas chromatography (GC) or liquid chromatography-mass spectrometry (LC-MS), may be used to quantify the preparations digested with octanoic acid. HPLC may be used to quantify free fatty acids. In short, the free fatty acids were measured at 40°C in a Waters Symmetry C 18 Column (3.5 μm, Free fatty acids (such as octanoic acid) were chromatographed on a 4.6 x 75 mm column equipped with a Phenomenex Security Guard Cartridge C 18 . Then use isocratic elution to measure free fatty acids at a flow rate of 1 mL / min. The mobile phase can include a mixture of 0.1% trifluoroacetic acid in water and methanol (40:60). Free fatty acids (e.g., octanoic acid) are detected at about 220 nm. The injection volume that can be used can be about 10 μL, the run time can be about 10 minutes, and the retention time of free fatty acids (e.g., octanoic acid) can be about 5.9 minutes. Free fatty acid stock solutions (such as octanoic acid stock solutions in water: methanol (40:60)) can be used to prepare a standard reference range of 0.01-2.00 mg / mL in the same solvent. Free fatty acid concentration can be calculated from peak area based on a calibration curve.

[0118] In some aspects, quantification of free fatty acids (e.g., caprylic acid) can be used to understand the digestion rate and / or digestion profile of the disclosed MCT preparations.

[0119] In some examples, to evaluate the release of caprylic acid and simulate gastric conditions, formulation AC-SD-03 (2.5 g) can be added to 20 mL of 50 mM phosphate buffer (at pH 1.2) and vortexed thoroughly. The formulation can be incubated in a 70-mL thermostatic glass container (37° C.) under constant magnetic stirring for 30 minutes. The sample (100 μL) can be collected into an Eppendorf tube containing 2.5 μL of lipase inhibitor (0.5 M 4-BPBA in methanol) and 147.7 μL of methanol and vortexed for 30 seconds at 0, 15, and 30 minutes. The pH of the mixture can then be adjusted to 6.800 ± 0.003 with 5 M NaOH solution. To simulate intestinal conditions, 2 mL of a 10-fold concentrated FaSSIF solution can be added to the container. Lipolysis can be initiated by adding a pancreatic lipase suspension (2 mL). The pancreatic lipase suspension can be prepared from a pancreatic extract using a previously described method

[26] . The pH of the sample in the digestion container can be maintained at 6.8 during digestion by using the 2MNaOH solution titrated into the digestion medium by pH-stat control module (Metrohm AG, Herisau, Switzerland). Before digestion (0 minute time point), samples (100 μ L) can be collected in the Eppendorf tube containing lipase inhibitor (0.5M4-BPBA in methanol) (2.5 μ L) and methanol (147.7 μ L), and vortexed for 30 seconds, then collected samples at intervals of 5 minutes for 60 minutes. The sample can then be centrifuged at 16142 g for 30 minutes. The gained supernatant can be transferred into the HPLC bottle for analyzing the released octanoic acid of digestion. The experiment can be repeated three times.

[0120] In some instances, the biopharmaceutical modeling system 100 can quantify the free fatty acids from the digestion sample. In some instances, the free fatty acids are caprylic acid. The digested preparation can be analyzed on a Shimadzu (Shimadzu, Kyoto, Japan) UPLC system, which is composed of a CBM-20A system controller, an LC-30AD solvent delivery module, a SIL-30AC automatic sampler, and a CTO-20AC column oven coupled to an SPD-M30A diode array. Caprylic acid can be chromatographically separated at 40°C on a Waters Symmetry C18 column (3.5 μm, 100A, 4.6x 75mm) equipped with a Phenomenex Security Guard Cartridge C18. Caprylic acid can be measured at a flow rate of 1 mL / min using isocratic elution. The mobile phase is composed of a mixture of 0.1% trifluoroacetic acid in water and methanol (40:60). Caprylic acid can be detected at 220nm. The injection volume can be 10 μ L, the running time can be 10 minutes, and the retention time of octanoic acid can be 5.9 minutes. The octanoic acid stock solution in water: methanol (40:60) can be used to prepare a standard reference range of 0.01-2.00 mg / mL in the same solvent. The detection limit and quantitative limit of the HPLC method can be 0.002 and 0.006 mg / mL respectively. Based on the calibration curve with R2>0.999, octanoic acid concentration can be calculated from the peak area. It will be appreciated by those of ordinary skill in the art that other free fatty acids can be used in the biopharmaceutical modeling system and method disclosed herein, and octanoic acid is only an example.

