Impact of obesity and roux‐en‐Y gastric bypass on the pharmacokinetics of (R)‐ and (S)‐omeprazole and intragastric pH

Abstract

This study employed physiologically‐based pharmacokinetic–pharmacodynamics (PBPK/PD) modeling to predict the effect of obesity and gastric bypass surgery on the pharmacokinetics and intragastric pH following omeprazole treatment. The simulated plasma concentrations closely matched the observed data from non‐obese, morbidly obese, and post‐gastric bypass populations. Obesity significantly reduces CYP3A4 and CYP2C19 activities, as reflected by the metabolic ratio [omeprazole sulphone]/[omeprazole] and [5‐hydroxy‐omeprazole]/[omeprazole]. The morbidly obese model accounted for the down‐regulation of CYP2C19 and CYP3A4 to recapitulate the observed data. Sensitivity analysis showed that intestinal CYP3A4, gastric pH, small intestine bypass, and the delay in bile release do not have a major influence on omeprazole exposure. Hepatic CYP3A4 had a significant impact on the AUC of (S)‐omeprazole, while hepatic CYP2C19 affected both (R)‐ and (S)‐omeprazole AUC. After gastric bypass surgery, the activity of CYP3A4 and CYP2C19 is restored. The PBPK model was linked to a mechanism‐based PD model to assess the effect of omeprazole on intragastric pH. Following 40 mg omeprazole, the mean intragastric pH was 4.3, 4.6, and 6.6 in non‐obese, obese, and post‐gastric bypass populations, and the daily time with pH >4 was 14.7, 16.4, and 24 h. Our PBPK/PD approach provides a comprehensive understating of the impact of obesity and weight loss on CYP3A4 and CYP2C19 activity and omeprazole pharmacokinetics. Given that simulated intragastric pH is relatively high in post‐RYGB patients, irrespective of the dose of omeprazole, additional clinical outcomes are imperative to assess the effect of proton pump inhibitor in preventing marginal ulcers in this population.

---

STUDY HIGHLIGHTS

• WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC?

Omeprazole is widely employed for the prevention and treatment of marginal ulcers following gastric bypass surgery. The efficacy of acid secretion attenuation is typically evaluated by assessing the average intragastric pH over 24 h and the duration of gastric pH above 4.

• WHAT QUESTION DID THIS STUDY ADDRESS?

Our study aimed to characterize the effect of the anatomical and physiological changes in the gastrointestinal tract post-bariatric surgery on the pharmacokinetics of omeprazole and the resulting effect on intragastric pH and time with pH above 4. The need for dose adjustments of proton pump inhibitors in this specific population remains uncertain.

• WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE?

The intragastric pH was slightly higher in obese patients compared with non-obese, mainly due to a dilution food effect, since the exposure to omeprazole was unchanged in this population. In post-gastric bypass patients, the daily time with pH >4 was consistently 24 h, regardless of the omeprazole dose (10–80 mg).

• HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE?

Low omeprazole doses (10–20 mg) may be considered for patients following a gastric bypass to maintain a high intragastric pH. The occurrence of marginal ulcers in the gastrojejunal anastomosis seems to be attributed to the increased vulnerability of the jejunum's mucosa to gastric acid secretion, thus requiring additional clinical outcomes to assess the effect of PPIs in the prevention of marginal ulcers.

---

INTRODUCTION

Omeprazole is widely employed to prevent and treat marginal ulcers following gastric bypass surgery.^1–3 By inhibiting the gastric parietal cell proton pump H⁺/K⁺-ATPase, omeprazole reduces basal and stimulates gastric acid secretion, increasing the intragastric pH.^4,5 The efficacy of acid secretion attenuation can be evaluated by assessing the average intragastric pH over 24 h and the duration of gastric pH above 4, which are clinically relevant outcomes of therapeutic success in patients with or at risk of marginal ulcers.^6,7 While generally well-tolerated, the use of omeprazole has been associated with the potential risk of bone damage,⁸ hypomagnesemia,⁹ and atrophic gastritis in Helicobacter pylori-infected patients.¹⁰ Recognizing the anatomical and physiological changes in the gastrointestinal tract post-bariatric surgery, such as modifications in the stomach volume and the number of parietal cells, alongside changes in pharmacokinetics (PK) induced by obesity and subsequent weight loss, dose adjustments of proton pump inhibitors (PPIs) in this specific population remains uncertain.

Obesity is associated with complex physiological alterations likely to affect PK.^11,12 The effect of obesity on drug distribution is well characterized and can vary depending on the drug's lipophilicity. Nevertheless, the effect of obesity on the expression and activity of drug-metabolizing enzymes and transporters remains unclear, with conflicting research findings.^12–15 Chronic inflammation in adipose tissue triggered by obesity may alter the expression of drug-metabolizing enzymes in the long term.^16,17 However, the extent of these changes compared with non-obese is largely unknown.^12–14 For instance, obese patients have shown reduced clearance for several CYP3A4 substrates and increased clearance for CYP1A2, but conflicting and unclear results were reported for CYP2E1, CYP2C9, and CYP2C19.^13,15,18

Bariatric surgeries, such as Roux-en-Y gastric bypass, lead to sustained weight loss and a significant improvement or remission of obesity-associated comorbidities. In the US, there has been a trend of increasing the number of bariatric surgeries in the last decade.¹⁹ The Roux-en-Y gastric bypass involves reducing the stomach to a small pouch, bypassing the duodenum and the initial part of the jejunum, and connecting the distal jejunum to the gastric pouch. These alterations affect stomach pH, gastric emptying time, and absorptive surface area.^20,21 However, it is not well understood whether the inflammation status and enzyme expression return completely to levels observed in non-obese individuals.

