Application of Mechanistic Pharmacokinetic Model for the Optimization of Metformin Delayed Release Dosage Form for Intestinal Targeting

Background: Metformin (MET) is a widely prescribed drug for managing Type 2 Diabetes Mellitus (T2DM). Despite the rich clinical experience and advantages, the clinical utility of MET in renal failure patients is limited because of the treatment-related side effects. A novel colon-targeted MET Delayed-Release (DR) dosage form could be a lucrative option for managing T2DM in renal failure patients. MET DR tablets have minimum systemic exposure and are believed to have the same efficacy as other formulations. Physiologically Based Pharmacokinetic (PBPK) modelling can be useful for drug product development, especially for new dosage forms for improving the safety, efficacy, and clinical applicability of off-patented generic drugs. Objectives: The current study aims to develop a PBPK model for Proto-type (PT) screening of MET DR tablets. Materials and Methods: MET PBPK model was developed based on the available literature data. Firstly, Intravenous (IV) and oral PBPK models of MET are developed and validated. The developed model was then used to predict the Bioavailability (BA) of PT MET DR tablets using Virtual Bioequivalence (VBE) trials. Results: The relative bioavailability of the MET DR tablet was about 20% when compared with other MET formulations. The results indicate that the DR formulation could be effective for reducing the BA and systemic exposure of the drug in chronic renal failure patients. Conclusion: PBPK modelling and VBE trials can successfully demonstrate the Pharmacokinetic (PK) parameters of PK formulations and the lower systemic exposure and site targeting of MET DR tablets could be useful for the management of T2DM in renal failure patients.


