Managing Drug Interactions in Cancer Therapy: A Guide for the Advanced Practitioner.

Mrs. P is a 30-year-old woman who presented to our bone marrow transplant program with myelodysplastic syndrome (MDS). She received a haploidentical allogeneic stem cell transplant with a conditioning regimen consisting of busulfan and cyclophosphamide. This treatment was followed by post-transplant immunosuppression for graft-versus-host disease (GVHD) with cyclophosphamide, mycophenolate mofetil (MMF), and tacrolimus (see Table 1 for medication list). Tacrolimus levels were monitored twice a week with adjustment to a goal range of between 5 and 10 ng/mL. We initiated tacrolimus at a dose of 0.03 mg/kg by mouth twice daily (rounded to 2 mg by mouth twice daily). Drug interactions were assessed by the clinical pharmacist prior to admission, routinely with medication changes, and then upon discharge.

Drug-interaction related risk factors include the use of drugs that are significantly impacted by inhibition or induction of drug metabolism (tyrosine kinase inhibitors [TKIs]), the use of drugs that have a significant inhibitory or inducing capacity of drug metabolism (certain antifungal medications), and the use of drugs with a narrow therapeutic window such as warfarin. Patient-specific risk factors include older age, renal or hepatic dysfunction, hematologic cancers, and the use of many prescribed medications (Panesar, 2011).

One retrospective drug review in cancer patients showed a high frequency of drug interactions. A total of 278 patients were reviewed, of which 40% of patients had reported potential drug interactions with their chemotherapy (van Leeuwen et al., 2011). Although this shows a high risk of drug interactions in cancer patients, it is unknown from this study what percentage of interactions were clinically relevant. In this article, we will introduce concepts and use clinically relevant examples to highlight the risk of drug interactions in patients with cancer.

Drug-drug interactions are common, not only in the oncology setting but also in the older adult population, and may be responsible for up to 4% of deaths in hospitalized oncology patients (Buajordet, Ebbesen, Erikssen, Brors, & Hilberg, 2001). A study by Van Leeuwen and associates (2013) noted that more than half of ambulatory patients with cancer had at least one potential drug interaction. One-third of ambulatory patients with cancer had a major potential drug interaction that could result in serious clinical consequences.

Identified risk factors for drug interactions are listed in Table 2. One universal identified risk factor is an age-related change, including changes in the gastrointestinal tract (increased or decreased absorption), decreases in body fat that may influence the length of time a drug remains in the body, and decreased hepatic and renal function. A study by Popa and colleagues (2014) examined records of 244 patients who were 70+ years of age and undergoing chemotherapy. This study found 75% of patients receiving chemotherapy had a potential for a serious drug interaction involving chemotherapeutic agents.

Table 2

Risk Factors for Drug Interactions

Other risk factors include polypharmacy, defined as the use of more medications than often medically required, and the increasing number of doses of a medication per day (Cope, 2013; Planton & Edlund, 2010; Popa et al., 2014; van Leeuwen et al., 2011). As many as 80% of oncology patients utilize over-the-counter medications (Van Leeuwen et al., 2011), and these agents are not often recorded in the patient’s medical record. Patients with cancer are at a higher risk due to the increasing number of daily medications—both oncologic drug(s) as well as supportive medications.

Many new agents approved for the treatment of cancer are orally administered, indicating they are under the influence of pharmacokinetic drug interactions including absorption, distribution, metabolism, and excretion (ADME), which can reduce their effectiveness or increase toxicity. In fact, 60% of new agents approved for cancer treatment by the US Food and Drug Administration (FDA) between 2012 and 2014 were orally administered (FDA, 2017a). Most of these drugs are significantly impacted by pharmacokinetic drug interactions.

In this article, we will focus on pharmacokinetic drug interactions, but it is important to understand that other types of drug interactions such as pharmacodynamic interactions may occur. Simply stated, a pharmacokinetic interaction is the effect of the body on the drug, and a pharmacodynamic interaction is the drug’s effect on the body (Beijnen & Schellens, 2004). Pharmacodynamic drug interactions are actually very common, and such examples include the use of multiple central nervous systems (CNS) depressants or the combination of nonsteroidal anti-inflammatory drugs (NSAIDs) and angiotensin-converting enzyme (ACE) inhibitors.

