Polypharmacy management among older adults with cancer requires multipronged approach

Management of polypharmacy — the concurrent use of five or more medications — is a major challenge for clinicians who care for older adults.

The number of prescribed medications directly correlates with potentially inappropriate medication (PIM) use and risk for harmful drug-drug interactions. Physical or cognitive function also can be compromised.

The problem is magnified among individuals with cancer due to their complex and frequently changing treatment and supportive care regimens.

More than half of older adults with cancer have at least one comorbidity — such as hypertension, cardiovascular disease, diabetes or arthritis — that requires medication management.

Cancer treatment-related toxicities can lead to hospitalization and prescribing cascades that further increase risk for polypharmacy and PIM use.

Provider and patient understanding of these risks among older adults with cancer is essential. However, a lack of guidelines and outcomes-driven studies requires clinicians to follow a multipronged approach that incorporates objective and subjective assessments.

Polypharmacy risks

More than one-third of older patients who take five to nine medications — and nearly all of those who take 10 or more — are hospitalized due to medication-related adverse events.

Outpatient visits are 88% more likely among those who take at least five medications than those who take two or fewer.

Approximately half of older adults who take five to nine medications experience at least one drug-drug interaction, and each additional medication prescribed increases the likelihood by about 12%.

Lindblad and colleagues conducted a survey and cross-sectional study to identify clinically important drug-drug interactions among 1,340 veterans aged 65 years or older. The most common involved use of first-generation calcium channel blockers by patients with congestive heart failure and aspirin use among those with peptic ulcer disease.

Polypharmacy also may contribute to risk for falls and worse fall-related outcomes.

Damián and colleagues conducted a systematic review that showed each additional medication prescribed increased fall risk among older adults by 7% (RR = 1.07; 95% CI, 0.95-1.21), with antidepressants associated with the highest risk.

Medication management

Data from randomized trials and observational studies suggest deprescribing — the identification and discontinuation of medications that are ineffective, inappropriate or unnecessary — can reduce polypharmacy, but few reported on clinical outcomes.

Reeve and colleagues conducted a population-based survey of 2,124 Medicare beneficiaries aged 65 years or older.

Results, published last year in JAMA Internal Medicine, showed more than 90% were willing to stop taking one or more of their medications if their provider considered it possible. More than two-thirds reported wanting to reduce the number of medicines they were taking.

However, providers may be hesitant to recommend deprescribing — particularly for those with cancer or other advanced illnesses.

Medication therapy management is a key component of the comprehensive geriatric assessment, but guidelines on the application of medication tools in geriatric oncology and outcomes-driven studies designed to examine the impact of deprescribing on clinical outcomes are lacking.

The OncPal Deprescribing Guideline has been validated in geriatric oncology for those with an estimated life expectancy of less than 6 months. Although useful, its application for the general geriatric oncology population — including those receiving curative treatment — is limited.

PIM identification tools

Frequently used PIM identification tools — such as American Geriatrics Society’s Beers Criteria — often identify medications as potentially inappropriate, although many of these medications are considered necessary to provide adequate supportive care for older patients with cancer.

A clinical assessment to determine medication appropriateness is needed; however, this is time consuming and may require specialists, including a pharmacist and/or geriatrician.

National Comprehensive Cancer Network guidelines recommend using Beers Criteria, the Medication Adherence Index — a series of 10 questions designed to assess medication appropriateness based on factors such as indication, efficacy, dosing, practicality of directions, drug-drug and drug-disease interactions, therapy duplication and cost — or Screening Tool to Alert Doctors to Right Treatment/Screening Tool of Older Person’s Prescriptions (START/STOPP), developed by a panel of experts in Europe. However, NCCN guidelines provide no guidance on whether to apply these alone or in combination.

American Geriatrics Society recommends Beers Criteria be used as a starting point and encourages clinicians to consider rationale when applying the criteria to individual patients. Importantly, Beers Criteria do not apply to end-of-life or palliative care due to differences in risk-benefit consideration.

