Oral Anticancer Therapy: Management of Drug Interactions

Britny G. Rogala, PharmD; Margaret M. Charpentier, PharmD; Michelle K. Nguyen, PharmD; Kaitlin M. Landolf, PharmD; Lamya Hamad, RPh; Kelly M. Gaertner, PharmD


J Oncol Pract. 2019;15(2):81-90. 

In This Article

Pharmacokinetic Interactions

CYP3A4 Enzyme Inhibition

Invasive fungal infections occur in approximately 40% of patients with hematologic malignancies, chemotherapy-induced neutropenia, and/or high-dose corticosteroid therapy, and are one of the primary contributors to cancer-related morbidity and mortality.[20–22] Antifungal prophylaxis is indicated for patients with an intermediate to high risk of infection.[23,24] Voriconazole, posaconazole, itraconazole, and ketoconazole are strong CYP3A4 inhibitors, thus requiring dose adjustments of coprescribed oral anticancer therapies that are major CYP3A4 substrates (Table 2).[5,25]

Idelalisib, a PI3K inhibitor used for the treatment of indolent lymphomas, is also a strong CYP3A4 inhibitor. When the major CYP3A4 substrate midazolam was coadministered with idelalisib during the initial pharmacokinetic DDI evaluation, the midazolam maximum concentration and area under the curve increased by 2.4-fold and 5.4-fold, respectively. To further emphasize the relevance of this interaction, a case report described a patient previously maintained with the CYP3A4 substrate diazepam, who had altered mental status and respiratory failure 10 days after idelalisib initiation.[26] Her respiratory failure quickly resolved after idelalisib discontinuation and conversion of diazepam to lorazepam. When azole antifungal medication or other strong CYP3A4 inhibitors are indicated in the setting of concomitant CYP3A4 substrates, consideration of dose or therapy modification (Table 2) should be made.

CYP3A4 Enzyme Induction

Antiepileptic drugs may be encountered in the population of patients with cancer as therapy for comorbid seizure disorders or in patients requiring treatment of new-onset seizures secondary to primary or metastatic brain tumors. Enzyme-inducing antiepileptic drugs include carbamazepine, phenobarbital, and phenytoin.[27] Concomitant use of certain oral anticancer therapies with these enzymeinducing antiepileptic drugs may require dose adjustment or should be avoided altogether (Table 2). The enzyme-inducing antiepileptic drugs are inducers of CYP 1A2, 2B6, 2C9, and 3A4/5, as well as Pgp. Nonenzymeinducing antiepileptic drugs, such as levetiracetam, topiramate, lamotrigine, pregabalin, and valproic acid, generally carry fewer DDIs and are preferred over enzymeinducing antiepileptic drugs, when appropriate.[28] In some cases, adjusting the dose of the antiepileptic drug or oral anticancer therapies can be sufficient to accommodate the DDI (Table 2).[8,9]

Transitional polytherapy describes the process by which a new antiepileptic drug is started and dose titrated, followed by a gradual discontinuation of the initial antiepileptic drug.[29] If an enzyme-inducing antiepileptic drug can be switched to an alternate antiseizure medication, transitional polytherapy executed over at least 4 weeks is the most efficacious and safe process for conversion between monotherapy antiepileptic drugs.[30] Consultation with the antiepileptic drug prescriber and a clinical pharmacist is strongly advised if dose or medication adjustments are necessary.

Apalutamide is a strong CYP3A4 and CYP2C19 inducer and a weak CYP2C9 inducer. Similarly, enzalutamide is a strong CYP3A4 inducer and moderate CYP2C19/9 inducer, and dabrafenib is a moderate inducer of CYP2C9 and CYP3A4. As enzyme inducers, enzalutamide, apalutamide, and dabrafenib may render many important medications ineffective. Atorvastatin (a CYP3A4 substrate), apixaban (a CYP3A4 and Pgp substrate), and warfarin are commonly used medications in which enzyme induction and lack of effect could lead to life-threatening consequences. In one report, for example, a 77-year-old woman was prescribed apixaban for atrial fibrillation while she was taking low-dose phenobarbital.[31] Six months later she had a cardioembolic stroke and was found to have a low apixaban trough concentration (89 ng/mL). Even after withdrawal of an enzyme-inducing drug, the enzyme can take 10 to 35 days to return to its normal level of function. This period does not account for the half-life of the enzyme-inducing medication. Importantly, exposure of prodrugs such as clopidogrel, which requires CYP2C19 metabolism for conversion into its active metabolite, can be increased with inducers of the enzyme required for bioactivation (ie, apalutamide).


Depression is more common in individuals with cancer as compared with the general population.[32] The incidence of depression in patients with cancer is estimated to range from 8% to 24%, depending on the assessment used, disease severity, and type of cancer. Subsequently, potential DDIs between anticancer agents and antidepressants pose a major safety concern. These interactions are primarily a result of inhibition or induction of the CYP450 enzymes.[33,34] Tamoxifen, for example, requires metabolism via CYP2D6 to be pharmacologically active.[33] Drugs that strongly inhibit CYP2D6, which include bupropion, paroxetine, and fluoxetine (Table 3), reduce the clinical benefit of tamoxifen by decreasing the levels of its active metabolites; therefore, concomitant use should be avoided.[34–37]