PHARMACOKINETIC DRUG INTERACTIONS

Absorption

The absorption of various oral chemotherapy agents is often influenced by multiple factors such as food and acid-suppressive agents. Ultimately, these factors can impact the solubility and bioavailability of chemotherapy agents (Halfdanarson & Jatoi, 2010). For example, many oral TKIs are influenced by gastric pH changes, as seen in Tables Table 3 and Table 4. Specifically, Table Table 4 illustrates how the pH-dependent solubility of dasatinib (Sprycel) decreases as pH increases (Bristol-Myers Squibb, 2008; Eley et al., 2009).

Table 3

Impact of pH on Select Tyrosine Kinase Inhibitorsa

Table 4

Dasatinib and Acid-Suppressive Agents

There are ways to mitigate the impact of acid suppression on drug absorption. They include the use of H2 blockers at specific times around administration of the TKI, as described in Table 3. Another reported option is to use a beverage that decreases the stomach pH for a short period such as a cola beverage (van Leeuwen et al., 2016). Furthermore, food can significantly alter the absorption of chemotherapy agents; however, the effect of food is not consistent among all chemotherapy agents, as illustrated in Table 5 (Koch et al., 2009; Reigner et al., 1998).

Table 5

Impact of Food on the Absorption of Select Chemotherapy Agents

Distribution

Specific drug characteristics such as high protein binding (> 90%), narrow therapeutic index, high hepatic extraction ratio, and intravenous dosage forms may increase the likelihood of altered distribution. In particular, the impact of plasma protein binding is often overemphasized in the literature and training (Rolan, 1994). For example, TKIs are highly protein bound, but there is minimal evidence of major interactions with drugs that have the ability to displace them from the protein-binding sites.

Metabolism

Metabolism primarily occurs in the liver involving cytochrome P450 enzymes. Multiple drugs (refer to Table 6) competitively inhibit or induce cytochrome P450 enzyme–binding sites. This can alter the metabolism of oral and intravenous chemotherapy agents, ultimately influencing their efficacy and safety (Guengerich, 2008; Zanger & Schwab, 2013).

Table 6

Select Cytochrome P450 Inhibitors and Inducers

For example, as illustrated in Table 6, antifungals such as voriconazole, posaconazole, and ketoconazole are very strong cytochrome P450 3A4 inhibitors, which interact with a large majority of TKIs. Certain TKIs such as ibrutinib (Imbruvica) and everolimus (Afinitor) could have profound toxicity if administered with strong inhibitors of CYP 3A4 (Kovarik et al., 2005; de Jong et al., 2015). On the other hand, rifampin or other strong inducers of CYP3A4 could significantly decrease the activity of many of the TKIs, as shown in Table 7. For these reasons, we recommend careful assessment for drug interactions any time a patient starts treatment with TKIs.

Table 7

Select Tyrosine Kinase Inhibitor CYP3A4 Drug Interactions

Elimination

Elimination occurs mainly via the kidneys or bile. A small portion of chemotherapy agents such as methotrexate and cisplatin are primarily removed via elimination from the kidneys. High-dose methotrexate treatment can cause severe harm and even death in patients who have difficulty eliminating methotrexate and/or active metabolites of methotrexate. Certain drugs such as specific antibacterials, proton pump inhibitors, and NSAIDs can reduce the elimination of methotrexate (Ferreri et al., 2004; Fjeldborg, Sorensen, & Helkjaer, 1986; Hammor & Hasan, 2013).

PREVENTION AND MANAGEMENT STRATEGIES FOR ADVANCED PRACTITIONERS

All oncology advanced practitioners (AP) have a vital role in the prevention, early detection, and prompt management of drug-drug interactions. As the number of oral oncologic agents increases, more safety issues and adherence issues will abound. The first step in this process of prevention and early detection of any adverse drug reaction is a full medication and health history review. The patient is instructed to bring any medication—prescribed or over-the-counter—to his clinic visit. The drug names, dosages, and schedule are documented. Any herbal supplements and/or vitamins should be documented along with dosages. Specific vernacular may be utilized to address sociocultural diversities (e.g., words such as "natural" products, folk medicine, or "home remedies"; salves; creams; and potions; Ben-Arye, Halabi, Attias, Goldstein, & Schiff, 2014).

