35.07.02 · health-medicine / pharmacology

Pharmacokinetics: ADME (absorption, distribution, metabolism, excretion), half-life, drug interactions

stub3 tiersLean: none

Anchor (Master): Rowland, M. and Tozer, T.N. — Clinical Pharmacokinetics and Pharmacodynamics, 5e (Wolters Kluwer, 2019)

Intuition Beginner

Pharmacokinetics is what the body does to a drug. Every medication takes a four-stage journey, remembered as ADME: absorption, distribution, metabolism, and excretion. Absorption is how a drug enters the bloodstream. An oral tablet must survive stomach acid, cross the intestinal wall, and pass through the liver before any of it reaches the rest of the body — a gauntlet called the first-pass effect. Distribution is where the drug goes once in the blood: some attach to circulating proteins, some leak into tissues, and a few cross the blood-brain barrier to act on the brain.

Metabolism, done mostly by the liver's cytochrome P450 enzymes, chemically transforms drugs into forms that are easier to eliminate. A few drugs, called prodrugs, are inactive until this step converts them — codeine becomes morphine this way. Excretion removes the drug and its breakdown products, usually through the kidneys into urine. The half-life — the time it takes for blood levels to fall by half — sets how often a drug must be dosed. And because the same P450 enzymes handle many drugs, one medication can speed up or slow down another's breakdown, producing a drug interaction.

Grapefruit juice inhibits the CYP3A4 enzyme and raises blood levels of many drugs, sometimes dangerously. Genetic differences in these enzymes — the field called pharmacogenomics — explain why the same dose helps one person and harms another. Someone who breaks down codeine very fast can overdose on a normal dose, while someone who breaks it down poorly gets no pain relief at all.

Visual Beginner

ADME stage	Question	Key organs	What sets it
Absorption	How does the drug get in?	Gut wall, liver (first-pass)	Route, gastric pH, food, transporters
Distribution	Where does it go?	Blood, tissues, brain	Volume of distribution, protein binding, BBB
Metabolism	How is it transformed?	Liver (P450), gut wall	CYP isoforms, induction, inhibition
Excretion	How does it leave?	Kidney, bile	Clearance, urine flow and pH, kidney function

The diagram ties the four ADME stages to a single concentration-versus-time trace. The slope of the trace is set by clearance and volume of distribution together; the point at which it crosses the minimum-effective line determines how soon the next dose is needed.

Worked example Beginner

Worked example: a half-life decay

A patient receives a drug whose half-life is 8 hours. The concentration just after the dose is 64 mg/L.

After 8 hours (one half-life): 64 drops to 32 mg/L.

After 16 hours (two half-lives): 32 drops to 16 mg/L.

After 24 hours (three half-lives): 16 drops to 8 mg/L.

After 32 hours (four half-lives): 8 drops to 4 mg/L.

After 40 hours (five half-lives): 4 drops to 2 mg/L — under 5 percent of the starting amount.

After about five half-lives, so little drug remains that it is considered effectively eliminated. This is why a drug with an 8-hour half-life is dosed several times a day, while a drug with a 40-hour half-life can be taken once daily.

Worked example: why an oral dose is bigger than the IV dose

A drug needs 400 mg to reach the target blood level when given intravenously. By mouth, only half the dose survives the first-pass effect, so its bioavailability is 50 percent (0.5).

To get 400 mg into the systemic circulation, the oral dose must be 400 / 0.5 = 800 mg. The intravenous dose, which bypasses absorption and the liver entirely, stays at 400 mg. Bioavailability is the fraction that actually reaches systemic circulation, and it is the reason oral and IV doses of the same drug often differ.

Check your understanding Beginner

Formal definition Intermediate+

Pharmacokinetics is the quantitative study of the time course of a drug and its metabolites in the body, decomposed into absorption, distribution, metabolism, and excretion. The discipline rests on compartment models (ordinary differential equations describing drug amounts moving between blood, tissues, and eliminating organs), the resulting concentration-time curves, and the derived parameters — clearance, volume of distribution, half-life, and bioavailability — that predict dosing.

Absorption, routes, and bioavailability

Absorption is the movement of drug from the site of administration into systemic circulation. Routes differ sharply in speed and completeness: intravenous (IV) delivers the full dose directly and is by definition 100 percent bioavailable; intramuscular, subcutaneous, transdermal, sublingual, rectal, inhaled, and topical routes each have characteristic onset times and bypass fractions. Oral absorption is the most variable, because the drug must survive gastric pH and digestive enzymes, cross enterocyte membranes (often through transporter proteins such as P-glycoprotein; see 17.02.*), and then survive hepatic first-pass metabolism delivered through the portal vein.

