Clinical Pharmacology

Overview

• Cancer therapy spans cytotoxic chemotherapy, hormonal therapy, small-molecule targeted agents, monoclonal antibodies, antibody-drug conjugates, and immunotherapy, often combined.
• Except for immunotherapy, most anticancer drugs have a narrow therapeutic window (small gap between toxic and therapeutic dose), requiring a careful balance of benefit against normal-tissue toxicity.

Principles of cytotoxic chemotherapy

• Three underlying principles:
  • Fractional cell kill hypothesis: a constant proportion (not number) of cells is killed per cycle.
  • A linear relationship between drug administered and cell kill.
  • Goldie-Coldman hypothesis: tumors acquire spontaneous mutations conferring drug resistance.
• These apply best to rapidly growing cancers (leukemias, lymphomas). They justify giving chemo for multiple cycles, at the highest tolerable dose, in combination with agents of different mechanisms and non-overlapping toxicities.
• Toxicity timing: acute (vomiting/diarrhea during or within 24 h) vs delayed (days to weeks). Myelosuppression is the classic dose-limiting delayed toxicity; other delayed and often irreversible toxicities include cardiotoxicity, pulmonary fibrosis, peripheral neuropathy, infertility, and secondary malignancy.
• Potentially curative in some malignancies (e.g., testicular cancer).

Pharmacokinetics (PK): what the body does to the drug

• ADME: absorption, distribution, metabolism, excretion.
• Bioavailability (F): fraction reaching systemic circulation relative to IV (defined as 100% for IV). Reduced by incomplete absorption and by first-pass metabolism (liver or GI tract); IV and sublingual routes bypass first pass; rectal only partially bypasses it (lower rectal veins drain systemically, upper rectal vein drains to the portal circulation). Important now that many targeted agents are oral.
• C_max and T_max: peak concentration and the time it occurs. C_max predicts acute toxicity (infusion reactions, ECG/QT changes, acute BP elevations); T_max predicts when they occur.
• Volume of distribution (Vd): relates amount of drug in the body to measured concentration. Vd ≤ 3 L means confined to plasma; Vd > 50 L means widely distributed to tissues (especially fat). Larger Vd means the drug is more in tissues (often the site of action).
• Clearance (CL): apparent volume of plasma cleared of drug per unit time (L/h or mL/min); elimination rate = CL x concentration; the single most important PK parameter because it drives dose and schedule. A function of renal and hepatic activity.
• Half-life (t½): time for concentration to fall by half; primarily determined by clearance. A drug is considered eliminated after 5 half-lives. Used to set dosing frequency and washout periods (e.g., erlotinib t½ ~36 h, but do not routinely wash out: switch EGFR TKIs without a gap to avoid disease flare (interrupt only for toxicity, individualized). Five-half-life washouts are research-protocol rules, not clinical practice).
• AUC (area under the curve): total systemic exposure over time; predicts response and chronic toxicity. Depends on dose, bioavailability, and clearance. Carboplatin is dosed by target AUC.
• Linear vs nonlinear (Michaelis-Menten): saturable elimination gives disproportionate exposure changes (e.g., phenytoin, 5-FU).
• Steady state: reached after 4 to 5 half-lives of regular dosing.

Pharmacodynamics (PD): what the drug does to the body

• Dose-response: sigmoidal E_max curve; ED50, EC50.
• Therapeutic index: ratio of toxic to therapeutic dose; many cytotoxics have a narrow index.
• Measuring anticancer PD directly is hard (would need serial biopsies/enzyme assays), so actionable somatic mutations serve as predictive biomarkers (PD biomarkers instead show a treatment-induced change, such as reduced target phosphorylation) (measurable in tumor or ctDNA). In some cases plasma PK predicts PD (carboplatin AUC predicts thrombocytopenia).
• Surrogate biomarkers of response/burden: PSA, CA-125, CEA, AFP, BCR-ABL transcript, ctDNA.
• PK/PD modeling (and physiologic models incorporating blood flow and organ volume) links exposure to response and informs trial design.

