BMI and mortality in hormone-negative breast cancer: A PLCO trial subgroup analysis.

person Reem Chamseddine Weill Cornell Medical College in Qatar, Doha, Qatar info_outline Reem Chamseddine, Stella Major, Arash Rafii, Hind Doha Elmalik

Authors person Reem Chamseddine Weill Cornell Medical College in Qatar, Doha, Qatar info_outline Reem Chamseddine, Stella Major, Arash Rafii, Hind Doha Elmalik Organizations Weill Cornell Medical College in Qatar, Doha, Qatar, Weill Cornell Medicine-Qatar, Doha, Qatar, Hamad Medical Corporation, Doha-Qatar, Qatar Abstract Disclosures Research Funding No funding received None. Background: Body Mass Index (BMI) is a significant prognostic factor in the survival of patients with hormone receptor-positive breast cancer (BC).[1-2] However, the relationship between BMI and survival is not well-established in hormone receptor-negative (HR-) breast tumors. [3] Emerging data is mixed: some studies show that BMI does not impact outcomes in HR- tumors [4], while others reveal worse survival in patients with elevated BMI.[5] In this study, our objective was to corroborate recent evidence about the relationship between BMI and mortality in hormone-negative breast cancer. Methods: We used the PLCO cancer screening trial, a publicly available database containing information about 155,000 participants enrolled in the United States between 1993 and 2001. Data were collected on cancer mortality through 2018 (median follow-up of 19.2 years post enrollment). The methodology of the PLCO trial has been described previously.[6] All female patients with a diagnosis of HR- BC diagnoses were selected for data retrieval, including HR-/HER2+ and HR-/HER2- tumors. Patients with equivocal receptor status were excluded. We used multivariate Cox regression to analyze mortality by BMI category, after adjusting for all significant variables at the bivariate level ( p <0.25). Results: 348 women with confirmed HR- BC at the time of study participation were included. Mean age was 69 years old, and 62 patients (17.8%) passed away due to breast cancer during study enrollment or follow-up. Patients who were of normal weight (BMI 25-<30) at age of trial enrollment were not significantly more likely to survive compared to overweight or obese patients (Hazard Ratio 1.274, p= 0.623). Conclusions: Our results show no significant relationship between BMI and mortality in patients with HR- BC tumors. This finding adds to the growing literature about weight status in hormone receptor-negative tumors. Our data suggests that the mechanism of tumor progression in HR- breast tumors is unlikely to involve signaling mechanisms in adipocytes. Clinical trial information: NCT00002540, NCT01696968 , NCT01696981 , NCT01696994 . Baseline characteristics and breast cancer-specific mortality. Baseline Characteristics P value for bivariate Cox regression Hazard ratio (95% CI) * P value for multivariate Cox regression Age at diagnosis 0.578 - - Race 0.297 - - Number of comorbidities 0.134 Diabetes:0.023 - 0.811 Diabetes: 0.034 Tumor Histology 0.172 - 0.405 Stage <0.001 - <0.001 HER2 Status 0.400 - - Underweight Category ** (BMI < 18.5) - 2.598 (0.220-30.645) 0.448 Normal Weight Category (BMI 18.5 - < 25) - 1.274 (0.485-3.342) 0.623 Overweight Category (BMI 25 - < 30) - 0.456 (0.199-1.046) 0.064 * Values are hazard ratios (HR) relative to the reference (obese category, BMI ³30). HRs were adjusted for history of diabetes and tumor stage, factors significant at p<0.25 at the bivariate level. ** Cell number < 5.

Clinical