We aimed to simulate the effect of obesity and weight loss progression after gastric bypass on the disposition of (R)- and (S)-omeprazole and on the intragastric pH using a physiologically-based pharmacokinetic/pharmacodynamic (PBPK/PD) modeling approach. The PBPK model was developed for each omeprazole enantiomer to increase the model's versatility in administering either omeprazole or esomeprazole. To verify our PBPK model, we utilized observed data from crossover studies designed to investigate the effect of bariatric surgery on omeprazole PK.^22,23 As a sensitive substrate for CYP3A4 and CYP2C19 activity, omeprazole is used in many phenotypic cocktails to explore the in vivo activity of CYP enzymes.^24,25 Therefore, our study provides valuable insights into how weight loss affects CYP3A4 and CYP2C19 activity.

METHODS

Clinical data

We assessed individual data from 39 subjects enrolled in a clinical study designed to investigate the effect of gastric bypass on the kinetic disposition of omeprazole.²² The study was conducted following approval from the research ethics committee METC LDD (NL42835.098.13) and approval as exempt by the University of Florida (IRB202301498). The study was conducted in patients from the Haaglanden Medisch Centrum and the Dutch Obesity Clinic, The Hague, the Netherlands. In summary, the clinical study involved subjects with a mean age of 48.7 (±10) years, 79% of females, with morbid obesity who underwent Roux-en-Y gastric bypass surgery. Exclusion criteria included previous bariatric surgery, malabsorption disorder, allergy to any of the study medications, increased risk of bleeding, or use of other medication known to interact with the study medication. All enrolled subjects were investigated in two different periods before and after surgery, and 34 subjects completed both study phases. The evaluations happened at least 2 weeks before the scheduled surgery and 8 weeks after surgery (58 days ± 18 days). All subjects received omeprazole 20 mg b.i.d. for 7 days.²² The main gene variants of CYP2C19 were genotyped using qPCR and translated to a CYP2C19-predicted phenotype according to CPIC's guidelines for PPIs.²⁶ Plasma concentrations were determined by LC-MS/MS, and the methods were linear in the ranges of 0.026–2.64, 0.011–2.117, and 0.010–0.946 μg/mL, with the lower limits of quantification of 0.016, 0.008, and 0.009 μg/mL for omeprazole, 5-hydroxy-omeprazole, and omeprazole sulphone, respectively. The metabolic ratios [5-hydroxy-omeprazole]/[omeprazole] and [omeprazole sulphone]/[omeprazole] were used to assess the in vivo activity of CYP2C19 and CYP3A4, respectively, and were calculated using plasma concentrations 4 h after drug administration.

Software and modeling strategy

We utilized Simcyp™ PBPK Simulator V21 Release 1 (Certara, Sheffield, UK) for joint model development, verification, and simulation. Leveraging previous PBPK modeling efforts,²⁷ drug-specific parameters were revised to create PBPK models for both (R)-omeprazole and (S)-omeprazole following a systematic, step-wise approach (Table S1). Initially, separate models were rigorously verified for each enantiomer using clinical data obtained from the administration of (S)-omeprazole and (R)-omeprazole separately, at 20 and 40 mg q.d., with PK assessments in the first and fifth days.^28,29 Notably, due to the autoinhibition of omeprazole on CYP2C19 enzyme,³⁰ with potential drug–drug interaction between the (R)- and (S)-enantiomers, our model underwent further refinement to account for (S)-omeprazole as a substrate and (R)-omeprazole as an inhibitor within the Simcyp simulator framework. The combined model, encompassing both (S)-omeprazole and (R)-omeprazole simultaneously, was then verified using clinical data that included the assessment of plasma concentrations of all chemical species,²⁹ following the administration of 20 and 40 mg racemic omeprazole q.d., with PK evaluations conducted after the first and fifth doses. Model verification also relied on a diverse set of subjects, including non-obese subjects genotyped or not for CYP2C19, with Caucasian and Japanese origins (Table S2).^28,29,31,32 The model was further enhanced to predict the effect of obesity and bariatric surgery on the PK of omeprazole enantiomers^22,23 (Figure S1) and linked to a mechanism-based pharmacodynamic model,³³ providing valuable guidance for individualized dosing strategies in these distinct populations.

Development and verification of a PBPK model in normal BMI subjects

We have refined a prior model utilized for describing the enantioselective PK of omeprazole,²⁷ enhancing it to account for a more mechanistic absorption process. The physicochemical properties from the omeprazole compound file available in the simulator library were used. The blood-to-plasma ratio of (S)-omeprazole and (R)-omeprazole was set at 0.59.³⁴ The fraction unbound was initially set as 0.05³⁰ and further optimized following a sensitivity analysis to fit the plasma concentration profile. To mechanistically predict the fraction absorbed for (R)- and (S)-omeprazole, we employed the advanced model for absorption, dissolution, and metabolism (ADAM). The experimental apparent permeability in Caco-2 cells (pH 6.5:7.4) was determined as 67.4 × 10⁻⁶ cm/s for omeprazole, using verapamil as a highly permeable reference compound.³⁵ The effective jejunal intestinal permeability was then predicted as 12 × 10⁻⁴ cm/s for (R)- and (S)-omeprazole. A minimal PBPK model was implemented to describe the systemic exposure in plasma. The Rodgers and Rowland method with ion permeability (Method 3) was employed to predict the steady-state volume of distribution (V_ss) for omeprazole, accounting for the permeability of the ionized fraction. The Kp scalar was fitted to the in vivo V_ss.³¹