Bioavailability $F$ is the fraction of the administered dose reaching systemic circulation intact, defined for oral versus IV dosing by the ratio of areas under the concentration-time curve:

F = \frac{A U C _{oral}}{A U C _{IV}}

Reduced oral bioavailability reflects incomplete absorption, gut-wall metabolism, hepatic first-pass loss, or degradation in the lumen. Food, gastric pH, gut motility, and transporter polymorphisms all move $F$ , sometimes by an order of magnitude.

Distribution: volume and protein binding

Once in systemic blood, a drug partitions between plasma and tissues according to physicochemical properties (lipophilicity, ionization, molecular size). The volume of distribution $V_{d}$ is the apparent volume that would contain the total body drug at the measured plasma concentration:

V_{d} = \frac{Amount in body}{C _{plasma}}

A $V_{d}$ near plasma volume (3–5 L) signals confinement to blood; a $V_{d}$ of hundreds of litres signals extensive tissue sequestration. Only the free (unbound) drug is pharmacologically active and available for filtration and metabolism. Albumin binding is therefore consequential: a highly bound drug (warfarin, roughly 99 percent bound; see 35.03.02) has a small free fraction, and displacement by a competitor can double or triple the active concentration. The blood-brain barrier — tight endothelial junctions plus P-glycoprotein efflux — excludes most drugs from the central nervous system (see 29.02.*), and the placenta is likewise a selective rather than absolute barrier, which is why teratogens such as thalidomide and isotretinoin reach the fetus (see 18.09.*).

Metabolism: Phase I and Phase II

Drug metabolism is functionally a solubilizing operation — converting lipophilic molecules into polar derivatives the kidney can excrete — and proceeds in two phases. Phase I reactions (oxidation, reduction, hydrolysis) are catalyzed chiefly by the microsomal cytochrome P450 superfamily. CYP3A4 handles roughly half of all clinically used drugs, CYP2D6 about a quarter, and CYP2C9 and CYP2C19 most of the remainder. Phase II conjugation reactions (glucuronidation, sulfation, acetylation, glutathione conjugation; see 17.04.*) attach polar moieties that mark the molecule for biliary or renal excretion.

Some drugs are prodrugs, inactive until metabolism activates them: codeine is O-demethylated by CYP2D6 to morphine, and enalapril is de-esterified to the active diacid enalaprilat (see 35.07.03). Two clinically dominant interaction patterns operate on this machinery. Enzyme induction (rifampin, phenytoin, carbamazepine, St. John's wort) upregulates CYP expression and accelerates clearance of co-administered substrates — the mechanism of oral contraceptive failure. Enzyme inhibition (ketoconazole, erythromycin, grapefruit, ritonavir) has the opposite effect, raising substrate concentrations toward toxicity (see 35.05.03, metabolism of drugs of abuse; 29.10.03, biological treatments in the elderly).

Excretion and clearance

Renal excretion proceeds by glomerular filtration of the free drug, active tubular secretion (organic anion and cation transporters), and pH-dependent passive reabsorption in the distal nephron — the basis of urine alkalinization to trap salicylate in overdose (see 18.08.*). Biliary excretion feeds the gut, and enterohepatic recirculation can prolong half-life substantially when gut bacteria deconjugate and the drug is reabsorbed (see 17.02.*).

Clearance $C L$ is the volume of plasma wholly cleared of drug per unit time, and it is the parameter that sets the maintenance dose rate. Renal drug clearance tracks glomerular filtration rate, which is why creatinine clearance (Cockcroft-Gault) or the estimated GFR guides dose adjustment in kidney disease.

Pharmacokinetic parameters

The four load-bearing parameters are tied together by a small set of identities. Elimination is modeled as first-order (a constant fraction cleared per unit time), giving exponential decay:

C (t) = C_{0} \cdot e^{- k t}

with elimination rate constant $k$ . The half-life is the time for concentration to halve:

t_{1/2} = \frac{ln 2}{k} = \frac{0.693 \cdot V _{d}}{C L}

This identity is nontrivial because it ties two independently measurable quantities — the distribution volume (a partitioning property) and clearance (an elimination property) — to the single clinically observable decay rate, and it predicts that a drug can have a long half-life either by being cleared slowly or by being distributed widely into tissues. At steady state under fixed dosing, intake equals elimination, and concentration stabilizes after roughly four to five half-lives:

C_{ss} = \frac{F \cdot Dose}{C L \cdot τ}

where $τ$ is the dosing interval. The loading dose $V_{d} \cdot C_{target} / F$ fills the volume immediately; the maintenance dose rate $C L \cdot C_{target} / F$ replaces what is cleared. These four formulae are the working toolkit of clinical pharmacokinetics.

First-order versus zero-order kinetics

Most drugs obey first-order elimination: a constant fraction is removed per unit time, so half-life is independent of dose. A few saturate their metabolic pathway at therapeutic concentrations and switch to zero-order kinetics, where a constant amount is removed per unit time. Ethanol and phenytoin are the classic examples (see 37.*, kinetic analysis). Saturation makes concentration rise disproportionately with dose — a small phenytoin dose increment can triple the blood level — and makes the apparent half-life concentration-dependent. The transition between regimes is captured by the Michaelis-Menten form (see the Key result below).

