Tailored Assays for Pharmacokinetic and Pharmacodynamic Investigations of Aliskiren and Enalapril in Children: An Application in Serum, Urine, and Saliva

According to the European Society of Hypertension, elevated blood pressure (BP) is more common in children than previously assumed.1 Despite extensive use, the efficacy and safety of most drugs successfully used to treat hypertension in adults are insufficiently investigated in children. The few medications that are specifically licensed for pediatric patients and the frequency of off-label use in children highlight this drawback.2–4 Pediatric regulations were initiated in the United States and Europe to increase knowledge about the safety and efficacy of drugs used in children. The U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA) both recommend that clinical investigations involving children should be associated with minimal risk; hence, analytic methods should require the smallest amount of blood possible. Both organizations also recommend alternative, non-invasive sampling procedures, such as urine or saliva collection.5,6 Although the FDA has no specific guidelines on the volume that should be drawn in pediatric patients, the EMA recommends a maximum trial-related blood loss of not more than 3% within 4 weeks. For any single sampling event, the amount of blood withdrawn should not exceed 1% of the total blood volume.6 For these reasons, low-volume assays are required for sophisticated investigations of pharmacokinetics and pharmacodynamics in children. Before implementing low-volume assays in pediatric trials, the usefulness of those assays should be demonstrate via a proof-of-concept study in healthy volunteers. Therefore, a study was conducted for aliskiren and enalapril, which are of high interest in pediatric pharmacotherapy and were selected based on information given at the expert group meeting on pediatric heart failure held at the EMA in London on November 29, 2010. Enalapril, an angiotensin-converting enzyme inhibitor (ACEI), is recommended as a first-line treatment for chronic heart failure and hypertension in children.7 Because aliskiren is the only available, direct renin inhibitor (DRI), it is listed under investigational drugs.8 

Aliskiren, the first marketed, orally available, DRI, has been shown to lower BP in humans. Recently, the first pediatric study evaluating the pharmacokinetics of aliskiren in children and adolescents aged 6–17 years was published by Sullivan et al.9 For pharmacokinetic determination, 1 mL of blood was drawn per sample, which appears inappropriate for dense sampling in younger children. Because of the high intraindividual and interindividual variability of aliskiren, sparse sampling is not advisable for gathering data in children.10 Therefore, a tailored method that required only 0.1 mL serum per sample was successfully developed to reliably determine aliskiren concentration.11 

The second investigated antihypertensive drug was enalapril, an ACEI. Enalapril is administered as a prodrug with a bioavailability of about 50% to 60% and is subsequently converted by hepatic carboxylesterases (CES1) to its active metabolite, enalaprilat. Enalapril and enalaprilat have been investigated in 3 pediatric pharmacokinetic trials.12–14 Although determination by high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) appears to be the most suitable assay, given its superior selectivity, each of the above studies determined the concentrations using radioimmunoassays.15 Therefore, a published HPLC-MS/MS assay16 was modified for high-throughput requiring 100 μL serum, which is at least as little as half of the amount necessary in other HPLC-MS/MS setups.17–23 

Following regulatory recommendations to elucidate alternative biological fluids for non-invasive sampling,5 salivary drug concentrations were determined. Because enalapril penetration into saliva had yet to be studied, we used a validated method for quantification of aliskiren in saliva.24 Preliminary results in 3 subjects suggested that salvia might be an acceptable alternative, but more in-depth investigation on saliva as quantification matrix is needed.

The renin-angiotensin-aldosterone (RAA) system is central in the regulation of BP and a target in current pharmacotherapy. The ability to measure biomarkers in the RAA system (e.g., angiotensin I and II, plasma renin activity [PRA]) is key to a holistic understanding of the effects of pharmacologic intervention and also to investigating potential differences between adults and children. Only sparse data are currently available on aliskiren and enalapril interactions with the RAA system in children. Sullivan et al9 determined the PRA by using 2 mL of blood per sampling point in a pediatric trial, which is the same amount of blood reported in adult studies. In other adult studies, 7 to 9 mL of blood was required to determine concentrations of angiotensin I, angiotensin II, and other RAA biomarkers.25,26 Collection of this volume of blood from a neonate or toddler in a single sampling event would be unethical. Even commercial, low-volume assays require 6 mL of blood to determine angiotensin I, angiotensin II, PRA, renin, and prorenin. This clearly indicates the need for child-appropriate analytic approaches for evaluating the pharmacokinetics of administered drugs, the resulting changes in the humoral parameters caused by the drug administration, and consequently, the pharmacodynamics.

In addition, clinical trials in adults would also profit from this low-volume assay approach because dense sampling associated with sophisticated pharmacokinetic and pharmacodynamics drug studies are often connected to trial-related blood loss. The availability of a low-volume assay would make these investigations better tolerated in both pediatric and adults populations. Moreover, the systematic data in a proof-of-concept study are useful for validating physiology-based models and simulations, the latter of which is a recommended tool in (pediatric) study optimization.27,28 

A comprehensive, bioanalytic approach for investigating the pharmacokinetics of aliskiren, enalapril, and enalaprilat, as well as their effects on humoral parameters of the RAA system, was assessed by low-volume assays in blood, saliva, and urine. Conducting this trial in adults was advisable before applying this bioanalytic approach to pediatric studies for both regulatory purposes and to examine the drug exposure and its effect. The primary objective was not to provide new insight on the pharmacokinetics of aliskiren, enalapril, and enalaprilat in adults, but rather to ensure the applicability of the assays and to assess the agreement of the results by comparing them with reported adult data in the literature. If the bioanalytic, low-volume assay proves acceptable for use in pediatric age groups, complementary phase II and III studies to investigate the pharmacokinetics and safety of enalapril and enalaprilat in children will be conducted within the EC-funded LENA project (grant 602295).

