Hydroxyurea and asparaginase represent two mechanistically distinct agents that occupy important niches in oncology and hematology: hydroxyurea as the prototype ribonucleotide reductase inhibitor used in myeloproliferative disorders, chronic myeloid leukemia, and sickle cell disease; and asparaginase as a unique enzyme-based cytotoxic that exploits the asparagine auxotrophy of certain leukemic cell populations.1
Hydroxyurea inhibits ribonucleotide reductase (RNR [ribonucleotide reductase]), the rate-limiting enzyme in de novo DNA (deoxyribonucleic acid) synthesis that catalyzes the reduction of ribonucleoside diphosphates to deoxyribonucleoside diphosphates (the precursors of dNTPs [deoxyribonucleoside triphosphates] required for DNA synthesis). RNR is a heterotetramer consisting of two large R1 (large regulatory) subunits and two small R2 (small catalytic) subunits; the R2 subunit contains a stable tyrosyl radical essential for catalytic activity, and hydroxyurea scavenges this radical, producing a stable tyrosyl radical-hydroxyurea complex that inactivates the enzyme. DNA synthesis is blocked, producing S-phase arrest. Cells in S phase at the time of hydroxyurea exposure undergo replication fork stalling and DNA strand breaks, ultimately triggering apoptosis. Because hydroxyurea is S-phase specific and has a short plasma half-life (approximately 2 to 3.5 hours), it can synchronize tumor cell populations at the G1 (gap 1)/S (synthesis phase) boundary when given intermittently, a property exploited in radiosensitization protocols. Hydroxyurea is administered orally and undergoes renal excretion; dose reduction is required when creatinine clearance falls below 60 mL/min.2
The clinical applications of hydroxyurea span hematologic malignancy and non-malignant hematologic disease. In oncology, hydroxyurea is used as cytoreductive therapy in essential thrombocythemia (ET), polycythemia vera (PV), primary myelofibrosis (PMF), and CML (chronic myeloid leukemia) when tyrosine kinase inhibitors are not immediately available or are contraindicated. In sickle cell disease (SCD), hydroxyurea increases fetal hemoglobin (HbF [fetal hemoglobin]) production by reactivating gamma-globin gene expression through mechanisms that are not fully defined but involve DNA (deoxyribonucleic acid) demethylation and histone modification at the gamma-globin promoter; elevated HbF reduces the proportion of HbS (sickle hemoglobin) available for polymerization, decreasing the frequency of vaso-occlusive crises, acute chest syndrome, stroke, and the need for red blood cell transfusions. Hydroxyurea is the most widely used disease-modifying therapy in SCD and is recommended for patients with frequent vaso-occlusive episodes. The most common toxicities are myelosuppression (neutropenia and thrombocytopenia, requiring periodic CBC [complete blood count] monitoring), macrocytosis (a consistent finding that does not require dose reduction), and mild mucositis at high doses. Long-term hydroxyurea use is associated with a small risk of lower extremity leg ulcers, particularly in patients with SCD or myeloproliferative neoplasms.2
Asparaginase is an enzyme that catalyzes the hydrolysis of L-asparagine to L-aspartic acid and ammonia, depleting circulating asparagine concentrations. Most normal mammalian cells synthesize asparagine endogenously through asparagine synthetase (ASNS [asparagine synthetase]) and are relatively protected from asparaginase cytotoxicity. Certain ALL (acute lymphoblastic leukemia) blasts, particularly T-cell ALL, express very low levels of ASNS and are dependent on exogenous asparagine for protein synthesis; depletion of circulating asparagine starves these cells of an essential amino acid, inhibiting ribosomal protein synthesis, inducing ER (endoplasmic reticulum) stress and the unfolded protein response, and ultimately triggering apoptosis. Asparaginase is therefore tumor-selective in a manner unique among conventional cytotoxics: it exploits an enzymatic deficiency of the malignant cells rather than a general property of rapid proliferation. Three formulations are in clinical use: native E. coli-derived L-asparaginase (no longer commercially available in many countries), pegaspargase (PEG [polyethylene glycol]-asparaginase, L-asparaginase conjugated to PEG), and Erwinia asparaginase (derived from Erwinia chrysanthemi, used as an alternative for patients allergic to E. coli-based preparations).3