Drug interactions

Drug interactions split into pharmacokinetic interactions (one drug alters the absorption, distribution, metabolism, or excretion of another — the CYP induction and inhibition cases above) and pharmacodynamic interactions (two drugs act at the same receptor or pathway, producing additive, synergistic, or antagonistic effects). Narrow-therapeutic-index drugs — warfarin (see 35.03.02), digoxin, lithium, the anti-epileptics — are the most dangerous hosts for either kind, because small concentration changes move them from ineffective to toxic. Polypharmacy in chronic disease and in the elderly multiplies these risks combinatorially (see 35.03.*, 29.10.03).

Pharmacogenomics

Inherited variation in metabolizing enzymes and drug transporters produces large interindividual differences in blood levels and response. The clinically important loci include CYP2D6 (codeine activation, tamoxifen, many antidepressants), CYP2C19 (clopidogrel activation — poor metabolizers risk stent thrombosis), CYP2C9 and VKORC1 (warfarin dosing), TPMT (thiopurine toxicity), and HLA-B*57:01 (abacavir hypersensitivity). Pharmacogenomics reframes dosing as genotype-conditional (see 35.08.*, genomic medicine; 31.04.03, human variation).

Key result: steady-state accumulation and the half-life-clearance identity Intermediate+

Two quantitative results do most of the clinical work in pharmacokinetics: the geometric accumulation factor that predicts steady state under repeated dosing, and the clearance-volume identity that ties half-life to measurable physiology.

Derivation of the accumulation factor

Consider a drug given as a fixed bolus $C_{0}$ every interval $τ$ , with first-order elimination $C (t) = C_{0} e^{- k t}$ . Define the fraction surviving one dosing interval as $R = e^{- k τ}$ .

Immediately before the second dose, the residual from the first is $C_{0} R$ ; the second dose adds $C_{0}$ , so the second peak is $C_{0} (1 + R)$ . Continuing, the peak after the $n$ -th dose is:

C_{peak, n} = C_{0} (1 + R + R^{2} + \dots + R^{n - 1}) = C_{0} \cdot \frac{1 - R ^{n}}{1 - R}

As $n \to \infty$ (with $0 < R < 1$ ), $R^{n} \to 0$ , and the geometric accumulation converges to the steady-state peak and trough:

C_{peak, ss} = \frac{C _{0}}{1 - R}, C_{trough, ss} = \frac{C _{0} R}{1 - R}

The factor $1/ (1 - R) = 1/ (1 - e^{- k τ})$ is the accumulation factor. It is the reason a drug dosed at intervals close to its half-life accumulates substantially before leveling off, and the reason steady state takes four to five half-lives regardless of dose: at $τ = t_{1/2}$ , $R = 0.5$ , and after four intervals the accumulation has reached $1 - 0. 5^{4} = 93.75$ percent of its asymptote.

The half-life-clearance-volume identity

Combining the definition of first-order clearance $C L = k \cdot V_{d}$ with the half-life definition $k = ln 2/ t_{1/2}$ gives:

t_{1/2} = \frac{0.693 \cdot V _{d}}{C L}

This identity is clinically load-bearing. In renal failure, $C L$ falls and half-life lengthens in direct proportion — the formal reason a renally cleared drug needs dose reduction. In obesity, $V_{d}$ rises for lipophilic drugs and half-life lengthens even with normal clearance. A loading dose is required when the target concentration must be reached before four to five half-lives elapse, as with antiarrhythmics or antibiotics in sepsis.

Nonlinear elimination: Michaelis-Menten

When metabolic capacity saturates, first-order kinetics gives way to capacity-limited elimination:

\frac{d C}{d t} = - \frac{V _{m a x} \cdot C}{K _{m} + C}

For $C ≪ K_{m}$ this reduces to $- (V_{m a x} / K_{m}) \cdot C$ , recovering first-order behavior with $k = V_{m a x} / K_{m}$ . For $C ≫ K_{m}$ it approaches $- V_{m a x}$ : a constant amount removed per unit time, the zero-order regime of ethanol and high-dose phenytoin. Near $C \approx K_{m}$ the system is acutely dose-sensitive, which is the formal reason phenytoin dosing near the therapeutic range is notoriously difficult and demands concentration-guided titration.

These results connect directly to drug-interaction management (CYP induction shifts $V_{m a x}$ and $K_{m}$ ; inhibitors raise substrate concentration into the saturating regime) and to the dose-individualization problems taken up in the Exercises.

Exercises Intermediate+

Exercise 3 (medium, short answer).

A patient on long-term warfarin is started on rifampin for tuberculosis. Several weeks later their INR drops sharply despite no change in warfarin dose. Explain the pharmacokinetic mechanism and the clinical consequence.

