Investigation of some factors contributing to negative food effects

Drug administration via feeding tubes-a procedure that carries risks: systematic identification of critical factors based on commonly administered drugs in a cohort of stroke patients

PURPOSE

Drug administration via feeding tubes is considered a process with many uncertainties. This review aimed to give a comprehensive overview of data available on feeding tube application and to carry out risk assessments for drug substances commonly administered to stroke patients.

METHODS

Drugs frequently administered via feeding tubes were identified through a retrospective analysis of discharge letters from a stroke unit. Physicochemical, pharmacokinetic, and stability properties of these drugs and data on drug-enteral nutrition interactions were systematically searched for in the European Pharmacopoeia, Hagers Handbook of Pharmaceutical Practice, Birchers clinical-pharmacological data compilation, and the Martindale Complete Drug Reference, as well as from databases including DrugBank, DrugDex, PubChem, Google Scholar, and PubMed.

RESULTS

Of the drugs most commonly administered via feeding tubes in the present stroke patient cohort, bisoprolol, candesartan, and ramipril could be considered the least critical due to their overall favourable properties. Acetylsalicylic acid, amlodipine, hydrochlorothiazide, omeprazole and esomeprazole, simvastatin, and torasemide pose risks based on pH or light-dependent instability or proposed food effects. The most critical drugs to be administered via feeding tubes are considered to be furosemide, levodopa, and levothyroxine as they show relevant instabilities under administration conditions and substantial food effects; the latter two even possess a narrow therapeutic index. However, little information is available on drug-tube and drug-formula interactions.

CONCLUSION

Feeding tube administration of medications turned out to be a highly complex process with several unmet risks. Therefore, investigations that systematically assess these risk factors using clinically relevant model systems are urgently needed.

Evaluating Utilization of Tiny-TIM to Assess the Effect of Food on the Absorptions of Oral Drugs and Its Application on Biopharmaceutical Modeling

Safety and efficacy are the most critical factors for the development of modern medications. For oral drugs, evaluating drug exposure under various conditions is one of the most important outcomes for clinical trials. These data will help to better understand the safety and efficacy of new drugs. Studies involving potential drug-drug interactions, proton pump inhibitors, and intake of food are often conducted to assess the above. Among the above, the influence of food on exposure to the drug is one of the key data sets for regulatory submission. Since food may have either a positive or negative effect on drug exposure, it is important to obtain an early assessment of the food effect. To better forecast and plan for clinical studies, substantial efforts have been made in the industry to develop modeling and in-vitro and in-vivo assays. Despite the efforts, predicting the effect of food on exposure without integrating the dynamic of the gastrointestinal tract in the assessment remains challenging. In this study, we evaluated the utilization of the dynamic Gastro-Intestinal Model (Tiny-TIM) for the food effect of over 20 drugs/formulations in development or on the market that covers all BCS classes. In general, the Tiny-TIM predicted food effects were in good agreement with the reported data in humans. This suggests that Tiny-TIM can successfully capture the impact of physicochemical properties on absorption under the influence of food.

Effect of Food, Crushing of Tablets, and Antacid Coadministration on Maribavir Pharmacokinetics in Healthy Adult Participants: Results From 2 Phase 1, Open-Label, Randomized, Crossover Studies

The effect of food composition, tablet crushing, and antacid coadministration on maribavir pharmacokinetics was assessed in 2 Phase 1 studies in healthy adults. In the first, a single maribavir 400-mg dose was administered under fasting conditions, with a low-fat/low-calorie or a high-fat/high-calorie meal. In the second, a single maribavir 100-mg dose was administered under fasting conditions, as a crushed tablet, or as a whole tablet alone or with an antacid. The 90% confidence intervals of the geometric mean ratios were within 80%-125% for area under the concentration-time curve (AUC), but not for maximum plasma concentration (Cmax) for low-fat/low-calorie and high-fat/high-calorie meals versus fasting or for whole tablet with antacid versus whole tablet alone. The 90% confidence intervals of the geometric mean ratios for AUC and Cmax were within 80%-125% for crushed versus whole tablet. Maribavir median time to Cmax value in plasma under fed conditions was delayed versus fasting conditions, but there was no statistical difference for crushed versus whole tablet or with versus without antacid. As the antiviral efficacy of maribavir is driven by AUC but not Cmax, findings suggest that maribavir can be administered with food or antacids or as a crushed tablet.

