Emergency Department Visits for Emesis Following Chemotherapy: Guideline Nonadherence, OP‐35, and a Path Back to the Future

Short abstract

Better adherence to supportive care guidelines is needed, especially for common challenges such as chemotherapy‐induced nausea and vomiting. This commentary remarks on recently published data on adherence to guidelines, considering treatment options and new strategies to reduce chemotherapy complications.

Practice variation is increasingly understood to be a contributor to the high cost of medical care and is as important in oncology as in other areas of medicine. Evidence‐based medicine, led by randomized controlled trials, has provided us with guidelines for the provision of care to our patients; it is no exaggeration to state that oncologists are inundated with such guidelines and recommendations for best practices. It is also true that, as oncologists, our attention is especially focused on those recommendations that relate to the choice of the optimal antineoplastic therapy. Which chemotherapy regimen is best in a given patient with a certain tumor of a given stage? Should radiation therapy be included in the treatment plan? How often should we repeat the tumor markers?

Other guidelines, equally based on evidence and randomized trials, guide our behavior in the arena of supportive care but are attended to with less rigor, as suggested in a paper by Navari et al. in this issue of The Oncologist [1]. Thus, the use of colony‐stimulating factors to prevent chemotherapy‐related neutropenia and infection, anticoagulation drugs to prevent and treat thromboembolic complications, and the management of dehydration, anemia, and pain stemming from cancer and its treatment are all well described in various recommendations, and these issues are a routine part of day‐to‐day cancer management.

Nonetheless, studies indicate that it is precisely these complications that account for the bulk of emergency department (ED) visits by patients undergoing chemotherapy and, secondarily, their subsequent hospital admission [2]. As a consequence, these issues are a major driver of cost to the health care system. Thus, it is important to understand why such acute care is necessary so often, and how we might prevent these complications. Increasingly, health care payors are paying attention, and Medicare, as discussed below, has launched OP‐35 as a strategy to reduce chemotherapy complications.

In this issue of The Oncologist, Navari et al. address the frequency of chemotherapy‐induced nausea and vomiting (CINV), one of the oldest supportive care challenges in oncology [1]. This complication, along with hair loss, is the one most familiar and most universally dreaded by patients naïve to chemotherapy [3]. Its significance was amplified with the introduction of high‐dose cisplatin and doxorubicin in the late 1970s, which were crippled in their use until improved antiemetics made their appearance.

Skipping ahead 4 decades, the sophistication of our pharmacopeia for CINV has dramatically advanced. With our current antiemetic agents, uncontrolled or refractory emesis should be relatively uncommon, even in the context of highly emetogenic chemotherapy (HEC), generally defined as chemotherapy regimens that entail a greater than 90% risk of emesis. Guidelines recommend, indeed mandate, the use of a neurokinin‐1 (NK1) receptor antagonist, such as aprepitant or fosaprepitant, together with a 5‐hydroxytryptamine (5HT3) receptor antagonist, such as ondansetron or granisetron, and dexamethasone for antiemetic prophylaxis based on randomized trials [4, 5, 6, 7]. With this regimen, ED visits and hospitalizations for refractory emesis should be relatively infrequent.

Navari et al. used a large database from 2012 to 2018 to look at acute care ED visits within 30 days of chemotherapy administration. They found that those undergoing HEC had an increased rate of acute care for CINV of 2.29 (95% confidence interval, 2.11–2.49) relative to those undergoing non‐HEC chemotherapy. The HEC patients were those receiving cisplatin, carboplatin, doxorubicin and cyclophosphamide (AC), or oxaliplatin [1]. This elevated hazard ratio for HEC versus non‐HEC chemotherapy was not surprising.

What was more consequential was the analysis of physician adherence to the CINV guidelines for the triple‐drug antiemetic therapy. They found adherence to guidelines in 66% of patients receiving cisplatin and 76% of those receiving AC, versus only 14% of those receiving carboplatin and 11% of those receiving oxaliplatin. These data imply that the increased rates of CINV‐related acute care, either ED visits or hospitalizations, are due at least in part to a lack of oncologist compliance with guidelines and are thus highly preventable. The data suggest that the key to solving this problem lies in our own compliance with the known guidelines. Other studies have confirmed these poor levels of physician adherence with antiemetic guidelines leading to poor outcomes regarding CINV [8, 9, 10, 11]. Mahendraratnam et al. [11] used data from the MarketScan database to show a 50% nonadherence rate with the recommended three drug regimen, mostly reflecting nonprescription of the NK1 receptor antagonist. There tends to be great variability among oncologists, with some achieving >90% adherence with guidelines whereas adherence by others is significantly less [9]. For those who were nonadherent, once again the drug most frequently omitted was from the NK1 class, such as aprepitant or fosaprepitant. Exactly why oncologists are nonadherent probably reflects a panoply of personal and systemic reasons, similar to patient nonadherence. It has been estimated that 48% of the cost of oncology care results from preventable toxicities such as CINV [12].