Hint

Rifampin is a potent hepatic enzyme inducer. Consider what induction does to warfarin (a CYP2C9 substrate) clearance.

Answer

Rifampin is a potent inducer of hepatic CYP2C9 (and CYP3A4). Induction raises enzyme expression and thereby raises warfarin clearance, lowering its plasma concentration and reducing anticoagulant effect — hence the falling INR and the loss of therapeutic anticoagulation, which risks thrombosis. Management requires substantial warfarin dose escalation (often two- to three-fold) during co-administration and a symmetric down-titration after rifampin is stopped, with frequent INR monitoring throughout. The episode illustrates why inducer and inhibitor interactions on narrow-therapeutic-index drugs demand proactive surveillance rather than reactive correction.

Exercise 4 (medium, short answer).

A patient genotyped as a CYP2D6 ultra-rapid metabolizer is prescribed codeine for post-operative pain. Predict the clinical consequence and explain the mechanism.

Hint

Codeine is a prodrug. Identify the enzyme that converts it to its active form.

Answer

Codeine is a prodrug that relies on CYP2D6 O-demethylation to generate morphine, its active analgesic metabolite. An ultra-rapid metabolizer converts codeine to morphine faster and more completely than a normal metabolizer, producing unexpectedly high morphine concentrations. The clinical consequences range from exaggerated sedation and respiratory depression to fatal overdose, particularly in children and in mothers who are ultra-rapid metabolizers and breastfeed infants also exposed through milk. The case is the canonical argument for CYP2D6 genotyping before codeine, tramadol, or other opioid prodrugs, and for preferring opioids not subject to CYP2D6 activation (morphine, hydromorphone) in ultra-rapid metabolizers.

Exercise 5 (hard, numeric).

A drug follows Michaelis-Menten elimination with $V_{m a x} = 50$ mg/hr and $K_{m} = 4$ mg/L. Compute the elimination rate at $C = 4$ mg/L and at $C = 40$ mg/L, and comment on how the apparent half-life changes between the two concentrations.

Hint

At $C = K_{m}$ the enzyme is half-saturated; at $C ≫ K_{m}$ the enzyme is nearly fully saturated.

Answer

At $C = K_{m} = 4$ mg/L: $\frac{d C}{d t} = - V_{m a x} \cdot 4/ (4 + 4) = - 25$ mg/hr — half-maximal, consistent with the definition of $K_{m}$ .

At $C = 40$ mg/L: $\frac{d C}{d t} = - 50 \cdot 40/ (40 + 4) \approx - 45.5$ mg/hr — near $V_{m a x}$ , the zero-order regime.

Doubling the concentration from 4 to 8 mg/L nearly doubles the rate (still first-order), but doubling from 40 to 80 mg/L barely changes it (saturated). The apparent half-life therefore increases with concentration: elimination that is first-order and predictable at low levels becomes constant-rate and dose-sensitive at high levels. This is why phenytoin, whose $K_{m}$ sits inside its therapeutic range, can produce disproportionate toxicity from a small dose increment and why its dosing must be concentration-guided rather than linear.

Exercise 6 (hard, short answer).

A patient has a measured creatinine clearance of 30 mL/min (normal approximately 100 mL/min). A drug is 80 percent renally cleared and 20 percent hepatically cleared, with no change expected in the hepatic component. Derive the dose-adjustment factor and recommend a dosing-regimen change.

Hint

Scale only the renal fraction of total clearance by the ratio of creatinine clearances, then express total clearance as a fraction of normal.

Answer

The renal fraction scales with creatinine clearance: $0.80 \times (30/100) = 0.24$ of normal. The nonrenal hepatic fraction is unchanged at $0.20$ . Total clearance is therefore $0.24 + 0.20 = 0.44$ of normal, giving a dose-adjustment factor of approximately $0.44$ . The maintenance dose should drop to about 44 percent of standard, achievable either by reducing each dose to 44 percent or (more conveniently) by lengthening the dosing interval by roughly $1/0.44 \approx 2.3$ -fold. Therapeutic drug monitoring is warranted if the drug has a narrow therapeutic index, and the regimen should be revisited if renal function changes further.

Advanced results Master

Compartment models and noncompartmental analysis

The body is not a single well-stirred beaker, and the choice of model is a choice about which simplifications are tolerable. The one-compartment model treats the body as a single uniform volume in which distribution is instantaneous; it is adequate for drugs that equilibrate rapidly and is the basis of the exponential decay used in the Intermediate formalism. The two-compartment model distinguishes a central compartment (blood and well-perfused organs) from a peripheral compartment (muscle, fat), producing a biexponential concentration-time curve: a steep distribution phase as drug leaves blood for tissues, followed by a slower elimination phase. Many drugs — lidocaine, digoxin, the aminoglycosides — require two compartments for honest prediction. Multi-compartment extensions add deep-tissue compartments for drugs with prolonged tissue retention (amiodarone, chloroquine).