Some studies conducted in the geriatric oncology population support the idea of a multipronged approach that utilizes multiple tools.

Maggiore and colleagues implemented Beers Criteria, Zhan Criteria, and the Healthcare Effectiveness Data and Information Set (HEDIS) to identify PIM use among 500 patients aged 65 years or older with cancer undergoing chemotherapy.

Zhan Criteria focuses only on medications to be avoided; it does not consider those to be used with caution or the potential for drug interactions. HEDIS was designed to determine reimbursement by payers.

Results revealed PIMs among 29% of patients based on Beers Criteria, 11% based on Zhan Criteria and 13% based on HEDIS. There was no clinical assessment in the study to evaluate medication appropriateness.

Nightingale and colleagues used a combination of Beers Criteria, STOPP and HEDIS to evaluate 234 patients (mean age, 80 years) with cancer at various stages. STOPP uniquely considers clinical scenarios and allows for situation-based assessment.

Results showed PIM use among 40% of patients based on Beers Criteria, 38% based on STOPP and 21% based on HEDIS. The study did not measure deprescribing or clinical outcomes.

Whitman and colleagues compared the use of Beers Criteria alone with a three-pronged approach that included Beers Criteria, START/STOPP and Medication Adherence Index.

Beers Criteria identified 38 PIMs, whereas the three-pronged approach identified 119 PIMs.

The multipronged approach led to deprescribing of 73% of PIMs. Investigators suggested this resulted in symptom reduction, although they performed no objective measurement. A downside of the multipronged approach is feasibility in a busy clinic where time is limited.

Pharmacist intervention can significantly reduce PIM use and may improve patient outcomes.

Nipp and colleagues conducted a pilot randomized trial to demonstrate the feasibility and efficacy of integrating pharmacists into the care of adults aged 65 years or older with breast, gastrointestinal or lung cancer receiving first-line chemotherapy.

Researchers randomly assigned patients to pharmacist-led medication management using Beers Criteria or usual care.

Results showed those assigned to pharmacist intervention had fewer discrepant medications (5.82 vs. 8.07) and PIMs (3.46 vs. 4.8) at 4 weeks; however, the differences did not reach statistical significance and clinical outcomes were not measured.

Optimal tool for oncology

What is the most appropriate tool in oncology?

Beers Criteria is the most widely studied, and it also is simple and quick. The inclusion of drug-disease interactions makes this tool particularly valuable.

The Medication Adherence Index requires clinical judgement and guides clinicians to assess all aspects of a medication to determine its appropriateness for each patient, potentially proving more valuable than Beers Criteria. The time requirement is a major limitation, however, as it takes up to 10 minutes per medication.

The ideal use of this index may be after application of Beers Criteria to assess questionable medications, such as those for supportive care that Beers may automatically exclude. However, no published studies have assessed the efficacy of this two-pronged approach in geriatric oncology.

Prospective studies needed

Most polypharmacy studies focus on PIM identification as a surrogate endpoint for clinical outcomes. The idea is PIM identification will lead to deprescribing, which hopefully will improve safety outcomes for patients and reduce health care costs.

Prospectively designed studies to assess the validity of tools created for PIM identification and deprescribing would guide providers to safe medication management for older adults with cancer.

The ideal prospective study design would include clinical endpoints — such as symptoms and safety — to establish superiority of a PIM identification method.

A multipronged approach would allow use of an objective medication assessment and subjective clinical judgement to trigger deprescribing.

Pharmacogenomics — the impact of genetics on drug response — may offer a personalized approach to drug prescribing and deprescribing among older patients.

Data suggest more than 90% of individuals carry a genetic variant that contributes to drug response. Effective integration of pharmacogenomics into oncology practice can help tailor supportive care and help determine the magnitude of drug-drug, drug-gene and drug-disease interactions.

Brixner and colleagues conducted an observational study that compared a prospective cohort of 205 patients aged 65 years or older who underwent pharmacogenetic testing with a propensity score-matched historical cohort of 820 untested patients in a claims database.