This study was conducted at the Department of Clinical Pharmacy and Pharmacotherapy, Heinrich-Heine-University (Düsseldorf, Germany), with approval from the university ethics committee (and following the Declaration of Helsinki and Good Clinical Practice recommendations). All subjects provided written, informed consent before participation. Volunteers were enrolled into 2 investigation groups. One group received aliskiren hemifumarate, whereas enalapril maleate was administered to volunteers in the second group.

The study was designed as an open-label, single-sequence, single-dose study in healthy subjects. After collection of a predose sample in the morning of the first investigation day, each participant received 1 tablet of 300-mg aliskiren hemifumarate (Rasilez, Novartis Europharm Limited, Frimley, UK) or 1 tablet of 20-mg enalapril maleate (EnaHEXAL, HEXAL, Holzkirchen, Germany). The participants fasted at least 10 hours before drug administration. The study medication was administered with 250 mL of water. No further fluid intake was allowed for 30 minutes before and 1 hour after dosing with enalapril or 4 hours after administration of aliskiren. The first meal was served 4 hours postdose. The participants went home 8 hours postdose and returned every 24 hours until the end of the study. Intake of xanthine-containing food or beverages and alcohol were not permitted for at least 48 hours after drug administration.

On day 1 of the study, an indwelling cannula was inserted into a peripheral arm vein for blood sampling. On the following investigation days, blood samples were taken by direct venipuncture. All participants rested in a supine position for at least 30 minutes before sampling to determine humoral parameters.

Blood samples were collected predose. Samples were then collected every 20 minutes over the next 3 hours followed by sampling every 30 minutes until 8 hours postdose. Further samples were taken 24, 48, and 72 hours postdose. Volunteers in the aliskiren group provided additional samples at 96, 120, 144, 168, and 192 hours postdose. Marked, multiple-peak phenomena had previously been noted in patients given aliskiren; hence, an intensive sampling strategy was used to make a precise determination of the multiple-peak behavior per individual. This was important because individual concentration-time profiles showing this phenomenon are lacking. To describe the pharmacokinetic parameters (e.g., T_max) as exactly as possible, the frequent sampling was also used for enalapril/enalaprilat.

Despite the dense sampling schedule, total blood volume of only approximately 105 mL per volunteer was taken for full pharmacokinetic determination in serum. This is less than one-quarter of the usual blood-donation volume and, consequently, was ethically acceptable in adults. Drawn samples were kept at room temperature until centrifuged at 2000 x g for 10 minutes at 4°C, and then the supernatant was frozen at −80°C until analysis.

Urine samples were collected predose. Samples were collected for up to 192 hours thereafter in the aliskiren group and up to 72 hours in the enalapril group. Every spontaneous urination was collected and analyzed as a discrete sample without pooling. Although a correction for creatinine concentration is sometimes used to minimize error in urinary concentrations because of the variability in fluid composition, this study did not require the correction because the excreted amount of the drug was calculated by multiplying the urine volume by the urinary drug concentration.

Saliva was examined as an alternative fluid for obtaining drug concentrations and, in turn, pharmacokinetic determination, which might constitute an advantage in pediatric research. Although some authors29 collect saliva from adults by having them spit the oral fluid into collection bowls, Hiremath et al30 concluded that collection of saliva by oral swabs offers methodological advantages over passive drool in the pediatric population. Preliminary investigations were performed with 2 commercially available devices. One swab was primarily designed to collect buccal cell samples for DNA testing (Omniswab, GE Healthcare, Germany), and the second was a chewing swab on a cotton base (Salivette, Sarstedt, Nuembrecht, Germany). We chose to use salivettes because they are routinely used to determine concentrations in saliva (e.g., cortisol) and have already been used in pediatric research.47 The saliva samples were taken without any stimulation every 20 minutes for the first 8 hours after drug administration and then every 24 hours. Immediately after blood collection, the participant was asked to chew slightly on the swab for 1 minute. The swab was stored on ice until centrifuged at 1000 x g for 2 minutes at 4°C, then frozen at −20°C or below until analysis.

All humoral parameters were determined pre-dose and again at 0.5, 1, 2, 3, 4, 5, 6, 7, and 8hours postdose. On subsequent days, samples were taken every 24 hours. In total, a minimum blood volume of 2.1 mL was required for analysis of all RAA system parameters per timepoint. Although blood withdrawal was performed with substantial excess, allowing for replicate determinations, overall, only 250 mL blood were collected for determination of the pharmacokinetics of drugs and humoral parameters. This is equal to about one-half of the usual blood-donation volume. Collection tubes intended for angiotensinI and II determination were equipped with an inhibitor cocktail consisting of 0.44 mM o-phenanthroline, 25 mM ethylenediaminetetraacetic acid (EDTA), 0.12 mM pepstatin A, and 1 mM p-hydroxy-mercuric benzoic acid (absent in angiotensin I tubes). In addition, the pharmacodynamics were characterized by systolic and diastolic BP and heart rate. The corresponding RAA system is illustrated in Figure 1.