INTRODUCTION
MET, an oral hypoglycaemic agent belonging to the class of biguanides is the first-line agent for managing T2DM.According to the American Diabetes Association, MET can be given alone or in combination with other agents for the management of T2DM along with lifestyle modification and diet. 1 MET is historically connected to Galega officinalis, a guanidine-rich herb used to treat diabetic symptoms in Europe in the 1700s.MET was first synthesized in 1922 and approved for treating T2DM in Europe in 1958.In 1995, MET was introduced into the US market and is still the most prescribed drug for T2DM worldwide. 2The anti-diabetic effect of MET is achieved by inhibiting hepatic glucose production, increasing peripheral glucose utilization, and increasing insulin-mediated glucose uptake.Additional benefits of MET therapy include cost-effectiveness, no weight gain, and non-hypoglycaemic effect. 3MET has several other pharmacological actions aside from glucose reduction, including cardioprotective effects, antitumor activity, immunosuppressive effects, and anti-aging properties, and it also reduces the hyperandrogenic symptoms of polycystic ovarian syndrome. 4e currently available MET oral solid dosage forms are in the form of Immediate-Release (IR) and Extended-Release (XR) tablets.IR tablets of MET are available in various doses like 500, 850, and 1000 mg, whereas XR tablets are in 500, 750, and 1000 mg doses.The dosage regimen for the management of T2DM is not fixed and can be individualized based on effectiveness and tolerance level without exceeding the maximum daily dose of 2000 mg. 5 Even though the significance of MET therapy is known, patients are showing poor adherence and compliance to the dosing regimens and treatment.The major barriers to patient compliance are swallowing difficulty (dysphagia) due to the large tablet size and rough coatings; and Gastrointestinal (GI) side effects such as abdominal discomfort, diarrhoea, cramps, nausea, and vomiting. 6MET use is limited in conditions characterized by high plasma concentrations, such as renal and hepatic impairment, because of higher plasma drug concentration; it leads to MET-associated Lactic Acidosis (LA) (a condition of elevated blood lactate level).The United States Food and Drug Administration (USFDA) recommends that before starting MET therapy, the glomerular filtration rate should be estimated and MET will contraindicate in patients with a glomerular filtration rate below 30 mL/min/1.73m 2 .The maximum feasible dose in such patients is between 250 to 500 mg. 7 MET administration has no direct effect on the fasting blood glucose level, insulin, and glucagon production in healthy individuals compared to oral administration.This could be linked with the hypothesis of local GI action of MET rather than systemic drug concentration. 8In a study on T2DM patients with MET on and off therapy, it was shown that withdrawal of MET is associated with reductions in gut hormones such as Glucagon-Like Peptide-1 (GLP-1), gastric inhibitory polypeptide, and peptide tyrosine-tyrosine, which confirms that MET has a gut-based local effect and could be used as a novel approach for the treatment of T2DM. 9 The ileum part of the intestine predominantly contains L cells, which are responsible for the secretion of incretin hormones.These hormones play an important role in blood sugar regulation by increasing insulin secretion and decreasing glucagon release. 10MET improves GLP-1 concentration in the gut either by a direct effect on L cells or by reducing dipeptidyl peptidase-4 activity, an enzyme responsible for the degradation of GLP-1.MET-associated alteration of the bile acid pool also results in GLP-1 secretion by stimulating bile acid receptors in the L cells. 11The absorption of MET is mainly taking place from the duodenum and jejunum.Targeting the drug release onto the ileum is achieved by formulating DR MET tablets.Site targeting can elevate its glucose-lowering efficacy with less systemic intolerance and GI side effects when compared with MET IR and XR formulations.Another important objective of delaying MET release is minimizing the systemic exposure, thereby reducing the incidence of LA in renal-impaired patients, and providing a diabetic treatment option for chronic kidney failure patients. 12e formulation and manufacturing of MET DR tablets are reported in the literature.The primary objective of MET DR is to reduce BA without compromising the therapeutic efficacy, thereby reducing GI intolerance and LA. 13 MET DR is studied in several clinical trials, including phases 1 and 2 studies.The phase 1 study consists of a comparative BA of MET DR with currently marketed formulations of IR and XR tablets in healthy volunteers.The results of the study proved that MET DR had a lower BA than IR and XR MET formulations.The phase 2 clinical trial was a placebo-controlled, dose-ranging study in patients with T2DM over 12 weeks.In this study, the patients were randomly assigned to receive either MET DR (600, 800, or 1000 mg Once Daily), or placebo, or MET XR (1000 or 2000 mg Once Daily).The treatment with MET DR produced a clinically significant reduction in the blood glucose level and was well tolerated when compared with MET XR tablets. 14Currently, a pivotal phase 3 clinical trial of MET DR is ongoing on T2DM patients with normal renal function and stage 3 chronic kidney disease to ensure the safety and efficacy of the product. 15PK models are mechanistic mathematical models, which describe the PK properties of drug substances and drug products. 16The main advantage of this model is that, by using physiologically integrated mathematical equations, it can predict the in vivo PK of the drug, which, subsequently minimizes the preclinical and clinical studies and is also useful for life cycle management and regulatory requirements.Other widely used applications of PBPK models are the prediction of drug-drug interactions, dose and dosage regimen, and human PK from preclinical PK data. 17PBPK models are composed of several compartments that represent the various organs and tissues of the body that are connected by circulating blood systems.The tissue volume and blood flow rate define each compartment.The drug's PK in these compartments is described by a series of differential equations. 18PBPK modelling for MET was developed to predict the PK profiles in diabetic patients, pregnant women, geriatrics, and impaired renal populations. 19,20The present study aims to develop a PBPK model for MET DR tablets for predicting the BA of the dosage form for managing T2DM in renal failure patients.

Software and Data Source
The PBPK model for MET was developed using the commercially available GastroPlus TM version 9.8.3 (Simulation Plus Inc.), with the Advanced Compartmental Absorption Transit (ACAT TM ) module, and the PBPKPlus TM module.The physicochemical properties of drug substances and clinical PK data were collected from literature using search engines such as Google, Web of Science, and PubMed and the keywords used are "Metformin", "Biopharmaceutics", "Chronic kidney failure", "Lactic acidosis", "PBPK", and "Clinical pharmacokinetics".The in silico physicochemical properties of the drug were predicted using the ADMET Predictor TM module of GastroPlus TM .Clinical PK study data were digitized using WebPlotDigitizer version 4.6.