In Vitro Lipolysis Model to Predict Food Effect of Poorly Water-Soluble Drugs Itraconazole, Rivaroxaban, and Ritonavir

It is desirable to predict positive food effect of oral formulations due to food mediated dissolution enhancement of lipophilic drugs. The objective was to assess the ability of in vitro lipolysis to anticipate a positive food effect. Tested formulations included rivaroxaban and itraconazole, where some formulations, but not all, exhibit a positive food effect in vivo in humans. Amorphous solid dispersion formulations of ritonavir, which exhibit a negative food effect in vivo in humans, were also studied. Fe-lipolysis and Fa-lipolysis media representing fed and fasted intestinal conditions were used. Results show frequent agreement between in vitro lipolysis predictions and in vivo human outcomes. For rivaroxaban, food effect of unformulated active pharmaceutical ingredient (API) and products were correctly predicted where 2.5 mg and 10 mg strengths did not show any food effect; however, 20 mg did show a positive food effect. For itraconazole, all four products were correctly predicted, with Sporanox, Sempera, and generic capsules having a food effect, but Tolsura not having a positive food effect. For ritonavir, lipolysis predicted a positive food effect for API and Norvir tablet and powder, but Norvir products have negative food effect in vivo in humans. Overall, the lipolysis model showed favorable predictability and merits additional evaluation.

Effect of High-Fat Meal on the Pharmacokinetics and Safety of Valsartan/Amlodipine Fixed Dose Combination Tablets in Healthy Subjects

Background

The objective of this study was to evaluate the effect of a high-fat meal on the pharmacokinetics and safety of 80/5 mg valsartan/amlodipine tablets in healthy subjects.

Subjects and Methods

These results were derived from a bioequivalence trial where subjects were randomly assigned to take valsartan/amlodipine 80/5mg under fed conditions or after a high-fat meal contained 978.6 kilocalories (54.6% from fat). The blood samples were collected and plasma concentrations of valsartan/amlodipine were measured using high-performance liquid chromatography-mass spectrometry. The non-compartmental module of Phoenix WinNonlin Version 8.2 was used to calculate pharmacokinetic parameters. The BE module of WinNonLin was used to analyze the statistics of the maximum plasma concentration (Cmax), the area under the concentration-time curve from zero to the last quantifiable time point (AUC0-t), and the area under the concentration-time curve from zero to infinity(AUC0-∞) in plasma. 88 healthy subjects were enrolled and divided into in a fasted group and a fed group.

Results

The Cmax, AUC0-t, and AUC0-∞ of valsartan in plasma under fed conditions were 51%, 56%, and 57% lower, respectively, than those under fasted conditions, and the 90% confidence interval (90% CI) were outside the 80.00-125.00% range. All the pharmacokinetic parameters for amlodipine under fed conditions were similar to those observed under fasted conditions, and the 90% CIs were within the 80.00-125.00% range. The incidence of treatment emergent adverse events (TEAE) was similar between the fasted group and the fed group, while adverse drug reaction (ADR) was more frequent in the fasted group which may be related to the higher blood concentrations of valsartan, but all were mild.

Conclusion

The result indicated that the high-fat meal had a significant effect on the pharmacokinetics of valsartan, but no effect on amlodipine. All treatments were safe and well tolerated in healthy subjects under fed and fasted conditions.