Pegaspargase is the standard asparaginase preparation in current pediatric and adult ALL protocols. PEGylation reduces immunogenicity of the E. coli asparaginase by shielding antigenic epitopes from the immune system, dramatically extending the plasma half-life from approximately 1.2 days (native asparaginase) to approximately 5.5 to 7 days; this allows less frequent dosing (every 2 to 4 weeks vs every 2 to 3 days for native asparaginase) and reduces the development of neutralizing antibodies. However, hypersensitivity reactions remain the most common clinically significant toxicity, occurring in 10 to 20% of patients at some point during therapy; reactions range from local infusion reactions to systemic anaphylaxis. Silent inactivation of pegaspargase by neutralizing antibodies without clinical hypersensitivity (silent hypersensitivity) is increasingly recognized and should be suspected when asparagine depletion (measured by serum asparagine or anti-asparaginase antibody titers) is inadequate despite apparent clinical tolerance. Beyond hypersensitivity, asparaginase-related toxicities include pancreatitis (1 to 18% incidence; ranges from biochemical elevation to severe acute pancreatitis), coagulopathy (depletion of coagulation factors including fibrinogen and antithrombin III, increasing both thrombotic and hemorrhagic risks), hyperglycemia (from insulin deficiency and insulin resistance related to reduced pancreatic function and reduced insulin secretion), and hepatotoxicity (elevated transaminases and hyperbilirubinemia). Therapeutic drug monitoring through trough asparaginase activity levels is increasingly used to guide dosing and detect silent inactivation.344
Asparaginase depletes not only asparagine but also other plasma proteins synthesized in the liver, including coagulation factors (fibrinogen, factors V, VIII, IX, X, and XI), natural anticoagulants (antithrombin III and protein C), and plasminogen. The resulting hemostatic imbalance favors both thrombosis and hemorrhage depending on which proteins are more severely depleted in a given patient. Central venous sinus thrombosis (CVST) is a particularly serious thrombotic complication, occurring in approximately 1 to 4% of ALL patients treated with asparaginase-containing regimens, presenting with headache, altered consciousness, focal neurological deficits, and seizures. Management involves systemic anticoagulation (unfractionated heparin or low molecular weight heparin), careful monitoring of coagulation parameters, and in cases of fibrinogen below 50 to 100 mg/dL, cryoprecipitate infusion. Antithrombin III supplementation has been used to restore the anticoagulant balance but evidence for definitive benefit is limited.
The immunomodulatory drugs (IMiDs) represent a class of thalidomide derivatives with pleiotropic anti-myeloma and immunomodulatory activity mediated primarily through their binding to the E3 (ubiquitin-protein ligase class 3) ubiquitin ligase adaptor protein CRBN (cereblon), which directs proteasomal degradation of specific transcription factors critical for myeloma cell survival and T-cell suppression. The section on CYP450 (cytochrome P450) drug interactions follows in Section 3.5
Thalidomide was originally developed as a sedative-hypnotic in the 1950s and caused one of the most catastrophic teratogenic disasters in pharmaceutical history when it was administered to pregnant women for morning sickness, producing approximately 10,000 children with severe limb malformations (phocomelia) globally. It was withdrawn from clinical use in 1961. Its anti-myeloma activity was discovered serendipitously in 1999 when Barlogie and colleagues observed responses in refractory multiple myeloma patients, and FDA approval for multiple myeloma followed in 2006. Thalidomide binds to CRBN, a substrate receptor of the CRL4 (cullin-RING ligase 4) E3 ubiquitin ligase complex, and recruits the transcription factors Ikaros (IKZF1 [Ikaros family zinc finger protein 1]) and Aiolos (IKZF3 [Ikaros family zinc finger protein 3]) as neo-substrates for ubiquitylation and proteasomal degradation. Ikaros and Aiolos are transcriptional activators of IRF4 (interferon regulatory factor 4) and MYC (MYC proto-oncogene) in myeloma cells; their degradation reduces myeloma cell proliferation and survival. Additionally, Ikaros and Aiolos are expressed in regulatory T cells and suppress IL-2 (interleukin-2) production; their CRBN-mediated degradation releases T-cell suppression, contributing to the immunostimulatory activity of IMiDs. Thalidomide also inhibits VEGF (vascular endothelial growth factor) and bFGF (basic fibroblast growth factor)-mediated angiogenesis, inhibits TNF-alpha (tumor necrosis factor-alpha) production from monocytes, and has direct anti-inflammatory properties through NF-kappa-B (nuclear factor kappa-B) pathway inhibition.5