To prevent renin cryoactivation, blood samples for renin and prorenin were collected in EDTA tubes and immediately centrifuged at 2000 x g for 10 minutes at room temperature. The supernatant was snap-frozen before being stored at −80°C until assayed. Before the analysis of prorenin by enzyme-linked immunosorbent assay (ELISA), the plasma samples (100 μL) were rapidly thawed. Renin required a minimum of 250 μL of plasma for determination with a chemiluminescent immunoassay based on monoclonal antibodies.

Plasma renin activity was determined using angiotensin I antibodies and circulating immunoreactive angiotensin I concentrations because that approach was expected to be more precise.31 The chosen ¹²⁵I radioimmunoassay (RIA) has a wide angiotensin I calibration range, even in children who have up to 15-fold greater PRA than adults have.32 Samples were drawn in cooled EDTA tubes without inhibitor. The tubes were then centrifuged at 2000 x g for 10 minutes at 4°C and subsequently snap-frozen and stored at −80°C.

Cooled and inhibitor-spiked EDTA tubes were used to collect samples for angiotensin I and II determination. During blood withdrawal, the tubes were kept on ice and were immediately centrifuged at 2000 x g for 10 min at 4°C. Obtained plasma was snap-frozen and stored at −80°C until analysis; 100 μL of plasma was used to obtain the angiotensin I concentration with a commercial ¹²⁵I RIA, whereas 500 μL was necessary for the angiotensin II measurement using a modified ELISA. During the entire analysis, all samples were kept between 2°C and 8°C, unless otherwise described.

Before this study, bioanalytic assays for determining aliskiren and enalapril drug concentrations were developed and validated in serum, urine, and saliva.11,16,24,33 In brief, sample purification of serum and saliva was conducted by solid-phase extraction (SPE) using Oasis MCX 96-well plates (Waters, Eschborn, Germany). The SPE of urinary samples used Oasis WAX and MCX 96-well plates. The HPLC separation was performed on Xselect C18 CSH columns (Waters), followed by MS/MS detection with an API 2000 (AB Sciex, Concord, Canada). The mass spectrometer operated in positive mode with electrospray ionization to determine the transitions of aliskiren (552.2 m/z → 436.2 m/z), enalapril (377.3 m/z → 234.2 m/z), enalaprilat (349.3 m/z → 206.1 m/z), and the internal standard benazepril (425.3 m/z → 351.2 m/z). For aliskiren, the calibration range was 0.15 (lower limit of quantification [LLOQ]) to 1200 ng/mL in serum, from 9.4 (LLOQ) to 9600 ng/mL in urine, and from 0.59 (LLOQ) to 1200 ng/mL in saliva. For enalapril and enalaprilat, the LLOQs were 0.78 ng/mL and 0.70 ng/mL, respectively, in serum, and 0.1 ng/mL in saliva. The calibration curve in urine was from 11.6 (LLOQ) to 12,000 ng/mL for enalapril and 8.8 (LLOQ) to 9000 ng/mL for enalaprilat.

Before analysis of the humoral parameters of interest, the available commercial assays were screened for their suitability in investigations of pediatric patients by assessing the assay's detection range, sensitivity, selectivity, and required blood volume. If necessary, the chosen assays were modified to fit to the enclosed concentration claims or required sample volumes. The selected assays covered the main parameters of the RAA system and enabled a comprehensive description of changes in the humoral parameters after administration of aliskiren, enalapril, and enalaprilat. All assays were validated at minimum for intrarun and interrun accuracy and precision according to EMA guidelines on bioanalytic method validation for ligand-binding assays and immunoassays.34 

Although blood volumes required for reliable determination of angiotensin I by commercial assays range from 20 to 500 μL of plasma, an assay (75 μL) with a broader range, instead of lowest possible volume, was chosen. This property allows the assay to be applied to all age groups because PRA varies significantly between adults and children.35 Angiotensin I and PRA were both determined with a commercial ¹²⁵I RIA (Immunotech, Prague, Czech Republic). The assay's intraday and interday precisions were 10.2% and 8.2% (coefficient of variance [CV]) with an angiotensin I calibration range of 0.2 to 30 ng/mL and were determined by 10 different runs. Calculation of the PRA was performed using equation 1:

where Ang I is angiotensin I, Conc is concentration, and PRA is plasma renin activity.

An ELISA was used to analyze prorenin concentration in 100 μL of plasma within a calibration range of 0.02 to 10 ng/mL. This assay required the lowest sample volume among those evaluated, detected prorenin directly, and had a broad calibration range (Molecular Innovations, Novi, USA). The CV was 8.2%, with an accuracy of 102.5% ± 8.4% (mean ± SD) and was determined by 10 different runs. The intraday precision was evaluated on 9 different days with a CV of 5.2%.

By contrast, commercial ELISAs were all inappropriate for precisely determining angiotensin II concentrations in low volumes. Therefore, the assay was optimized by combining SPE (in-house) and a modified ELISA (IBL, Hamburg, Germany). The assay required 500 μL of plasma, whereas common ELISAs required about 1 to 2 mL. The purification was performed on Oasis HLB 96-well plates, which allowed angiotensin II concentrations to be reliably determined with a calibration range of 1 to 125 pg/mL. The LLOQ was as small as possible because a decline in angiotensin II was expected with both drugs. The assay had an accuracy of 95.5% to 106.4% and an interday precision of 6.7% (CV), which was determined by 7 different runs on 4 different days.

For renin determination, effort was successfully made to halve the required plasma volume of the chemiluminescence immunoassay (CLIA) used (Diasorin, Saluggia, Italy). Renin was evaluated using 250 μL plasma with a LLOQ of 2.5 micro international units/mL and a calibration range of 2.53 to 500 microunits/mL. The intraday and interday precisions were 2.5% and 3.9% (CV). Both were evaluated by 10 different runs.