Medical records from other health-care providers should also be examined, including clinic notes, hospitalization records, and emergency department (ED) reports. It is also important to note any potential drug absorption issues due to previous surgeries, feeding tubes, or diseases such as Crohn’s disease.

It is estimated that up to 90% of patients use some sort of alternative or complementary medicines or therapies (Arslan, Tural, & Akar, 2013; Mao, Palmer, Healy, Desai, & Amsterdam, 2011; Naing et al., 2011; Yates et al., 2005). Yet the majority of these patients do not disclose this information to their health-care providers (Mao et al., 2011; Oh et al., 2010; Yates et al., 2005; Yildirim, 2010).

The FDA ensures the safety and efficacy of a drug released to the public. Nutritional products and supplements are exempt from this review process, however (Vogel, 2011). Patients do not understand this concept and often assume the lack of FDA regulation makes these products "safe." Unfortunately, ingredients in the products can be variable and unspecified, and there can be a lack of quality control—meaning there can be meaningful differences in the amounts of the product between different batches (Arslan et al., 2013). The AP must ensure patients understand the importance of disclosing any and all alternative/complementary therapies.

There are risk assessment tools to assist APs to prevent, monitor for, and/or allow early identification of symptoms (Table 8). These tools may prompt APs to prescribe an appropriate medication or prevent the prescription of an inappropriate prescription. Other tools can aid APs in evaluating a medication’s potential effect on a patient’s functional and disease status. Cope (2013) noted 10 essential assessment elements to evaluate medications in older adults (Table 9).

Table 8

Risk Assessment Tools for Prescribing Appropriate Medications

Table 9

Essential Elements of Medication Assessment in Older Adults

The AP risk assessment not only includes a thorough medication review, but also the documentation of any side effects experienced by a patient. Patients should be questioned about any previous adverse events from any therapy. Assessment for substance abuse is also important, as drug metabolism may be affected. For example, smoking can induce drug-metabolizing enzymes of cytochrome P450, thus decreasing the efficacy of some oncologic agents as well as other categories of drugs (O’Malley, King, Conte, Ellingrod, & Ramnath, 2014; Sohn et al., 2015). Subcutaneous medications may have suboptimal absorption due to smoking effects. The stimulatory effects of smoking could lessen the effects of drugs such as the benzodiazepines and opioids. If a patient should suddenly quit smoking, the practitioner must maintain alertness to the possibility of a drug overdose due to increased drug exposure, such as with methadone (Zevin & Benowitz, 1999).

Thorough patient education, including proper dosing and scheduling instructions, is imperative to decrease potential drug interactions. Written information as well as verbal instructions are needed. Reminder devices such as a smartphone or an alarm clock could be helpful. Pill boxes may be useful, but many oral oncologic agents should not be placed in pill boxes, but left in the original container protected from light and moisture (Drug Information Service, University of Utah Hospitals and Clinics, 2016). The prescribing information may be consulted for details about medication storage and handling.

Ideally, a drug-drug interaction is prevented. Up to 30% of adverse drug events in the outpatient setting are preventable (Gurwitz et al., 2003). Any prescription is carefully evaluated. Limiting the number of medications in older adults can reduce the risk of drug-drug interaction. Six or more medications increases the risk of an adverse drug event times four (Pretorius, Gataric, Swedlund, and Miller, 2013). Each new medication prescribed adds more than one adverse drug event a year. Multiple prescribers also increase adverse drug events; each additional prescriber increases the risk of an adverse drug event by 30% (Pretorius et al., 2013).

Every prescriber should share records and medication lists. Prescribers should avoid treating every side effect with another medication, considering if the dose of the offending medication could be decreased or changed to another medication (Pretorius et al., 2013). When a new medication is prescribed, a follow-up visit 2 to 4 weeks after initiating the medication is in order. In older adults, the Beers criteria should be observed (American Geriatrics Society, 2012).