Drug metabolism and the CYP450 system

• Most metabolism occurs in the liver (and, for oral drugs, the GI tract). Drugs can be substrates, inhibitors, or inducers of a metabolizing enzyme; CYP3A4 is most commonly involved in oral targeted-therapy interactions.
• CYP3A4: major metabolizer of many TKIs and cytotoxics (docetaxel, vinca alkaloids, ibrutinib, venetoclax); paclitaxel is metabolized primarily by CYP2C8 and secondarily by CYP3A4.
• CYP2D6: activates tamoxifen prodrug to endoxifen.
• CYP2C19: PPIs, voriconazole.
• Strong CYP3A4 inhibitors (itraconazole, posaconazole, voriconazole, clarithromycin, grapefruit juice; fluconazole, isavuconazole and erythromycin are moderate, azithromycin minimal) → ↑ TKI/substrate levels and toxicity.
• Strong CYP3A4 inducers (rifampin, phenytoin, carbamazepine, phenobarbital, St. John's wort) → ↓ substrate levels and efficacy.
• Metabolic interactions can change exposure by an order of magnitude, forming toxic or active metabolites; manage by avoiding co-administration or dose adjustment (dose changes are often in the label, e.g., reduce erlotinib with strong CYP3A inhibitors).

Drug interactions: absorption (food and pH)

• Food (especially high-fat) can increase, decrease, or not affect oral bioavailability; the FDA requires a food-effect study for new oral agents. Manage by counseling on timing relative to meals.
• Acid suppression raises gastric pH and impairs absorption of pH-dependent drugs. Duration of effect: antacids ~2 h, H2 antagonists ~12 h, PPIs ≥ 24 h (all can raise pH > 6).
• pH-dependent absorption is not always a class effect: bosutinib, dasatinib, and nilotinib have reduced absorption at elevated gastric pH, whereas imatinib is pH-independent.
• Polyvalent cations in antacids (Mg²+, Ca²+) can form insoluble chelates that limit absorption.
• Practical rule for pH-dependent TKIs: avoid PPIs (dose separation does not overcome their prolonged acid suppression); if acid suppression is needed, stagger an H2RA per label (e.g., erlotinib 10 h after and ≥ 2 h before the H2RA) or separate an antacid by ~2 h.

High-yield drug-drug interactions

• Methotrexate + NSAIDs / PPIs → ↑ MTX levels (acute renal failure risk).
• Warfarin + anticancer agents: fluoropyrimidines (5-FU/cape) markedly raise INR; targeted-agent effects are drug-specific (e.g., dabrafenib lowers warfarin exposure and INR). Review each agent and the anticoagulation indication before switching to a DOAC (DOACs also have CYP3A/P-gp interactions).
• Tamoxifen + strong CYP2D6 inhibitors (paroxetine, fluoxetine, bupropion) → ↓ active endoxifen; weak inhibitors (venlafaxine, citalopram, escitalopram) are preferred; sertraline is a moderate inhibitor (use with caution).
• PPIs + pH-dependent TKIs (erlotinib, dasatinib, pazopanib) → ↓ absorption.
• Allopurinol + 6-MP/azathioprine → ↑↑ toxicity (shared xanthine oxidase); reduce 6-MP dose by ~75%.
• Ibrutinib / venetoclax: strong CYP3A4 substrates; ibrutinib: reduce to 280 mg once daily with a moderate CYP3A inhibitor; venetoclax: reduce by at least 50%; avoid strong inducers. Ibrutinib: avoid strong CYP3A inhibitors (reduced 140 or 70 mg doses allowed with voriconazole or posaconazole). Venetoclax with strong inhibitors: contraindicated during ramp-up in CLL; otherwise reduce to 100 mg (70 mg with posaconazole, standard in AML).

Dose finding and calculations

• BSA (Mosteller): square root of (Ht × Wt / 3600). Some protocols cap at 2.0 m², but ASCO recommends full (uncapped) actual-weight BSA dosing in obese adults.
• Calvert formula (carboplatin): dose = AUC × (GFR + 25). Cap GFR at ~125 (Cockcroft-Gault). Carboplatin AUC predicts thrombocytopenia.
• Body weight options: total, ideal, adjusted, or lean, depending on drug and obesity status.

Therapeutic drug monitoring (TDM)

• Methotrexate (HD): levels at 24, 48, 72 h; leucovorin rescue per nomogram; glucarpidase for severe delayed clearance.
• Busulfan (HSCT): PK-guided dosing.
• 5-FU PK-guided dosing: improves outcomes but not routine in the US.
• Imatinib trough: monitor adherence in CML.
• Vancomycin, aminoglycosides, anticonvulsants, immunosuppressants (cyclosporine, tacrolimus, sirolimus): TDM is standard.