A retrograde approach was employed to estimate the enantioselective^36,37 in vitro metabolism parameters based on clinical-rich PK data obtained from healthy subjects genotyped as CYP2C19 normal metabolizers ante treated with intravenous omeprazole³¹ or (R)-omeprazole oral solution.²⁹ The intrinsic clearance obtained from this approach was combined with in vitro enantioselective K_m values³⁸ to calculate their respective V_max. (R)- and (S)-omeprazole are mainly metabolized by CYP2C19 to their respective inactive metabolites, (R)- and (S)-5-hydroxy-omeprazole, and by CYP3A4 to the non-chiral metabolite omeprazole sulphone.^36–38 The fractional metabolism of CYP2C19 (fm_CYP2C19) for (R)- and (S)-omeprazole was 95.8% and 67.9%, while the fm_CYP3A4 was 4.0% and 32.0% for (R)- and (S)-omeprazole, respectively. It is noteworthy that the plasma levels of (S)-5-hydroxy-omeprazole are significantly lower compared with (R)-5-hydroxy-omeprazole,³¹ underscoring the need for models capable of distinguishing between omeprazole enantiomers. In addition, omeprazole's renal clearance, which was determined at 0.034 L/h in non-obese individuals, makes a relatively minor contribution to the overall elimination of the drug.³⁹

To account for the nonlinear PK of omeprazole,^29,32 a mechanism-based inhibition was implemented in the model, based on prior modeling efforts.²⁷ Omeprazole and omeprazole sulphone are inhibitors of CYP2C19 in a time-dependent fashion. Omeprazole also inhibits CYP3A4,^27,30 and the same inhibition parameters were applied for both (R)- and (S)-omeprazole.

Simulated PK parameters and profiles were compared with observed PK data in non-obese subjects (Table S3). Ten virtual trials were carried out for each study, replicating the demographics in the respective clinical trial. The ratios between simulated and observed PK parameters (AUC, C_max, and T_max) were calculated. Model performance was deemed adequate when simulated PK parameters were within a twofold error range (0.5 ≤ ratio ≤2). Adult PK profiles after intravenous³¹ and oral²⁹ administration were used for model development as well as mass balance information.^36,37,40

Development and verification of a PBPK model in morbidly obese subjects

Systems-parameters for the obese and morbidly obese populations are available in the Simcyp simulator and were used with minor modifications. To account for the delayed transit time observed in the morbidly obese,^22,23 the mean gastric residence time was set to 0.94 h in this population after a local sensitivity analysis. The down-regulation of CYP3A4 activity in obese subjects, as demonstrated in clinical studies using CYP3A4-sensitive substrates dextromethorphan⁴¹ and 4β-hydroxycholesterol,⁴² is currently implemented in the morbidly obese model in Simcyp by the reduced hepatic CYP3A4 abundance, which is set at 137 pmol/mg of protein in the liver in non-obese, and 65 pmol/mg of protein in the liver in morbidly obese virtual subjects, resulting in total mean abundances of 7.12 and 5.3 μmol to express differences in enzyme activity, respectively. Since CYP2C19 activity is also reduced in obese patients and increases after weight loss, as demonstrated for [5-hydroxy-omeprazole]/[omeprazole],⁴³ the morbidly obese population model was adjusted to account for the reduced hepatic abundance (Table S4). The CYP2C19 abundance was set to 4.4 and 2.5 pmol/mg of protein in the liver of non-obese and morbidly obese through parameter estimation, resulting in total mean abundances of 0.25 and 0.2 μmol. Simulations were carried out for the administration of 20 mg of omeprazole b.i.d. for 7 days and mimicked the trial design in terms of age, body mass index, and % of females.²² For all simulations, drug-specific parameters of omeprazole enantiomers remained unchanged.

Development and verification of a PBPK model in post-gastric bypass surgery subjects

The healthy volunteers population served as a base model and adjustments were made to the system-model parameters to account for the modifications in the GI tract that resulted from gastric bypass surgery.⁴⁴ In particular, we implemented a bypass in the duodenum-jejunum-I region by reducing the transit time of the drug in these compartments to near-zero value. Additionally, the gastric emptying time was reduced to 7 min, and the small intestine transit time was reduced to 3.5 h.⁴⁴ The volume of concomitant fluid intake administered with the dose was set to 30 mL. Other changes included the reduced initial stomach fluid volume to 9.9 mL, the delay in the bile release to the jejunum-II compartment, and the reduced gut abundance of CYP3A4/5 to values close to zero in the duodenum and a 38% reduction in the jejunum-I. Please refer to the Supplementary material (Table S4) for a detailed list of all changes in system parameters for this model.

We conducted a local sensitivity analysis to assess the impact of gastric bypass on the exposure to omeprazole enantiomers. In this analysis, we tested independent variations in factors, such as body weight, bile release delay, gastric pH, gastric emptying time, small intestine transit time, small intestine length, CYP3A4 abundance in the small intestine, and CYP3A4 and CYP2C19 abundance in the liver, while keeping all other parameters constant. These physiological parameters were varied within a range representing the variations resulting from the surgery. Line plots were used to depict the effect of each input variable in AUC_0–∞, C_max, and T_max, and the model was considered sensitive to the effect when simulated PK parameters were not within twofold the typical value (0.5 ≤ ratio ≤2).

The gastric bypass model was verified using clinical data conducted to evaluate omeprazole PK in patients submitted to gastric bypass (n = 34),²² receiving 20 mg of racemic omeprazole b.i.d. For model verification, only patients with a translated CYP2C19 normal metabolizer phenotype were included (n = 10). Since the clinical data available did not distinguish between omeprazole enantiomers, total drug concentration was calculated by combining the concentrations of the enantiomeric mixture.