[0121] In some cases, the biopharmaceutical modeling system 100 can perform extraction and analysis of literature data to generate training data 122 for the PBB model 102. Plasma concentration-time curves selected from the literature can be extracted using WebPlotDigitizer (v4.3, Ankit Rohatgi, USA). For pharmacokinetic analysis, model design, and data simulation, MonolixSuiteD4 2020R1 (Lixoft, Antony, France) can be used, such as Figure 2 Depicted.

[0122] Clinical pharmacokinetic studies can be conducted with MCT preparations (e.g., to generate model input 106). For example, a clinical pharmacokinetic study was conducted on the tricaprylin product AC-SD-03, which was approved by the local institutional review board and conducted in accordance with the Good Clinical Practice guidelines of the International Conference of Harmonization (ICH-GCP). It was an open-label, randomized, two-way crossover study. Twenty healthy male volunteers (Caucasian and Chinese) ranging in age from 18 to 50 years were recruited. All subjects underwent a supervised overnight fast of at least 8 hours until 1 hour before their scheduled morning dosing. The breakfast for all subjects was standardized and consisted of 2 slices of buttered toast, 2 strips of bacon, 2 scrambled eggs, 113 grams of hash browns, and 240 mL of whole milk. AC-SD-03 (50 g, equivalent to 42,500 μmol tricaprylin) dispersed in 240 mL of water was administered to each subject 30 minutes after the end of the standard breakfast. The dosage cup was rinsed with another 60 mL of water and administered to the subject. Water was restricted only 1 hour before and 1 hour after drug administration, but free drinking of water was allowed at all other times. During the entire restriction period, food and other fluid intakes were restricted. After oral administration of a single dose of the preparation, blood tricaprylin, caprylic acid and ketone body concentrations were quantified. For tricaprylin, blood samples were collected at 0, 1.5, 2.0, 2.5 and 8h. For caprylic acid and ketones (β-hydroxybutyric acid and acetoacetic acid), blood samples were collected at 0, 0.5, 1.0, 1.5, 2.0, 2.5, 4.0 and 8h after administration, respectively. It will be appreciated by those of ordinary skill in the art that other MCT preparations can be used in the biopharmaceutics modeling system and method disclosed herein, and AC-SD-03 is only an example of such a preparation.

[0123] Fig.11 An example system 1100 for implementing the biopharmaceutics modeling system 100 with the PBB model 102 is shown. The system 1100 can include one or more computer devices 1102 that can implement the systems 100-1000 and / or perform the methods discussed herein. In one embodiment, the one or more computing devices 1102 include a clinical device of the biopharmaceutics modeling system 100, or any other device discussed in the present disclosure.

[0124] In some examples, computing device(s) 1102 include computers, personal computers, desktop computers, laptop computers, terminals, workstations, cellular or mobile phones, mobile devices, smart mobile devices, tablet computers, wearable devices (e.g., smart watches, smart glasses, smart skin devices, etc.), multimedia consoles, televisions, Internet of Things (IoT) devices, smart home devices, medical devices, virtual reality (VR) or augmented reality (AR) devices, vehicles (e.g., smart bikes, car computers, etc.), combinations thereof, etc. It should be understood that the specific implementations of these devices may have different possible specific computing architectures, and although not all of these architectures are specifically discussed herein, those of ordinary skill in the art will understand.

[0125] The computing device 1102 may be a computing system capable of executing a computer program product to perform a computer process. The PBB model 102 may be stored on (multiple) computing devices 1102 and executed thereon (e.g., as one or more software components, algorithm modules, etc.). Data and program files may be input to the computing device 1102, which reads the files and executes the programs therein to provide the various components of the PBB model 102 and generate the predictions discussed herein. Some of the elements of the computing device 1102 include one or more hardware processors 1104, one or more storage devices 1106, and / or one or more ports, such as (multiple) input / output (IO) ports 1108 and (multiple) communication ports 1110. The various elements of the computing device 1102 may communicate with each other via (multiple) communication ports 1110 and / or one or more communication buses, point-to-point communication paths, or other communication means.

