Chemotherapy-induced nausea and vomiting risk predicted before each cycle
medwireNews: A model that incorporates eight readily available risk factors accurately predicts a patient’s likelihood for experiencing grade 2 or above chemotherapy-induced nausea and vomiting (CINV) on a cycle-by-cycle basis, researchers report.
They say: “The primary use of this tool as we see it is to modify current antiemetic prophylaxis in order to avoid unnecessary nausea and vomiting, identify patients who require additional education regarding CINV management, and to monitor current antiemetic protocols.”
The repeated measures cycle-based model was developed using data from 1198 patients, participating in five non-interventional CINV prospective studies, who underwent 4197 chemotherapy cycles. The patients had a median age of 58 years, were mostly female (75%), and generally had early stage disease (76%).
During chemotherapy, 42.2% of patients experienced grade 2 or above CINV, defined as two or more vomiting episodes or a decrease in oral intake due to nausea.
Patients were significantly more likely to experience this level of CINV if they had nausea or vomiting in the previous cycle (odds ratio [OR]=5.17), used non-prescribed antiemetics at home (OR=2.70), were receiving platinum or anthracycline-based chemotherapy (OR=1.94), were aged 60 years or younger (OR=1.41), and had anticipatory nausea and vomiting (OR=1.41), less than 7 hours of sleep the night before chemotherapy (OR=1.34), or a history of morning sickness (OR=1.30). Patients in the first two cycles of chemotherapy were also more likely to have CINV compared with those in later cycles.
George Dranitsaris (The Ottawa Hospital Regional Cancer Centre, Ontario, Canada) and team used these eight variables to develop a risk score that ranged from 0 to 32, where higher scores were associated with an increased risk for grade 2 or worse CINV during the first 5 days of chemotherapy.
Area under the receiver operating characteristic curve analysis showed that the model predicted CINV with 69% accuracy, which the researchers describe as “acceptable.”
The team also determined that a score of 16 or higher prior to each cycle of therapy could be used to identify patients at high risk for grade 2 or above CINV. At this threshold, the model had a sensitivity of 87.4% and a specificity of 38.4%.
Dranitsaris et al accept that this specificity was “low” and could “potentially lead to over treatment of patients at a lower risk for CINV.”
However, they add: “[I]t is important to realize that these risk score thresholds are not fixed and can vary based on the patient or oncologist’s risk tolerance.” They also say that external validation of the model could improve specificity.
Writing in the Annals of Oncology, the researchers conclude that their tool “will be an important source of individual patient risk information for the oncology clinician and may enhance patient care by optimizing the use of the antiemetics in a proactive manner.”
By Laura Cowen
medwireNews is an independent medical news service provided by Springer Healthcare. © 2017 Springer Healthcare part of the Springer Nature group