Application of the PBPK model to predict intragastric pH after a gastric bypass

Simulated plasma concentration profiles from our PBPK model were linked to a mechanism-based pharmacodynamic (PD) model to predict the intragastric H⁺ concentration³³ following repeated omeprazole doses using Pumas-AI 2.0 (Pumas-AI, Baltimore, USA). The primary goal was to explore if the impact of anatomical and physiological changes in the GI tract combined with changes in omeprazole PK would result in an altered response in patients after a gastric bypass. The PD model was previously developed and verified in non-obese subjects with accurate predictions of gastric pH over 24 h, accounting for the irreversible inhibition of H⁺/K⁺ATPase, the food dilution effect in the intragastric concentration of H⁺, and the circadian rhythm resulting in increased gastric pH at night.³³ In summary, the semi-mechanistic model was adjusted to account for an increased food intake in morbidly obese, the effect of gastric bypass surgery on the anatomy and physiology of the stomach, and changes in the food intake after the surgery (increased frequency and reduced volume). Please refer to the Supplementary material for a detailed description of the semi-mechanistic PD model (Table S5 and Semi-mechanistic model to predict intragastric pH section). The average 24-h intragastric pH and the time above pH 4 were computed for each population.

Statistics and non-compartmental analysis

Statistic tests were performed in the software R version 4.3.0 (R Core Team, ). Non-compartmental analysis was performed with the R package NonCompart, version 0.6.0.

RESULTS

In this study, we have enhanced our comprehension of the kinetic disposition and response to (R)- and (S)-omeprazole in patients undergoing Roux-en-Y gastric bypass surgery, achieved by integrating a mechanistic absorption model and enantioselective enzyme kinetics data into preexisting compound files. Our developed model incorporates the distinct metabolic pathways of each enantiomer, particularly the enantioselective CYP2C19 metabolism (Table S1).

Using a retrograde approach guided by clinical data,^28,29,31 we successfully integrated in vitro data^36–38 to accurately replicate the differential contributions of CYP2C19 and CYP3A4 with good predictions for both omeprazole enantiomers (Figures 1 and 2; Figure S2). By considering the autoinhibition,³⁰ our model effectively captures the nonlinear exposure observed in the plasma concentration–time profiles of healthy subjects on the first and fifth days after administration of 20 mg q.d. or 40 mg q.d. racemic omeprazole (Figure 2), also shown for a broader range of doses (Figure S3).

[IMAGE OMITTED. SEE PDF]

[IMAGE OMITTED. SEE PDF]

The clinical findings indicate the activity of CYP3A4 and CYP2C19 increases in patients undergoing a gastric bypass, as evidenced by the increasing metabolic ratios [5-hydroxy-omeprazole]/[omeprazole] and [omeprazole sulphone]/[omeprazole] with the weight loss following gastric bypass surgery (Figure 3; Figure S4). The increase in CYP2C19 activity following the surgery was observed for different phenotypes, but only two patients were poor metabolizers (*2/*2), and one was an ultrarapid metabolizer (*17/*17), limiting our assessment for these subgroups (Figure 3). While the Simcyp simulator already accounted for the reduced abundance of CYP3A4 in the obese population (Table S4), adjustments were made solely to the CYP2C19 hepatic abundance to better align with the clinical observations. The model was verified using individual plasma concentration data and PK parameters for omeprazole in morbidly obese patients phenotyped as CYP2C19 homozygous normal metabolizers (*1/*1, n = 10). In the morbidly obese model, when accounting for the reduced hepatic abundance for CYP2C19 and CYP3A4, the exposure to total omeprazole was effectively captured by the PBPK model, as illustrated in Figures 4 and 5 (Table S3).

[IMAGE OMITTED. SEE PDF]

[IMAGE OMITTED. SEE PDF]

[IMAGE OMITTED. SEE PDF]

Following the implementation of significant gastrointestinal anatomical changes in a gastric bypass population, a parameter sensitivity analysis was carried out to show how sensitive omeprazole exposure is to system parameters. The analysis revealed that omeprazole exposure is not sensitive to increased gastric pH, bypass in the upper small intestine, reduced gut CYP3A4, and bile delay. Contrastingly, omeprazole exposure exhibited high sensitivity to variations in hepatic abundance/activity of CYP2C19 and CYP3A4. Specifically, (R)-omeprazole displayed high sensitivity to changes in CYP2C19, while (S)-omeprazole demonstrated sensitivity to both CYP2C19 and CYP3A4 (Figure S5). This observation underscores omeprazole's high sensitivity to these hepatic enzymes, reinforcing its role as a valuable substrate for evaluating enzyme activity.

Simulations of 20 mg of (R)-, (S)-, or racemic omeprazole q.d. for 7 days in non-obese, obese, and post-gastric bypass populations demonstrated that the AUC of (S)-omeprazole is decreased in the post-gastric bypass population compared with morbidly obese (Figure S6). By linking our PK predictions to a previously verified PD semi-mechanistic model, we were able to assess the effect of omeprazole or esomeprazole on intragastric pH in non-obese, obese, and post-gastric bypass patients (Figure 6). Our predictions showed that patients undergoing bariatric surgery indicated a mean predicted intragastric pH of 6.6, which is significantly elevated compared with non-obese (4.3) and obese (4.6) following the administration of 40 mg racemic omeprazole q.d. for 7 days (Figure 7). Overall, after gastric bypass surgery, patients will exhibit a daily time with pH >4 close to 24 h irrespective of the omeprazole or esomeprazole dosing.