Noncompartmental analysis (NCA) sidesteps model selection entirely. From the raw concentration-time curve it extracts the area under the curve $A U C$ (by the trapezoidal rule) and derived moments, from which clearance, mean residence time, and bioavailability follow without assuming any particular compartment structure. NCA is the regulatory standard for bioequivalence testing of generic drugs (the 80–125 percent confidence-interval rule on $A U C$ and $C_{m a x}$ ; see 30.06.*).

Population pharmacokinetics and physiologically-based modeling

Individual pharmacokinetics varies with age, weight, organ function, genetics, and co-medications. Population PK (Sheiner's NONMEM framework; see 29.01.03) uses mixed-effects statistics to estimate both fixed-effect covariate relationships and the random inter-individual residual, yielding both a population model and a Bayesian posterior estimate of each patient's parameters from sparse monitoring samples. This is the engine behind individualized dosing of vancomycin, anti-epileptics, and immunosuppressants.

Physiologically-based pharmacokinetic (PBPK) models replace abstract compartments with anatomically faithful organs connected by blood flow, each with drug-specific partition coefficients. PBPK integrates in vitro metabolism data (recombinant CYP kinetics), tissue composition, and physiology to predict concentration-time profiles a priori — the basis of modern pediatric and pregnancy dosing extrapolations, drug-drug interaction prediction, and first-in-human dose selection (see 43.*, numerical methods for ODE systems; 33.07.*, computing and simulation). Bayesian estimation, population PK, and PBPK increasingly converge into model-informed precision dosing.

Pharmacokinetics in special populations

Pediatric pharmacokinetics is not adult pharmacokinetics shrunk by weight: the neonatal liver has immature CYP expression and the neonatal kidney has low glomerular filtration, so half-lives are longer than in adults during the first weeks of life; by infancy, hepatic clearance per kilogram often exceeds the adult rate, requiring higher weight-adjusted doses (see 29.06.*). Geriatric pharmacokinetics is shaped by declining renal function (often masked by a normal serum creatinine because muscle mass has also fallen), reduced hepatic blood flow, altered body composition, and the polypharmacy of chronic disease (see 29.06.04). Pregnancy increases GFR and plasma volume, changes protein binding, and raises cardiac output, often raising clearance enough to risk undertreatment of epilepsy or HIV; the placenta is a selective rather than absolute barrier, so teratogenic potential constrains every choice (see 18.09.*). Hepatic impairment (graded by Child-Pugh) and critical illness (altered perfusion, edema, third-spacing) each demand their own dose reconsideration (see 35.03.02).

Therapeutic drug monitoring

For drugs with a narrow therapeutic index and a concentration-response relationship that varies between patients, dosing to a target plasma concentration outperforms dosing to a population average. The classical TDM drugs — digoxin, lithium, vancomycin, the aminoglycosides, phenytoin, the calcineurin inhibitors, and many anti-epileptics — share this profile. Bayesian feedback methods combine a population prior with measured trough (or peak-and-trough) samples to estimate individual clearance and recommend the next dose, formalizing what experienced clinicians once did by intuition (see 35.07.*, 35.08.03).

Drug development: ADME, microdosing, and bioequivalence

ADME studies run throughout drug development, beginning with in vitro CYP panels and permeability assays in discovery, advancing to radiolabelled human ADME studies and thorough QT and drug-drug interaction trials in later phases (see 33.04.*, drug design; 35.07.03, drug classes). Microdosing (Phase 0) administers sub-pharmacological doses to obtain human pharmacokinetics earlier and more cheaply than a classical Phase I, trading the certainty of pharmacological dosing for speed (see 35.02.04, clinical trial design). For generics and biosimilars, the regulatory question is bioequivalence rather than re-demonstration of efficacy: a generic must deliver $A U C$ and $C_{m a x}$ within 80–125 percent of the reference product, a criterion that is adequate for most small molecules but contested for narrow-therapeutic-index drugs and for biologics, where biosimilars face a more demanding characterization (see 35.03.03).

Pharmacogenomics at scale and model-informed precision dosing

Preemptive genotyping — sequencing CYP2D6, CYP2C19, CYP2C9, HLA-B, DPYD, TPMT, and related loci once, before any prescription — converts pharmacogenomics from a reactive test to an electronic-health-record resident decision-support tool. The CPIC (Clinical Pharmacogenetics Implementation Consortium) guidelines translate genotypes into actionable dose recommendations for dozens of drug-gene pairs; the limiting factor is implementation, not evidence. Combined with Bayesian TDM and PBPK, genotype-guided prescribing defines model-informed precision dosing (see 35.08.03).

