What Is Known
- In 2003 the Food and Drug Administration published guidance on exposure-response relationships and their role in pediatric drug approval.
- Oral and IV esomeprazole were approved in adults based on endoscopic assessment, which is not practical in children despite being the standard clinical endpoint.
What Is New
- This is the first example for proton pump inhibitors in which exposure-response analyses for efficacy supported approval in children by extrapolating efficacy from adults to inform pediatric dose selection.
Children often differ from adults in disease progression, pharmacokinetics (PK), response to treatment, and in the ability to conduct clinical trials (1,2). In pediatric drug development, these differences have limited the availability of approved products for use in children and emphasize the unmet clinical need in children in which no FDA-approved therapies exist (1). The failure rate of efficacy trials for drugs approved in adults for use in children is approximately 50% (2). Additional pediatric challenges include clinical trials face difficult recruitment, smaller sample sizes, and clinical endpoints in adults that may not be practical to perform in children (3).
In response to some of these challenges in 1994 the FDA proposed the use of extrapolation of efficacy from adult population to children when certain criteria hold true (4). First, when the pathophysiology of the disease progression and response to intervention are similar between adults and children, evidence of effectiveness may be extrapolated from adults to children. A third criterion is useful for dose selection in pediatrics—if the exposure-response relationships are also similar between children and adults. This is discussed in further detail in the FDA Guidance for Industry: Exposure-Response Relationships—Study Design, Data Analysis, and Regulatory Applications document and is depicted in the FDA pediatric decision tree shown in Supplemental Digital Content 1 (http://links.lww.com/MPG/A846) (5,6). In this case we can make use of exposure matching analyses: identifying the therapeutic range of concentrations from the adult trials and selecting a dose in pediatrics that would produce the corresponding drug exposure. When the exposure-response is not similar between adults and children additional pharmacodynamics (PD) measures in pediatrics are necessary to identify the appropriate dose. The advantage of extrapolation of evidence of effectiveness from adults, which can be substantiated on pharmacometric approaches such as exposure-response for PD measures of efficacy and/or PK exposure matching to adults, is that approval and dose selection can be performed without the need for additional efficacy trials (4–7). Pharmacometrics refers to quantitative measures of drug action, and may include population PK models, interpatient factors of drug response, understanding of drug pharmacology and the relevant PD data, and how these factors contribute to the assessment of drug effect.
Esomeprazole (Nexium) is approved for use as an oral and IV formulation in adults and children ages 1 month to 17 years for treatment of gastroesophageal reflux disease (GERD). Oral esomeprazole was first approved in the United States in 2001 in adults, for the treatment of GERD. An esomeprazole pediatric clinical program was designed to investigate the potential of esomeprazole to treat GERD in children from birth through 17 years of age, inclusive. Label extensions for esomeprazole use in patients 12 years of age or older were approved in the United States in 2006. Eventually, label extensions for esomeprazole use in patients 1 to 11 years of age and a new oral dosage form (sachet) of Nexium for children who require a 10 mg dose of esomeprazole were approved in the United States in 2008. The original clinical trial for oral esomeprazole in children, however, failed to meet the primary efficacy endpoint in the age range of 1 month to <1 year. The clinical endpoint in that trial was time to discontinuation due to worsening of symptoms. A numerical trend in favor of esomeprazole was observed (31% longer time to discontinuation due to worsening of symptoms); however, the finding was statistically insignificant (P = 0.275, 53 children). Given the clinical endpoint's effect size (31%), high variability in the endpoint in this population, and the number that completed the double-blind phase (n = 53, 80 were enrolled in this phase). It is quite likely that the trial was not powered sufficiently to establish significance (P = 0.275). The limitations of conducting a sufficient trial prevented the determination of statistical significance. The approved doses in adults are 20 and 40 mg once daily. For children, the approved oral doses are 10 or 20 mg for 1- to 11-year-old patients and 20 or 40 mg for 12- to 17-year-old patients (8).
In the case of IV esomeprazole for children ages 1 to 17 years, during the investigational new drug stage the agency agreed to extrapolate evidence of efficacy from adults after IV esomeprazole and from children 1 to 17 years of age after oral esomeprazole. Oral esomeprazole was approved for use in children ages 1 to 17 years at the time of the IV submission. The required therapeutic exposures could be identified from both the oral formulation data in children and the IV formulation in adults to select an IV dose that matches these exposures. Consequently, with the new drug application submission, the applicant was only required to submit a PK study in children for dose selection by matching exposures to adults and any additional safety data that were required. The first part of this manuscript is a discussion of the dose selection based on these data.