[IMAGE OMITTED. SEE PDF]

[IMAGE OMITTED. SEE PDF]

DISCUSSION

This study advances our understanding of the exposure and response to the PPIs omeprazole and esomeprazole in patients undergoing gastric bypass. By establishing individual models for (R)- and (S)-omeprazole, the model's versatility extends beyond the study of racemic omeprazole formulations by making it equally applicable to formulations containing the single enantiomer esomeprazole. By linking the predicted PK profiles with a previously verified PD model,³³ we could simulate the intragastric pH following omeprazole or esomeprazole administration in patients submitted to gastric bypass surgery.

Omeprazole is a biomarker for assessing the in vivo activity of CYP3A4 and CYP2C19. Leveraging individual plasma concentration data for omeprazole and its major metabolites, 5-hydroxy-omeprazole and omeprazole sulphone²² enabled us to investigate the impact of weight loss on in vivo activity of CYP3A4 and CYP2C19. Obese subjects showed decreased activity of CYP3A4 and CYP2C19 as compared with subjects after gastric bypass surgery, as reflected by the decreased metabolic ratios [5-hydroxy-omeprazole]/[omeprazole] and [omeprazole sulphone]/[omeprazole], irrespective of the CYP2C19 phenotype (Figure 3). A post-translational regulation of CYP3A4 activity has been suggested in obese patients as a consequence of nonalcoholic steatosis.⁴⁵ Longitudinal PK studies in obese patients before and after surgery^22,23 serve as a well-placed study design to evaluate the effects of obesity on the activity of drug-metabolizing enzymes due to their crossover nature as compared with a parallel study design, where obese subjects are compared with non-obese. Kvitne et al.⁴³ examined the short and long-term effects of body weight and gastric bypass on CYP2C19 activity. Remarkably, significant changes in CYP2C19 were observed at weeks 3, 9, and year 2 after surgery, with maximum changes at week 9. In the study conducted by Mitrov-Winkelmolen et al.,²² patients were investigated at least 6 weeks (mean 58 ± 18 days) after surgery with an average reduction in BMI from 44.9 to 39.8 kg/m². Thus, we can infer that the Roux-en-Y gastric bypass (RYGB) population is representative of the post-surgery condition, assuming that patients following a gastric bypass will experience substantial weight loss.⁴³

The reduced abundance of CYP3A4 in obese subjects was already implemented in the Simcyp simulator, aligning with the clinical observations for enzyme activity (Figure 3) and protein expression.⁴⁶ As CYP2C19 is a major enzyme in omeprazole metabolism,^36,37 we implemented a reduced CYP2C19 abundance in the obese population as suggested by the reduced metabolic ratio [5-hydroxy-omeprazole]/[omeprazole] in obese (Figure 3). This alteration is supported by previous clinical data that showed that the clearance of voriconazole, a sensitive substrate for CYP2C19, was lower in obese versus non-obese pediatric subjects.⁴⁷ The apparent clearance of pantoprazole, which is predominantly metabolized by CYP2C19, was reduced by 50% in obese children compared with non-obese.⁴⁸ The activity of CYP2C19 was lower in obese patients, and weight loss due to calorie restriction or gastric bypass resulted in increased activity of CYP2C19.⁴³ A clinical study in 101 subjects receiving a single dose of omeprazole demonstrated that obese patients show reduced clearance of omeprazole compared with normal weight patients.⁴⁹ It is important to note that for further application of the current model to simulate PK/PD for patients with varying BMIs following gastric bypass surgery, selecting an appropriate non-obese, obese, or morbidly obese population-base model for the RYGB population is essential, ensuring the incorporation of obesity effects on CYP abundance as appropriate.

The decreased activity of CYP3A4 and CYP2C19 in obesity with further restoration after weight loss resulted in decreased AUC values only for (S)-omeprazole in post-bariatric populations compared with morbidly obese (Figure S6). In general, despite the critical obesity-induced changes in CYP3A4 and 2C19 activity, simulated exposure to omeprazole was unchanged in patients with morbid obesity compared with non-obese after multiple doses. This finding seems to contrast with previous results, which showed a reduced clearance for omeprazole in obese⁴⁹; however, the earlier study only examined omeprazole PK after single doses and did not differentiate between omeprazole enantiomers. In contrast, the similar exposure observed in non-obese and obese subjects after multiple doses of omeprazole can be attributed to the combined effects of obesity and CYP3A4 and CYP2C19 autoinhibition by omeprazole.^27,29 The time-dependent inhibition of these enzymes induced by omeprazole in both non-obese and obese populations likely masks the obesity effect on enzyme activity.

The PBPK model in post-gastric bypass describes the restoration of enzyme activity following bariatric surgery, thus capturing the dynamic changes in enzyme activity associated with obesity and subsequent recovery in post-bariatric surgery. The reduced gastric emptying post-gastric bypass affects primarily T_max and C_max (Figure S5). The simulated effect of gastric bypass in the gastrointestinal transit probably reflects a long-term effect. In patients evaluated 12 months after a gastric bypass, the gastric emptying reduced from 70 to 24 min for solid meals and from 15 to 7 min for liquid meals as compared with non-obese.⁵⁰ This would explain why the PK assessed in the short time after surgery (1–6 months) displays a similar profile as in obese subjects with an underprediction of T_max and an overprediction of C_max (Figure 4a,b). However, because the response to omeprazole is mainly linked to exposure (AUC) rather than C_max, our model verification was deemed adequate. While our sensitivity analysis indicated that omeprazole exposure is not sensitive to variations in gastric pH, it is noteworthy that the active pharmaceutic ingredient omeprazole undergoes rapid degradation in acid pH.⁴⁰ However, current formulations utilize enteric-coated tablets or extended-release capsules containing enteric-coated granules, and variations in the intragastric pH are not expected to affect omeprazole degradation in these formulations.