Lenalidomide is a thalidomide analog with substantially greater potency for CRBN-mediated Ikaros/Aiolos degradation, greater immunostimulatory activity, less neurotoxicity, and with the additional ability to recruit a third neo-substrate: CK1-alpha (casein kinase 1 alpha), the activity of which is required for survival of del(5q) myelodysplastic syndrome (MDS) cells. This third substrate specificity explains lenalidomide's remarkable and nearly specific activity in lower-risk MDS with isolated deletion 5q (del[5q] MDS), where it produces red blood cell transfusion independence in approximately 67% of patients and complete cytogenetic responses in approximately 45%.6 In multiple myeloma, lenalidomide in combination with dexamethasone (Rd) is a standard regimen for newly diagnosed and relapsed/refractory disease. Lenalidomide is primarily renally excreted and requires dose adjustment for renal impairment; unlike thalidomide, it does not require CYP450 (cytochrome P450) metabolism for clearance. The dose-limiting toxicity is myelosuppression (neutropenia and thrombocytopenia). Lenalidomide significantly increases the risk of venous thromboembolism (VTE), particularly when combined with dexamethasone or doxorubicin; thromboprophylaxis with aspirin (low-risk patients), low molecular weight heparin (LMWH), or warfarin (high-risk patients) is mandatory during lenalidomide-based therapy. Long-term lenalidomide maintenance after autologous stem cell transplantation in myeloma significantly prolongs progression-free survival but is associated with a small but statistically significant increased risk of second primary malignancies (SPMs), particularly hematologic malignancies including MDS and acute leukemia.12
Pomalidomide is a third-generation IMiD with the greatest potency for CRBN-mediated Ikaros/Aiolos degradation and retained anti-myeloma activity in lenalidomide-refractory disease. It is approved for relapsed/refractory multiple myeloma after at least two prior therapies including lenalidomide and a proteasome inhibitor. Like lenalidomide, pomalidomide requires thromboprophylaxis and is teratogenic; unlike lenalidomide, it undergoes extensive CYP1A2 (cytochrome P450 1A2) and CYP3A4 (cytochrome P450 3A4) metabolism, making it subject to drug interactions with CYP1A2 inducers (cigarette smoke, carbamazepine). All three IMiDs are absolute contraindications in pregnancy and are distributed only under REMS (Risk Evaluation and Mitigation Strategy) programs (Thalomid REMS for thalidomide, Revlimid REMS for lenalidomide, Pomalyst REMS for pomalidomide) that require mandatory pregnancy testing, contraception counseling, and prescriber and pharmacy registration to prevent fetal exposure.5
Thalidomide, lenalidomide, and pomalidomide are all absolutely contraindicated in pregnancy, with thalidomide carrying the highest known human teratogenic risk of any pharmaceutical agent — a single dose at a critical developmental window is sufficient to cause severe limb malformations. Every prescriber of IMiDs must be registered in the applicable REMS program, confirm a negative pregnancy test within 10 to 14 days before starting therapy for women of childbearing potential, ensure two simultaneous forms of contraception are in use throughout therapy and for 4 weeks after the last dose, and document monthly pregnancy testing in women of childbearing potential. Male patients must also use condoms throughout therapy and for at least 4 weeks after the last dose because thalidomide is present in semen. These requirements are not institutional policies but federally mandated REMS conditions; prescribing without REMS compliance is unlawful and exposes the prescriber and institution to regulatory sanction.