Angiotensin-converting enzyme (ACE) activity was assessed by determining the ratio of plasma angiotensin II to angiotensin I concentrations in blood. The change in activity was a percentage and was calculated according to the equation 2:

where Ang I and Ang II are angiotensin I and II, respectively, and t is the timepoint for determining angiotensin concentration.

The pharmacokinetics of the compounds in serum were characterized by maximum serum concentration (C_max), time to maximum concentration (T_max), area under the curve from zero to timepoint t (AUC _0–t), AUC from zero to infinity (AUC_0–∞), half-life (t½), elimination rate constant (k_e), and oral clearance of a drug (CL_F). The AUC was calculated by applying the trapezoidal rule with infinity extrapolation. The oral clearance was determined by dividing the administered dose by the AUC_0–∞. The dose was adopted for enalaprilat according to the molar equivalent of enalapril maleate. Furthermore, the conversion ratio of enalapril and enalaprilat was calculated using the ratio of the AUC_0–∞ of enalaprilat to the AUC_0–∞ of enalapril. The apparent volume of distribution (Vd_F) was determined by a non-compartmental analysis as follows: Vd_F = CL_F/k_e, whereas CL_F was defined as Dose/AUC_0–∞. In urine, the unchanged amount of the compounds excreted into this fluid (A_e) and the renal clearance (CL_R) were evaluated. Saliva concentrations were examined for C_max, T_max, AUC_0–t, and the corresponding ratio between serum and saliva. Pharmacokinetic parameters were determined by non-compartmental analysis using a Phoenix 6.3.0.395 (Certara, St. Louis, MO). For humoral parameters, C_max and T_max were analyzed. Because the primary objective of this study was not to provide new insights into the pharmacokinetics of the 3 drugs in adults, but rather to ensure the appropriateness of the low-volume analytic method and to assess the agreement of the results with those reported in the literature, a non-compartmental analysis, rather than a compartmental analysis, was used.

The patient's health status was evaluated in personal interviews at least 3 days before and on day1 of the study. Based on the interview, only subjects who were 18 years and older with a body weight greater than 50 kg and with no known organ diseases were allowed to participate. In addition, subjects with angioedema, urticaria, known allergies, or low BP values in the past (<90/60 mm Hg) were excluded for safety reasons. The hypotensive value limit was defined according to the recommendations of the National Heart, Lung, and Blood Institute. Before drug administration on day 1, all female subjects were asked to perform a pregnancy test. For the duration of the study, BP and heart rate measurements were conducted regularly. The BP was measured after at least 5minutes of rest in a supine position using an automatic upper arm oscillometric BP monitor (Omron M5, Omron, Mannheim, Germany). Adverse events were also monitored, their duration was recorded, and, if necessary, any concomitant drug given was documented.

All 22 volunteers enrolled were white and healthy. A group of 13 volunteers (6 males, 7 females), aged 19 to 31 years, were enrolled to the aliskiren group, and 6 men and 3 women (aged 19–30 years) were enrolled to the enalapril group. Baseline characteristics for all participants are shown in Table 1.

Aliskiren, enalapril, and enalaprilat were investigated in serum, urine, and saliva by highly dense sampling of the drug concentrations in the corresponding biological fluid for 5 elimination half-lives of the drugs. Data are presented as means ± SD.

Eleven of 13 subjects had double peaks of aliskiren in serum. The extent and timepoint of the occurrence of the double peak differed among subjects. However, the first peak occurred with a median T_max of 0.7 hours and reached a lower maximum concentration compared with the second peak. The T_max of the second peak ranged between 0.8 and 4.3 hours, with the median of 2.7 hours. Individual concentration-time profiles are arranged in Figure 2. The mean AUC_0–∞ of aliskiren was 2076.1 ± 1340.9 ng × hr/mL with a C_max of 502.5±503.1 ng/mL. The mean terminal half-life of aliskiren was 75.6 hours. The total amount recovered in urine was 0.7% ± 0.4% of the administered dose. The mean calculated CL_R was 1.18 L. Corresponding urinary profiles are arranged in Figure 3. Pharmacokinetic parameters obtained in serum and urine are listed in Table 2.

The concentration-time profiles of aliskiren in saliva showed 2 different distribution phases over time with a comparable profile shape up to 5 hours, whereas it changed substantially at later timepoints when compared with the serum concentration. The mean C_max for 0 to 9 hours for aliskiren in saliva was 7.1 ± 8.4 ng/mL, whereas the C_max for 0 to 192 hours was 7.3 ± 7.1 ng/mL. The AUC for 0 to 4.9 hours was 11.0 ± 10.5 ng × hr/mL and changed to 352.2 ± 367.4 ng × hr/mL from 0 to 192 hours. The T_max during the investigation period up to 5 hours was 2.7 hours, whereas it changed to 6 hours when the entire study period (192 hours) was considered. The ranged between 0.84 saliva-to-serum ratio of k_e and 1.31 for 0 to 4.9 hours and clearly altered for 0 to 192 hours (0.2–4.4). If compared with serum, in saliva—up to first 5 hours after the mean C_max administration—was about 1.4%, and the AUC reflected 1.5% of the serum concentration. In contrast, the AUC for 0 to 192 hours was about 18.2% of that seen in serum. Comparative individual concentration-time profiles of aliskiren in serum and saliva are presented in Figure 2, and a comparative data set is compiled in Table 3. Table 4 provides individual ratios of AUC, T_max, and C_max in saliva and serum.