For infants younger than 1 year PK and PD data are critical for the approval of esomeprazole for the treatment of endoscopically diagnosed erosive esophagitis associated with acid-related GERD. For this more severe, acid-specific indication, PD data consist of intragastric measures of pH and the corresponding time course of response to drug administration. A Gastrointestinal Drugs Advisory Committee (GIDAC) meeting on November 5, 2010 concerning proton pump inhibitors in patients with GERD younger than 1 year of age concluded that it is reasonable to extrapolate adult proton pump inhibitor efficacy data to pediatric patients 1 month to <1 year of age for the treatment of endoscopically diagnosed erosive esophagitis secondary to acid-related GERD provided that there are supporting PK and PD data (9). This was not established for GERD in general, and potential physiological differences between infants and adults were discussed that lead to similar symptoms. Rather, for erosive esophagitis the assumption that the disease pathology and response to intervention are similar between children and adults was predicated on the fact that the symptoms are acid-related for both the pediatric and adult populations (9).
The agency reviewed available PK and PD data for infants following oral esomeprazole in infants younger than 1 year. The primary efficacy endpoint in children older than 1 year requires an endoscopic assessment and is not practical in infants. As a result, esomeprazole was approved for the erosive esophagitis indication in the age range of 1 month to younger than 1 year for IV use on April 29, 2011 and for oral use on December 15, 2011. The agency's review of PK/PD data from infants after oral esomeprazole administration was also used as an additional evidence to support the exposure-matching approach between children 1 to 17 years of age and adults because the exposure-response relationships were similar between the pediatric and adult populations. This indicated that the response to treatment was the same and therapeutic exposures were identified for pediatric dose selection.
The approval and dose selection of esomeprazole in children for IV and oral use is an example of approval of pediatric indications based on PK data or biomarker PK/PD and PK data, without the need for additional efficacy clinical trials. This article discusses the approval rationale and dose optimization of esomeprazole for oral and IV use in children.
MATERIALS AND METHODS
Pediatric Pharmacokinetic and Pharmacodynamic Studies
The applicant conducted 3 phase 1 and 1 phase 3, randomized, clinical studies in children with GERD (ages 1–17 years) or symptoms of GERD (ages <1 month–17 year) that provided data for assessment of multiple dose PK of once-daily oral esomeprazole (10–13). Oral esomeprazole doses evaluated in children and adolescents are shown in Table 1. Steady-state PK samples were collected on the last day of each study as outlined in Supplemental Digital Content 2 (http://links.lww.com/MPG/A847). Single-dose samples were also collected on the first day of dosing for adolescents (12–17 years old). In patients younger than 1 year of age, 24-hour intragastric-pH PD data were collected on day 7.
The applicant conducted a phase 1, open-label, randomized, multinational study to define the multiple-dose PK of once-daily IV esomeprazole as 3-minute injections in hospitalized pediatric patients 0 to 17 years of age, who required acid suppression therapy (14). Doses of IV esomeprazole evaluated in children and adolescents are shown in Table 1. The steady-state PK of esomeprazole was determined from blood samples collected on day 4 of once-daily dosing (Supplemental Digital Content 2, http://links.lww.com/MPG/A847).
Adult Pharmacokinetic and Pharmacodynamic Studies
The applicant conducted a double-blind, randomized, 2-way cross-over comparative study of 40 mg esomeprazole given intravenously during a period of 3 or 30 minutes for 10 days in healthy subjects older than 18 years (15). Serial PK samples and 24-hour intragastric-pH PD data were collected on day 1 and day 10 in each treatment period (Supplemental Digital Content 2, http://links.lww.com/MPG/A847).
Oral and IV Esomeprazole
The applicant conducted an open, randomized, 2-way crossover comparative study of 20 mg oral and intravenous esomeprazole for 5 days in healthy subjects older than 18 years (16). Serial PK samples and 24-hour intragastric-pH PD data were collected on day 1 and day 5 in each treatment period (Supplemental Digital Content 2, http://links.lww.com/MPG/A847).
PK data were available from 117 and 50 pediatric patients receiving oral and IV (3-minute injections) esomeprazole, respectively, with 6 to 21 individuals in each dose group. In adults, data were used from 24 and 42 patients receiving oral and IV (3-minute injections or 30-minute infusions) esomeprazole at either 20 or 40 mg doses. A predose sample and between 4 and 20 postdose samples were collected per individual. A detailed breakdown of the dosing and PK sampling by age group and formulation are shown in Table 1.