Drug-drug interactions mediated by the CYP450 (cytochrome P450) enzyme system are among the most clinically consequential pharmacological hazards in oncology. Most cytotoxic drugs have narrow therapeutic indices; modest changes in exposure from CYP (cytochrome P450)-mediated interactions can precipitate life-threatening toxicity or unacceptable loss of efficacy. The most clinically relevant isoforms in oncology are CYP3A4 (cytochrome P450 3A4), CYP2C8 (cytochrome P450 2C8), CYP2D6 (cytochrome P450 2D6), and CYP1A2 (cytochrome P450 1A2).7
CYP3A4 is the most abundant hepatic CYP isoform and metabolizes the largest fraction of clinically used drugs, including nearly all taxanes, vinca alkaloids, imatinib, erlotinib, gefitinib, cyclophosphamide (to its active metabolite 4-hydroxycyclophosphamide), docetaxel, cabazitaxel, and many antiemetics (ondansetron, aprepitant). CYP3A4 inhibitors reduce metabolism of these substrates, increasing plasma exposure and toxicity risk. Major clinical CYP3A4 inhibitors encountered in oncology patients include azole antifungals (fluconazole, itraconazole, voriconazole, posaconazole), macrolide antibiotics (clarithromycin, erythromycin; azithromycin does not inhibit CYP3A4), HIV (human immunodeficiency virus) protease inhibitors (ritonavir, atazanavir, lopinavir), cobicistat (a pharmacokinetic booster used in HIV antiretroviral regimens), and grapefruit juice (which irreversibly inhibits intestinal CYP3A4 via furanocoumarin compounds). CYP3A4 inducers increase metabolism, reducing plasma concentrations and potentially causing treatment failure; major inducers include rifampin (rifampicin), carbamazepine, phenytoin, phenobarbital, and St. John's wort (Hypericum perforatum). St. John's wort is widely used as an over-the-counter supplement for depression and patients do not reliably disclose its use; direct questioning about herbal and dietary supplement use is essential before initiating any CYP3A4 substrate chemotherapy.78
CYP2C8 is the primary metabolic pathway for paclitaxel (converting it to 6-alpha-hydroxypaclitaxel) and also metabolizes imatinib (minor pathway) and several tyrosine kinase inhibitors. The CYP2C8 inhibitor gemfibrozil (a fibrate used for hypertriglyceridemia) produces a clinically significant interaction with paclitaxel that increases paclitaxel AUC (area under the concentration-time curve) by approximately 2-fold in pharmacokinetic studies; gemfibrozil should be discontinued before paclitaxel chemotherapy when possible. Trimethoprim and montelukast are moderate CYP2C8 inhibitors. Rifampin is also a potent CYP2C8 inducer in addition to CYP3A4, and concurrent rifampin use significantly reduces paclitaxel and imatinib exposure. CYP2D6 metabolizes tamoxifen (the active metabolite endoxifen is generated by CYP2D6 from tamoxifen), ondansetron, and codeine. CYP2D6 inhibitors (paroxetine, fluoxetine, bupropion, and cinacalcet) reduce conversion of tamoxifen to endoxifen, potentially reducing the efficacy of adjuvant tamoxifen therapy in hormone receptor-positive breast cancer; paroxetine and fluoxetine are the most potent CYP2D6 inhibitors and should be avoided in patients receiving tamoxifen for breast cancer when alternative antidepressants are available. Venlafaxine, citalopram, and escitalopram have low CYP2D6 inhibitory activity and are preferred antidepressants for patients on tamoxifen.78
Aprepitant and fosaprepitant, the NK1 (neurokinin 1) receptor antagonists used for prevention of chemotherapy-induced nausea and vomiting (CINV), are both moderate CYP3A4 inhibitors and CYP3A4 substrates, creating complex interactions with co-administered chemotherapy. Aprepitant increases exposure of docetaxel, vinorelbine, ifosfamide, and cyclophosphamide when co-administered; dosage adjustment is required in some regimens. Aprepitant also induces CYP2C9 (cytochrome P450 2C9) after completion of the 3-day CINV prophylaxis course, transiently reducing warfarin levels; INR (international normalized ratio) should be monitored in patients on chronic warfarin receiving aprepitant-containing CINV prophylaxis. Netupitant (in fixed combination with palonosetron as NEPA) is a more potent CYP3A4 inhibitor than aprepitant. Palonosetron itself does not inhibit CYP3A4 and is the preferred 5-HT3 (5-hydroxytryptamine type 3) receptor antagonist in patients at high risk for drug interactions.78
The most dangerous CYP interactions in oncology practice are those that dramatically increase exposure to drugs with narrow therapeutic indices. Voriconazole or itraconazole with vincristine: the azole inhibits CYP3A4-mediated vincristine metabolism, causing potentially fatal vincristine neurotoxicity at standard doses; this combination requires dose reduction of vincristine (to 50 to 75%) or substitution of a less potent CYP3A4 inhibitor. Rifampin with imatinib: rifampin induces CYP3A4 and reduces imatinib AUC by approximately 70%, leading to treatment failure; imatinib dose doubling (to 800 mg/day) or substitution of rifampin with an alternative anti-tuberculosis drug is required. Paroxetine or fluoxetine with tamoxifen: endoxifen plasma concentrations fall by 65 to 75%, potentially reducing long-term tamoxifen efficacy; use a CYP2D6-sparing antidepressant instead. These interactions are frequently encountered because oncology patients commonly receive antifungal prophylaxis, anti-tuberculosis therapy, and antidepressants concurrently with chemotherapy.