To explore the impact of obesity and bariatric surgery on omeprazole response in the intragastric pH, we linked the PK of omeprazole to a semi-mechanistic PD model.³³ Overall, the simulated intragastric pH was slightly higher in obese than in non-obese individuals, irrespective of omeprazole dose (Figure 7). Simulations indicated that the daily time with pH >4 was 14.7 h in non-obese treated with 40 mg omeprazole q.d. for 7 days, and this duration increased to 16.4 h (p < 0.0001) in obese patients. This effect can be attributed to the greater dilution impact on intragastric pH resulting from increased food intake in obese patients. Consequently, no dosing recommendations for omeprazole were proposed for obese patients.

For individuals undergoing gastric bypass, simulations focused on the intragastric pH in the small gastric pouch, given the vagotomy of the vagus nerve branch in the remnant stomach that significantly changes the effect of gastric hormones on this organ. Model predictions in post-gastric bypass patients could not be verified due to this population's lack of available data. Although marginal ulcers in the gastrojejunal anastomosis are a recognized complication of bariatric surgery,^1,2 their cause may not be directly associated with the intragastric pH in the gastric pouch. Instead, the primary factor may be the limited buffering capacity of the jejunum, even with low concentrations of protons released by the gastric pouch. This vulnerability may be exacerbated by bile release absence in the proximal jejunum in patients who have undergone a gastric bypass. Given that the daily time with pH >4 was consistently 24 h, regardless of the omeprazole dose, a low omeprazole dose (10–20 mg) may be considered for patients following a gastric bypass to maintain a high intragastric pH. This exploratory work suggests that marginal ulcers in the gastrojejunal anastomosis primarily stem from the increased vulnerability of the jejunum's mucosa to gastric acid secretion, even with a gastric secretion that is less acidic than in non-operated patients. However, a broader understanding of clinical outcomes associated with the efficacy of PPIs in treating marginal ulcers in post-gastric bypass patients warrants further investigation.

This comprehensive PBPK/PD model for omeprazole enantiomers successfully captures the PK parameters and concentration–time profiles in diverse populations, including healthy subjects, morbidly obese, and post-gastric bypass individuals. The model accounts for enantioselective CYP-mediated metabolism and nonlinear PK associated with mechanism-based inhibition of CYP2C19. Notably, the hepatic activity of CYP3A4 and CYP2C19 are essential contributors to the variability in omeprazole PK. By implementing the down-regulation of CYP2C19 and CYP3A4 and its restoration 6 months after bariatric surgery, we were able to capture the observed metabolic ratios, which are well-recognized biomarkers when assessing the in vivo activity of CYP3A4 and 2C19. Once omeprazole metabolic ratios are good predictors of the in vivo activity of CYP2C19 and CYP3A4 enzymes, the current modeling approach will improve dose optimization of CYP3A4 and 2C19 substrates in obese and post-gastric bypass patients.

The current work has some limitations. Although the frequency of CYP2C19*17 allele is high in Caucasians,²⁶ we did not assess the effect of CYP2C19 rapid and ultrarapid metabolizers, which could increase the interpatient variability in omeprazole PK and PD. The lack of available intragastric pH throughout 24 h prevented the validation of the PD model in post-gastric bypass patients. Additionally, certain risk factors associated with increased susceptibility to marginal ulcers, such as the use of nonsteroidal anti-inflammatory drugs (NSAIDs), smoking, and alcohol consumption,³ were not considered in our PD model. Future iterations of the model could benefit from incorporating these variables known to influence the risk of marginal ulcers in this specific patient population. The model can be further extended to predict responses to additional drugs metabolized by CYP2C19 and CYP3A4. Additionally, incorporating other significant CYP enzymes influenced by obesity and weight loss will amplify the utility of this model.

AUTHOR CONTRIBUTIONS

L.F.P., V.V., and N.M. designed the research. L.F.P., A.S., V.V., R.C., W.S.J., and N.M. analyzed the data. L.M.W. and D.T. performed the clinical research. L.F.P. and N.M. wrote the manuscript.

ACKNOWLEDGMENTS

We acknowledge Priscila A. Yamamoto for the insightful and fruitful discussions regarding clinical pharmacokinetics in patients undergoing gastric bypass surgery.

FUNDING INFORMATION

The clinical study was funded by the Central Hospital Pharmacy, The Hague; Medical Center Haaglanden, The Hague; and the Dutch Obesity Clinic West, The Hague. The data analysis and preparation of this article were performed independently of this grant.

CONFLICT OF INTEREST STATEMENT

The authors declared no competing interests for this work.

References

Sapala JA, Wood MH, Sapala MA, Flake TM. Marginal ulcer after gastric bypass: a prospective 3‐year study of 173 patients. Obes Surg. 1998;8:505‐516.

Ying VWC, Kim SH, Khan KJ, et al. Prophylactic PPI help reduce marginal ulcers after gastric bypass surgery: a systematic review and meta‐analysis of cohort studies. Surg Endosc. 2015;29:1018‐1023.

Salame M, Jawhar N, Belluzzi A, et al. Marginal ulcers after roux‐en‐Y gastric bypass: etiology, diagnosis, and management. J Clin Med. 2023;12: [eLocator: 4336].

