Evaluating response to interventions for vasomotor symptoms in patients with breast cancer: A patient-centered approach.
Journal Articles
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
e24126 Background: Vasomotor symptoms (VMS), such as hot flashes, are a common reason for early discontinuation of endocrine therapy for patients (pts) with breast cancer (BC). The optimal intervention for VMS remains unknown. Using a novel symptom change analysis and regression trees, we examined whether clinically important change in VMS severity depends on baseline symptomatology. Methods: In this prospective study, pts with BC experiencing VMS chose either a lifestyle (LS), or non-LS (i.e. complementary and alternative therapies [CAM], prescription medications [PM], or endocrine therapy modification [ETM]) intervention. The primary outcome was change in symptom severity using the 10-point Hot Flush Rating Scale (HFRS). Patients declining interventions were included as controls. At the end of the 6-week intervention, participants rated the effectiveness of their chosen intervention on a five-point Likert scale. A logistic sigmoid function and Bayesian optimization were used to weight the change in VMS score based on baseline VMS severity. A regression tree was trained to predict which intervention resulted in the greatest improvement in VMS at 6 weeks. Results: 100 baseline and follow-up questionnaires from 85 pts were included in an intention to treat analysis. The median baseline HFRS was 5.0 (IQR 3.33 ,7.00). Selected interventions included LS (27%), CAM (25%), PM (11%) and ETM (8%), with 29% declining interventions. After removal of missing data, 59 individuals provided responses for both the HFRS score and the effectiveness score. Higher baseline HFRS scores had greater impact on symptom change scores. The largest difference in VMS improvement was noted between the non-LS (median symptom change –0.82, IQR –2.00, –0.01) and LS/control groups (median symptom change – 0.38, IQR –1.46, 0.00) (Table 1). Of non-LS interventions, CAM had the greatest improvement in VMS severity. Pre-menopausal individuals who pursued LS interventions had the least improvement in VMS severity. Conclusions: Interventions for VMS were most impactful among pts with greater baseline symptoms. While LS interventions were the most commonly selected intervention, the resulting improvement in symptoms at 6 weeks was negligible, and alternative strategies should be encouraged. Future studies integrating pt preferences and accounting for baseline symptom severity are needed. [Table: see text]