Evaluation and prediction of oral drug absorption and bioequivalence with food-drug interaction

This article reviews the impacts on the in vivo prediction of oral bioavailability (BA) and bioequivalence (BE) based on Biopharmaceutical classification systems (BCS) by the food-drug interaction (food effect) and the gastrointestinal (GI) environmental change. Various in vitro and in silico predictive methodologies have been used to expect the BA and BE of the test oral formulation. Food intake changes the GI physiology and environment, which affect oral drug absorption and its BE evaluation. Even though the pHs and bile acids in the GI tract would have significant influence on drug dissolution and, hence, oral drug absorption, those impacts largely depend on the physicochemical properties of oral medicine, active pharmaceutical ingredients (APIs). BCS class I and III drugs are high soluble drugs in the physiological pH range, food-drug interaction may not affect their BA. On the other hand, BCS class II and IV drugs have pH-dependent solubility, and the more bile acid secretion and the pH changes by food intake might affect their BA. In this report, the GI physiological changes between the fasted and fed states are described and the prediction on the oral drug absorption by food-drug interaction have been introduced.

Machine learning predicts the effect of food on orally administered medicines

Food-mediated changes to drug absorption, termed the food effect, are hard to predict and can have significant implications for the safety and efficacy of oral drug products in patients. Mimicking the prandial states of the human gastrointestinal tract in preclinical studies is challenging, poorly predictive and can produce difficult to interpret datasets. Machine learning (ML) has emerged from the computer science field and shows promise in interpreting complex datasets present in the pharmaceutical field. A ML-based approach aimed to predict the food effect based on an extensive dataset of over 311 drugs with more than 20 drug physicochemical properties, referred to as features. Machine learning techniques were tested; including logistic regression, support vector machine, k-Nearest neighbours and random forest. First a standard ML pipeline using a 80:20 split for training and testing was tried to predict no food effect, negative food effect and positive food effect, however this lead to specificities of less than 40%. To overcome this, a strategic ML pipeline was devised and three tasks were developed. Random forest achieved the strongest performance overall. High accuracies and sensitivities of 70%, 80% and 70% and specificities of 71%, 76% and 71% were achieved for classifying; (i) no food effect vs food effect, (ii) negative food vs positive food effect and (iii) no food effect vs negative food effect vs positive food effect, respectively. Feature importance using random forest ranked the features by importance for building the predictive tasks. The calculated dose number was the most important feature. Here, ML has provided an effective screening tool for predicting the food effect, with the potential to select lead compounds with no food effect, reduce the number of animal studies, and accelerate oral drug development studies.

Impact of presence of excipients in drug analysis in fed-state gastric biorelevant media

In this study, the influence of the presence of excipients in sample preparation and clean-up steps required prior to drug analysis in milk-based media which simulate the in vivo properties of the fed state stomach was investigated. 15 excipients, normally present in solid dosage forms of five APIs tested (atenolol, paracetamol, furosemide, nifedipine and propafenone hydrochloride) were mixed (one at a time) with the active pharmaceutical ingredient of interest either via vortexing, co-grinding or shaking of the physical mixture and dissolved in Fed State Simulated Gastric Fluid (FeSSGF). The objective of the study was the assessment of the extraction efficiency of three protein precipitation protocols (using MeOH, ΑCN and 10% w/v TCA), typically used in drug analysis, in milk-based biorelevant media in the presence of the excipients. The mixing technique, fat content of the medium and excipient and solvent effects were investigated. The efficiency of three different protein precipitation reagents in drug extraction when dissolved as API:excipient mixtures in the fed-state medium was compared against the equivalent drug amount recovered in the absence of the excipient in FeSSGF. Most excipients had a significant negative effect (p < 0.05) on drug recovery in the milk-based medium as indicated by the multiple linear regression (MLR) analysis performed. For magnesium stearate and HPMC, the % recovery values were the lowest in four out of the five drugs studied, with a range of 10-100% depending on the API, mixing technique and protein precipitation protocol selected. The negative excipient-dependent effect was more profound in nifedipine and propafenone hydrochloride, the most lipophilic compounds of the study. Acetonitrile was the most effective extraction reagent for most drugs in the presence of excipients, followed by methanol and 10% w/v trichloroacetic acid. Data analysis also revealed a dependence of the extraction method efficiency on the medium lipid content. Application of the above extraction protocols in commercially available formulations highlighted the need for assessment of the effect of excipients in extraction efficiency, before transferring the method directly to dissolution studies of formulations in milk-based fed gastric media. In conclusion, the presence of excipients and the selection of protein precipitation protocol are parameters which can affect significantly the efficiency of protein precipitation when FeSSGF is used as dissolution medium and need to be taken into consideration when developing a quantitative method based on the above sample clean-up technique.