[0126] Processor 1104 may include, for example, a central processing unit (CPU), a microprocessor, a microcontroller, a digital signal processor (DSP), and / or one or more internal cache levels. There may be one or more processors 1104, such that processor 1104 includes a single central processing unit, or multiple processing units capable of executing instructions and performing operations in parallel with each other, commonly referred to as a parallel processing environment.

[0127] Computing device 1102 may be a conventional computer, a distributed computer, or any other type of computer, such as one or more external computers that may be provided through a cloud computing architecture. The presently described techniques may optionally be implemented in software stored on (multiple) data storage devices (such as (multiple) storage device 1106), and / or communicate through one or more of ports 1108 and 1110 to communicate with each other. Fig.11 The computing device 1102 in is transformed into a dedicated machine for performing operations to provide a computer-generated PBB model 102 of the 3D space.

[0128] The one or more storage devices 1106 may include any non-volatile data storage device capable of storing data generated or employed within the computing device 1102, such as computer executable instructions for executing a computer program, which may include instructions of an application program and an operating system (OS) that manages the various components of the computing device 1102. The (multiple) storage devices 1106 may include, but are not limited to, magnetic disk drives, optical disk drives, solid-state drives (SSDs), flash drives, etc. The (multiple) storage devices 1106 may include removable data storage media, non-removable data storage media, and / or external storage devices made available through a wired or wireless network architecture with such computer program products, including one or more database management products, World Wide Web server products, application server products, and / or other additional software components. Examples of removable data storage media include compact disk read-only memory (CD-ROM), digital versatile disk read-only memory (DVD-ROM), magneto-optical disk, flash drive, etc. Examples of non-removable data storage media include internal hard disks, SSDs, etc. The one or more storage devices 1106 may include volatile memory (eg, dynamic random access memory (DRAM), static random access memory (SRAM), etc.) and / or nonvolatile memory (eg, read only memory (ROM), flash memory, etc.).

[0129] A computer program product containing mechanisms for implementing systems and methods according to the presently described techniques may be stored in (multiple) storage devices 1106, which may be referred to as machine-readable media. It should be understood that a machine-readable medium may include any tangible, non-transitory medium capable of storing or encoding instructions to perform any one or more of the operations of the present disclosure for execution by a machine, or capable of storing or encoding data structures and / or modules utilized by or associated with such instructions. A machine-readable medium may include a single medium or multiple media (e.g., a centralized or distributed database, and / or associated caches and servers) that store one or more executable instructions or data structures.

[0130] In some embodiments, computing device 1102 includes one or more ports, such as I / O port 1108 and communication port 1110, for communicating with other computing, network or media devices. It should be understood that I / O port 1108 and communication port 1110 can be combined or separated, and more or fewer ports can be included in computing device 1102.

[0131] I / O port 1108 may be connected to an I / O device or other device through which information may be input to or output from computing device 1102. Such I / O devices may include, but are not limited to, one or more input devices, output devices, and / or environmental sensor devices.

[0132] In one embodiment, the input device converts human-generated signals (such as human voice, physical movement, physical touch or pressure, etc.) into electrical signals as input data entering the computing device 1102 through the I / O port 1108. Similarly, the output device can convert electrical signals received from the computing device 1102 through the I / O port 1108 into signals that can be perceived by humans as output, such as sound, light and / or touch. The input device can be an alphanumeric input device, including alphanumeric and other keys for communicating information and / or command selections to the processor 1104 through the I / O port 1108. The input device can be another type of user input device, including, but not limited to: direction and selection control devices, such as a mouse, a trackball, cursor direction keys, a joystick and / or a scroll wheel; one or more sensors, such as a camera, a microphone, a position sensor, a direction sensor, an inertial sensor and / or an accelerometer; and / or a touch-sensitive display screen ("touch screen"). The output device can include, but is not limited to, a display, a touch screen, a speaker, a tactile and / or haptic output device, etc. In some embodiments, the input device and the output device may be the same device, such as in the case of a touch screen.