This nonadherence with the antiemetic guidelines is, therefore, a major issue and strategies to rectify or reduce noncompliance are essential. Whatever the underlying cause or causes, the Centers for Medicare and Medicaid Services has sought ways to reduce chemotherapy‐related expenses and, for this purpose, promulgated rule OP‐35 (OP refers to Outpatient). This rule, “Admissions and Emergency Department Visits for Patients Receiving Outpatient Chemotherapy,” was activated in 2020 (https://www.cms.gov/Medicare/Quality‐Initiatives‐Patient‐Assessment‐Instruments/HospitalQualityInits/HospitalOutpatientQualityReportingProgram). OP‐35 will require that hospitals monitor as a quality outcome measure the rates of ED visits, hospitalizations, and rehospitalizations for patients with cancer (leukemia is excluded from these reporting requirements) within 30 days of chemotherapy for 10 toxicities or symptoms considered preventable (anemia, dehydration, diarrhea, emesis, fever, nausea, neutropenia, pain, pneumonia, and sepsis). The rates of these acute care outcomes will be publicly reported. This metric can be used by patients for the purposes of hospital comparison, and eventually, it may be utilized in hospital and physician reimbursements. CINV represents 2 of the 10 toxicities included in OP‐35.

OP‐35 is a laudable attempt to improve guideline adherence for CINV and the other included preventable toxicities. However, one wonders to what extent hospital reporting requirements will influence individual practitioners. Strategies to enhance physician adherence with antiemetic guidelines should be at the forefront of efforts to improve these outcomes. Such measures include appropriate antiemetic regimens incorporated into electronic or written chemotherapy order sets and physician education regarding current recommended drugs for antiemesis. The identification of patients at particularly high risk for CINV who require special monitoring and care could also be beneficial [13]. Handley et al. have identified a number of evidence‐based strategies for reducing the toxicities included in the OP‐35 mandate [2]. Such efforts, and a reduction in these ED visits and hospitalizations, would not only produce an economic benefit, but would also improve the health and welfare of our patients.

Our view is that guideline‐based management of supportive care for chemotherapy toxicities makes sense and streamlines care. However, supportive care guidelines need to evolve as rapidly as the rest of oncology. Olanzapine, an atypical antipsychotic drug, has been shown to be efficacious as an antiemetic in a Cochrane review [14] and has been proposed as a replacement for the NK1 receptor antagonist as part of the three‐drug regimen for HEC to maintain the same level of antiemetic efficacy [15]. Furthermore, a recent randomized trial demonstrated enhanced efficacy when it was added as a fourth drug [16]. Thus, the most recent guidelines [7, 17] suggest that future HEC antiemetic regimens may contain four drugs, with the addition of olanzapine, and even now is listed as an option. At Columbia/New York Presbyterian, olanzapine has already been added to standard order sets for HEC.

Concurrently, concerns regarding the use of high‐dose corticosteroids in conjunction with immune checkpoint inhibitors have been raised [18]. As discussed in an accompanying editorial by Janowitz et al. [19], there are real concerns about the ability of dexamethasone and other glucocorticoids to reduce the immune response generated by checkpoint inhibitors. Emerging data suggest dexamethasone may be harmful in the setting of checkpoint inhibitors, and one has to ask whether there are any other settings in oncology in which an immune response is necessary to achieve full chemotherapy benefit. For now, guidelines continue to recommend the continued use of dexamethasone in regimens that contain immune checkpoint inhibitors [7], but with the addition of olanzapine to the supportive care armamentarium, it is time to re‐evaluate the role of dexamethasone and consider its elimination. At a recent analysis, there were 4,720 immuno‐oncology drugs in the pipeline and 6,281 trials testing immuno‐oncology agents [20]. It is difficult to understand how a field so committed to the induction of an immune response in virtually every tumor type can ignore the possibility that in HEC, dexamethasone may have outlived its usefulness.

Thus, in highlighting oncologist noncompliance, Navari et al. open a Pandora's box: clearly better adherence to supportive care guidelines is needed. OP‐35 is intended to enhance that adherence. But it is not clear that traditional antiemetic regimens for highly emetogenic chemotherapy are optimized for regimens that require an immune response.

Disclosures

Alfred I. Neugut: Pfizer, Teva, Otsuka, Esai, United Biosource Corp. Hospira (C/A), EHE International (SAB). Susan E. Bates indicated no financial relationships.

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board