After administration of 20 mg enalapril male-ate, the prodrug was rapidly absorbed (mean T_max, 0.9 hours) with maximum concentration of 159.7 ± 67.8 ng/mL. The AUC_0–∞ was 231.6 ng × hr/mL; therefore, it was about half that of the enalaprilat. Calculated mean elimination half-life for enalapril was 0.9 hour. In total, 17% of enalapril was found in urine after 72 hours, with a renal clearance of 15.1 ± 3.5 L/hr. The active metabolite enalaprilat reached a mean C_max of 52.5 ± 25.3 ng/mL after 4 hours. The observed T_max ranged between 2.8 to 4.4 hours. The mean elimination half-life was 10 hours. The mean AUC_0–72h was 491.4 ± 148.0 ng × hr/mL. The conversion rate of enalapril and enalaprilat, calculated by the corresponding AUC_0–∞, ranged from 1.66 to 4.92, with the median of 2.0. The amount of excreted enalaprilat was about 27% of the dose administered. The calculated renal clearance for enalaprilat was 7.5 ± 2.1 L/hr. Figure 4 shows the urinary excretion amount of enalapril and enalaprilat. Pharmacokinetic parameters in serum and urine are listed in Table 2.

The correlations between serum and saliva concentration of enalapril and enalaprilat were investigated in 7 volunteers. In 5 of 7 volunteers, salivary concentration-time profiles were obtained. One subject was excluded because the mouth was not washed after drug administration and thus detected salivary drug concentrations represented a combination of the residual of the tablet itself and the penetration into saliva. The second subject developed gum bleeding, and the concentrations obtained cannot, therefore, safely be accounted for saliva only. The evaluable concentration-time profiles revealed low concentrations of enalapril and in particular very low concentrations of enalaprilat in saliva. For enalapril, the C_max in saliva was 5.5 ± 6.1 ng/mL and varied between 1% and 14% of that found in serum. The median T_max was at 1 hour in both fluids. The AUC _0–72hr of enalapril was 6.6 ± 6.5 ng × hr/mL, and the individual ratios of saliva to serum for that period ranged between 0.01 and 0.1. The determined AUC_{0–72 hr} of enalaprilat was 2.34 ± 1.9 ng × hr/mL, with a mean maximum concentration of 0.3 ± 0.1 ng/mL at 4 hours (T_max). Although visual inspection confirmed a comparable profile shape for the prodrug enalapril in serum and saliva, that cannot be claimed for the active metabolite enalaprilat. The penetration was less than the 1 of that for enalapril and was occasionally below the LLOQ. The evaluated correlation between saliva and serum concentrations was not good because the shape of the concentration time profiles and the corresponding elimination rates were different in both fluids. All individual ratios of AUC, T_max, and C_max are arranged in Table 4. Because of the variable ratios, it is not advisable to apply a matrix factor to the saliva AUC to estimate the serum AUC. The corresponding comparative individual profiles in serum and saliva are compiled in Figure 2. Further comparative pharmacokinetic parameters are arranged in Table 2.

In addition, the closeness of the agreement in the measured concentrations of aliskiren, enalapril, and enalaprilat were assessed with Passing and Bablok regression scatter plots. The corresponding plots for aliskiren, enalapril, and enalaprilat are shown in Figure 5.

Based on published results in adults, and as shown in Figure 1, aliskiren directly blocks the enzyme renin and subsequently affects the PRA, which was confirmed by the study results. Mean PRA was maximally reduced 34-fold with a T_max at 1 hour. Moreover, the PRA did not return to predose values within the observation period. Because of aliskiren administration, we expected the RAA system biomarkers angiotensin I and angiotensin II to be influenced. However, no noticeable change in angiotensin I was observed. All mean predose values were already borderline with or at the LLOQ, and subsequently, no drop in angiotensin I concentrations was seen as a result of aliskiren administration. Because a more-sensitive assay requiring the same small amount of blood was not available, a reanalysis could not be performed. In contrast, the optimized assay was able to measure concentrations of angiotensin II at all time points. The angiotensin II concentrations evaluated had already decreased 30 minutes after administration of aliskiren and reached their minimum around 2 hours postdose. After 48hours, at least the pre-dose values had again been reached, whereas, in the meantime, there was a mean reduction of 70% from the baseline observed. Renin concentration increased approximately 19-fold compared with initial values, with a Tmax at 5 hours. On day 9, the renin in the aliskiren group was still about 4.7-fold higher than it was for the predose value. Renin concentration was highly variable, as indicated by the SD of 434microunits/mL for aliskiren at C_max. Prorenin was only determined in participants taking aliskiren hemifumarate. The mean predose value for prorenin was 0.71 ng/mL and varied between 0.21 and 1.83ng/mL. The mean predose prorenin concentrations were about 45-fold higher than the measured renin concentrations. Within the first 8 hours, the concentrations started to increase slightly and reached their maximum mean concentration 48 hours after dosing with aliskiren (156% of mean predose value).