Chemotherapy-induced peripheral neuropathy (CIPN) is one of the most common and functionally debilitating toxicities of cancer treatment, caused predominantly by taxanes, platinum compounds, vinca alkaloids, bortezomib, and thalidomide. It affects up to 40 to 60% of patients receiving neurotoxic chemotherapy and can persist long after treatment ends, compromising quality of life and limiting the ability to administer full therapeutic doses.9
The pathological mechanisms of CIPN differ between drug classes. Taxane-induced neuropathy results primarily from axonal transport disruption caused by microtubule stabilization; stabilized, non-dynamic microtubules impair the retrograde and anterograde transport of mitochondria, neurotrophic factors, and other cargo along sensory axon lengths, leading to distal axonal degeneration in a length-dependent (stocking-glove) pattern preferentially affecting the longest axons (those innervating the distal extremities). Platinum-induced neuropathy (from oxaliplatin and cisplatin) targets the dorsal root ganglia (DRG) neurons directly; platinum compounds accumulate in DRG cell bodies, forming platinum-DNA (deoxyribonucleic acid) adducts that trigger DRG apoptosis. Oxaliplatin produces two distinct neuropathy syndromes: an acute cold-triggered sensory syndrome (paresthesias and dysesthesias in the hands, feet, and perioral region triggered by cold exposure, occurring in up to 85 to 95% of patients after each infusion, mechanistically distinct from classic CIPN and caused by altered sodium channel kinetics) and a cumulative sensory neuropathy indistinguishable from taxane neuropathy that accumulates over multiple cycles. Bortezomib-induced neuropathy involves direct proteasome inhibition in DRG neurons, impairing protein quality control and triggering apoptosis; subcutaneous administration reduces neuropathy incidence by approximately 50% compared to intravenous bolus administration.91010
Assessment of CIPN for clinical decision-making requires both patient-reported outcomes and objective neurological examination. The CTCAE (Common Terminology Criteria for Adverse Events) grading scale divides CIPN into: Grade 1 (asymptomatic, clinical or diagnostic findings only), Grade 2 (moderate symptoms, limiting instrumental ADL [activities of daily living]), Grade 3 (severe symptoms, limiting self-care ADL), and Grade 4 (life-threatening, urgent intervention required). Patient-reported outcomes instruments such as the FACT/GOG-Ntx (Functional Assessment of Cancer Therapy/Gynecologic Oncology Group Neurotoxicity) subscale and CIPN-20 (the EORTC [European Organisation for Research and Treatment of Cancer] CIPN-specific questionnaire) capture sensory, motor, and autonomic neuropathy symptoms not always apparent on brief neurological examination. The minimum neurological assessment should include documentation of deep tendon reflexes (particularly the ankle jerk, the earliest indicator of length-dependent axonopathy), vibration sense at the distal lower extremities, and two-point discrimination in the fingertips.910
For prevention of CIPN, the evidence base for pharmacological neuroprotection is limited. Calcium/magnesium infusions before and after oxaliplatin were investigated based on theoretical sodium channel modulation but failed to prevent cumulative neuropathy in a randomized trial; they are no longer recommended. Vitamin E, glutamine, glutathione, and acetyl-L-carnitine have not demonstrated consistent benefit in randomized trials and are not recommended as standard prophylaxis. Dose reduction or schedule modification (switching from 3-weekly to weekly paclitaxel, which paradoxically reduces neuropathy severity despite similar cumulative dose in some breast cancer protocols) represents the most consistently effective neuropathy-sparing strategy where oncologically acceptable. For treatment of established CIPN, duloxetine is the only agent with Level I evidence from a randomized controlled trial (the ACCRU Alliance trial) demonstrating significant reduction in CIPN pain and sensory symptoms; a 30 mg daily starting dose titrated to 60 mg daily is the standard approach.11 Other agents including gabapentin, pregabalin, amitriptyline, and topical compounded lidocaine/amitriptyline gels have been investigated with equivocal results; the ASCO (American Society of Clinical Oncology) guidelines suggest these may be considered as alternatives when duloxetine is ineffective or not tolerated but acknowledge the limited evidence.91011
CIPN management requires individualized dose decisions balancing oncological necessity against neurological harm. For curative-intent regimens, tolerance for persistent Grade 2 or Grade 3 CIPN is lower than for palliative regimens where quality of life is the primary endpoint. As a general framework: Grade 1 CIPN does not require dose modification but warrants careful documentation for trend monitoring; Grade 2 CIPN in a curative regimen warrants a 25% dose reduction and reassessment; Grade 2 CIPN persisting after dose reduction or worsening to Grade 3 warrants strong consideration of discontinuation of the offending agent with substitution of a less neurotoxic alternative if the regimen permits; Grade 3 or 4 CIPN requires discontinuation of the neurotoxic agent. In functional terms, neuropathy that prevents safe ambulation, causes falls, or produces painful paresthesias interfering with sleep constitutes functionally Grade 3 neuropathy regardless of how it scores on brief clinical examination, and should be managed accordingly.