Application of the MechPeff model to predict passive effective intestinal permeability in the different regions of the rodent small intestine and colon

A major component of physiologically based pharmacokinetic (PBPK) models is the prediction of the rate and extent of absorption of orally dosed drugs for which knowledge of effective passive intestinal permeability (P_eff ) is essential. Single-pass intestinal perfusion (SPIP) studies are used to establish effective permeability in vivo but are difficult to perform in rodents, while mechanistic models to predict drug P_eff in rat and mouse have not been published. This work evaluates the predictive performance of the 'MechPeff' model to predict P_eff in the rodent intestine based upon knowledge of regional gut physiology and drug-specific physicochemical parameters. The 'MechPeff' model, built-in to the Simcyp Rat and Mouse Simulators, predicts transcellular, paracellular and mucus layer permeabilities and combines these to give the overall P_eff . The jejunal and/or ileal P_eff was predicted for 12 (4) acidic, 13 (12) basic, 10 (8) neutral and 2 (0) ampholytic drugs in the rat (mouse), spanning a wide range of MW and logP_o:w , and compared with experimental P_eff obtained using SPIP. A key input is the intrinsic transcellular permeability (P_trans,0 ) which can be derived from modelling of appropriate in vitro permeability experiments or predicted from physicochemical properties. The P_eff predictions were reasonably good when experimentally derived P_trans,0 was used; from 42 P_eff,rat values, 24 (57%) were within 3-fold, and of 19 P_eff,mouse values, 12 (63%) were within 3-fold, of observed P_eff . Considering the lack of alternative models to predict P_eff in preclinical species, and the minimal drug-specific inputs required, this model provides a valuable tool within drug discovery and development programmes. Copyright © 2017 John Wiley & Sons, Ltd.

A Food Effect Study of an Oral Thrombin Inhibitor and Prodrug Approach To Mitigate It

LB30870, a new direct thrombin inhibitor, showed 80% reduction in oral bioavailability in fed state. The present study aims to propose trypsin binding as a mechanism for such negative food effect and demonstrate a prodrug approach to mitigate food effect. Effect of food composition on fed state oral bioavailability of LB30870 was studied in dogs. Various prodrugs were synthesized, and their solubility, permeability, and trypsin binding affinity were measured. LB30870 and prodrugs were subject to cocrystallization with trypsin, and the X-ray structures of cocrystals were determined. Food effect was studied in dogs for selected prodrugs. Protein or lipid meal appeared to affect oral bioavailability of LB30870 in dogs more than carbohydrate meal. Blocking both carboxyl and amidine groups of LB30870 resulted in trypsin Ki values orders of magnitude higher than that of LB30870. Prodrugs belonged to either Biopharmaceutical Classification System I, II, or III. X-ray crystallography revealed that prodrugs did not bind to trypsin, but instead their hydrolysis product at the amidine blocking group formed cocrystal with trypsin. A prodrug with significantly less food effect than LB30870 was identified. Binding of prodrugs to food components such as dietary fiber appeared to counteract the positive effect brought with the prodrug approach. Further formulation research is warranted to enhance the oral bioavailability of prodrugs. In conclusion, this study is the first to demonstrate that the negative food effect of LB30870 can be attributed to trypsin binding. Trypsin binding study is proposed as a screening tool during lead optimization to minimize food effect.

