Polypharmacy is common among people with cancer, particularly older adults. Some treatments prescribed for supportive oncology care or for coexisting medical conditions are high-risk medications with potential harms that outweigh their benefits for certain patients. These treatments are identified as potentially inappropriate medications (PIMs). Review of PIMs is now included in major oncology guidelines and in regulatory policies, but until recently the primary tool used to measure PIMs in older patients—the Geriatric Oncology Potentially Inappropriate Medications (GO-PIMs) scale—was only validated in patients with hematologic malignancies. Research published in the Journal of the National Comprehensive Cancer Network evaluated the ability of GO-PIMs to identify high-risk medications and their association with outcomes in older patients with hematologic or solid malignancies.
The retrospective cohort study used data from the United States national Veterans Affairs (VA) Cancer Registry and the VA Corporate Data Warehouse (N = 388,113 patients). The study found that high-risk medications were prevalent in patients with solid and hematologic cancers. The most common classes of PIMs were selective serotonin reuptake inhibitors, opioids, benzodiazepines, and corticosteroids. At the time of cancer diagnosis, at least 1 GO-PIM was prescribed for 38% of all patients and 56% of those with moderate-to-severe frailty. Each additional GO-PIM was associated with increasing the odds of being mildly or moderately-to-severely frail at diagnosis by 66%. Higher numbers of GO-PIMs were also associated with increased rates of unplanned hospitalization and death during follow-up.
This was the first study to demonstrate that increasing numbers of GO-PIMs are associated with greater health care utilization and higher mortality among patients both with solid and liquid malignancies, independent of cancer type and stage, comorbidity, and other covariates. The authors recommend that PIMs should be reviewed and optimized at an individual level to enhance care for patients with cancer.
High level
The findings of this study suggest that the prognostic utility of the GO-PIMs scale extends beyond the hematologic malignancies setting to patients with solid tumors. This supports its use in structured review of PIMs and optimization of care in patients initiating cancer treatment. On a population level, GO-PIMs evaluation can help oncology teams adhere to guidelines and improve outcomes in the aging cancer population. Integrating the GO-PIMs scale into electronic health records-based applications has potential to assist oncology and research teams in identifying high-risk medications among their patients. Further research is recommended to examine PIMs in frail patients with cancer, including studies designed to explore prospective measures of frailty in order to determine whether the association between GO-PIMs and frailty reflects causality or correlation.
Ground level
Although somewhat limited by the inability to determine causality, the results of this study reiterate the need to scrutinize indications, weigh benefits vs harms, and consider safer alternatives for high-risk medications on an individual patient level. Patients who are frail and facing the added physiologic stressor of cancer and its treatment are at higher risk of medication side effects. These patients require careful consideration of whether the harms of continued prescribing of PIMs outweigh their benefits. There may be instances where a PIM is appropriate to manage a comorbidity or for supportive oncology care, so clinicians must carefully assess the benefits vs risks of PIMs on an individual level and monitor these patients carefully to determine the best course of action.