Enalaprilat, the active metabolite of enalapril, binds competitively to the ACE and blocks the conversion of angiotensin I to angiotensin II. The angiotensin I concentrations rose with enalaprilat and reached maximum concentrations in all 9 subjects in 5 to 7 hours after oral administration of the prodrug enalapril. The mean increase was 3.7-fold when compared with baseline concentrations. In contrast, angiotensin II concentrations were reduced and reached a maximum reduction between 3 and 7 hours postdose. The mean inhibition was 64% of the predose value, showing a broad range of 30% to 95% reduction. When comparing aliskiren and enalapril, the inhibition of angiotensin II synthesis was reduced for 24 to 48 hours with aliskiren, whereas with enalapril treatment, the value was reduced for 8 hours only. The ACE activity was specifically inhibited by enalaprilat and was, therefore, calculated in volunteers taking enalapril only. The activity was maximum reduced by 76% between 3 and 8 hours postdose. The median Tmax was 6 hours. Within 24 to 48 hours postdose, almost all patients had returned to at least predose values. Renin concentrations increased with enalapril approximately 16-fold from initial values, with a T_max of about 6 hours after enalapril administration. The concentrations slowly went back to a 2.7-fold increase by day 3 of the study. The variability in renin concentration is indicated by an SD of 89 microunits/mL (464 microunits/mL if the volunteer with the short-term hypotensive values is included) at the Cmax. With enalapril, the PRA remained nearly unchanged for the first 3 hours after dosing. The activity started to increase at 3 hours after administration and reached a maximum concentration 5 hours postdose. The mean activity was increased by up to 19-fold compared with the mean predose values. Figure 6 represents the mean pharmacokinetics of aliskiren and enalapril/enalaprilat and the corresponding changes in the humoral parameters within the RAA system. Details on the pharmacokinetic parameters of the RAA are listed in Table 5.

Both drugs were generally well tolerated. Three adverse events of dizziness were reported by subjects taking 20 mg enalapril maleate. Subjects who received 300 mg aliskiren hemifumarate also reported 3 adverse events: 2 complained about headaches, 1 about dizziness. All adverse events were mild or moderate and discontinued within 12 hours after the first incidence. No notable changes in BP values were detected in the healthy volunteers of either group. The ratios of the mean systolic and diastolic BP values to heart rate remained nearly unchanged. Details are summarized in Table 6. Nevertheless, 1 subject taking enalapril experienced approximately 30-minutes of hypotensive BP values of 80/44 mm Hg RR. The timing of the hypotensive values occurred around the T_max of the active metabolite Enalaprilat; however, the volunteer was clinically asymptomatic.

In summary, this proof-of-concept study confirmed the applicability of the child-appropriate assays for determining pharmacokinetics of aliskiren, enalapril, and enalaprilat and the corresponding changes in the humoral parameters in the RAA system. These assays reliably quantified blood, urine, and saliva concentrations of both drugs and the corresponding humoral parameters over the course of 5 elimination half-lives.

Tailoring drug assays to determine drug concentrations and biomarkers of the RAA system in small-volume samples fully succeeded with the required blood volumes. Despite the very dense sampling (in total, 40 pharmacokinetic sampling points for 1 pharmacokinetic profile), the tailored assays applied here allowed us to reliably determine the drug concentrations and the subsequent changes in humoral parameters in only 45 mL of blood per subject. Although the drug concentrations were determined in a maximum of 100 μL of serum per sampling point, all 5 humoral parameters of the RAA system were analyzed in 2.1 mL blood. Compared with Sullivan et al9 who investigated only 1 humoral parameter in children and required 2 mL blood, this optimization in sample volume by developing the low-volume assays should be a huge benefit in pediatric pharmacotherapy. By applying this bioanalytic platform, evaluation of pharmacokinetics and pharmacodynamics is even possible in neonates and infants. This novel option facilitates obtaining sufficient data points per child and circumvents the population pharmacokinetic approach with its possible imprecisions. The results obtained by the low-volume assays proved their reliability by comparing them with values in the literature. This legitimizes the chosen bioanalytic settings to be applied for pediatric populations and provides systematic data sets for enalapril/enalaprilat. For aliskiren, all bioanalytic assays were suitable for a precise determination, except the angiotensin I assay. Because of the low angiotensin I baseline in the aliskiren group, the expected decrease in concentration was not detectable. Consequently, its suitability for determining those decreases are not yet possible, and further optimization is required for future investigations.

The findings regarding the pharmacokinetics of aliskiren in adults were in line with those in the literature and reflected high interindividual variability in serum concentration and low urinary excretion. Aliskiren displayed a biphasic absorption behavior in varying degree among all healthy volunteers, that could be precisely described by applying the dense sampling used in this study. Currently, mean concentration-time profiles for Aliskiren are primarily published, which only vaguely suggest this behavior. The reason for this double-peak phenomenon can only be assumed because it has not been previously reported in detail. Comparable behavior with a first small peak followed by a second, much larger peak can be seen with phenazopyridine,36 but the reason for that phenomenon has not been reported either. In the literature, double peaks are generally correlated with enterohepatic recirculation, changes in salt, changes in bile solubility, with the formulation, and with p-glycoprotein, among others. For aliskiren, a change in salt, an additional absorption site along the gastrointestinal tract, or precipitation is likely but needs to be investigated further. The mean variability in AUC of 65% in this study is in line with data reported by Vaidyanathan et al8 who claimed a range between 40% to 70%. Because of the confirmed high intraindividual and interindividual variability, population pharmacokinetics and spare sampling are not advisable for aliskiren. Moreover, the elimination half-life obtained by this study differed from those compared in previous publications. In studies with an investigation period of 96 hours, a half-life of 40 hours was claimed,10 but this study, with its investigation period of 192 hours, revealed a mean elimination half-life of 75.6 hours (median, 65.9 hours). These results should be considered during the planning and conduction of future studies; in particular, the calculation of the wash-out period in crossover studies should be made carefully because aliskiren elimination is much slower than previously assumed. It was possible to determine drug concentrations even 3 weeks after a single oral administration. The other pharmacokinetic parameters, such as C_max and T_max as determined by the low-volume LC-MS/MS assay, were generally comparable to those found by Vaidyanathan et al.10 Likewise, the amount recovered in urine (0.7%) was comparable with to the Waldmeier et al37 data of 0.6% aliskiren. The mean CL_R was similar to that reported by Vaidyanathan et al.38 As mentioned, weeks after administration of aliskiren, drug concentrations were still detectable in concentrated urine samples, which accounts for a more-prolonged elimination phase than previously assumed.