Pharmacokinetics, pharmacodynamics and food effect of LB30870, a novel direct thrombin inhibitor, after single oral doses in healthy men

1. The safety, tolerability, pharmacokinetics, pharmacodynamics, and food effect of LB30870, a new selective thrombin inhibitor, were studied in 16 healthy men. 2. A double-blind, placebo-controlled single ascending dose study was done at oral doses of 5, 15, 30, 60, 120, and 240 mg under fasting conditions. An open, randomized, balanced cross-over food effect study was done at 60 mg dose. Plasma and urinary concentrations were measured up to 48 h post-dose. Coagulation and thrombin activity markers were measured at selected time points. 3. Cmax of LB30870 was at 1.3-3.0 h post-dose with a mean apparent terminal half-life (t1/2) of 2.8-4.1 h. AUC after doses above 15 mg appeared greater than dose-proportional. In fed state, AUC showed 80% reduction relative to fasting condition. 4. At doses 60 and 120 mg, peak activated partial thromboplastin time (aPTT) increased by 1.5- and 2-fold, respectively, from baseline. The aPTT and international normalized ratio (INR) were concentration-dependent, with less within-individual variability than ecarin clotting time (ECT), prothrombin time (PT), or thrombin time (TT). 5. Single oral doses of LB30870 up to 240 mg were well tolerated. The food effect must be overcome if LB30870 is to be used as an oral anti-coagulant.

A novel ritonavir paediatric powder formulation is bioequivalent to ritonavir oral solution with a similar food effect

BACKGROUND

A novel ritonavir oral powder formulation has been developed to eliminate the alcohol and propylene glycol contents in the current ritonavir oral solution for paediatric use. Two clinical studies were conducted to assess the bioequivalence of the powder formulation to the marketed oral solution and to evaluate the effect of food and vehicles on bioavailability.

METHODS

Study 1 was a randomized, partial-crossover, 4-period study in 48 subjects. Regimens included: oral solution under moderate-fat conditions, powder formulation in water under fasting, moderate-fat or high-fat conditions, and powder formulation in chocolate milk or pudding under moderate-fat conditions. Study 2 was a randomized, crossover, 4-period study in 24 subjects. Subjects were randomized to a sequence of the oral solution and powder formulation in water, infant formula and apple sauce, all under moderate-fat conditions. Bioavailability comparisons were assessed by the 90% CIs for the geometric least-squares mean ratios.

RESULTS

Ritonavir powder formulation in water was found to be bioequivalent to the marketed oral solution. Ritonavir powder formulation administered in chocolate milk, pudding, infant formula or apple sauce was bioequivalent to the powder formulation administered in water. Compared with fasting conditions, moderate-fat and high-fat meals were associated with approximately 25-40% and 35-50% reduction in ritonavir concentrations, respectively.

CONCLUSIONS

The novel ritonavir powder formulation is bioequivalent to marketed ritonavir oral solution under moderate-fat conditions with a similar effect of meals. None of the vehicles tested negatively affected the bioavailability, which suggests the potential for use of a broad range of vehicles for dose preparation.

Effects of green tea extract and (-)-epigallocatechin-3-gallate on pharmacokinetics of nadolol in rats