Changes to any antidepressant regimen should be performed only after consultation with the primary prescriber. Antidepressants may cause withdrawal symptoms when abruptly stopped or rapidly reduced.[38] Several strategies are available for switching among antidepressants. Consideration should be given to the risk of drug interactions, withdrawal symptoms, and breakthrough depressive symptoms. When initiating a switch between two interacting antidepressants, potentially dangerous drug interactions can be avoided by gradually tapering the initial antidepressant dose, followed by a washout period equivalent to five half-lives of the drug (ie, if a drug's half-life is 14 hours, the washout period should be 70 hours, or approximately 3 days).[39] Otherwise, a direct switch or cross-tapering are generally preferable to a gradual taper and wash-out period, because of the risk of withdrawal and depressive symptoms, as well as suicidal ideation.[36]

If switching a selective serotonin reuptake inhibitor (SSRI) to another SSRI or a serotonin norepinephrine reuptake inhibitor (SNRI), a transition to an approximately equipotent dose may be considered, because of similar serotonergic effects.[40] Similarly, SNRIs can be switched to an approximately equipotent SSRI or SNRI dose without cross-tapering. An exception is when the current antidepressant is at the higher end of the dosing range; in this instance, cross-tapering may be desired. Cross-tapering generally is performed over 1 to 2 weeks but may be prolonged when the initial antidepressant has a long half-life (ie, fluoxetine) or in patients who have had withdrawal symptoms in the past.[34,39] Switching bupropion to another antidepressant is a less-studied change that would likely be best accomplished with crosstapering. In patients receiving tamoxifen, antidepressants should be switched to a mild or minimal inhibitor of CYP2D6 (Table 3). In patients with cancer, antidepressant selection should be based on the adverse effect profile and the potential for treatment of comorbid conditions (Table 3).[35–37,41]

Acid-suppression Therapy

Proton pump inhibitors (PPIs), histamine H2-receptor antagonists, and antacidsmay alter the gastric pH and interfere with the absorption of oral anticancer therapies. Antacids not only affect acidity but can also bind drugs, due to cations contained in the antacid, resulting in decreased absorption. Pantoprazole is an inhibitor of drug transporters Pgp and breast cancer resistance protein, which can also adversely affect drug concentrations.[42] Because TKIs are weakly basic, they are optimally absorbed in an acidic environment, where they are more ionic and, therefore, more soluble.[5]

PPIs are contraindicated for use with acalabrutinib, dacomitinib, dasatinib, erlotinib, neratinib, and pazopanib.[42] Acalabrutinib, erlotinib, and nilotinib can be taken 10 hours after or 2 hours before taking a histamine H2-receptor antagonist. A histamine H2-receptor antagonist can be administered 6 hours after or 10 hours before dacomitinib. In addition, over-the-counter antacids should be administered 2 hours before or after the administration of acalabrutinib, dasatinib, erlotinib, gefitinib, bosutinib, nilotinib, and ponatinib, and 3 hours before neratinib.

Studies evaluating TKI absorption are often limited and may have conflicting results. For example, erlotinib is optimally absorbed in an acidic environment; therefore, coadministration with PPIs should be avoided. In a study of advanced lung cancer, researchers reported that administering erlotinib with an acidic beverage (ie, cola) increased the bioavailability by approximately 40% in patients receiving erlotinib concomitantly with a PPI and, thus, may minimize the PPI drug interaction.[43] In addition, in a recent review conducted in a Japanese population of 130 consecutive patients with lung cancer who were receiving either erlotinib or gefitinib, researchers concluded that acid suppressive therapy (taken by 47 patients) did not affect the overall response rates.[44]

Although results of in vivo studies indicate nilotinib relies on an acidic environment for absorption, clinical evidence suggests concomitant acid suppressive use does not affect outcomes. A retrospective evaluation of patients receiving nilotinib (n 5 748) for chronic myeloid leukemia demonstrated that acid suppressive therapy did not affect nilotinib efficacy (ie, major molecular or cytogenetic response) or pharmacokinetics.[45]

A secondary analysis of a randomized clinical trial evaluating capecitabine and oxaliplatin with or without lapatinib in gastroesophageal cancer raised concern for coadministration of PPIs with capecitabine.[46] PPI-treated patients had worse progression-free survival and overall survival in the placebo group and a shorter overall survival in a multivariate analysis of the lapatinib arm. Given that a significant pharmacokinetic interaction was not observed in an earlier trial when capecitabine was coadministered with magnesium hydroxide and aluminum hydroxide, data are inconsistent to make a strong recommendation at this time.[47]

There is conflicting evidence regarding the use of pazopanib with PPIs. Phase I and II studies did not reveal a clinically significant interaction between pazopanib and a PPI; however, a more recent pharmacokinetic evaluation of pazopanib demonstrated that administration of pazopanib in the morning and esomeprazole in the evening resulted in a 40% decrease in the area under the curve.[48,49]

Specifically timed coadministration of PPIs with pazopanib and the use of less-potent PPIs (ie, once-daily dosing of omeprazole or pantoprazole)may mitigate this interaction.[50,51] Enteric-coated PPIs administered once daily have a delayed onset of approximately 3 to 4 hours, resulting in breakthrough gastric acidity during this time.[51] If concomitant PPI therapy is necessary with a TKI with pH-dependent solubility, the goal is to target this window of acidity. This may be achieved by administering both medications in the morning, with the TKI administered 2 hours prior to the enteric-coated PPI.

Until more data are available, avoiding PPIs in patients prescribed TKIs with pH-dependent absorption, including acalabrutinib, bosutinib, crizotinib, dabrafenib, dasatinib, erlotinib, gefitinib, neratinib, and pazopanib, is recommended. If this is not possible, then use of a less-potent PPI and morning administration with the PPI administered 2 hours after the TKI may be reasonable.[50,51] If histamine H2-receptor antagonists or antacids are administered with a TKI, appropriate timing of the TKI with respect to acid suppressive therapy, as previously discussed, is recommended.