For enalapril, successful implementation of assays to determine serum and urinary concentrations were demonstrated. Mean C_max values of 160 ng/mL were obtained for the prodrug enalapril, as well as 53 ng/mL for the active metabolite enalaprilat, which are comparable to those of Najib et al,23 who reported 124 ng/mL and 54 ng/mL also using LC-MS/MS. Determined elimination half-lives of enalapril (0.9 hour) and enalaprilat (10 hours) were consistent with Najib et al.23 The bioavailability of enalapril is about 50%, which is primarily excreted renally. This study found a mean total renal excretion of 44% of the administered dose, which differs slightly from data reported by Rippley et al,39 who found 50% excretion, and Noormohamad et al,40 who reported 51%. The difference might be explained by the variability in bioavailability of 21% to 61%,40 which, consequently, affects the amount excreted into urine. The amount of enalaprilat in urine was 27%, which is comparable to other reported values of 30% to 36%.39–41 The conversion rate of enalapril and enalaprilat (61%) was within the ratios determined by Till et al42 (67%) and Rippley et al39 (60%). Observed CL_R of enalapril (15 L/hr) and enalaprilat (8 L/hr) were also in line with the literature.43,44 All findings confirmed the suitability of the preparation of urinary samples, avoided extensive degradation, and proved the applicability of the applied low-volume assay.

The replacement of invasive pharmacokinetic sampling using saliva seems to be inappropriate for the drugs investigated. The evaluation by a validated assay did not result in saliva drug profiles comparable to profiles obtained in serum, which prohibits a meaningful interpretation of the data. The aliskiren concentrations obtained were substantially less than the expected 50% penetration from serum to saliva, as calculated based on physicochemical properties. There was a change in the penetration at 5 hours, which persisted for the next days. The shift in penetration rate was observed in nearly all volunteers taking aliskiren and was evaluated by a posteriori analysis. The subsetting for 0 to 5 hour and 0 to 192 hours was performed after visual inspection of the data and was strengthened by the estimations of pharmacokinetic parameters using a non-compartmental analysis. The persistence points toward additional drug penetration from other compartments as the cause of the change because ingestion resulted in a temporary change only. However, a similar finding has not been reported in literature, and the reason for such a shift requires further research. Regarding enalapril and enalaprilat, the expected low penetration rates, based on theoretical assumptions about the unbound fraction in blood and the pk_a values, were confirmed. Nevertheless, although salivary enalapril concentrations are comparable to concentrations in serum, the pharmacokinetics of the active metabolite enalaprilat was not covered sufficiently by saliva when compared with serum. The reason for the poorer penetration of enalaprilat into saliva might be due to the drug's ability to form a zwitterion, which could contribute to insufficient penetration from serum into saliva.

Because of the low drug concentration observed in saliva, some swabs were belatedly and randomly re-eluted with ammonia in acetonitrile/methanol solution. No additional drug substance was obtained by this approach, which suggests that the low concentrations were based on the drug distribution in the body and not on a bad recovery from the swabs. Furthermore, the low concentration cannot be explained as artificial dilution because of stimulated salivation. Only oral swabs (Salivette) without any stimulants (e.g., citric acid) or other adjuvants were used. Because only the free fraction can penetrate into saliva, a probable reason for the low concentrations found might be that the previously reported protein binding of about 50% of the DRI aliskiren was too low. This assumption can be supported by findings in marmosets, in which the binding was 92%.45 

The chosen saliva collection device or method was already successfully used to evaluate cortisol and cytokines in all pediatric age groups. Gunnala et al46 investigated salivary cortisol concentrations in children between 4 weeks and 18 years and found a correlation in critically ill children. Davis et al47 used cotton swabs in combination with the salivette device to collect saliva from 18 premature infants to determine betamethasone exposure. Glover et al48 measured the cortisol concentrations in newborns and infants up to 1 year old with salivettes. Using the same saliva collection device was considered promising for establishing a comparable method for cardiovascular drugs. In summary, we cannot recommend saliva as biological fluid for determining the pharmacokinetics of these 2 drugs, based on our findings.

Both drugs showed marked and sustained inhibition of the RAA system in healthy volunteers. The measured changes in all humoral parameters were in temporal concordance with serum drug concentrations. They confirmed the current knowledge about the RAA system, with the exception of angiotensin I in the aliskiren group, because of the low predose values. Although the first parameters changed within 0.5 to 1 hour after administration of aliskiren, change took about 3 to 4 hours in the enalapril group because of the bioactivation of enalapril to enalaprilat. Although the T_max of aliskiren in this study was about 2.7 hours after administration, most humoral parameters were altered earlier. Aliskiren showed a double peak phenomenon, which suggests that the first peak of the double peak was sufficient to intensively interact with the RAA system. This first peak appeared between 0.3 and 1.5 hours after oral administration. Additionally, this assumption is supported by the aliskiren concentrations of the first peak (162 ng/mL), which exceeds the 50% inhibitory concentration of renin (0.6 nmol/L).49 Even on day 9 of the investigation, the evaluated mean drug concentration in serum was 1.2 ng/mL and, therefore, still accounted for at least 50% inhibition of the target structure.