[0133] In one embodiment, the communication port 1110 can be connected to a network, and the computing device 1102 can receive network data that can be used to perform the methods and systems described herein and transmit information and network configuration changes determined thereby. In other words, the communication port 1110 can connect the computing device 1102 to one or more communication interface devices that are configured to transmit and / or receive information between the computing device 1102 and other devices via one or more wired or wireless communication networks or connections. Examples of such networks or connections include, but are not limited to, Universal Serial Bus (USB), Ethernet, Wi-Fi, Near Field Communication (NFC), etc. One or more such communication interface devices can be used to communicate with one or more other machines through the communication port 1110, which is directly through a point-to-point communication path, through a wide area network (WAN) (e.g., the Internet), through a local area network (LAN), through a cellular network (e.g., third generation (3G), fourth generation (4G), long term evolution technology (LTE), fifth generation (5G), etc.) or through another communication method. In addition, the communication port 1110 can communicate with an antenna or other link for electromagnetic signal transmission and / or reception.

[0134] In an example implementation, the PBB model 102 may be implemented by instructions stored on the storage device 1106 and executed by the processor 1104 .

[0135] exist Fig.11 The system 1100 described in the foregoing is only one possible example of a computer system that can be employed or configured according to aspects of the present disclosure. It should be understood that other non-transitory tangible computer-readable storage media can be utilized, which store computer-executable instructions for implementing the currently disclosed technology on a computing system. In the present disclosure, the disclosed methods can be implemented as an instruction set or software readable by a computing device 1102.

[0136] Fig.12 An example method for predicting a pharmacokinetic parameter following administration of a formulation is shown, which method may be performed by any of the systems 100 - 1100 disclosed herein.

[0137] In some embodiments, at operation 1202, method 1200 can generate training data indicating the human metabolic results of the preparation, the human metabolic results include at least one of gastric emptying parameters, digestion rate, absorption constant or conversion rate, in the case where the training target variable is plasma free fatty acid (such as octanoic acid) concentration or ketone concentration. At operation 1204, by one or more computer devices executing computer-readable instructions, method 1200 can train the PBB model with training data. At operation 1206, as part of the parameter sensitivity analysis of the PBB model, method 1200 can change the input parameters of the training data according to the simulated average plasma free fatty acid (e.g., octanoic acid) curve or the simulated average ketone plasma curve. At operation 1208, method 1200 can receive patient data corresponding to the patient who orally receives the preparation through one or more computer devices. At operation 1210, method 1200 can use the patient data to predict at least one of the target plasma free fatty acid (such as octanoic acid) concentration or the target ketone concentration with the PBB model.

[0138] It should be understood that Fig.12 The specific order or hierarchy of steps in the depicted method 1200 or in the present disclosure is an example of an embodiment scheme and may be rearranged while remaining within the disclosed subject matter. Fig.12 Any of the operations depicted in or in the present disclosure may be omitted, repeated, performed in parallel, performed in a different order, and / or performed in a different order than described in Fig.12 or any other operations in this disclosure.

[0139] Although the present disclosure has been described with reference to various embodiments, it should be understood that these embodiments are illustrative, and the scope of the present disclosure is not limited to them. Many variations, modifications, additions and improvements are possible. More generally, embodiments according to the present disclosure have been described in the context of specific embodiments. Functionality can be separated or differently combined in various embodiments of the present disclosure, or described with different terms. These and other variations, modifications, additions and improvements may fall within the scope of the present disclosure as defined in the appended claims.

Claims

1. A method for predicting pharmacokinetic parameters after administration of a formulation comprising medium chain triglycerides (MCT), the method comprising: generating training data indicative of human metabolic outcomes of the agent; training a physiologically based biopharmaceutics (PBB) model using the training data via one or more computer devices executing computer readable instructions; As part of a parameter sensitivity analysis of the PBB model, varying an input parameter of the training data according to a simulated mean plasma free fatty acid profile or a simulated mean ketone plasma profile; receiving, by the one or more computer devices, patient data corresponding to a patient orally receiving the formulation; and At least one of a target plasma free fatty acid concentration or a target ketone concentration is predicted with the PBB model using the patient data.