The Beers criteria give a comprehensive list of medications to be avoided or used with caution in older adults. Drugs that are on this list include benzodiazepines, diphenhydramine, ibuprofen, megestrol acetate, metoclopramide, promethazine, sliding-scale insulin, and zolpidem among others. Special attention should be given to those with a history of an adverse drug event, those who are nonadherent, those who have cognitive impairment or psychiatric disease, those who have substance abuse problems, and those who live alone.

Before increasing a dose of a medication due to seemingly suboptimal effect, APs must consider whether nonadherence is an issue. Any unnecessary medications should be discontinued. Recommended lab monitoring for certain medications should be followed according to the FDA prescribing information. When prescribing a medication, it is recommended to avoid starting more than one medication at a time (Pretorius et al., 2013).

Although prescribers and patients must be knowledgeable about potential drug adverse reactions, the office staff must also be educated about oral oncologic agents and drug-drug interactions. Telephone triage personnel must be educated about the signs of a potential drug interaction and promptly intervene if one is suspected. The medical oncology office staff should have an oral medication adherence assessment protocol and dedicated nursing staff for oral regimens (Moody & Jackowski, 2010). Clinical decision support systems may improve prescribing quality as well by alerting the prescriber to potential drug interactions or dosages that might be incorrect. However, APs should beware of "alert fatigue," which can occur when there is poor alert specificity (Seidling et al., 2014; Weingart, Zhu, Young-Hong, Vermilya, & Hassett, 2014).

The oncology AP should maintain a high index of suspicion for a drug-drug interaction. Some common signs of an adverse drug event might include a fall; orthostatic hypotension; heart failure; delirium or cognitive impairment; a change in daily functioning; a hospital admission; or exaggerated common adverse events (Pretorius et al., 2013). Notation should be made of the timing of symptoms after a new medication starts or after a dose change.

If a drug interaction is noted, the AP should evaluate the clinical significance of the event. The number of drugs involved should be noted. Options for management should then be reviewed and may include removal of the offending agent, removal of the affected agent, potential dose adjustments of medication, or prescription of an alternative agent(s).

CASE STUDY

Patients with leukemia are at a heightened risk of drug interactions due to the frequent use of medications that interact with cytochrome P450 enzymes, such as antifungal medications. Posaconazole, for example, is a broad-spectrum azole antifungal and a strong inhibitor of the CYP3A4 isoenzyme. Multiple medications the patient in our case study received were metabolized at least partially by CYP3A4, including cyclophosphamide and tacrolimus. Posaconazole was not initiated until day +5 after stem cell transplantation, to reduce the risk of a potential drug interaction with cyclophosphamide, which was administered on days –3, –2, +3, and +4. Although posaconazole has not been studied in combination with cyclophosphamide, itraconazole has been shown to increase levels of the potentially more toxic metabolites (Marr et al., 2004).

Tacrolimus is an immunosuppressant used to decrease the risk of graft-versus-host disease (GVHD), a common complication of stem cell transplantation. Tacrolimus is metabolized predominately by CYP3A4; therefore, dosing requirements are significantly decreased (> 50%) when it is used concomitantly with strong inhibitors of CYP3A4 (El-Dahshan, Bakr, Donia, Badr, & Sobh, 2004). It is important to monitor levels meticulously, as subtherapeutic levels may increase the risk of GVHD, whereas supratherapeutic levels may increase the risk of kidney dysfunction. Table 10 includes tacrolimus levels and doses throughout the inpatient admission and in the clinic.

Table 1

Medication List for Case Study Patient

Tacrolimus Adjustments due to Drug Interaction

Tacrolimus levels initially were subtherapeutic. Over time, however, the posaconazole decreased tacrolimus metabolism through inhibition of CYP3A4. The full impact of changes in cytochrome P450 enzyme activity and a clinical interaction may not be immediately apparent, due to a delay in hepatic enzyme inhibition caused by the posaconazole and a lag in the increase in tacrolimus drug levels. This example highlights the pharmacovigilance necessary and the role APs play in monitoring patients for critical drug interactions.