Treatment-related secondary malignancies are a well-established long-term consequence of curative cytotoxic chemotherapy, representing the price paid for cancer cure. The two most clinically important forms are treatment-related acute myeloid leukemia (t-AML) and treatment-related myelodysplastic syndrome (t-MDS), which together constitute treatment-related myeloid neoplasm (t-MN) in the 2022 WHO classification. Their pathogenesis, latency, cytogenetics, and prognosis differ systematically according to the class of causative agent.12
Alkylating agent-related t-MN follows a characteristic pattern: the latency from exposure to t-MN diagnosis is long (typically 5 to 10 years, with a modal peak at 5 to 7 years), and the disease typically presents as t-MDS with a hypocellular bone marrow and a prolonged MDS (myelodysplastic syndrome) phase of months to years before transformation to frank AML (acute myeloid leukemia). The cytogenetic abnormalities are characteristic: monosomy 5 or deletion 5q, monosomy 7 or deletion 7q, and complex karyotype are the most frequent findings, reflecting loss of tumor suppressor genes on chromosomes 5 and 7 that normally regulate hematopoietic stem cell self-renewal. The prognosis of alkylating agent-related t-MN is poor: median survival from diagnosis is approximately 8 to 12 months, response to standard AML induction is low (complete remission rates of 20 to 40%), and the only potentially curative approach is allogeneic hematopoietic stem cell transplantation (alloHSCT) in eligible patients. The cumulative risk of t-MN is dose-dependent and is approximately 1 to 2% at 10 years for standard-dose cyclophosphamide regimens and up to 5 to 10% for high-dose conditioning regimens.12
Topoisomerase II inhibitor-related t-MN (caused by anthracyclines and etoposide) has a strikingly different pattern: the latency is short (1 to 5 years, most cases within 3 years of exposure), the disease typically presents de novo as frank AML without a preceding MDS phase, and the cytogenetic hallmark is a balanced chromosomal translocation involving the KMT2A (formerly MLL [mixed lineage leukemia]) gene at chromosome 11q23 (for etoposide and some anthracyclines) or the RUNX1T1 (formerly AML1/ETO) gene at chromosome 21q22 (for etoposide). KMT2A-rearranged t-AML from topoisomerase II inhibitors arises because these drugs stabilize topoisomerase II cleavable complexes preferentially at specific genomic loci including the AT (adenine-thymine)-rich break cluster region of KMT2A, directly generating translocations that fuse KMT2A to partner genes (including AF4 [ALL1-fused gene 4], AF9 [ALL1-fused gene 9], AF10 [ALL1-fused gene 10], ENL [eleven-nineteen leukemia protein], and others). Despite their de novo presentation, KMT2A-rearranged t-AML cases have intermediate prognosis and can respond to intensive induction chemotherapy, with alloHSCT in first remission recommended for most patients.12
Regarding chemotherapy in pregnancy, the clinical challenge arises most commonly when cancer is diagnosed in the first or second trimester. No chemotherapy is safe in the first trimester: organogenesis (weeks 3 to 8 post-conception) is the critical window for major structural teratogenicity, and essentially all cytotoxic chemotherapy is contraindicated during this period; spontaneous abortion rates are also substantially elevated. After the first trimester, many cytotoxic agents have been administered in the second and third trimesters with acceptable maternal and neonatal outcomes, particularly ABVD (doxorubicin [Adriamycin], bleomycin, vinblastine, dacarbazine) for Hodgkin lymphoma and R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone) for diffuse large B-cell lymphoma; platinum compounds and taxanes have also been used in the second and third trimesters in breast cancer and ovarian cancer. Antimetabolites (particularly methotrexate and 5-fluorouracil) remain relatively contraindicated throughout pregnancy because of their anti-folate mechanisms and ongoing fetal organ development requirements. All chemotherapy should be stopped at least 3 to 4 weeks before anticipated delivery to allow maternal and fetal drug clearance and reduce the risk of neonatal bone marrow suppression.
