Risk Stratification for Antiemetics May Prevent Overtreatment

Identifying breast cancer patients who are at low risk for chemotherapy-induced nausea and vomiting before the start of treatment may help avoid unnecessary use of antiemetic medications without compromising quality of care, according to results of a recent trial published in JAMA Oncology.

“Despite multiple variables being associated with the risk of chemotherapy-induced nausea and vomiting, patient-centered factors are rarely considered when making antiemetic recommendations,” wrote authors led by Mark Clemons, MD, of the division of medical oncology at the Ottawa Hospital, Ottawa, Canada.

The study randomized 324 women with breast cancer to receive antiemetics according to standard guidelines (control group) or based on established risk factors for nausea and vomiting (intervention group), such as low alcohol use, history of motion/morning sickness, or younger age. All patients in the intervention group received dexamethasone and a 5HT3 receptor antagonist prior to the first round of chemotherapy, but only those considered at high risk were also given aprepitant, a neurokinin 1 antagonist, which is currently recommended in guidelines.

During the acute treatment period, reports of nausea (54% vs 42%) and vomiting (92% vs 82%) were significantly lower in the intervention group compared with the control group. Since 82% of women in the intervention group were classified as high risk and received aprepitant, it could be argued that all patients should receive aprepitant at cycle 1, the authors noted. However, an exploratory analysis revealed that acute and delayed symptom control was similar among all patients in the intervention group, suggesting aprepitant might not be warranted before the first round of treatment.

Nausea and vomiting are among the most dreaded adverse effects of chemotherapy, often leading to treatment delays and even discontinuation of therapy, the authors note. While risk factors have been identified, they have not been incorporated into guidelines. This risk model could potentially offer a way to avoid the unnecessary cost of additional medications without compromising outcomes.

However, an accompanying editorial by David Warr, MD, FRCPC, of the Princess Margaret Cancer Centre, Toronto, Canada and coauthors found flaws in the stratification model. Only 19% of the patients in the intervention group were designated as low risk prior to the first cycle of chemotherapy, and only about 5% of those did not receive aprepitant before the second cycle.

“Thus, a large majority of supposedly low-risk patients were deemed to need better antiemetic therapy for their second cycle of chemotherapy,” wrote the editorial authors. “The classification scheme was not accurate.”

The study demonstrates that adding a low dose of olanzapine in later rounds of chemotherapy may help some women at higher risk for nausea, according to the editorial.  In the risk managed group, the 40% of patients who received 2.5 mg of olanzapine in cycles 3 and 4 experienced fewer incidences of nausea in both the acute and delayed treatment phases. Other trials have reported high incidences of sedation when olanzapine was administered at a dose of 10 mg.

Although the study does not provide enough evidence to change current practice, there is reason to refine current guidelines, according to the editorial authors. For example, guidelines do not currently distinguish between different types of cyclophosphamide chemotherapy regimens but there is data to suggest that not all chemotherapy containing anthracycline and cyclophosphamide combinations are equally emetogenic.

For the moment, however, choosing antiemetics based on emetogenicity of the chemotherapy as opposed to personal risk factors should remain the gold standard in clinical practice, the editorial concluded. “Even if one can identify an unusually high- or low-risk group for emesis based on patient factors, would clinicians spend the time to elicit this information, calculate the risk of emesis, and change the antiemetics that may be prespecified on paper or electronic orders?” wrote the editorial authors. “Unless this is a single, easily ascertained risk factor such as sex, the answer is ‘probably not.’”