Intragastric pH was evaluated in a total of 52 neonates and infants, between the ages of birth and 1 year and in the 66 adults patients mentioned above. Intragastric pH data were not available in children between 1 and 18 years of age. PD endpoints used to evaluate the exposure-response relationship for gastric pH include the percentage of time in the first 24 hours postdose in which the intragastric pH was >4 and the median pH during the first 24 hours postdose. Missing PK and/or PD data were not imputed or included in the analysis.
Two population PK modeling analyses were used to describe the PK of either oral esomeprazole or IV esomeprazole and estimate the clearance of esomeprazole in pediatric patients (17–19). In brief, the population PK model for the oral data was developed with data from 117 subjects birth to 17 years of age using the nonlinear mixed effects modeling software NONMEM (version V). The oral esomeprazole PK model used for simulations was a linear 1-compartment model with first-order absorption. The IV esomeprazole population PK model was also developed with NONMEM using data from 50 children 1 month to 17 years of age and is a linear 2-compartment model. Each individual's Bayesian post-hoc estimate of area under the curve (AUC) and maximal concentration (Cmax) were used to support selection of a dose in pediatric patients. Simulations were performed using the modeled pediatric demographic data and individual Bayesian post-hoc PK parameters to identify a dosing regimen that matched exposures (AUC and Cmax at steady-state) in adults after oral or IV esomeprazole http://www.fda.gov/downloads/Drugs/DevelopmentApprovalProcess/DevelopmentResources/UCM258684.pdf). AUC was matched because this was a measure of overall esomeprazole exposure and often correlates with efficacy. To ensure safety, pediatric esomeprazole Cmax values were matched to adults Cmax values to limit the maximal esomeprazole exposures being dosed to children. For the simulations the Bayesian post-hoc estimates of esomeprazole PK parameters and the demographics of the evaluated pediatric population were used.
The PD endpoints of change from baseline in the percentage of the first 24 hours postdose in which the gastric pH >4.0 and change from baseline in the median gastric pH during the first 24 hours postdose were plotted against esomeprazole AUC to establish an exposure-response relationship for effectiveness in children and in adults separately. These 2 plots were then overlaid to provide a visual comparison of the exposure-response relationship between adults and children.
RESULTS AND DISCUSSION
Pediatric Exposures Can Be Matched to Adult Exposures to Support Selection of the Esomeprazole Dose
Figure 1 indicates that the PK-PD relationships for both pH measures are similar between infants and adults. This conclusion supports the extrapolation of adult efficacy to children (for either IV or oral esomeprazole administration) and the selection of a dose based on exposure-matching children to adults. This finding is also consistent with the GIDAC's conclusion that efficacy can be extrapolated from adults to pediatric patients younger than 1 year of age for the treatment of erosive esophagitis secondary to acid-related GERD (9). It is critical to note here that this extrapolation was not recommended for GERD in general, rather a specific population in which the disease is acid related in both children and adults and the disease is more severe and there is unmet medical need, justifying treatment to reduce intragastric pH.
Pediatric Exposures Should Match the Adult Exposures Following 20 mg Esomeprazole
Esomeprazole is approved at 2 dose levels in adults (20 and 40 mg). The current esomeprazole labeling indicates no therapeutic advantage of 40 mg compared with 20 mg esomeprazole for the healing of erosive esophagitis and sustained resolution of heartburn in adult clinical trials (8). In addition at the time of review, oral esomeprazole was approved for use in children 1 to 17 years of age. Because the exposure-response relationships were similar between children and adults, the exposures from the 20 mg IV dose in adults were chosen as a target when selecting the pediatric dose. Supplemental Digital Content 3 (http://links.lww.com/MPG/A848) is a graph showing that the target range (90% confidence interval for observed adult 20 mg data) of esomeprazole AUCs produces a response, defined as change from baseline in percentage time with pH >4 in 24 hours, that falls within the plateau of the PK-PD relationship, demonstrating that the target response is achieved with less overall exposure than the 40 mg dose in adults.
Selection of Esomeprazole IV Dosing in Children
Pediatric Dosing Should Be Weight Based
Based on the population PK analyses, esomeprazole clearance and volume of distribution increased with body weight. Esomeprazole clearance increased 10-fold over the range of 3 to 75 kg. The inclusion of body weight as a covariate reduced the between subject variability in clearance by 24%. No other covariates were found to correlate with esomeprazole clearance or volume of distribution. These results suggest that dosing should be based on body weight to produce similar exposures across the pediatric population.
Supplemental Digital Content 4 (http://links.lww.com/MPG/A849) is a box plot showing that a simulated dosing regimen in children with a body weight cut point at 55 kg for children older than 1 year and a dosing regimen of 0.5 mg/kg for infants older than 1 month to younger than 1 year reasonably match observed esomeprazole exposures from the 20 mg adult dose. None of the pediatric exposures exceeded those from the 40 mg dose in adults.
