Cancer Chemotherapy for Pediatric Patients

Absorption

Medications are typically administered via the gastrointestinal tract (mainly orally), intravenously (IV), intramuscularly, and via the skin (subcutaneous or transdermal). Orally administered agents must surmount multiple barriers to reach the systemic circulation ( Figs. 46-2 and 46-3 ). Gastric pH, gastrointestinal motility, transporter-mediated uptake and efflux, and metabolism before a drug enters the systemic circulation play important roles in the absorption of many drugs. Although chemotherapy is rarely administered intramuscularly or transdermally, the lack of muscle and fatty tissue in newborns and infants and their thin skin would cause erratic absorption and increase the risk of toxicity.

Gastric pH is neutral during the first few weeks of life and does not approximate adult values until the age of 2 years, which may affect the bioavailability of compounds. Higher pH delays the absorption of weak acids and increases the absorption of weak bases. Methotrexate absorption, for example, is significantly reduced by coadministration with milk, which may further increase gastric pH. The solubility of tyrosine kinase inhibitors (e.g., dasatinib and nilotinib) is also dependent on pH. Further, pH-dependent drug degradation in the stomach affects the quantity of intact drug that reaches the small intestine. Absorption of lipophilic drugs requires their micellar solubility, which is dependent on biliary acids, whose secretion in turn is affected by food intake. Bile acid secretion is low at birth and increases to the adult level at approximately 6 months.

Gastric emptying time varies with gestational age and is typically longer in premature infants and neonates than in older children. Low gastrointestinal motility persists until 6 to 8 months of age and may cause delayed absorption or increased absorption because of greater contact time. In addition, drugs may be administered in altered forms to small children. For example, tablets may be crushed and added to food or slurries/suspensions and sometimes may be given via nasogastric tube, which may alter their rate and extent of absorption.

Transporter proteins are involved in drug uptake and movement across the intestinal epithelium ( Fig. 46-3 ). Intestinal epithelial cells contain several uptake trans­porters, including one or more members of the organic anion-transporting polypeptide (OATP) family, peptide transporter (PEPT)–1 (PEPT1; solute carrier [SLC]–15A1 [SLC15A1]), ileal apical sodium/bile acid cotransporter (SLC10A2), and monocarboxylic acid transporter 1 (SLC16A1) in their apical (luminal) membranes ( Fig. 46-4, A ). Adenosine triphosphate–binding cassette (ABC) transporters such as ABCB1 (P-glycoprotein, multidrug resistance protein [MDR]-1), ABCC1 (multidrug resistance-associated protein [MRP]–1), ABCC2 (MRP2), and ABCG2 (breast cancer resistance protein [BCRP]) are located on the apical membranes of intestinal epithelial cells, where they secrete substrates back into the intestinal lumen, thereby limiting intestinal absorption and decreasing bioavailability. The basolateral membranes of intestinal epithelial cells contain organic cation transporter–1 (OCT1; SLC22A1), heteromeric organic solute transporter α-β, and ABCC3 (MRP3). Some drugs are eliminated by hepatocytes into the bile before they are metabolized ( Figs. 46-3 and 46-4, B ). Phase I (e.g., cytochrome P [CYP]–450) and phase II (e.g., glutathione S -transferase) drug-metabolizing enzymes in the gastrointestinal tract can metabolize drugs before they enter the systemic circulation (see Fig. 46-3 ). Although the data are limited, some studies have shown that developmental changes in at least some of these enzyme activities and transporter proteins can affect drug absorption.

Volume of Distribution

The volume of distribution (Vd) of a systemically absorbed drug is the extent to which the drug penetrates extravascular tissues (see Fig. 46-2 and Table 46-2 ). Vd is affected by body composition (water and fat) and plasma proteins. Changes in body composition between birth and adolescence can alter pharmacokinetic parameters. The proportions of body water compartments, particularly extracellular fluid volume, change dramatically between birth and adulthood; extracellular fluid volume represents 50% of body weight in premature infants, 35% in infants 4 to 6 months old, and 20% in adolescents and adults. Thus polar drugs, which are distributed primarily in body water, will have a larger volume of distribution in infants than in older children and adults. The net result of an isolated increase in Vd is a lower peak concentration and prolonged terminal half-life, assuming that drug clearance remains unchanged.

Newborns have less skeletal muscle mass and subcutaneous fat than do older infants. Fat components gradually increase after birth, but they differ in boys and girls after the onset of puberty. During adolescence, boys actually lose body fat (reaching a mean of 12% of body weight), whereas girls gain body fat (reaching a mean of 25% of body weight). Therefore gender-related differences in Vd and clearance of lipophilic drugs may be more prominent during adolescence than in preadolescent children or older adults.