Global pharmacology and the pharmacokinetics of access

Pharmacokinetic principles shape global drug policy in ways that extend beyond individual dosing. The WHO Essential Medicines List encodes a global minimum standard of access (see 31.06.*). Counterfeit and substandard drugs — common in low- and middle-income markets — deliver bioavailability far from the labeled value, producing both treatment failure and resistance selection (see 30.07.03, 30.06.*). Traditional and herbal medicines carry active constituents that induce or inhibit CYP enzymes (St. John's wort the textbook case), generating herb-drug interactions that ethnopharmacology documents and clinical practice often ignores (see 31.06.02). Antibiotic environmental contamination selects resistance through sub-therapeutic exposure of environmental bacteria — a pharmacokinetic phenomenon acting at population scale (see 35.07.01, drug ethics).

Connections Master

Foundation-of pharmacology and therapeutics

Pharmacokinetics is the foundational substrate on which the rest of clinical pharmacology rests (see 35.07.01): every statement about efficacy, toxicity, dosing interval, or drug interaction presupposes a quantitative model of absorption, distribution, metabolism, and excretion. The unit generalises directly into the classification of drug classes by mechanism and clinical use (see 35.07.03), where ADME properties — first-pass vulnerability, half-life, CYP substrate status — explain why drugs in the same class are dosed differently.

Membrane transport and metabolism

P-glycoprotein, the OAT/OCT transporters, and other membrane carriers are the molecular substrate of absorption and excretion (see 17.02.*, membrane proteins); the Phase I and Phase II enzyme families are the substrate of metabolism and detoxification (see 17.04.*). The recurrence is structural: the same transporters and enzymes reappear in gut, kidney, liver, and blood-brain barrier, so a single polymorphism or interaction can perturb absorption, distribution, and excretion simultaneously.

Renal physiology and the basis of clearance

Clearance is, operationally, a renal-physiology concept exported to the whole body. Glomerular filtration, tubular secretion, and pH-dependent reabsorption in the distal nephron define both drug elimination and the creatinine-clearance estimate that guides dose adjustment (see 18.08.*). The Cockcroft-Gault and eGFR formulae bridge nephrology and pharmacology.

Development, aging, and the placenta

The pharmacokinetic profile is not constant across the lifespan. Neonatal enzyme immaturity, infant hypermetabolism, geriatric renal decline, pregnancy-related hemodynamic change, and placental transfer of teratogens each impose their own dosing logic (see 29.06.*, 29.06.04, 18.09.*). The recurrence across life stages is the bridge between pharmacology and developmental physiology.

Neuroscience and the blood-brain barrier

The blood-brain barrier — tight endothelial junctions reinforced by P-glycoprotein efflux — is the limiting distribution problem for central nervous system pharmacology. It explains why first-generation antihistamines sedate and second-generation ones do not, why many antibiotics fail in meningitis, and why CNS drug design must engineer BBB penetration as a first-class constraint (see 29.02.*).

Cardiovascular pharmacology and the narrow therapeutic window

Warfarin, digoxin, and the antiarrhythmics are the canonical narrow-therapeutic-index drugs, and their dosing is where pharmacokinetics, pharmacogenomics (CYP2C9 and VKORC1 for warfarin), and drug interactions converge most dangerously (see 35.03.02). The same logic governs clopidogrel activation by CYP2C19 in coronary stent patients.

Addiction, mental health, and drug interactions in the elderly

Substance-use pharmacology rests on the same metabolic machinery: opioid prodrug activation, alcohol's zero-order kinetics, and the CYP interactions of psychiatric polypharmacy (see 35.05.03). Geriatric psychopharmacology layers age-related clearance decline onto polypharmacy, making the elderly the population at greatest risk of interaction-mediated harm (see 29.10.03).

Genomic medicine, human variation, and precision dosing

Pharmacogenomics is the operational arm of precision medicine: inherited variation in metabolizing enzymes, transporters, and targets makes genotype a covariate in every dosing equation (see 35.08.*, 31.04.03). The dual recurrence — population structure in CYP2D6 copy-number variation, individual genotype in dose selection — ties pharmacokinetics to both population genetics and clinical decision support.

Numerical analysis, probability, and simulation

Compartment ODEs, the trapezoidal AUC, mixed-effects population models, and PBPK simulation draw on numerical analysis (see 43.*), probabilistic and Bayesian reasoning (see 37.*, 29.01.03), and scientific computing (see 33.07.*). The concentration-time curve is, mathematically, the trajectory of a dynamical system whose estimation is a problem in inverse modelling — the same formal structure that appears in 38.* (dynamics).

Epidemiology, clinical trials, and bioequivalence

The clinical-trial framework that generates pharmacokinetic evidence (first-in-human, thorough drug-drug interaction, bioequivalence) is the domain of epidemiology and trial design (see 35.02.04). The 80–125 percent bioequivalence rule is a statistical decision procedure as much as a pharmacokinetic one, and the debate over narrow-therapeutic-index generics is a debate about when average bioequivalence fails to guarantee individual equivalence.

