Commonly Encountered Issues in Pediatric Drug Dosing.

One of the most prevalent problems in pediatrics is the high incidence of adverse drug reactions associated with the use of off-label or unlicensed drugs. In 2000, it was reported that about 70% of the drugs prescribed to pediatric patients in five European countries are off-label or unlicensed drugs (Conroy et al., 2000). It is not surprising that at the time of approval of new drugs, they have seldom been tested in children, even when there might be pediatric applications. Of the formulas that have been used to scale adult doses to children, rules such as Clark’s body weight and body surface area, which are purely based on allometric scaling (Anderson and Meakin, 2002), are typically used. However, such simple extrapolation methods usually fail to predict pharmacokinetic behavior in younger pediatric patients (Johnson, 2008). This is mainly because of developmental changes in organ function (including the ontogeny of drug disposition pathways) and variations in body composition across the different age ranges of children (Kearns et al., 2003).

Physiologically Based Pharmacokinetic (PBPK) Models in Pediatric Populations.

Development of PBPK models in pediatrics has facilitated the evaluation of pediatric population exposure to drugs and xenobiotics. These models take into consideration drug-related data such as biochemical, demographic, and physiologic data—specifically, the ontogeny of drug disposition and elimination pathways. The models can be used to assist in the extrapolation of in vitro data to predict the in vivo behavior of drugs in any age group, including children (Barrett et al., 2012).

Robust PBPK models require good data, in particular, quantitative data on pediatric drug metabolizing enzymes (cytochrome P450 or UDP-glucuronosyltransferases and transporters). Furthermore, it is not sufficient to take a single snapshot of the childhood drug metabolizing capability because it changes with age and needs to be described as an age-dependent ontogeny function. Hence, measurements of the levels of these proteins over the whole childhood period are necessary for in vitro to in vivo extrapolation (Prasad and Unadkat, 2014).

Drug Transporters.

According to the direction in which transporters flux their drug substrates through membranes, they can be grouped as efflux or uptake transporters. Efflux transporters drive their substrates out of cells, whereas uptake transporters transfer them into cells. Alternatively, transporters may belong to the ATP-binding cassette, solute carrier transporter, or organic solute transporter families.

Liver Transporters.

Liver transporting proteins are crucial factors for the uptake and efflux of various drugs and endogenous substances (Klaassen and Lu, 2008; Klaassen and Aleksunes, 2010). Therefore, they are major determinants of drug efficacy and toxicity; they affect drug concentrations in plasma through their roles in metabolic or biliary clearance (Borst and Elferink, 2002). Figure 1 shows the most important hepatic uptake and efflux transport proteins and their locations in the hepatocytes.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

The most important hepatic uptake and efflux transporters and their locations in the hepatocytes: BCRP, BSEP, MATE1, P-gp, MRP, NTCP, OAT, OATP, OCT, and OSTα/β.

Ontogeny of Liver Transporters.

Although there are some metabolizing enzymes known to be overexpressed in the early days of infant life compared with adulthood (for example, CYP3A7), drug transporter expression is expected to be fairly low at birth because of the observation that infants’ disposition machineries are unable to handle toxic xenobiotics. The liver matures rapidly after birth from simply an organ for formation of blood cells to an organ holding the major metabolism and elimination machinery for drugs and xenobiotics. Therefore, liver transporters maturation is important for the proper flux of xenobiotics across the cells (Cui et al., 2012). Despite improved knowledge of drug transporters in humans, knowledge of the developmental patterns of individual drug transporters remain incomplete, particularly in relation to transporter developmental and ontogenic expression in the pediatric population. This is largely because of the scarcity of pediatric clinical studies in this area (Wei et al., 2014).