Results showed lower rates of hospitalization (9.8% vs. 16.1%) and ED visits (4.4% vs. 15.4%) in the pharmacogenomics group than the control group.

Elliott and colleagues conducted a randomized controlled trial to assess the clinical impact of pharmacogenetic-guided therapy compared with standard therapy for 110 patients aged 50 years or older.

At 60 days, the pharmacogenetic-guided approach significantly reduced rehospitalizations per patient (0.33 vs. 0.7; RR = 0.48; 95% CI, 0.27-0.82) and mean number of ED visits per patient (0.39 vs. 0.66; RR = 0.58; 95% CI, 0.34-0.99).

These data suggest pharmacogenomics may be informative for medication management among older patients, and that larger randomized and pragmatic clinical trials to validate its use in practice — particularly for patients with cancer — is warranted.

Conclusion

PIM identification and polypharmacy management in the geriatric oncology population is challenging due to unique patient factors, including multiple comorbidities, complex disease states and physiologic changes associated with aging.

Available medication tools have not been validated in oncology, and evidence to guide their use is limited.

A multipronged approach to medication regimen assessment that includes objective measurements, such as Beers Criteria, and subjective measurements — such as the Medication Adherence Index — may be ideal, although it is time consuming.

A genomics-driven approach may further personalize medication management in this high-risk population.

A concerted effort should be made to validate these approaches prospectively to ensure the optimal management of polypharmacy among older adults with cancer.

References:

2019 American Geriatrics Society Beers Criteria Update Expert Panel. J Am Geriatr Soc. 2019;doi:10.1111/jgs.15767.

Bourgeois FT, et al. Pharmacoepidemiol Drug Saf. 2010;doi:10.1002/pds.1984.

Brixner D, et al. J Med Econ. 2016;doi:10.3111/13696998.2015.1110160.

Damián J, et al. BMC Geriatr. 2013;doi:10.1186/1471-2318-13-6.

Elliott LS, et al. PLoS One. 2017;doi:10.1371/journal.pone.0170905.

Fried TR, et al. J Am Geriatr Soc. 2014;doi:10.1111/jgs.13153.

Jyrkka J, et al. Pharmacoepidemiol Drug Saf. 2011;doi:10.1002/pds.2116.

Koroukian SM, et al. J Clin Oncol. 2006;24:2304-2310.

Lindblad CI, et al. Clin Ther. 2006;doi:10.1016/j.clinthera.2006.08.006.

Lindsay J, et al. Support Care Cancer. 2015;doi:10.1007/s00520-014-2322-0.

Maggiore RJ, et al. J Am Geriatr Soc. 2014;doi:10.1111/jgs.12942.

National Comprehensive Cancer Network. NCCN Guidelines for Older Adult Oncology (Version 1.2019). Available at: www.nccn.org/professionals/physician_gls/pdf/senior.pdf.

Nightingale G, et al. J Clin Oncol. 2015;doi:10.1200/JCO.2014.58.7550.

Nipp RD, et al. Oncologist. 2019;doi:10.1634/theoncologist.2018-0408.

Patel JN, et al. Oncologist. 2018;doi:10.1634/theoncologist.2017-0599.

Reeve E, et al. JAMA Intern Med. 2018;doi:10.1001/jamainternmed.2018.4720.

Roden DM, et al. Clin Pharmacol Ther. 2018;doi:10.1002/cpt.1035.

Sharma M, et al. J Geriatr Oncol. 2016;doi:10.1016/j.jgo.2016.07.010.

Whitman A, et al. Support Care Cancer. 2018;doi:10.1007/s00520-018-4281-3.

For more information:

Jai N. Patel, PharmD, BCOP, is chief of pharmacology research and associate professor in the division of hematology/oncology at Levine Cancer Institute at Atrium Health. He also is a HemOnc Today Editorial Board Member. He can be reached at jai.patel@atriumhealth.org.

Disclosure: Patel reports no relevant financial disclosures.