Chemistry, drug design, and the future of the field

ADME properties are now designed in, not discovered late: medicinal chemistry optimizes clearance, permeability, and metabolic stability in parallel with target potency (see 33.04.*). RNA therapeutics, antibody-drug conjugates, and biologics extend the ADME framework beyond small molecules — biologics distribute, are metabolized (by proteolysis rather than CYP), and are excreted by distinct pathways, requiring their own pharmacokinetic theory (see 35.08.01, 35.03.03).

Historical and philosophical context Master

The coinage of pharmacokinetics: Dost and the post-war German school

The term Pharmakokinetik was introduced by the German pediatrician Hartmut Dost in his 1953 monograph Der Blutspiegel: Kinetik der Konzentrationsabläufe in der Kreislaufflüssigkeit. Dost framed the discipline as the quantitative study of the time course of drug concentration in blood, derived the basic relationships between dose, volume, and elimination, and argued that pharmacology without kinetics was a science of effects without a science of means. His framing — that the therapeutic action of a drug is conditioned throughout by the rates at which the body absorbs, distributes, and eliminates it — defined the field's vocabulary and its ambition, though his monograph was slow to reach the Anglophone mainstream.

Teorell and the founding compartment analysis

The mathematical origin of the field predates its naming. In 1937 the Swedish physiologist Torsten Teorell published a two-part analysis in Archives Internationales de Pharmacodynamie et de Thérapie modelling the body as a central (blood) compartment exchanging drug with a peripheral (tissue) compartment, with elimination from the central pool. Teorell's two-compartment model produced the biexponential curve that remains the working description of drugs that distribute before they are eliminated, and his treatment of clearance as a flow-limited process anticipated the formal definition used today. The paper was neglected for two decades; its rediscovery in the 1960s established Teorell, with Dost, as a co-founder of the discipline.

Cytochrome P450 and the molecular basis of metabolism

The discovery that drug metabolism is catalyzed by a specific hepatic microsomal enzyme system belongs to Ronald Estabrook, Tsuneo Omura, and Ryo Sato. Omura and Sato's 1964 Journal of Biological Chemistry papers characterized the carbon-monoxide-binding pigment of liver microsomes as a hemoprotein with a reduced-CO difference spectrum peaking at 450 nm — the signature that named it P450. Estabrook, Cooper, and Rosenthal showed in 1963 that this enzyme system catalyzes the C-21 hydroxylation of steroids, establishing P450 as the terminal oxidase of microsomal drug metabolism. The molecular identification of the CYP superfamily, and the realization that a handful of isoforms — CYP3A4, CYP2D6, CYP2C9, CYP2C19 — metabolize the great majority of clinically used drugs, transformed drug-interaction prediction from empirical observation into enzymology.

Pharmacogenetics from Motulsky to Kalow

The idea that genetic variation underlies interindividual differences in drug response was articulated by Arno Motulsky in a 1957 paper (Heredity and the reaction to drugs) and given its name by Friedrich Vogel in 1959 (Moderne Probleme der Humangenetik). Werner Kalow's 1962 monograph Pharmacogenetics: Heredity and the Response to Drugs consolidated the field, drawing on the classical examples of pseudocholinesterase deficiency (prolonged apnea after succinylcholine) and the slow- and fast-acetylator polymorphism of isoniazid. The later identification of CYP2D6 copy-number variation — from poor metabolizers carrying no functional allele to ultra-rapid metabolizers carrying multiple gene duplications — gave the field its most clinically consequential locus and reframed codeine and tamoxifen dosing as genotype-conditional decisions.

The grapefruit effect

The CYP3A4-inhibition interaction that bears its name was discovered by accident. David Bailey and colleagues, investigating the effect of alcohol and grapefruit juice on felodipine in a 1989 interaction study, found that grapefruit (but not orange) juice raised felodipine blood levels several-fold. Their 1991 Lancet report established the effect, traced it to irreversible mechanism-based inhibition of intestinal CYP3A4 by furanocoumarins in grapefruit, and established food-drug interactions as a serious pharmacokinetic concern rather than a curiosity.

Population pharmacokinetics and the Sheiner revolution

Lewis Sheiner's work in the 1970s and 1980s, culminating in the NONMEM software, transformed pharmacokinetics from a discipline of individual two-stage estimation into one of population mixed-effects modelling. Sheiner's contribution was the recognition that sparse, routine clinical monitoring data — one or two trough concentrations per patient across hundreds of patients — contained enough information to estimate both a population model and each individual's parameters jointly, by treating inter-individual variation as a random effect to be estimated rather than as noise to be averaged away. The Bayesian feedback methods that grew from this work underlie modern therapeutic drug monitoring and model-informed precision dosing.

The philosophical content: population averages and the individual patient

Pharmacokinetics encodes a persistent tension in medicine between the population average and the individual patient. A standard dose is a compromise optimal for no one, designed for a representative member of a population that does not contain the specific patient in front of the clinician. Clearance, volume, CYP expression, renal function, and genotype each move an individual away from the population mean, and the discipline's quantitative apparatus exists to turn that movement into a dose recommendation. The shift from dosing by population average to dosing by individual estimate — through TDM, pharmacogenomics, and Bayesian estimation — is the technical expression of a deeper commitment: that therapy is calibrated to the person, not the demographic.