In order to determine the ontogenic profile of liver transporters, a literature search was conducted to identify studies dealing with major drug transporters in the liver [i.e., breast cancer resistance protein (BCRP), bile salt export pump (BSEP), multidrug and toxin extrusion protein (MATE), multidrug resistance-associated protein (MRP), Na⁺-taurocholate cotransporting polypeptide (NTCP), organic anion transporter (OAT), organic anion transporting polypeptide (OATP), organic cation transporter (OCT), organic solute transporter (OST), and P-glycoprotein (P-gp)]. A search of relevant publications on the abundance data of the efflux, uptake, and bidirectional transporters was done through PubMed (http://www.ncbi.nlm.nih.gov/pubmed) using the following keyword combinations: “hepatic/liver” plus “MATE1, BSEP, BCRP, MRP2, MRP3, MRP4, MRP6, NTCP, P-gp, OATP1B1, OATP1B3, OATP2B1, OAT2, OCT1, OAT7, OSTα/β” plus “uptake or influx or efflux or flux or transport” plus “abundance, ontogeny, correlation of expression, quantification.” Searches were limited to humans (children: birth to 18 years). Titles and abstracts were reviewed to keep the search centered on the levels of transporter expression in pediatric subjects. The literature was reviewed to assess the development of drug transporter expression with age, and the following data were obtained regarding each of the relevant liver transporter ontogeny.

Hepatic Uptake Transport Proteins.

Hepatic Efflux Transport Proteins.

These may be classified into either canalicular or basolateral efflux transport proteins.

Canalicular Transport Proteins.

The role of canalicular transport proteins is to excrete drugs and their metabolites through the hepatic apical membrane to the bile (canalicular membrane).

Basolateral Efflux Transport Proteins.

These transporters are a major class of export proteins that mediate xenobiotic excretion from the liver into the sinusoidal blood across the basolateral membrane.

OSTα-OSTβ.

OSTα-OSTβ are basolateral hepatic bidirectional liver transporters mediating bile acid flux. The mRNA for both transporters was detected at low levels in pediatric livers (Chen et al., 2008), but there is little reported data on their ontogeny.

Drugs with Known Pediatric Applications that Are Substrates for Transporters.

To appreciate the importance (or lack of relevance) of transporter ontogeny for pediatric drug treatment, it seems natural to assess the overlap between the sets of drugs that are used in pediatric drug treatment and the drugs acting as substrates or modulators of drug transporting proteins. However, as will be described subsequently, this approach might be misleading.

Management of preterm infants and children of all ages often involves using a variety of drugs that are frequently transported by one or more transporter proteins. In order to identify drugs with important uses in pediatric populations that are substrates for liver transporters, a review of the literature was carried out in PubMed using the keywords combination (hepatic or liver) plus (NTCP or OATP or OAT or OCT or P-gp or P-glycoprotein or MDR or BSEP or MRP or BCRP or MATE1 or OST) plus (substrate); only human studies were taken into account and titles/abstracts were looked into for relevant information. References in each report were scrutinized for further sources of published data on drug transporter substrates. The data collected are shown in Table 1. Drugs known to be substrates of liver transporters were then compared with drugs used in pediatrics from the British National Formulary for Children 2014/2015, and the matching drugs are indicated in bold in Table 1.

According to Table 1, there are about 175 drug substrates for liver transporters, 104 of which are of pediatric application. This suggests that about 60% of the drugs prescribed to children may be affected by the function of one or more liver transporters (see Fig. 2). The accuracy of this calculated proportion may, of course, be compromised by the fact that there are many off-label or unlicensed drugs that are used. However, it seems likely that the proportion of off-label drugs that are substrates for transporters is similar to the value obtained for the British National Formulary for Children drugs, and that the calculated value might be taken as a rough estimate for the scale of the relevance of transporters in pediatric drug treatment. Obviously, the involvement of transporters does not necessarily translate to a crucial impact for them in the drug disposition; assessing the significance of transporter involvement is an area that has only started to mature (see the subsequent sections).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

The number of drugs of pediatric application that are substrates for liver transporters.

Based on the increasing number of drugs used in pediatric populations that are substrates of or modulators for liver transporters, it is therefore apparent that studies about developmental changes of these transport proteins should be improved. Most of the available studies on liver transporters have concentrated only on snapshots of gene or protein expression and few have focused on the determination of age-dependent transporter activities (Fattah et al., 2015).