Fellenius E, Berglindh T, Sachs G, et al. Substituted benzimidazoles inhibit gastric acid secretion by blocking (H+ + K+)ATPase. Nature. 1981;290:159‐161.

Lindberg P, Nordberg P, Alminger T, Brändström A, Wallmark B. The mechanism of action of the gastric acid secretion inhibitor omeprazole. J Med Chem. 1986;29:1327‐1329.

Howden CW, Ballard ED, Koch FK, Gautille TC, Bagin RG. Control of 24‐hour intragastric acidity with morning dosing of immediate‐release and delayed‐release proton pump inhibitors in patients with GERD. J Clin Gastroenterol. 2009;43:323‐326.

Funaki Y, Tokudome K, Izawa S, et al. Comparison of the effect of a single dose of omeprazole or lansoprazole on intragastric pH in Japanese participants: a two‐way crossover study. J Chin Med Assoc. 2013;76:131‐134.

Lespessailles E, Toumi H. Proton pump inhibitors and bone health: an update narrative review. Int J Mol Sci. 2022;23: [eLocator: 10733].

Florentin M, Elisaf MS. Proton pump inhibitor‐induced hypomagnesemia: a new challenge. World J Nephrol. 2012;1:151‐154.

Kuipers EJ, Lundell L, Klinkenberg‐Knol EC, et al. Atrophic gastritis and helicobacter pylori infection in patients with reflux esophagitis treated with omeprazole or fundoplication. N Engl J Med. 1996;334:1018‐1022.

Srinivas NR. Influence of morbid obesity on the clinical pharmacokinetics of various anti‐infective drugs: reappraisal using recent case studies‐issues, dosing implications, and considerations. Am J Ther. 2018;25:e224‐e246.

Gouju J, Legeay S. Pharmacokinetics of obese adults: not only an increase in weight. Biomed Pharmacother Biomed Pharmacother. 2023;166: [eLocator: 115281].

Brill MJE, Diepstraten J, van Rongen A, van Kralingen S, van den Anker JN, Knibbe CAJ. Impact of obesity on drug metabolism and elimination in adults and children. Clin Pharmacokinet. 2012;51:277‐304.

Tchernof A, Lévesque É, Beaulieu M, et al. Expression of the androgen metabolizing enzyme UGT2B15 in adipose tissue and relative expression measurement using a competitive RT‐PCR method. Clin Endocrinol. 1999;50:637‐642.

Kotlyar M, Carson SW. Effects of obesity on the cytochrome P450 enzyme system. Int J Clin Pharmacol Ther. 1999;37:8‐19.

Aitken AE, Richardson TA, Morgan ET. Regulation of drug‐metabolizing enzymes and transporters in inflammation. Annu Rev Pharmacol Toxicol. 2006;46:123‐149.

Mathieu P, Lemieux I, Després J‐P. Obesity, inflammation, and cardiovascular risk. Clin Pharmacol Ther. 2010;87:407‐416.

Zarezadeh M, Saedisomeolia A, Shekarabi M, Khorshidi M, Emami MR, Müller DJ. The effect of obesity, macronutrients, fasting and nutritional status on drug‐metabolizing cytochrome P450s: a systematic review of current evidence on human studies. Eur J Nutr. 2021;60:2905‐2921.

American Society for Metabolic and Bariatric Surgery. Estimate of bariatric surgery numbers, 2011–2022. Am Soc Metab Bariatr Surg. 2022. https://asmbs.org/resources/estimate‐of‐bariatric‐surgery‐numbers

Edwards A, Ensom MHH. Pharmacokinetic effects of bariatric surgery. Ann Pharmacother. 2012;46:130‐136.

Kral JG, Näslund E. Surgical treatment of obesity. Nat Clin Pract Endocrinol Metab. 2007;3:574‐583.

Mitrov‐Winkelmolen L, van Buul‐Gast MCW, Swank DJ, Overdiek HWPM, van Schaik RHN, Touw DJ. The effect of roux‐en‐Y gastric bypass surgery in morbidly obese patients on pharmacokinetics of (acetyl)salicylic acid and omeprazole: the ERY‐PAO study. Obes Surg. 2016;26:2051‐2058.

Portolés‐Pérez A, Paterna ABR, Sánchez Pernaute A, et al. Effect of obesity and roux‐En‐Y gastric surgery on omeprazole pharmacokinetics. Obes Facts. 2022;15:271‐280.

Giri P, Patel H, Srinivas NR. Use of cocktail probe drugs for indexing cytochrome P450 enzymes in clinical pharmacology studies ‐ review of case studies. Drug Metab Lett. 2019;13:3‐18.

Lenoir C, Terrier J, Gloor Y, et al. Impact of SARS‐CoV‐2 infection (COVID‐19) on cytochromes P450 activity assessed by the Geneva cocktail. Clin Pharmacol Ther. 2021;110:1358‐1367.

Lima JJ, Thomas CD, Barbarino J, et al. Clinical pharmacogenetics implementation consortium (CPIC) guideline for CYP2C19 and proton pump inhibitor dosing. Clin Pharmacol Ther. 2021;109:1417‐1423.

Wu F, Gaohua L, Zhao P, et al. Predicting nonlinear pharmacokinetics of omeprazole enantiomers and racemic drug using physiologically based pharmacokinetic modeling and simulation: application to predict drug/genetic interactions. Pharm Res. 2014;31:1919‐1929.

Hassan‐Alin M, Andersson T, Bredberg E, Röhss K. Pharmacokinetics of esomeprazole after oral and intravenous administration of single and repeated doses to healthy subjects. Eur J Clin Pharmacol. 2000;56:665‐670.