Antidotes for chemotherapy overdose

• Glucarpidase (for IV and intrathecal methotrexate): recombinant bacterial enzyme cleaving MTX into inactive metabolites; > 97% reduction in MTX within 15 min. Give within 48 to 60 h of starting high-dose MTX (defined as ≥ 500 mg/m² over 2 to 36 h). Do not repeat within 48 h. Because leucovorin is also a substrate, separate leucovorin from glucarpidase by at least 2 h; for the first 48 h keep the pre-glucarpidase leucovorin dose, then dose by methotrexate level and continue until methotrexate stays below threshold for at least 3 days. Glucarpidase falsely elevates immunoassay MTX levels for ~48 h (use a chromatographic assay).
• Uridine triacetate (for 5-FU or capecitabine overdose, or early severe toxicity such as cardiomyopathy, altered mental status, mucositis, severe N/V/diarrhea, neutropenia): competes with fluorouridine triphosphate for RNA incorporation. Adult dose 10 g PO every 6 h for 20 doses, starting within 96 h of the end of 5-FU/capecitabine.
• Methylene blue (for ifosfamide-induced encephalopathy): inhibits monoamine oxidase, preventing generation of the neurotoxic chloroacetaldehyde metabolite; 50 mg IV or PO every 4 to 8 h until symptoms resolve. Caution in G6PD deficiency (hemolysis).

Pharmacogenomics

• Study of variability in drug response due to inherited (germline) polymorphisms in drug-metabolizing enzymes or targets, mostly single nucleotide polymorphisms (population frequency ≥ 1%). Genotype-phenotype concordance is incomplete. Guidelines: Clinical Pharmacogenetics Implementation Consortium (CPIC).
• Thiopurines (TPMT and NUDT15): 6-mercaptopurine and 6-thioguanine are S-methylated by TPMT; reduced activity causes severe toxicity. NUDT15 variants (which normally inactivate thioguanine triphosphate) also cause thiopurine toxicity. Testing is routine and drives dose reduction in intermediate/poor metabolizers.
• DPYD (DPD) and fluoropyrimidines: DPD is the rate-limiting enzyme converting 5-FU to dihydrofluorouracil. Reduced-function variants → ↓ clearance, ↑ half-life, and severe myelosuppression, neurotoxicity, and GI toxicity (capecitabine is a 5-FU prodrug and behaves the same). ESMO recommends DPD testing before fluoropyrimidines; In the US, fluorouracil and capecitabine labels now carry a boxed warning (Feb 2026): test for DPYD variants before starting unless immediate treatment is necessary, and avoid in complete DPD deficiency.
• UGT1A1 and irinotecan: SN-38 (active metabolite, topo I inhibitor) is inactivated by glucuronidation via UGT1A1. A TATA-box promoter dinucleotide-repeat polymorphism reduces UGT1A1 expression (the common cause of Gilbert syndrome). Homozygous UGT1A1*28 patients have much lower SN-38 glucuronidation and higher rates of grade 4 to 5 neutropenia and severe diarrhea; the FDA label recommends dose reduction. UGT1A1 is also relevant to sacituzumab govitecan (SN-38 payload).
• CYP2D6 and tamoxifen: tamoxifen is a prodrug activated by CYP2D6 to endoxifen; poor metabolizers may have inadequate activation. Data on survival impact are conflicting, but guidelines suggest an alternative agent (or ovarian suppression + AI) in patients with little/no CYP2D6 activity, and avoiding strong CYP2D6 inhibitors (fluoxetine, paroxetine).
• Barriers to routine testing (except thiopurines and, now, DPYD): limited trial evidence for dose reduction, insurance coverage, lack of alternatives, physician preference, and turnaround time.

Renal and hepatic dose adjustment

• Renal-cleared chemo: cisplatin, methotrexate, carboplatin (Calvert incorporates GFR), bleomycin. Assess renal function before treatment (NCCN, SIOG); Cockcroft-Gault is common but no equation is consensus-preferred, and it was derived in 249 hospitalized men (no women) and does not account for body composition. Renal impairment can also alter absorption, hepatic metabolism, protein binding, and distribution, even for non-renally cleared drugs.
• Hepatically cleared: anthracyclines, taxanes, vinca alkaloids, irinotecan. Hepatic impairment can cause drug accumulation or failure to form an active metabolite, altering both PK and PD.
• Albumin: low albumin → ↑ free fraction of highly protein-bound drugs (e.g., docetaxel).

Special populations

• Geriatrics: usually no significant PK difference from age alone; age is a poor predictor of tolerability. The main issue is renal clearance (renal mass falls ~25 to 30% over a lifetime; GFR ~0.75 mL/min/yr after age 40, though ~1/3 have no change). Dose renally cleared agents cautiously but do not reduce dose based on age alone, except high-dose conditioning regimens. Use a geriatric assessment (G8, CGA); watch polypharmacy.
• Obesity: altered adipose-to-lean ratio increases the Vd of lipid-soluble drugs (fat acts as a depot). Use actual body weight and actual BSA for weight-based dosing (ASCO); underdosing contributes to higher mortality, and full-weight dosing does not increase adverse effects. Exception: high-dose conditioning regimens.
• Pediatric: BSA-based; renal/hepatic immaturity in neonates.