Bibliography Master

Katzung, B. G. (ed.) (2021). Basic and Clinical Pharmacology (15th ed.). McGraw-Hill. Units I-II, pharmacokinetics, drug absorption, distribution, metabolism, elimination, and pharmacogenomics. ^{[source pending]}
Rowland, M. and Tozer, T. N. (2019). Clinical Pharmacokinetics and Pharmacodynamics: Concepts and Applications (5th ed.). Wolters Kluwer. Compartment models, clearance, half-life, nonlinear kinetics, steady state, population PK. ^{[source pending]}
Wilkinson, G. R. (2005). "Drug metabolism and variability among patients in drug response." New England Journal of Medicine, 352(21), 2211-2221. Pharmacogenomic basis of interindividual variation in CYP-mediated metabolism. ^{[source pending]}
U.S. Food and Drug Administration. Drug Development and Drug Interactions: Table of Substrates, Inhibitors and Inducers. Center for Drug Evaluation and Research. Regulatory table of CYP3A4/2D6/2C9/2C19/1A2 interactions and labeling guidance. ^{[source pending]}
Dost, F. H. (1953). Der Blutspiegel: Kinetik der Konzentrationsabläufe in der Kreislaufflüssigkeit. Georg Thieme, Leipzig. The monograph that named pharmacokinetics and derived its foundational relationships.
Teorell, T. (1937). "Kinetics of distribution of substances administered to the body, I and II." Archives Internationales de Pharmacodynamie et de Thérapie, 57, 205-240. The founding two-compartment analysis of drug distribution and elimination.
Omura, T. and Sato, R. (1964). "The carbon monoxide-binding pigment of liver microsomes." Journal of Biological Chemistry, 239(7), 2370-2385. The characterization of cytochrome P450.
Estabrook, R. W., Cooper, D. Y. and Rosenthal, O. (1963). "The light reversible carbon monoxide inhibition of the steroid C21-hydroxylase system of the adrenal cortex." Biochemische Zeitschrift, 338, 741-755. P450 as the terminal oxidase of microsomal and adrenal hydroxylation.
Motulsky, A. G. (1957). "Drug reactions, enzymes, and biochemical genetics." Journal of the American Medical Association, 165(7), 835-837. The paper that argued heredity conditions drug response.
Kalow, W. (1962). Pharmacogenetics: Heredity and the Response to Drugs. W. B. Saunders, Philadelphia. The monograph that consolidated pharmacogenetics as a field.
Bailey, D. G., Spence, J. D., Munoz, C. and Arnold, J. M. (1991). "Interaction of citrus juices with felodipine and nifedipine." Lancet, 337(8736), 268-269. The discovery of the grapefruit–CYP3A4 interaction.
Sheiner, L. B., Rosenberg, B. and Melmon, K. L. (1972). "Modelling of individual pharmacokinetics for computer-aided drug dosing." Computers and Biomedical Research, 5(5), 441-459. The founding paper of population mixed-effects pharmacokinetic modelling.

Prerequisites

35.07.01

Used in

35.07.03

Tier anchors

beginner: Katzung, B.G. — Basic and Clinical Pharmacology, 15e (McGraw-Hill, 2021)
intermediate: Katzung, B.G. — Basic and Clinical Pharmacology, 15e (McGraw-Hill, 2021)
master: Rowland, M. and Tozer, T.N. — Clinical Pharmacokinetics and Pharmacodynamics, 5e (Wolters Kluwer, 2019)

References

Katzung, B.G. (ed.) — Basic and Clinical Pharmacology, 15th ed. (McGraw-Hill, 2021) · Units I-II: Pharmacokinetics, drug absorption, distribution, elimination, metabolism, and pharmacogenomics
Rowland, M. and Tozer, T.N. — Clinical Pharmacokinetics and Pharmacodynamics: Concepts and Applications, 5th ed. (Wolters Kluwer, 2019) · Ch. 2-6, 9-11: compartment models, clearance, half-life, nonlinear kinetics, steady state, population PK
Wilkinson, G.R. — Pharmacogenomics of drug-metabolizing enzymes and the mechanistic basis of interindividual variation in drug response · CYP2D6/2C19/2C9 polymorphisms, Phase I/II variation, genotype-guided dosing
U.S. Food and Drug Administration — Drug Development and Drug Interactions: Table of Substrates, Inhibitors and Inducers (FDA, Center for Drug Evaluation and Research) · CYP3A4/2D6/2C9/2C19/1A2 substrate, inhibitor, and inducer tables; labeling guidance

Estimated time

beginner: 15m
intermediate: 30m
master: 45m