Hassan‐Alin M, Andersson T, Niazi M, Röhss K. A pharmacokinetic study comparing single and repeated oral doses of 20 mg and 40 mg omeprazole and its two optical isomers, S‐omeprazole (esomeprazole) and R‐omeprazole, in healthy subjects. Eur J Clin Pharmacol. 2005;60:779‐784.

Shirasaka Y, Sager JE, Lutz JD, Davis C, Isoherranen N. Inhibition of CYP2C19 and CYP3A4 by omeprazole metabolites and their contribution to drug‐drug interactions. Drug Metab Dispos. 2013;41:1414‐1424.

Shiohira H, Yasui‐Furukori N, Yamada S, Tateishi T, Akamine Y, Uno T. Hydroxylation of R(+)‐ and S(−)‐omeprazole after racemic dosing are different among the CYP2C19 genotypes. Pharm Res. 2012;29:2310‐2316.

Rost KL, Roots I. Nonlinear kinetics after high‐dose omeprazole caused by saturation of genetically variable CYP2C19. Hepatology. 1996;23:1491‐1497.

Liu D, Yang H, Jiang J, et al. Pharmacokinetic and pharmacodynamic modeling analysis of intravenous esomeprazole in healthy volunteers: esomeprazole in healthy volunteers. J Clin Pharmacol. 2016;56:816‐826.

Regårdh CG. Pharmacokinetics and metabolism of omeprazole in man. Scand J Gastroenterol Suppl. 1986;118:99‐104.

Kamiya Y, Takaku H, Yamada R, et al. Determination and prediction of permeability across intestinal epithelial cell monolayer of a diverse range of industrial chemicals/drugs for estimation of oral absorption as a putative marker of hepatotoxicity. Toxicol Rep. 2020;7:149‐154.

Andersson T, Hassan‐Alin M, Hasselgren G, Röhss K, Weidolf L. Pharmacokinetic studies with esomeprazole, the (S)‐isomer of omeprazole. Clin Pharmacokinet. 2001;40:411‐426.

Andersson T, Weidolf L. Stereoselective disposition of proton pump inhibitors. Clin Drug Investig. 2008;28:263‐279.

Äbelö A, Andersson TB, Antonsson M, Naudot AK, Skånberg I, Weidolf L. Stereoselective metabolism of omeprazole by human cytochrome P450 enzymes. Drug Metab Dispos. 2000;28:966‐972.

Ogilvie BW, Yerino P, Kazmi F, et al. The proton pump inhibitor, omeprazole, but not lansoprazole or pantoprazole, is a metabolism‐dependent inhibitor of CYP2C19: implications for coadministration with clopidogrel. Drug Metab Dispos. 2011;39:2020‐2033.

Cederberg C, Andersson T, Skånberg I. Omeprazole: pharmacokinetics and metabolism in man. Scand J Gastroenterol Suppl. 1989;166:33‐40; discussion 41‐42.

Rodríguez‐Morató J, Goday A, Langohr K, et al. Short‐ and medium‐term impact of bariatric surgery on the activities of CYP2D6, CYP3A4, CYP2C9, and CYP1A2 in morbid obesity. Sci Rep. 2019;9: [eLocator: 20405].

Kvitne KE, Hole K, Krogstad V, et al. Correlations between 4β‐hydroxycholesterol and hepatic and intestinal CYP3A4: protein expression, microsomal ex vivo activity, and in vivo activity in patients with a wide body weight range. Eur J Clin Pharmacol. 2022;78:1289‐1299.

Kvitne KE, Krogstad V, Wegler C, et al. Short‐ and long‐term effects of body weight, calorie restriction and gastric bypass on CYP1A2, CYP2C19 and CYP2C9 activity. Br J Clin Pharmacol. 2022;88:4121‐4133.

Darwich AS, Pade D, Ammori BJ, Jamei M, Ashcroft DM, Rostami‐Hodjegan A. A mechanistic pharmacokinetic model to assess modified oral drug bioavailability post bariatric surgery in morbidly obese patients: interplay between CYP3A gut wall metabolism, permeability and dissolution. J Pharm Pharmacol. 2012;64:1008‐1024.

Jamwal R, de la Monte SM, Ogasawara K, Adusumalli S, Barlock BB, Akhlaghi F. Nonalcoholic fatty liver disease and diabetes is associated with decreased CYP3A4 protein expression and activity in human liver. Mol Pharm. 2018;15:2621‐2632.

Ulvestad M, Skottheim IB, Jakobsen GS, et al. Impact of OATP1B1, MDR1, and CYP3A4 expression in liver and intestine on interpatient pharmacokinetic variability of atorvastatin in obese subjects. Clin Pharmacol Ther. 2013;93:275‐282.

Takahashi T, Smith AR, Jacobson PA, Fisher J, Rubin NT, Kirstein MN. Impact of obesity on Voriconazole pharmacokinetics among pediatric hematopoietic cell transplant recipients. Antimicrob Agents Chemother. 2020;64: [eLocator: e00653‐20].

Shakhnovich V, Abdel‐Rahman S, Friesen CA, et al. Lean body weight dosing avoids excessive systemic exposure to proton pump inhibitors for children with obesity. Pediatr Obes. 2019;14. https://onlinelibrary.wiley.com/doi/10.1111/ijpo.12459

Chen K, Luo P, Yang G, et al. Population pharmacokinetics of omeprazole in obese and normal‐weight adults. Expert Rev Clin Pharmacol. 2022;15:461‐471.

Horowitz M, Cook DJ, Collins PJ, et al. Measurement of gastric emptying after gastric bypass surgery using radionuclides. Br J Surg. 1982;69:655‐657.