Targeted therapy, monoclonal antibody, ADC, and immunotherapy principles

• Small-molecule targeted agents: small enough to cross membranes and hit intracellular targets (kinases, proteasome, heat-shock proteins). Usually oral and daily (short half-life), CYP450-metabolized with food/drug interactions; can hit multiple receptors (off-target toxicity). Second/third-generation agents improve potency, CNS penetration, or overcome resistance (e.g., osimertinib covers EGFR T790M). Benefit is regimen-specific: several targeted-plus-chemo combinations are standards (e.g., osimertinib + platinum/pemetrexed in EGFR-mutant NSCLC; midostaurin + chemo in FLT3-mutated AML).
• Monoclonal antibodies: large proteins targeting overexpressed/unique tumor antigens (EGFR, VEGF, ERBB2, CD20). Long half-life means slower toxicity reversal; toxicities relate to the target (e.g., cetuximab skin rash from EGFR on normal skin).
• Antibody-drug conjugates (ADCs): antibody + linker + cytotoxic payload. Non-cleavable linkers (e.g., ado-trastuzumab emtansine) rely on lysosomal degradation and maximize plasma stability; cleavable linkers (pH, glutathione, proteases) enable a bystander effect in heterogeneous or low-antigen tumors (e.g., trastuzumab deruxtecan). Payload released early into blood causes off-target effects: peripheral neuropathy, hepatotoxicity, visual changes, cytopenias; infusion reactions occur during and 24 to 48 h after infusion (premedication is product-specific per the individual ADC label (e.g., sacituzumab govitecan: infusion-reaction + antiemetic premed; T-DXd: antiemetic prophylaxis; T-DM1: none routine)).
• Immunotherapy (checkpoint inhibitors): reactivate antitumor immunity rather than targeting cancer cells directly. CTLA-4 (first checkpoint identified; ipilimumab blocks it) and PD-1/PD-L1 inhibitors. Biomarkers: PD-L1 expression, MSI status, tumor mutational burden (pembrolizumab approved for high TMB, Jun 2020), though many biomarker-negative patients still respond. Immune-related adverse events can affect any organ (GI, hepatic, endocrine, pulmonary), with onset at any time during treatment and even months after discontinuation (no fixed upper limit).

Biomarkers for targeted-therapy selection

• Cancer accumulates mutations; most are neutral passengers, but a few driver mutations give a selective advantage and serve as predictive biomarkers. Targeting drivers with small molecules is effective (e.g., sotorasib for KRAS-G12C; there are now multiple EGFR inhibitors for EGFR-mutant NSCLC).
• Resistance commonly develops (e.g., a binding-pocket mutation blocks EGFR TKI binding); next-generation agents bind alternative sites or cover resistance mutations.

Novel clinical trial designs

• Umbrella trial: multiple mutations within one tumor type are matched to different drugs (e.g., MET, PI3K, CDK4/6 in second-line squamous lung cancer).
• Basket trial: one mutation targeted across different tumor types with the same therapy, regardless of tissue of origin (e.g., NCI-MATCH).

Special pharmacology topics

• Antibody-drug conjugates (ADCs): mAb (targeting) + linker + cytotoxic payload. Examples: T-DM1, T-DXd, sacituzumab, datopotamab, mirvetuximab, enfortumab, telisotuzumab.
• Bispecific T-cell engagers (BiTEs): blinatumomab (CD3 × CD19), tarlatamab (CD3 × DLL3), tebentafusp (ImmTAC: soluble TCR recognizing gp100 peptide-HLA-A*02:01, fused to anti-CD3; a T-cell engager, not a conventional BiTE).
• CAR-T: ex vivo modified autologous T-cells expressing a chimeric antigen receptor.
• Radiopharmaceuticals: radioligands 177Lu-DOTATATE and 177Lu-PSMA-617; radium-223 (bone-seeking calcium mimetic, not ligand-targeted); 131I-MIBG (Azedra discontinued in the US in 2024).