The detected rise in renin concentration after oral administration of aliskiren and enalapril was comparable to the observations of Nussberger et al.25 They mentioned an up to 20-fold dose-dependent increase after administration of 640 mg aliskiren, whereas this study showed a 19-fold rise in renin following a 300-mg aliskiren administration (maximum approved dose). Furthermore, after administration of 20 mg enalapril, a 16-fold up-regulation of renin was confirmed through the child-appropriate assay. The time to reach the highest mean renin concentration under aliskiren treatment differed between Nussberger et al25 and this study (10 vs. 5 hours). The difference might be explained by the much denser sampling in this study, which allowed a more-precise determination of the T_max. Furthermore, by comparing the results obtained by Nussberger et al25 with those of the current study, the data from Nussberger et al25 showed a C_max of renin with enalapril after 3 hours, whereas, with aliskiren, it was reached 10 hours postdose. With regard to the known pharmacokinetic parameters of both compounds, an earlier C_max with aliskiren and a later one with enalapril treatment would be more rational. This study found the C_max of renin after administration at about the same time postdose (5 hours aliskiren vs 6 hours enalapril). Finally, an additional benefit for future pediatric studies is the low required volume of 250 μL plasma.

The alteration of PRA with enalapril/enalaprilat and aliskiren, respectively, was comparable to that found by Nussberger et al.25 Nevertheless, an appreciable improvement was gained by the reduction in required blood volume for a reliable determination. Although Sullivan et al9 required 2 mL of blood per sampling point, only 400 μL of blood was necessary for reliable determination of PRA in children in this study.

The low-volume assay developed reliably determined angiotensin II concentrations. Because available commercial assays to determine angiotensin II were less sensitive, a combination of high sample purification and sample concentration by SPE and an ELISA with monoclonal antibodies was established to measure the low concentrations adequately. This enabled detection in 500 μL plasma, which is a reduction of up to 75% compared with the required volumes of kits available on the market. The decreased angiotensin II concentrations with enalapril were comparable to the ones reported by Brunner et al.50 Our study found a decline of 64%, whereas the Brunner et al50 results showed a decline of 58%. The slight deviation in angiotensin II concentrations is best explained by the different analytic methods and the high variability observed. The reduction in angiotensin II results with aliskiren of 70% were comparable to those found by Nussberger et al,25 who found about a 70% decline with 160 mg aliskiren.

Finally, this comprehensive data set of drug concentrations and their effects on humoral parameters allow us to validate the physiologic-based computer models for pharmacokinetics and pharmacodynamics. These models are of increasing interest for optimal planning and conduction of pediatric clinical trials. The modeling approach is highly recommended by the EMA and the FDA for sophisticated pediatric research.27,28 The combination of both tools enables high-quality clinical trials to be conducted in future pediatric populations with a maximum of knowledge gained with a minimum blood volume required. Such an approach is particularly useful for future marketing authorization in the pediatric population.

The pharmacokinetics of aliskiren and enalapril and the changes in humoral parameters of the RAA system after a single dose of both drugs were systematically investigated, and the low-volume assays used enabled a comprehensive determination. This proof-of-concept study confirmed the applicability of the low-volume assays for determining drug concentrations and enabled the reliable investigation of pharmacokinetics and pharmacodynamics after administration of aliskiren hemifumarate and enalapril maleate. This comprehensive data set of drug concentrations and their effects on humoral parameters allow us to validate physiologic-based computer models that are of increasing interest for the optimal planning and conduction of pediatric clinical trials. Saliva seemed inappropriate as a replacement for invasive PKa sampling of aliskiren and enalapril. Moreover, these findings mean low-volume assays can be applied in pediatric research, and their applications will allow researchers to increase the important information required on drug exposure and its effect on the RAA system in children.

The research leading to these results received funding from the European Union Seventh Framework Programme (FP7/2007–2013) under grant agreement 602295 (LENA). The authors would like to thank Labor Dr. Spranger (Ingolstadt, Germany) for its contribution during the bioanalysis of the humoral parameters investigated.

ACE

angiotensin converting enzyme

ACEI

angiotensin converting enzyme inhibitor

A_e

amount excreted into urine

Ang I

angiotensin I

Ang II

angiotensin II

AUC

area under the concentration-time curve

BP

blood pressure

CL_F

oral clearance

CL_R

renal clearance

CLIA

chemiluminescence immunoassay

C_max

maximum concentration

CV

coeffcient of variation

DRI

direct renin inhibitor

EDTA

ethylenediaminetetraacetic acid

ELISA

enzyme-linked immunosorbent assay

EMA

European Medicines Agency

FDA

Food and Drug Administration

g

earth's gravitational acceleration

hr

hour

HPLC

high-performance liquid chromatography

k_e

elimination constant

LLOQ

lower limit of quantification

mM

millimolar

mm Hg

millimeters of mercury

MS/MS

tandem mass spectrometer

m/z

mass-to-charge ratio

PRA

plasma renin activity

RAA

renin-angiotensin-aldosterone

RIA

radio immunoassay

SPE

solid-phase extraction

t½

half-life

T_max

time of maximum concentration

Vd_F

apparent volume of distribution