Physicochemical Properties of the Drug Substance
MET is a basic drug that exists as a salt of hydrochloric acid.It is an odourless, white, crystalline, hygroscopic powder with the molecular formula C 4 H 11 N 5 .
HCl. 21 The physicochemical properties of a drug substance are the important input parameters to build the PBPK modelling of the drug.Lipophilicity and molecular weight are the major parameters that determine the membrane permeability of the compound.In contrast, aqueous solubility determines the compound's absorption from the GI tract, which in turn depends on the pKa of the compound. 22The physicochemical parameters of the drug were summarised in Table 1. 23,24

Biopharmaceutics and PK
MET has a high pH-independent aqueous solubility of >100 mg/ mL at room temperature.The highest single dose of 1000 mg is soluble in 250 mL of aqueous media over the physiological pH range.The maximum therapeutic dose of 3000 mg is also soluble across the physiological pH range of 1.2 to 7.2.Hence, it is designated as a highly soluble drug in accordance with the Biopharmaceutical Classification System (BCS).The predicted pKa value of MET is 10.39 and experimental data demonstrates two acid dissociation constants at pH 2.8 and 11.5.In silico and experimental pKa values indicate that the drug has a greater extent of ionization and hence high solubility across the physiological pH range.The log p value of MET is -1.83, which indicates that the drug has poor compound lipophilicity and passive diffusion across the biological membrane.The delayed Tmax of 3 hr after oral IR formulation implies that the drug has a slow absorption rate constant and this could be correlated with its physicochemical properties.MET is poorly absorbed from the stomach.The drug has an absolute oral BA of 50-60% and the rest of the drug is excreted as an unchanged drug in faeces.MET has a longer urinary excretion rate and a long terminal half-life of about 20 hr.MET shows dose-dependent absorption and hence lacks dose proportionality from 500 to 1500 mg.The lack of dose proportionality could be lined with carrier-mediated absorption process and saturation kinetics of the transporter systems, which results in less extent of absorption with high dose.The Caco2 cell permeability data of MET implies that the drug has poor permeability (i.e., lower than metoprolol).Food reduces the rate and extent of MET absorption, with 40% and 25% reductions in C max and AUC, respectively.MET is a substrate for the organic cation transporters, plasma membrane monoamine transporter, thiamine transporter-2, serotonin transporter, and high-affinity choline transporter.The organic cation transport system plays an important role in the absorption, distribution, and elimination of MET. 25,26The biopharmaceutical and PK characteristics of MET were summarised in Table 1.

PBPK Modelling
Sixteen clinical PK studies published in the literature were collected and eleven were used for the development and validation of the PBPK model for MET.Four IV PK studies were considered for model development from the reported literature.Oral PK data were extracted from different pharmaceutical dosage forms such as IR, XR, and DR dosage forms.The patient demographics of all the clinical studies were summarised in Table 2. 14,[27][28][29][30][31][32][33][34][35][36][37][38][39]

IV PK Model Development and Validation
IV PK data were used to understand the distribution and elimination parameters.Demographic information such as age, body weight, sampling times, and methodology were obtained from the literature.IV PK data were used to select the suitable PK and PBPK models using Gastroplus 9.8.3 software.Based on the best fit and prediction errors, the optimized model was selected.The selected model was then used for the development of the oral absorption model.

Oral Absorption Model Development and Validation
The oral PK absorption model was built based on the PK data reported in the literature.Numerical data were obtained from the published graph using WebPlotDigitizer.

VBE trials for the proposed MET DR tablets
The developed PBPK model was then applied to predict the PK parameters of the proposed DR MET tablets for the management of T2DM in chronic renal failure patients.The dissolution data obtained from the literature were integrated with oral PBPK model data and used to build the PBPK model for DR tablets.The patient demographics and other input parameters for these models were taken from the published phase 2 clinical data.The developed model was then validated by using PK data reported in phase 2 clinical studies.Four VBE trials were conducted to estimate the in vivo PK and BE of the MET DR tablets.MET 1000 mg IR tablets (twice daily) were used as a reference product, and three PT MET 900 DR tablets with different drug release profiles were used as test products.The description of the test and reference group is given in Table 3.Each trial was designed in two period, two sequence, two treatments, randomized, single-dose, cross-over, VBE simulations with PK endpoints in healthy subjects under fed conditions.
Another VBE trial was conducted using PT 3 (Target profile) MET DR tablets as the test group and MET 500 mg IR tablets as the reference group.The objective of this study is to compare the PK of PT 3 MET DR tablets vs MET 500 mg IR tablets.The maximum daily dose of MET for the management of T2DM in chronic renal failure patients is 500 mg.The dose of 500 mg in the reference group was chosen based on the recommendations of the American Diabetes Association (i.e., a maximum dose of 500 mg MET for renal failure patients).The number and demographics of virtual subjects were taken similarly to the in vivo Bioequivalence (BE) studies.Average BE analysis was performed for the two sequence, two treatment, cross-over, VBE simulation.The sensitivity for determining the formulation differences was accessed by comparing the PK parameters such as C max , AUC 0-t, and AUC 0-∞.The 90% confidence intervals for the geometric mean ratios of test and reference are between 80% and 125%.