In the next section, the theoretical aspects related to the relative importance of transporters, as opposed to enzymes, with age are discussed. It cannot be assumed that the transporter effect in a certain drug’s disposition (as a prominent determinant, regardless of absolute value) does not vary with age.

Problems Associated with Estimating Drug Transporter Relevance to Pediatric Drugs.

There is a general deficiency of data on the developmental changes in transporters in humans. However, from the available studies it is clear that there is great variability in the developmental scenarios between individual transporters. This is also known for enzymes (Salem et al., 2013). Although several efflux and uptake transporters were found to be expressed in the fetal liver, some transporters show some developmental maturation in expression from fetal to neonatal and adulthood periods such as BSEP, MRP3, MRP6, MATE1, and OAT2, while the expression of other transporters such as MRP2, MDR1, OAT1B1, and OAT1B3 increases from the neonatal period to some point in childhood and then stabilizes at adult levels. On the other hand, the maturation of some transporters is not related to age; these include BCRP, OCT1, and NTCP. These results are broadly in line with a recently published review that assessed the ontogeny of human transporters in intestine, liver, and kidney (Brouwer et al., 2015).

Of the available transporter ontogeny data, some are in the form of mRNA and others are in the form of protein data. Furthermore, most of these data still need to be correlated with activity, and the relative significance of the transporters in the distribution of drugs and the ways in which this significance changes with age requires further investigation. For a transporter to be of relevance to the disposition of a certain drug, the following range of factors should be considered:

• First, consider the availability of this transporter and its abundant expression in the tissue of interest;

• Second, consider the relative contribution from this transporter in drug distribution compared with the contribution from other transporters or from passive diffusion (i.e., the fraction of drug transported, f_T, for each transporter);

• Third, consider the degree of the drug affinity to the transporter and the proportion of the drug transported by this specific transporter; and

• Finally, consider the modulation of this transporter by induction or inhibition through endogenous or exogenous substances.

Drugs may be affected by transporters to varying degrees according to the fraction of drug dose being absorbed, distributed between body tissues, or cleared out of the body. Therefore, the concept of fraction transported is of great importance in the determination of the transporter effect on the concentration of drugs in any organ by estimating the transporter’s role in the absorption, distribution, metabolism, and excretion profile of drugs. When this fraction of drug transported is high, then the transporters influencing this drug and their modulation by drug-drug interactions is very important to the drug absorption, distribution, metabolism, and excretion.

Prasad and Unadkat (2015) summarized the central nervous system affecting drugs and the antiretroviral agents that are at the same time substrates for P-gp, BCRP, or MRP in the blood-brain barrier, together with their f_T values. Drugs with high f_T values (0.67–0.98) are set in italics in Table 1. However, this picture becomes more complex when the dimension of age is added to it, where the fraction of drug transported may not be constant across various age groups.

To indicate the varying relative contribution of different transporters with age, two hypothetical transporters (T1 and T2) were assumed to have relative importance in adults, f_T1 and f_T2 of 0.1 and 0.9, respectively. In case A, both transporters were assumed to be expressed at birth with relative values of 0.6 and 0.5 for T1 and T2, respectively, compared with the adults. In case B, T2 was assumed not to be expressed at birth, while the T1 relative value was the same as in case A. The relative abundance values for each transporter were assumed to follow a different trajectory (Fig. 3B), unlike the first scenario when the ontogeny was similar for both transporters (Fig. 3A). Under these circumstances it can be shown that the relative importance of transporters (f_T) can be age dependent for the scenario described for case B (Fig. 3D) while the relative importance may remain the same in adults despite the ontogeny of transporters in case A (Fig. 3C).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3.

