Is there an association between body mass index and 21-gene recurrence score?. (September 2020)

• Record Type: Journal Article
• Authors: Pomponio, Maria; Burkbauer, Laura; Goldbach, Macy; Keele, Luke; Allison, Kelly C.; Li, Yun R.; Nazarian, Susanna M.; Tchou, Julia
• Is Part Of: Surgical oncology. Volume 34(2020)
• Page Start: 74
• Page End: 79
• Publication Date: 2020-09
• Subjects: Breast neoplasms -- Oncotype DX -- Obesity and breast cancer; Cancer -- Surgery -- Periodicals; Neoplasms -- surgery -- Periodicals; Cancer -- Chirurgie -- Périodiques; Electronic journals; 616.994059
• DOI: 10.1016/j.suronc.2020.01.007
• Languages: English
• ISSNs: 0960-7404

Abstract

Abstract: Purpose: The 21-gene recurrence score (RS) is an established predictor of recurrence for early stage, hormone receptor positive breast cancer. The association between RS and other risk factors such as obesity has not been fully explored. We hypothesized that patients with obesity may present with primary breast cancers with higher recurrence scores. Methods: We identified 1546 patients who have body mass index (BMI) recorded around the time of RS assay. Obesity was classified as per CDC definitions of overweight (BMI 25–30 kg/m 2 ) and obesity (BMI >30 kg/m 2 ). RS was assessed as a continuous variable and according to pre- and post-TAILORx classifications. Kaplan Meier survival analysis was employed to assess the interaction between RS and BMI on overall survival (OS) and disease-free survival (DFS). Results: In univariate analyses, the median RS in patients with overweight was 15, which was significantly lower than the median RS (16) of patients with normal weight (p = 0.03). The overall recurrence rate of patients with obesity was 4.1%, which was significantly worse than the overall recurrence rate of patients with normal and overweight of 2.6% and 1.5%, respectively (p = 0.05). In multivariate analyses using the inverse probability weighted regression adjustment (IPWRA) method to adjust for imbalances between subgroups, patients with overweight or obesity had significantly lower RS than patients with normal weight, correlating to an average decrease in RS value of 2.37 and 1.71, respectively (both p < 0.01). A similar relationship was seen between BMI categories and RS as a categorical variable stratified according to pre- or post-TAILORx categories. This inverse effect was predominantly seen in post-menopausal patients. Despite the generally lower RS in patients with obesity, a high RS in these patients is associated with diminished DFS (p = 0.04). Conclusion: Tumors in post-menopausal women with higher BMI generally have lower RS. DFS is significantly worse in women with obesity whose RS ≥ 30. The reasons for poor outcomes for postmenopausal patients with obesity despite lower presenting RS merits further study.