Green tea catechins have been shown to affect the activities of drug transporters in vitro, including P-glycoprotein and organic anion transporting polypeptides. However, it remains unclear whether catechins influence the in vivo disposition of substrate drugs for these transporters. In the present study, we investigated effects of green tea extract (GTE) and (-)-epigallocatechin-3-gallate (EGCG) on pharmacokinetics of a non-selective hydrophilic β-blocker nadolol, which is reported to be a substrate for several drug transporters and is not metabolized by cytochrome P450 enzymes. Male Sprague-Dawley rats received GTE (400 mg/kg), EGCG (150 mg/kg) or saline (control) by oral gavage, 30 min before a single intragastric administration of 10 mg/kg nadolol. Plasma and urinary concentrations of nadolol were determined using high performance liquid chromatography. Pharmacokinetic parameters were estimated by a noncompartmental analysis. Pretreatment with GTE resulted in marked reductions in the maximum concentration (Cmax) and area under the time-plasma concentration curve (AUC) of nadolol by 85% and 74%, respectively, as compared with control. In addition, EGCG alone significantly reduced Cmax and AUC of nadolol. Amounts of nadolol excreted into the urine were decreased by pretreatments with GTE and EGCG, while the terminal half-life of nadolol was not different among groups. These results suggest that the coadministration with green tea catechins, particularly EGCG, causes a significant alteration in the pharmacokinetics of nadolol, possibly through the inhibition of its intestinal absorption mediated by uptake transporters.

Quantitative prediction of formulation-specific food effects and their population variability from in vitro data with the physiologically-based ADAM model: a case study using the BCS/BDDCS Class II drug nifedipine

Quantitative prediction of food effects (FE) upon drug pharmacokinetics, including population variability, in advance of human trials may help with trial design by optimising the number of subjects and sampling times when a clinical study is warranted or by negating the need for conduct of clinical studies. Classification and rule-based systems such as the BCS and BDDCS and statistical QSARs are widely used to anticipate the nature of FE in early drug development. However, their qualitative rather than quantitative nature makes them less appropriate for assessing the magnitude of FE. Moreover, these approaches are based upon drug properties alone and are not appropriate for estimating potential formulation-specific FE on modified or controlled release products. In contrast, physiologically-based mechanistic models can consider the scope and interplay of a range of physiological changes after food intake and, in combination with appropriate in vitro drug- and formulation-specific data, can make quantitative predictions of formulation-specific FE including the inter-individual variability of such effects. Herein the Advanced Dissolution, Absorption and Metabolism (ADAM) model is applied to the prediction of formulation-specific FE for BCS/BDDCS Class II drug and CYP3A4 substrate nifedipine using as far as possible only in vitro data. Predicted plasma concentration profiles of all three studied formulations under fasted and fed states are within 2-fold of clinically observed profiles. The % prediction error (%PE) in fed-to-fasted ratio of Cmax and AUC were less than 5% for all formulations except for the Cmax of Nifedicron (%PE=-29.6%). This successful case study should help to improve confidence in the use of mechanistic physiologically-based models coupled with in vitro data for the anticipation of FE in advance of in vivo studies. However, it is acknowledged that further studies with drugs/formulations exhibiting a wide range of properties are required to further validate this methodology.

Effect of Contrasted Sodium Diets on the Pharmacokinetics and Pharmacodynamic Effects of Renin–Angiotensin System Blockers

Dietary sodium, the main determinant of the pharmacodynamic response to renin–angiotensin system blockade, influences the pharmacokinetics of various cardiovascular drugs. We compared the effect of contrasted sodium diets on the pharmacokinetics of single oral doses of 8 mg candesartan cilexetil, 160 mg valsartan, 10 mg ramipril, and 50 mg atenolol administered to 64 (16 per group) normotensive male subjects randomly assigned to sodium depletion (SD) or sodium repletion (SR) in a crossover study. Pharmacodynamic response was assessed as the increase in plasma renin concentration for renin–angiotensin system blockers and electrocardiographic changes in PR interval duration for atenolol. The area under the curve (AUC) for plasma candesartan and atenolol concentrations was significantly lower for SR than for SD (respective ratios of AUC 0–∞ : 0.74; [90% CI, 0.66–0.82] and 0.69 [90% CI, 0.54–0.88], respectively), indicating a lack of bioequivalence between SR and SD. SR did not affect the pharmacokinetics of valsartan or ramipril. The increase in plasma renin concentration with the 3 renin–angiotensin system blockers was 10 times lower during the SR than the SD period. In the multiple regression analysis, the AUC 0–24 of plasma drug concentration explained <1% and 21% of the variance of the AUC 0–24 of delta plasma renin concentration for candesartan ( P =0.8882/ P =0.0368) during the SR and SD periods, respectively. The atenolol-induced lengthening of PR interval was fully reversed by SR. Thus, sodium balance modulates the pharmacokinetics of candesartan cilexetil and atenolol, with measurable effects on the selected pharmacodynamic end points.