Drug distribution may also be affected by plasma proteins. Theoretically, drug displacement from blood components by lower protein binding increases the apparent distribution volume. Although the resulting increase in the free drug fraction may make the drug more available to its target, it also enhances metabolic and renal elimination. Protein binding may be reduced because of persistence of fetal albumin or by low plasma protein content, particularly albumin, α1-acid glycoprotein, and gamma globulin. Although the protein binding of acidic drugs may reach adult levels between 1 and 2 years of age, adult gamma-globulin levels are not reached until age 7 to 12 years.

One unique aspect of infancy is the immaturity of specific organs. For example, the myelin content of the brain is lower in newborns. Because of incomplete maturation of the blood-brain barrier (which reaches adult level by age 3 years), membrane permeability is greater in the infant brain, and the brain-to-plasma ratio of some drugs has been shown to vary with age.

Hepatic Metabolism

The liver plays an important role in drug metabolism (see Fig. 46-2 ). Drug-metabolizing enzymes are divided into two classes: phase I (oxidation, reduction, hydrolysis, or demethylation) and phase II (conjugation or coupling of endogenous molecules, including glucuronide, sulfate, acetate, amino acid, methyl, and glutathione moieties, to the parent drug or a phase I metabolite; Table 46-3 ). Immature organ systems in infants and young children can alter the disposition of many classes of drugs, including antibiotics, anticonvulsants, and antineoplastic agents. Phase I enzymes (e.g., CYP450 microsomal enzymes) are highly expressed in the liver. They can diminish the pharmacologic activity of a drug but can also convert a prodrug to its biologically active form (e.g., cyclophosphamide and codeine). CYP3A4 is the phase I enzyme involved in the metabolism of the largest number of drugs, followed in approximate order by CYP2D6, CYP2C8/9/19, CYP1A2, CYP2B6, CYP2A6, and CYP2E1. CYP3A7 is highly expressed in the fetus and then decreases rapidly after birth. The metabolic activity of other CYP450 enzymes (e.g., CYP3A4, CYP2D6, and CYP2C9) is low in neonates but increases dramatically between age 2 weeks and 3 years. Clearance rates by CYP450 enzymes rise from 20% of that in adults to two to six times that in adults between infancy and adolescence, and then they decline gradually to adult values during puberty. Low hydrolysis rates have been reported for drugs such as procaine in premature infants and neonates, but normal activity is measurable by 12 months. Phase II glucuronidation is almost exclusively a detoxification mechanism that inactivates a compound and facilitates its excretion by enhancing its hydrophilicity and its recognition by biliary canicular efflux proteins. The activity glucuronidation enzyme rises to 30% of adult activity by 3 months of age and then reaches adult levels between 6 and 12 months of age. Esterase activity, which is important for the metabolisms of agents such as irinotecan, gradually increases during the first year of life.

Renal and Hepatic Excretion

The kidney is predominantly responsible for excretion of drugs and their metabolites (see Fig. 46-2 ). The immature kidney has a low glomerular filtration rate (GFR). In full-term infants, the GFR is 40 mL/min/1.73 m ² and varies substantially among individuals. The GFR gradually increases, reaching adult values between age 6 and 12 months. Tubular function and passive resorption may also be significantly lower in the neonate. Because tubular resorption matures more slowly than does glomerular filtration, toddlers may have remarkably high clearance rates for compounds that undergo tubular resorption in older children.

GFR, as assessed by technetium-99m diethylenetriamine pentaacetic acid filtration, was related to body surface area (BSA) in children 2 months to 17 years of age, but GFR was not related to age when normalized to BSA. Premature and term infants require dose adjustments based on GFR measurement or estimation for drugs eliminated primarily by glomerular filtration, such as carboplatin. Formulas are available for estimation of creatinine clearance in infants and older children on the basis of serum creatinine and height or on another endogenous surrogate marker of GFR, such as cystatin C.

Many drugs are excreted in bile as the parent compound or as a metabolite. Biliary excretion favors compounds that have a molecular weight greater than 300 and contain both polar and lipophilic groups. Conjugation, particularly with glucuronate, increases biliary excretion. After the bile empties into the intestine, a fraction of the drug may be reabsorbed and eventually return to the liver (enterohepatic cycling). The remainder is excreted into the feces. Biliary excretion is low at birth and reaches adult levels at approximately 6 months of age.