2. The method of claim 1, wherein the formulation comprises medium chain triglycerides (MCTs) composed of 8-carbon fatty acids having a water solubility of 0.4 mg / L, and the target plasma free fatty acid concentration is a target plasma caprylic acid concentration.

3. The method of claim 1, wherein the training data comprises a data set based on a gastric emptying model.

4. The method of claim 3, wherein the gastric emptying model comprises a sinusoidal algorithm corresponding to a rhythmic emptying pattern.

5. The method of claim 3, wherein the gastric emptying model comprises a first route of solid food and a second route of liquid food.

6. The method of claim 3, wherein the gastric emptying model comprises a gastric amplitude parameter, a complete emptying parameter, and a gastric cycle parameter.

7. The method of claim 1, wherein predicting at least one of the target plasma free fatty acid concentration or the target ketone concentration comprises determining a digestion rate constant obtained by fitting a release curve of the formulation in the presence of a lipase while assuming first order kinetics.

8. The method of claim 1, wherein the training data comprises an in vitro parameter which is a digestion rate or a bioavailability digestion fraction.

9. The method of claim 1, wherein the training data comprises an in vivo parameter that is at least one of a metabolic fraction, an absorption constant, a conversion rate, an elimination rate, a volume distribution of free fatty acids, or a volume distribution of ketone bodies.

10. The method of claim 1, further comprising performing a dispersibility study on the MCT to determine the physical stability of the MCT, predicting at least one of the target plasma free fatty acid concentration or the target ketone concentration based on the dispersibility study.

11. The method of claim 1, further comprising: The release profile of the formulation was used to model the digestion rate; and A digestion rate constant is obtained based on modeling the digestion rate, wherein predicting at least one of the target plasma free fatty acid concentration or the target ketone concentration comprises using the digestion rate with the PBB model.

12. The method of claim 11, wherein predicting at least one of the target plasma free fatty acid concentration or the target ketone concentration comprises predicting the bioavailability of free fatty acids in the patient based on the digestion rate using the PBB model.

13. The method of claim 1, wherein training the PBB model comprises instructing the PBB model to consider the enzymatic conversion of free fatty acids to plasma ketones as a non-saturated first order process.

14. A method for predicting a pharmacokinetic parameter after administration of a formulation, the method comprising: generating training data indicative of one or more gastric emptying parameters, digestion rates, absorption constants, or conversion rates, wherein the training target variable is ketone concentration; training a physiologically based biopharmaceutics (PBB) model using the training data via one or more computer devices executing computer readable instructions; receiving, by the one or more computer devices, patient data corresponding to a patient receiving the formulation; and The target ketone concentration is predicted using the PBB model using the patient data.

15. The method of claim 14, wherein predicting the target ketone concentration comprises predicting the patient's plasma free fatty acid concentration using the PBB model and the patient data.

16. The method of claim 14, further comprising validating the PBB model using clinical trial data quantifying free fatty acids and ketone bodies from volunteer plasma.

17. A method for predicting a pharmacokinetic parameter following administration of a formulation comprising medium chain triglycerides (MCTs), the method comprising: generating training data indicative of a gastric emptying parameter, a digestion rate, an absorption constant, or a conversion rate, wherein the training target variable is a plasma free fatty acid concentration; training a physiologically based biopharmaceutics (PBB) model using the training data via one or more computer devices executing computer readable instructions; receiving, by the one or more computer devices, patient data corresponding to a patient receiving the formulation; and The target plasma free fatty acid concentration is predicted using the PBB model using the patient data.

18. The method of claim 17, further comprising predicting ketone concentrations with the PBB model using the patient data, the target plasma free fatty acid concentration, and the conversion rate estimate.

19. The method of claim 17, wherein training the PBB model comprises providing a gastric absorption emptying model with a sinusoidal algorithm, the model having an amplitude variable, a period variable, and a complete emptying time variable.

20. The method of claim 17, wherein the training data comprises: A first cluster includes physiologically based parameters, a second cluster includes population parameters, and a third cluster includes in vitro parameters.

Images