High-yield pharmacology pearls

• Calvert formula for carboplatin dosing.
• CYP3A4 interactions (azoles, grapefruit): careful with TKIs.
• DPYD test before any 5-FU/cape (FDA boxed warning).
• UGT1A1 for irinotecan (and sacituzumab).
• Allopurinol + 6-MP/azathioprine: dose-reduce 6-MP by ~75%.
• CYP2D6 inhibitors with tamoxifen.
• PPI + TKI interaction: avoid PPIs with pH-dependent TKIs; follow the individual label: erlotinib permits the H2RA schedule, but dasatinib and pazopanib should avoid H2 blockers and PPIs entirely (antacids may be separated from dasatinib by at least 2 h).
• Glucarpidase for HD-MTX delayed clearance; uridine triacetate for 5-FU/cape overdose; methylene blue for ifosfamide encephalopathy.
• AUC, half-life, and 5 half-lives fundamentals (exposure, washout, steady state).

2026 update: new therapeutic modalities and pharmacogenomics

ADC class pharmacology

• ADCs = mAb + linker + cytotoxic payload. Recent payloads: deruxtecan (Dxd) exatecan-derived topo I (T-DXd, Dato-DXd, HER3-DXd); SN-38 (sacituzumab); MMAE / MMAF auristatins (enfortumab, tisotumab, telisotuzumab, polatuzumab, brentuximab); DM1/DM4 maytansinoids (T-DM1, mirvetuximab); calicheamicin (gemtuzumab, inotuzumab); PBD.
• Bystander effect: membrane-permeable payload (deruxtecan, SN-38) kills antigen-negative neighbors: activity in heterogeneous / low-antigen tumors.
• Class safety: deruxtecan → ILD; enfortumab → hyperglycemia/DKA + SJS/TEN; mirvetuximab / tisotumab → keratopathy; polatuzumab / brentuximab → neuropathy.

Bispecific T-cell engagers (TCEs)

• Blinatumomab (CD3×CD19), tarlatamab (CD3×DLL3, SCLC), tebentafusp (CD3×gp100/HLA-A*02:01, uveal melanoma), teclistamab / elranatamab / talquetamab (MM), epcoritamab / glofitamab (CD3×CD20, B-NHL); non-T-cell-engaging bispecifics: amivantamab (EGFR×MET), zanidatamab (biparatopic HER2), ivonescimab (PD-1×VEGF).
• Step-up dosing + CRS/ICANS prophylaxis for T-cell engagers.

Radioligand / theranostics

• 177Lu-PSMA-617 (Pluvicto): VISION mature; PSMAfore (FDA Mar 2025): pre-taxane mCRPC post-ARPI.
• PSMAddition (Tagawa Lancet 2026): 177Lu-PSMA-617 + ARPI in PSMA+ metastatic hormone-sensitive prostate cancer; FDA Aug 2026 approval.
• 177Lu-DOTATATE (Lutathera): NETTER-2 (Singh Lancet 2024): 1L grade 2/3 GEP-NET, PFS benefit; the FDA Apr 2024 label action extended Lutathera to pediatric patients ≥ 12 yr.

Pharmacogenomics: DPYD as SOC in EU/UK

• DPYD pretest mandate: UK NHS/MHRA (2020) requires DPD testing before all systemic fluoropyrimidines (IV fluorouracil, capecitabine, tegafur), not just high-dose; most of EU followed. Genotype 4 common variants (*2A, *13, D949V, HapB3); intermediate metabolizer → 50% dose; poor → avoid.
• UGT1A1: genotype-based risk relevant for irinotecan and sacituzumab govitecan (SN-38 payload); known reduced activity warrants closer monitoring, but routine pretreatment testing is not a universal label requirement (DPWG recommends it for irinotecan).

Drug interaction updates

• Ibrutinib / venetoclax: strong CYP3A4 substrates: ibrutinib: reduce to 280 mg once daily with a moderate CYP3A inhibitor; venetoclax: reduce by at least 50%; avoid strong inducers; with strong inhibitors, venetoclax is reduced to 100 mg (70 mg with posaconazole) after ramp-up (contraindicated during CLL ramp-up).
• Osimertinib: no dose change with CYP3A inhibitors (advantage vs 1st/2nd-gen EGFR TKIs), but avoid strong CYP3A4 inducers (if unavoidable, increase to 160 mg daily).
• Enfortumab vedotin: hyperglycemia esp with pre-existing diabetes/high BMI: monitor glucose.

Andexanet alfa nuance (ANNEXA-I 2024)

• ANNEXA-I (Connolly NEJM 2024): superior hemostatic efficacy vs usual care in FXa-i-related ICH BUT higher thrombotic events (10.3% vs 5.6%). Andexxa US commercial sales ended Dec 2025 after FDA thromboembolic safety concerns; use 4F-PCC for factor Xa inhibitor reversal where clinically indicated.