IV PK Model Development and Validation
MET IV PBPK model was developed using the reported plasma concentration-time profile from the literature.The model was optimized and validated for better predictability.The PK data after 250 mg, 500 mg, and 1000 mg of MET IV dose was screened for various models.Two compartment models was selected as the best-fitted model.The predicted and observed PK parameters of various MET IV doses after simulation were reported in Table 4. Table 5 summarizes the PK parameters of the validated model.Figure 1 represents the predicted PK data along with the observed data of MET IV after population simulation.

Oral Absorption Model Development and Validation
The oral PBPK model was developed using IR dosage forms.500 mg and 1000 mg IR tablets were chosen as dosage forms.
Both once-daily and twice-daily dosage regimens were modelled to understand the effect of the residual unabsorbed drug on subsequent dosage administration.All four oral data were validated using the oral absorption model and IV disposition kinetics model.Table 6 illustrates the PK parameters of all oral IR absorption models and Figure 2 (a-d) shows the plasma concentration profile of all IR doses after simulation.Population PK trials were performed between two 500 mg IR tablets (Sr     No: 1 and 2 of Table 6).The population simulation results were compared with reported values and summarized in Table 7.The %T/R ratios for all PK parameters are well within the limit.The predictive outcomes of the oral PK model reveal that the model is robust and prediction errors are well within the regulatory limit.
Dissolution data obtained from the mechanical convolution method was used to predict the PK properties of XR and DR formulations.All other model parameters are fixed, except the dissolution profiles.Figure 3a and 3b show the convoluted % in vitro drug release of both MET XR and DR, respectively.
Oral   Table 6 represents the PK parameters of developed models of XR and DR tablets, and the plasma concentration profile of XR and DR tablets after population simulation is given in Figure 4 (a-d).The model validation data demonstrates oral PK models for all the dosage forms (IR, XR, and DR) could be used for predicting the BA of new PT formulations using pharmaceutical and dissolution data.

VBE Trials for the Proposed MET DR Tablets
MET DR 900 mg tablets are designed to target the drug release in the ileum for localized therapeutic activity in the GI tract.
The localized therapy reduces the systemic exposure of the drug with a targeted relative BA of 20% when compared with MET IR tablets.The dosage form with reduced BA shall be used as an alternative therapeutic option for the management of T2DM in chronic renal failure patients.The hypothetical dissolution data was integrated with the validated oral PK model.MET being a pH-independent high soluble drug, the minimum influence of pH on drug dissolution shall be anticipated.However, DR formulations are usually fabricated with pH-dependent polymers.Hence, the influence of pH on drug release from DR formulation is based on the buffer species, buffering capacity, and pH of the dissolution medium.The proposed dissolution method for hypothetical dissolution data is derived from product-specific guidance documents from the Office of Generic Drugs, USFDA. 40able 8 shows the dissolution profiles of three PT formulations.Population simulations were carried out using 20 healthy subjects and Table 9 shows the results of the population simulation.The PK profiles of three MET DR PT formulations were predicted using the validated PK model reported in the above-mentioned section.Figure 5 (a-c) shows the plasma concentration-time profile of all the 3 PT MET DR tablets after population simulation.PT        mg DR tablets twice daily.The 500 mg IR once daily was chosen based on the maximum daily recommendation dose for type 2 diabetic patients with renal failure.The results of VBE studies are illustrated in Figure 6 and Table 10.MET 900 mg DR tablets twice daily have 58.5% relative BA when compared with MET 500 mg IR tablets once daily.The results obtained from the VBE study demonstrate that the MET DR tablets, even with an 1800 mg daily dose have significantly lower BA than the dose of 500 mg IR formulation.Lower systemic exposure could be beneficial for renal failure patients in real-time clinical settings.However, Phase 3 pivotal clinical studies are required to prove the efficacy and local action of MET DR tablets.The positive phase 3 clinical outcomes in the near future change the therapeutic strategy for diabetic management not only in renal failure patients but also in other diabetic subjects.