Many SLC family transporters and ABC family transporters have been identified; these transporters mainly facilitate cellular drug uptake and elimination in kidney and liver (see Fig. 46-4, B and C ). The apical (luminal) membranes of renal proximal tubules contain organic anion transporter (OAT)–4 (OAT4, SLC22A11), urate transporter–1 (SCL22A12), PEPT1 (SLC15A1) and PEPT2 (SLC15A2), ABCC2 and ABCC4 (MRP4), multidrug and toxin extrusion (MATE) protein–1 (MATE1, SLC47A1), ABCB1, organic cation/ergothioneine transporter (OCTN1; SLC22A4), and organic cation/carnitine transporter (OCTN2; SLC22A5). The basolateral membranes of the proximal tubules contain the uptake transporters OATP4C1 (SLCO4C1); OCT2; and OAT1, OAT2 (SLC22A7), and OAT3 (SLC22A8).

Uptake transporters in the hepatocyte basolateral (sinusoidal) membrane include the sodium/taurocholate cotransporting peptide (SLC10A1), three members of the OATP family (OATP1B1 [SLCO1B1], OATP1B3 [SLCO1B3], and OATP2B1 [SLCO2B1]), OAT2 and OAT7 (SLC22A9), and OCT1. Efflux pumps in the hepatocyte basolateral membrane include ABCC1 (MRP1), ABCC3, ABCC4, ABCC5 (MRP5), and ABCC6 (MRP6). Apical (canalicular) efflux pumps of the hepatocyte comprise ABCB1, ABCC2, ABCB3 (transporter associated with antigen processing 2 [TAP2]), ABCB11 (bile salt export pump [BSEP] or sister of P-glycoprotein [SPGP]), ABCG2, and MATE1.

ABCB1 activity in lymphocytes peaks at birth, decreases between ages 0 and 6 months, and stabilizes at the adult level between the ages of 6 months and 2 years. Developmental changes in the functions of other transporters have not been fully defined, but many are under investigation.

Influence of Underlying Disease on Metabolism

Pathophysiologic changes associated with specific pediatric malignancies may alter drug disposition. For example, the clearance of antipyrine and lorazepam was observed to increase between diagnosis and the end of remission induction therapy in children with ALL. The clearance of unbound teniposide is lower in children with relapsed ALL than during first remission. Because leukemic infiltration of the liver is common at the time of diagnosis of ALL, drugs metabolized by the liver may have reduced clearance, as has been documented in preclinical models.

In mouse models, certain tumors elicited an acute-phase response that coincided with downregulation of human hepatic CYP3A4 and of the mouse orthologue Cyp3a11. The reduction of murine hepatic Cyp3a gene expression in tumor-bearing mice resulted in decreased Cyp3a protein expression and consequently a significant reduction of Cyp3a-mediated metabolism of midazolam. These findings support the possibility that tumor-derived inflammation may alter the pharmacokinetic and pharmacodynamic properties of CYP3A4 substrates, leading to reduced drug metabolism in humans.

Unique Drug Effects in Pediatric Patients

Experience has shown that children often tolerate the conventional toxicities of chemotherapy better than adults, and that children (with the exception of neonates) experience very low regimen-related mortality. This finding may be partly explained by relatively low drug exposure caused by higher drug clearance rather than by altered pharmacodynamics. Lower exposure may explain briefer periods of neutropenia, less mucositis, and rare hepatic veno-occlusive disease. Children's greater ability to tolerate intensive therapy may partly explain why children with ALL have a higher cure rate compared with adults.

However, the immature organ systems of the very young may be more susceptible to certain aspects of toxicity, including late cardiomyopathy, neuron developmental disorders, delayed puberty, growth plate arrest, and infertility. Children appear to be more sensitive to the cardiotoxic effects of anthracyclines (e.g., doxorubicin, daunorubicin, idarubicin, and mitoxantrone). Follow-up studies have shown that months or years after intrathecal therapy for central nervous system (CNS) leukemia, with or without cranial irradiation, children treated for ALL may experience a chronic demyelinating encephalopathy and/or cognitive late effects. Further, the latent effects of chemotherapy on fertility appear to vary with the developmental stage of the patient at the time of treatment. Treatment of leukemia and lymphoma in prepubertal boys, for example, does not appear to produce sustained testicular damage unless large cumulative doses of alkylating agents (e.g., cyclophosphamide >6 g/m ² ) are used. Treatment during puberty is more likely to permanently damage the germinal epithelium, although sexual maturation may proceed on a normal schedule. Most prepubertal girls treated for leukemia with antimetabolites achieve menarche and progress through puberty. Ovarian failure is associated mainly with alkylating agents and gonadal radiotherapy and is directly related to cumulative dose and age at exposure. Nearly 30% of survivors treated with alkylating agents and abdominal-pelvic irradiation experienced nonsurgical primary ovarian failure (defined as cessation of menses before age 40 years). Procarbazine exposure at any age, cyclophosphamide exposure from age 13 to 20 years, and ovarian radiation doses greater than 10 Gy were independent risk factors for acute ovarian failure (defined as loss of ovarian function within 5 years after diagnosis). Most girls treated for brain tumors with procarbazine and/or nitrosourea and with neuraxis radiation showed evidence of ovarian hormonal failure.