Dose adjustment for organ impairment requires knowing the primary elimination pathway of each drug. Renally cleared drugs (carboplatin, bleomycin, topotecan, methotrexate, etoposide, capecitabine, hydroxyurea) require dose reduction proportional to the degree of renal dysfunction, typically calculated using the Calvert formula for carboplatin (targeting AUC based on GFR [glomerular filtration rate]) or percentage dose reductions for other drugs based on creatinine clearance thresholds. Hepatically metabolized drugs requiring biliary excretion (doxorubicin, vincristine, vinblastine, docetaxel, paclitaxel, irinotecan, ixabepilone) require dose reduction in hepatic impairment; the bilirubin-based dose reduction guidelines for anthracyclines (50% dose for bilirubin 1.2 to 3.0 mg/dL; 25% dose for bilirubin above 3.0 mg/dL) are the most widely applied. Cisplatin is particularly nephrotoxic and also renally cleared; pre-hydration with 1 to 2 liters of normal saline before and after cisplatin administration, combined with forced diuresis, is mandatory to reduce renal tubular platinum accumulation and prevent nephrotoxicity. The threshold of creatinine clearance below 60 mL/min is generally considered the boundary for cisplatin dose reduction, and below 45 mL/min carboplatin substitution should be strongly considered.
Chapter 33 has covered the foundational cytotoxic agents of oncology pharmacology across six modules, from the principles of cell cycle specificity and pharmacokinetic determinants of toxicity to the organ-specific toxicities, pharmacogenomic interactions, and long-term consequences that define the clinical practice of medical oncology. The following synthesis connects the major conceptual threads across the chapter.12
The most fundamental organizing principle of cytotoxic pharmacology is cell cycle specificity. S-phase-specific agents (antimetabolites, hydroxyurea, camptothecins) are most effective against rapidly dividing tumors where a large fraction of cells are in S phase at any given time, and their activity is duration-dependent rather than peak-concentration-dependent, explaining why continuous infusion or prolonged oral dosing schedules are used for fluorouracil, capecitabine, and etoposide. M-phase-specific agents (vinca alkaloids, taxanes, epothilones) arrest cells in mitosis and are most effective against tumors with high mitotic indices; their activity is schedule-independent in that a single infusion can arrest a large fraction of dividing cells. Phase-nonspecific agents (alkylating agents, anthracyclines, bleomycin, platinum compounds) can damage DNA (deoxyribonucleic acid) or otherwise kill cells regardless of cell cycle phase, and their activity is more linearly related to total dose administered. Understanding these distinctions is essential for designing rational combination regimens that exploit non-overlapping mechanisms and schedule-dependent activity windows.7
The second major organizing principle is organ toxicity specificity. Each drug class carries a characteristic and reproducible pattern of dose-limiting and cumulative toxicities: anthracyclines produce cumulative cardiomyopathy; bleomycin produces pulmonary fibrosis; platinum compounds produce cumulative sensory neuropathy and nephrotoxicity; vinca alkaloids produce peripheral and autonomic neuropathy; alkylating agents produce gonadal toxicity, hemorrhagic cystitis (cyclophosphamide and ifosfamide, preventable with mesna [sodium 2-mercaptoethanesulfonate]), and secondary leukemia; taxanes produce cumulative sensory neuropathy. Recognizing these patterns allows proactive surveillance (echocardiographic monitoring with anthracyclines, DLCO (diffusing capacity for carbon monoxide) monitoring with bleomycin, creatinine and urine protein monitoring with platinum compounds), appropriate dose adjustment before irreversible damage accumulates, and selection of less organ-toxic alternatives when cumulative thresholds approach. The concept of lifetime cumulative dose limits, particularly for anthracyclines, is a defining feature of oncology practice that requires documentation and tracking across all prior treatment exposures.79