The relative abundance values (A) and (B), and the age-related changes in the relative importance of two transporters T1 (C) and T2 (D) in different age groups. (A) The relative values of T1 and T2 are 0.6 and 0.5, respectively, at birth compared with the adult values. (B) T1 relative value at birth is 0.6 compared with the adult value, while T2 is not expressed at birth. (C) Both T1 and T2 have a constant relative importance across the different age groups. (D) T1 has a higher relative importance than T2 in neonates compared with adults.

This concept is a general case and can be considered not just between two transporters but also in terms of the relative importance of a nontransporter route versus a transporter-related route. The parallels between the aforementioned case with the results of another recent investigation by Salem et al. (submitted manuscript) on the selection of covariates and their age dependence are obvious. This study by Salem et al. (F. Salem, K. Abduljalil, Y. Kamiyama, and A. Rostami-Hodjegan, submitted manuscript) discussed the validity of the commonly known fact that the hepatic extraction ratio, E_H, is a characteristic property of the drug. Salem et al. (F. Salem, K. Abduljalil, Y. Kamiyama, and A. Rostami-Hodjegan, submitted manuscript) concluded that caution should be taken before assuming that the extraction ratio in pediatric populations is the same as in adults because classification of a drug as a high or low extraction ratio does not take into account the variation of the E_H calculation parameters with age. In conclusion, it is clear that the relative ontogeny of transporters and enzymes may not follow the same trajectory and they can differ in their relative importance to a specific drug and its disposition.

Progress with Experimental Methods for the Study of Transporter Expression.

The majority of data about the abundance of drug transporters originates from in vitro methods based on, for example, primary hepatocytes or cell lines with transfected human transporting protein (Hirano et al., 2004; Kitamura et al., 2008). Nevertheless, these in vitro models are not available for the pediatric population. Preclinical animal studies are of value in assessing the ontogenic profile of transporters, and much of that information seems to be in agreement with clinical findings in adults. Nonetheless, most of these data are in the form of transcriptional information limited to the gene expression level and most of the abundance data are obtained through western blot analysis and immunohistochemistry, which suffer from poor reproducibility (Al Feteisi et al., 2015).

Transcript levels (as measured by conventional mRNA methods) have been shown to have only a weak association with the expression levels of proteins. However, liquid chromatography–tandem mass spectrometry experiments can be designed to give sensitive characterization and quantification of proteins. Unlike relative quantification methods that rely on comparing the protein concentrations in two samples relative to each other, absolute quantification is highly recommended to aid in the measurement of the absolute values of proteins in samples, and consequently it facilitates interlaboratory comparisons.

Mass Spectrometry-Based Absolute Quantification.

Because of the high sensitivity and selectivity of mass spectrometry, it is of crucial importance in quantitative analysis of pediatric samples, especially because of the minute size and the limited availability of these samples. Mass spectrometry-based absolute quantification is mainly based on isotope dilution, where a predetermined amount of a heavy and isotope-labeled internal standard is mixed with the analyte protein as a reference. Comparison of the signal intensity of labeled and unlabeled peptides leads to the concentration of the protein of interest.

Internal standards may be isolated peptides labeled with stable isotopes (sometimes known as absolute quantification peptides) (Brun et al., 2007); this technique has been successfully used in the hepatic CYP2D6 quantification in humans (Langenfeld et al., 2009). Alternatively, marker quantotypic peptides of various proteins of interest may be expressed (using an artificial gene) concatenated in an artificial protein and released on proteolytic digestion. This approach (known as the quantification concatemer method) allows the quantification of up to 50 proteins using a single standard (Beynon et al., 2005; Pratt et al., 2006). The robustness of the quantification concatemer approach for simultaneous quantification of a few tens of proteins has been demonstrated by Al-Majdoub et al. (2014), and the approach has been used in the quantification of enzymes and transporters (Russell et al., 2013).

Label-free quantification of transporters and enzymes by liquid chromatography–tandem mass spectrometry is also possible and is especially useful for establishing an initial overview (Kito and Ito, 2008). Labeled standards allow for more direct measurements. They are dependent upon fewer assumptions about the relationship of the peptide signal to protein abundance and are normally preferred for repeated precise measurements.