Although the AUC values from the weight-based regimen, in children from 1 month to 17 years of age, matched the AUC values from the 20 mg adult dose, the geometric mean Cmax values for the weight-based 0.5 mg/kg, 10 mg, and 20 mg pediatric dose groups are, 3.2-, 2.6-, and 3.1-fold higher than the geometric mean Cmax value for the adult 20 mg dose given as a 3-minute injection (Supplemental Digital Content 5, http://links.lww.com/MPG/A850). Cmax values are higher in children owing to a smaller volume of distribution of esomeprazole. That AUC values match those in adults and Cmax does not suggest that the dose amount is appropriate; however, the infusion duration could be increased to reduce the maximum concentration in children. The next section discusses what infusion durations in children would match exposures in adults.
Infusion Durations Should Be 10 to 30 Minutes in Children
Increasing the infusion duration to 10 minutes matches the steady-state Cmax values to that for the 40-mg, 3-minute injection in adults (Fig. 2, left panel). Increasing the infusion time to 30 minutes matches the Cmax values with that for the 20-mg, 3- or 30-minute infusion in adults (Fig. 2, right panel). Although exposures in children are expected to be higher than the 20 mg dose in adults for infusions <30 minutes, these exposures are short-lived, and are projected to decline to concentrations comparable to the 20 mg adult dose within 2.4 minutes postdose for a 10-minute infusion.
Approval of Oral and IV Esomeprazole in Children Younger Than 1 Year
Before the GIDAC meeting of November 5, 2010, oral esomeprazole was approved in adults and children older than 1 year of age. The GIDAC meeting on November 5, 2010 concluded that it is reasonable to extrapolate efficacy of proton pump inhibitor products in pediatric patients 1 month to younger than 1 year of age for the treatment of endoscopically diagnosed erosive esophagitis secondary to acid-related GERD (9).
Following the advisory committee and approval of intravenous esomeprazole to treat GERD in children 1 month to younger than 18 years of age, the applicant submitted data from 2 PK-PD studies to support approval of oral esomeprazole in children 1 month to younger than 1 year of age. The following weight-based dosing regimen in patients 1 month to younger than 2 years of age was proposed: 2.5 mg once daily for patients with body weight of 3 to 5 kg; 5 mg once daily for patients with body weight >5 to 7.5 kg; and 10 mg once daily for patients with body weight >7.5 to 12 kg. Utilizing the PK data in birth to 24 months of age along with data from older children (2–17 years of age), the applicant developed a population PK model, which reasonably described the plasma concentration time course of esomeprazole after oral administration in patients with symptoms of GERD. The FDA modified this model to include age as a covariate on clearance (18). The model's individual post-hoc estimates of clearance exhibited a clear correlation between esomeprazole clearance and body weight (Supplemental Digital Content 6, http://links.lww.com/MPG/A851). The agency used these individual post-hoc estimates of clearance from the adjusted population PK model to determine the AUCs of individual patients to obtain an overall measure of esomeprazole exposure (AUC) for the exposure-response and exposure-matching analyses.
As seen from Figure 3, the proposed weight-based dosing produced exposures in pediatric patients 1 month to younger than 1 year of age that are similar to the exposures observed with the lowest approved dose in adults (ie, 20 mg once daily). In addition, the exposures in patients 1 month to younger than 1 year appear to be similar to pediatric patients 1 to 17 years of age with substantial overlap in exposures. Thus, based on exposure matching, the proposed weight-based dosing regimen was considered appropriate for treating GERD in pediatric patients 1 month to younger than 1 year of age (8). On April 29, 2011, the agency approved the use of IV esomeprazole in infants (>1 month to 1 year of age) for the treatment of GERD with erosive esophagitis (20). And on December 15, 2011, the agency approved the use of oral esomeprazole for use in infants for the treatment of GERD (8).
The FDA's Guidance for Industry: Exposure-Response Relationships—Study Design, Data Analysis, and Regulatory Applications document indicates that when disease progression and response to intervention are similar between adults and children, evidence of effectiveness may be extrapolated from adults to children without the need for further efficacy trials (5). Approval of esomeprazole in children 1 to 17 years of age was granted on this basis. Although, in infants approval required the evaluation PK/PD in infants as outlined by the November 5, 2010 GIDAC meeting. Evaluation of the PK/PD relationship in infants and adults revealed a similar relationship between the 2 populations suggesting that exposure matching was a reasonable approach to select the dose across all pediatric populations.