NSAIDs: take with food or after fasting?

OBJECTIVE

Published and regulatory advice is to take NSAIDs with fluids and/or food irrespective whether NSAIDs are taken over the counter or long-term. The basis for this recommendation is not clear and we sought to establish the reasons for it through a search of published literature and personal files.

RESULTS

Results from experimental animals show that fasting increases the gastric side effects of NSAIDs while food increases small bowel damage, but this has not been tested in humans. The possible effects of food in modifying the gastric damage caused by NSAIDs are complex, as food quantity and composition modify the responses substantially. Food usually delays peak levels of NSAIDs (and hence onset of action) without affecting total bioavailability. This may not be important when a steady state is achieved, but rapid onset of action is highly relevant for over-the-counter use of NSAIDs. The safety of over-the-counter use of ibuprofen and naproxen appears to be excellent and comparable with paracetamol.

CONCLUSION

The rapid onset of action of NSAIDs is most important during over-the-counter use, in which case it may be more appropriate to take the drugs on a fasting stomach.

The effect of food on the oral bioavailability of drugs: a review of current developments and pharmaceutical technologies for pharmacokinetic control

Here we review the mediation of the food effects on drugs by pharmaceutical technologies. The pharmacokinetics of drugs are affected by the interaction of drugs with food, which changes drug physicochemical and physiological properties (food effects). Several pharmaceutical technologies may be used to control food effects. Drugs exhibit different patterns of solubilization depending on release formulations. Formulations such as nanoparticle, solid dispersion and cyclodextrin systems, may control the solubility and release of insoluble drugs. Other controlled-release technologies, such as osmotic-controlled release or colon-specific delivery systems may also control food effects. As the structure of drug candidates becomes more complex, different methods of investigation, such as in vitro and in vivo correlation and in silico simulation will be required to predict drug characteristics and food effects.

Predictive models for drugs exhibiting negative food effects based on their biopharmaceutical characteristics

CONTEXT

A drug is defined to exhibit food effects if its pharmacokinetic parameter, area under the curve (AUC₀₋∞) is different when co-administered with food in comparison with its administration on a fasted stomach. Food effects of drugs administered in immediate release dosage forms were classified as positive, negative, and no food effects.

OBJECTIVE

In this study, predictive models for negative food effects of drugs that are stable in the gastrointestinal tract and do not complex with Ca²⁺ are reported.

METHODS

An empirical model was developed using five drugs exhibiting negative food effects and seven drugs exhibiting no food effects by multiple regression analysis, based on biopharmaceutical properties generated from in vitro experiments. An oral absorption model was adopted for simulating negative food effects of model compounds using in situ rat intestinal permeability.

RESULTS

Analysis of selected model drugs indicated that percent food effects correlated to their dissociation constant, K (K(a) or K(b)) and Caco-2 permeabilities. The obtained predictive equation was: Food effect (%)=(2.60 x 10⁵·P(app))--(2.91 x 10⁵·K)--8.50. Applying the oral absorption model, the predicted food effects matched the trends of published negative food effects when the two experimental pH conditions of fed and fasted state intestinal environment were used.

CONCLUSION

A predictive model for negative food effects based on the correlation of food effects with dissociation constant and Caco-2 permeability was established and simulations of food effects using rat intestinal permeability supported the drugs? published negative food effects. Thus, an empirical and a mechanistic model as potential tools for predicting negative food effects are reported.