Combination Chemotherapy

The vast majority of pharmacologic studies of anticancer agents have modeled the effects of a single drug. However, as a consequence of somatic mutations and/or selection of genetically resistant clones, tumor cytotoxicity tends to diminish over subsequent courses of single-agent treatment. Therefore cancer chemotherapy is most frequently given as a combination of drugs that individually have shown activity against the specific cancer to be treated but that have different pharmacologic mechanisms of action. The selection of agents that exert maximal anticancer effects and minimal adverse effects requires a clear understanding of the biochemical, molecular, and pharmacokinetic mechanisms of action and of individual and overlapping toxicities of the agents. The treatment for ALL provides an example of excellence. Most of the drugs used were developed before 1970. However, their dosage and schedule of administration have been optimized on the basis of leukemic-cell biologic features, response to therapy (e.g., minimal residual disease), and patient pharmacodynamic and pharmacogenomic findings. The 5-year survival rate for pediatric ALL has been approximately 90% in recent trials.

Optimal combination chemotherapy theoretically should meet the following criteria:

• Drugs that show single-agent activity should be selected. Because of primary resistance, rates of complete response to a single agent rarely exceed 20%.

• Drugs with different mechanisms of action should be combined. The various classes of anticancer agents have different cellular targets. The use of multiple agents with different mechanisms of action enables independent killing of cells by each agent. Cells resistant to one agent may be sensitive to another drug or drugs in the regimen. Known patterns of cross-resistance must be taken into consideration.

• Drugs with different mechanisms of resistance should be combined. Resistance to many agents may be the result of mutations selected by those agents. However, a single mutational change can also cause resistance to multiple drugs. The number of known mechanisms of resistance is continuously increasing and is partly drug dependent. Because drug-resistant mutants may be present at the time of clinical diagnosis, the earliest possible use of non–cross-resistant drugs is recommended to avoid the selection of double mutants. Adequate cytotoxic doses of drugs must be administered as frequently as possible to achieve maximal killing of sensitive and moderately resistant cells.

• If possible, drugs with different dose-limiting toxicities should be combined, because it is more likely that drugs with nonoverlapping toxicities can be used at the full dosage, optimizing their potential effectiveness.

As noted, multidrug therapy can give rise to clinically important drug-drug interactions, which typically occur when the pharmacokinetic behavior of one drug is altered by the other ( Fig. 46-5 and Table 46-1 ). These interactions are important in the design of drug combinations, because diminished therapeutic efficacy or increased toxicity of one or more of the administered agents can occur. Combinations of drugs may also show pharmacodynamic interactions that cannot be explained by altered pharmacokinetic profiles. Some of these interactions are at the cellular level or are related to the cell cycle. Interactions can be classified as synergistic (i.e., the effect of the combination is greater than the sum of the individual effects), additive (the effect of two drugs is the sum of the effects of each), or antagonistic (the effect of two drugs is less than the sum of their individual effects). Provided that the drugs used are active against a particular disease, knowledge of cellular kinetics can be used to consider the use of non–cell cycle phase-specific agents (e.g., alkylating agents, platinum agents, and anthracyclines), first to reduce tumor bulk and second to recruit slowly dividing cells into active deoxyribonucleic acid (DNA) synthesis. After the latter is achieved, treatment can be continued with cell-cycle phase-specific agents (e.g., methotrexate or fluoropyrimidines), which affect cells mainly during DNA synthesis. Similarly, repeated courses of S-phase–specific drugs, such as cytarabine and methotrexate, are most effective if administered during the rapid rebound recovery of DNA synthesis that follows suppression of DNA synthesis in the previous course. If pharmacokinetic and pharmacodynamic interactions exist, the drug doses and sequence of administration that allow safe administration of combination chemotherapy are typically defined during early clinical trials (i.e., phase I and phase II studies).

Coadministration of Non-Anticancer Drugs

Many prescription and over-the-counter medications are used during chemotherapy to improve quality of life or to treat other diseases, but they may interact with anticancer agents, thus altering their pharmacokinetic characteristics, clinical effectiveness, and/or toxicities. More than 100,000 deaths per year in the United States alone are attributed to drug-drug interactions, placing these interactions between the fourth and sixth leading cause of death. Clearly, all aspects of pharmacokinetics (i.e., absorption, distribution, metabolism, and excretion) may be affected when a drug is given in combination with another drug. The generally narrow therapeutic index of cytotoxic chemotherapy means that drug-drug interactions would predispose some persons to excessive toxicity or inadequate efficacy.