The exposure-matching analysis for esomeprazole permitted approval of a dosing regimen not directly studied in pediatric efficacy trials. Exposure-response analyses for intragastric pH permitted approval for the treatment of GERD in children in whom it was not possible to evaluate the adult primary endpoint, mucosal healing assessed by endoscopy.
The authors would like to thank Drs Lynne Yao and Vikram Sinha for their thoughtful guidance regarding the completion of this manuscript.
1. Addressing the Barriers to Pediatric Drug Development: Workshop Summary. Institute of Medicine (US) Forum on Drug Discovery, Development, and Translation. Washington, DC: National Academies Press (US); 2008.
2. Benjamin DK Jr, Smith PB, Jadhav P, et al. Pediatric antihypertensive trial failures: analysis of end points and dose range. Hypertension
4. Dunne J, Rodriguez WJ, Murphy DM, et al. Extrapolation of adult data and other data in pediatric drug-development programs. Pediatrics
5. Guidance for industry: Exposure-Response Relationships—Study Design, Data Analysis, and Regulatory Applications, US FDA, April 2003. http://www.fda.gov/ucm/groups/fdagov-public/@fdagov-drugs-gen/documents/document/ucm072109.pdf
. Accessed January 27, 2017.
6. Draft Guidance for industry: General Clinical Pharmacology Considerations for Pediatric Studies for Drugs and Biological Products, US FDA, December 2014. http://www.fda.gov/ucm/groups/fdagov-public/@fdagov-drugs-gen/documents/document/ucm425885.pdf
. Accessed January 27, 2017.
7. Tornøe CW, Tworzyanski JJ, Imoisili MA, et al. Optimising piperacillin/tazobactam dosing in paediatrics. Int J Antimicrob Agents
8. Nexium (esomeprazole magnesium) Label. http://www.accessdata.fda.gov/drugsatfda_docs/label/2006/021957lbl.pdf
. Published January 20, 2012. Accessed December 28, 2015.
9. Official Meeting Transcript for the November 5, 2011 Gastrointestinal Drugs Advisory Committee Meeting.
10. Zhao J, Li J, Hamer-Maansson JE, et al. Pharmacokinetic properties of esomeprazole in children aged 1 to 11 years with symptoms of gastroesophageal reflux disease: a randomized, open-label study. Clin Ther
11. Li J, Zhao J, Hamer-Maansson JE, et al. Pharmacokinetic properties of esomeprazole in adolescent patients aged 12 to 17 years with symptoms of gastroesophageal reflux disease: a randomized, open-label study. Clin Ther
12. Omari T, Lundborg P, Sandström M, et al. Pharmacodynamics and systemic exposure of esomeprazole in preterm infants and term neonates with gastroesophageal reflux disease. J Pediatr
13. Omari T, Davidson G, Bondarov P, et al. Pharmacokinetics
and acid-suppressive effects of esomeprazole in infants 1-24 months old with symptoms of gastroesophageal reflux disease. J Pediatr Gastroenterol Nutr
14. Sandström M, Davidson G, Tolia V, et al. Phase I, multicenter, randomized, open-label study evaluating the pharmacokinetics
and safety profile of repeated once-daily doses of intravenous esomeprazole in children 0 to 17 years of age. Clin Ther
15. Wilder-Smith C, Röhss K, Bondarov P, et al. Intravenous esomeprazole 40 mg is effective for control of intragastric acid levels whether given as a 3-minute injection or a 30-minute infusion. Clin Drug Investig
16. Wilder-Smith CH, Bondarov P, Lundgren M, et al. Intravenous esomeprazole (40 mg and 20 mg) inhibits gastric acid secretion as effectively as oral esomeprazole: results of two randomized clinical studies. Eur J Gastroenterol Hepatol
17. FDA Clinical Pharmacology Review for Nexium IV (Esomeprazole Sodium). http://www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/ucm049872.htm
. Published December 30, 2014. Accessed December 28, 2015. In press.
18. FDA Clinical Pharmacology Review for Nexium (Esomeprazole Magnesium). Published December 30, 2014. http://www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/ucm049872.htm
. Accessed December 28, 2015.
19. Van der Walt J-S, Standing J, Karlsson MO, et al. A population model describing the pharmacokinetics
of iv esomeprazole in patients aged 0 to 17 years, inclusive. PAGE
2010; 19: Abstract 1905. www.page-meeting.org/?abstract=1905
. Accessed January 27, 2017.
20. Nexium IV (esomeprazole sodium) Label. http://www.accessdata.fda.gov/scripts/cder/drugsatfda/
. Published June, 30, 2011. Accessed December 28, 2015.