Association of body mass index with clinicopathological features and survival in patients with primary invasive lobular breast cancer
Karen Van Baelen,Ha-Linh Nguyen,Anne-Sophie Hamy-Petit,François Richard,Maria Margarete Karsten,Guilherme Nader Marta,Peter Vermeulen,Aullene Toussaint,Fabien Reyal,Anne Vincent-Salomon,Luc Dirix,Adam David Dordevic,Evandro de Azambuja,Denis Larsimont,Ottavia Amato,Marion Maetens,Maxim De Schepper,Tatjana Geukens,Sileny N Han,Thaïs Baert,Kevin Punie,Hans Wildiers,Ann Smeets,Ines Nevelsteen,Giuseppe Floris,Elia Biganzoli,Patrick Neven,Christine Desmedt
DOI: https://doi.org/10.1016/j.ejca.2023.112988
Abstract:Purpose: Invasive lobular carcinoma (ILC) represents up to 15% of all breast carcinomas. While the proportion of women with overweight and obesity increases globally, the impact of body mass index (BMI) at primary diagnosis on clinicopathological features of ILC and the prognosis of the patients has not been investigated yet. Patients and methods: We performed a multicentric retrospective study including patients diagnosed with non-metastatic pure ILC. The association of BMI at diagnosis with clinicopathological variables was assessed using linear or multinomial logistic regression. Univariable and multivariable survival analyses were performed to evaluate the association of BMI with disease-free survival (DFS), distant recurrence-free survival (DRFS), and overall survival (OS). Results: The data of 2856 patients with ILC and available BMI at diagnosis were collected, of which 2570/2856 (90.0%) had oestrogen receptor (ER)-positive and human epidermal growth factor receptor (HER2) not amplified/overexpressed (ER+/HER2-) ILC. Of these 2570 patients, 80 were underweight (3.1%), 1410 were lean (54.9%), 712 were overweight (27.7%), and 368 were obese (14.3%). Older age at diagnosis, a higher tumour grade, a larger tumour size, a nodal involvement, and multifocality were associated with a higher BMI. In univariable models, higher BMI was associated with worse outcomes for all end-points (DFS: hazard ratio (HR) 1.21, 95CI 1.12-1.31, p value<0.01; DRFS: HR 1.25, 95CI 1.12-1.40, p value<0.01; OS: HR 1.25, 95CI 1.13-1.37, p value<0.01). This association was not statistically significant in multivariable analyses (DFS: HR 1.09, 95CI 0.99-1.20, p value 0.08; DRFS: HR 1.03, 95CI 0.89-1.20, p value 0.67; OS: HR 1.11, 95CI 0.99-1.24, p value 0.08), whereas grade, tumour size, and nodal involvement were still prognostic for all end-points. Conclusion: Worse prognostic factors such as higher grade, larger tumour size, and nodal involvement are associated with higher BMI in ER+/HER2- ILC, while there was no statistical evidence for an independent prognostic role for BMI. Therefore, we hypothesise that the effect of BMI on survival could be mediated through its association with these clinicopathological variables.