The third organizing principle is pharmacogenomics and drug interaction vigilance. Clinically meaningful pharmacogenomic variants that affect cytotoxic drug toxicity include DPYD (dihydropyrimidine dehydrogenase) deficiency (for fluoropyrimidines), TPMT (thiopurine methyltransferase) and NUDT15 (nudix hydrolase 15) deficiency (for thiopurines used in ALL maintenance), and uridine diphosphate-glucuronosyltransferase 1A1 polymorphism (UGT1A1*28) (for irinotecan SN-38 glucuronidation). CYP3A4 (cytochrome P450 3A4)-mediated interactions with azole antifungals are the most commonly encountered dangerous drug interactions in oncology practice, with the azole-vincristine combination representing a potentially fatal interaction requiring mandatory dose adjustment. The IMiD-specific requirement for REMS (Risk Evaluation and Mitigation Strategy) programs and the oxaliplatin-specific acute cold-triggered neuropathy requiring patient counseling before each infusion are drug-specific management requirements that are not intuitive and must be systematically transmitted during transitions of care. The prevention and early treatment of secondary malignancies through awareness of the distinct alkylating agent and topoisomerase II inhibitor t-MN phenotypes represents the long-term pharmacovigilance obligation of the oncologist to their survivors.12
Five clinical decisions recapitulate the most consequential knowledge points in Chapter 33: (1) Never allow intrathecal administration of a vinca alkaloid under any circumstances; dispense only in minibags, separate temporally and spatially from intrathecal chemotherapy. (2) Before starting any anthracycline-containing regimen, calculate lifetime cumulative doxorubicin-equivalent dose from all prior exposures and document baseline LVEF (left ventricular ejection fraction). (3) Before prescribing paroxetine or fluoxetine for a patient on tamoxifen, recognize the CYP2D6 interaction that reduces endoxifen and potentially compromises breast cancer outcomes; prescribe a CYP2D6-sparing antidepressant instead. (4) Before any surgery in a patient with prior bleomycin exposure, inform the anesthesiologist to minimize FiO₂ (inspired oxygen fraction); there is no safe time threshold. (5) Before starting lenalidomide, register with Revlimid REMS, confirm negative pregnancy test in women of childbearing potential, and initiate mandatory thromboprophylaxis; none of these steps is optional.
Platt OS, Brambilla DJ, Rosse WF, et al. Mortality in sickle cell disease: life expectancy and risk factors for early death. N Engl J Med. 1994;330(23):1639–1644.
doi:10.1056/NEJM199406093302303Charache S, Terrin ML, Moore RD, et al. Effect of hydroxyurea on the frequency of painful crises in sickle cell anemia. N Engl J Med. 1995;332(20):1317–1322.
doi:10.1056/NEJM199505183322001Tram Nguyen T, Mattingly M, Ghassemieh B, Bhatt DL, Kim BK. Asparaginase in pediatric and adult acute lymphoblastic leukemia: pharmacology and clinical relevance. Ann Pharmacother. 2016;50(5):374–387.
doi:10.1177/1060028016634014Vrooman LM, Stevenson KE, Supko JG, et al. Postinduction dexamethasone and individualized dosing of Escherichia coli L-asparaginase each improve outcome of children and adolescents with newly diagnosed acute lymphoblastic leukemia. J Clin Oncol. 2013;31(9):1202–1210.
doi:10.1200/JCO.2012.43.2070Kronke J, Udeshi ND, Narla A, et al. Lenalidomide causes selective degradation of IKZF1 and IKZF3 in multiple myeloma cells. Science. 2014;343(6168):301–305.
doi:10.1126/science.1244851List A, Kurtin S, Roe DJ, et al. Efficacy of lenalidomide in myelodysplastic syndromes. N Engl J Med. 2005;352(6):549–557.
doi:10.1056/NEJMoa041668Scripture CD, Figg WD. Drug interactions in cancer therapy. Nat Rev Cancer. 2006;6(7):546–558.
doi:10.1038/nrc1887Beijnen JH, Schellens JH. Drug interactions in oncology. Lancet Oncol. 2004;5(8):489–496.
doi:10.1016/S1470-2045(04)01528-1Hershman DL, Lacchetti C, Dworkin RH, et al. Prevention and management of chemotherapy-induced peripheral neuropathy in survivors of adult cancers: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol. 2014;32(18):1941–1967.
doi:10.1200/JCO.2013.54.0914Staff NP, Grisold A, Grisold W, Windebank AJ. Chemotherapy-induced peripheral neuropathy: a current review. Ann Neurol. 2017;81(6):772–781.
doi:10.1002/ana.24951Smith EM, Pang H, Cirrincione C, et al. Effect of duloxetine on pain, function, and quality of life among patients with chemotherapy-induced painful peripheral neuropathy: a randomized clinical trial. JAMA. 2013;309(13):1359–1367.
doi:10.1001/jama.2013.2813Godley LA, Larson RA. Therapy-related myeloid leukemia. Semin Oncol. 2008;35(4):418–429.
doi:10.1053/j.seminoncol.2008.04.012