CONCLUSION
MET DR tablets could be an important attempt for the management of T2DM in renal failure patients.This dosage form was designed to offer minimum systemic exposure and associated side effects.The PBPK model is the lucrative in silico tool in pharmaceutical product development, which minimizes the cost and time of development and improves the quality of the products being developed.The PBPK model for MET DR tablets was developed and validated to facilitate the PT selection.Three PT formulations of MET DR 900 mg tablets with hypothetical dissolution data were subjected to VBE trials against MET IR 1000 mg as a reference product.The relative BA of PT formulations was close to the targeted theoretical BA of 20%.Furthermore, these formulations were subjected to VBE using 500 mg once daily IR tablets as reference product, based on the current standard of care therapy, and were found to be suitable for the management of T2DM in renal failure patients.PBPK modeling and VBE trials demonstrate that the current PT formulations meet the desired PK profiles and future clinical trial outcomes validate the PK outcomes of VBE trials.

Figure 1 :
Figure 1: Plasma concentration time profile of MET IV infusion after population simulation.

Figure 2 :
Figure 2: (a) Plasma concentration time profile of MET 500 mg IR tablet after population simulation, (b) Plasma concentration time profile of MET 500 mg IR tablet after population simulation, (c) Plasma concentration time profile of MET 1000 mg IR tablets after population simulation, (d) Plasma concentration time profile of MET 1000 mg IR BID tablets after population simulation

Figure 3 :
Figure 3: (a) In vitro dissolution profile of 500 mg MET XR tablets, (b) In vitro dissolution profile of 500 mg MET DR tablets

Figure 5 :
Figure 5: (a) Plasma concentration time profile of PT 1 after population simulation, (b) Plasma concentration time profile of PT 2 after population simulation, (c) Plasma concentration time profile of PT 3 after population simulation.

Figure 4 :Figure 6 :
Figure 4: (a) Plasma concentration time profile of MET1000 mg XR tablet after population simulation, (b) Plasma concentration time profile of MET 500 mg XR BID tablet after population simulation, (c) Plasma concentration time profile of MET 500 mg DR tablet after population simulation, (d) Plasma concentration time profile of MET 1000 mg DR tablet after population simulation

Table 1 : Physicochemical and biopharmaceutical characteristics of the drug.
The study demographic parameters such as age, sex, and body weight were incorporated Malayandi, et al.: PBPK modelling for metformin DR tablets for model building.MET is generally administered with food; hence, a high-fat, high-calorie meal menu was chosen to build the model.PK rate constants obtained from the validated IV PK model were used to predict the oral profile of MET dosage forms.Opt log D Model SA/V 6.1 was used as an absorption scale factor for human-fed state physiologies.The developed oral PK model was fine-tuned with pharmaceutical characterization data such as mean particle size of active pharmaceutical ingredient, particle OD: Once daily; BID: Twice daily; T: Test; R: Reference; GR-6: Gastro-retentive 90 % drug release in 6 hours; GR-9: Gastro-retentive 90 % drug release in 9 hours.

Table 3 : VBE trials of MET DR tablets.
B/P ratio: Blood to plasma ratio; Cl: Clearance; T 1/2 : Biological half-life; K 12 and K 21 : Elimination rate constants; Vc and V 2 : Volume of distribution.

Table 6 : Oral IR, XR, and DR PK parameters after PBPK simulation.
3 was targeted to get 20% of the relative BA.The simulation results demonstrate that the relative predicted BA of 19.9% when compared with MET IR 1000 mg tablets administered twice daily.The results indicate that the DR formulation could be effective for reducing the BA and systemic exposure of the drug in chronic renal failure patients.Clinical trials are required to demonstrate the safety and efficacy of the DR formulation in the management of T2DM in renal failure patientsThe second VBE trials were conducted to establish the comparative PK between the 500 mg IR tablets once daily vs 900 OBS: Observed data; PRE: Predicted data.

Table 8 : In vitro dissolution data of proposed DR tablets.
BID: twice daily.

Table 9 : PK parameters of IR tablets and proposed DR tablets after population simulation
(n=20).Malayandi, et al.: PBPK modelling for metformin DR tablets