Metabolic abnormalities and survival among patients with non-metastatic breast cancer.

ABSTRACT

BACKGROUND: Research on the impact of metabolic abnormalities on breast cancer prognosis is limited by small samples and assessment of laboratory values at a single time point, often prior to cancer diagnosis and treatment. In this population-based cohort, time-updated laboratory values were adjusted for cancer treatment to assess the association between metabolic risk factors (glucose, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglycerides) and breast cancer survival. METHODS: 13,434 women diagnosed with stage I-III breast cancer from 2005-15 at Kaiser Permanente were included. All outpatient fasting glucose, HDL-C, LDL-C, and triglyceride values from diagnosis through 2019 or death were extracted from electronic medical records. Risk of breast cancer-specific mortality was evaluated with Cox proportional hazards models adjusted for metabolic labs, demographics, body mass index, diabetes, dyslipidemia and anti-hypertensive medications, tumor characteristics (stage, ER and HER2 receptor status) and cancer treatment (use of chemotherapy, tamoxifen, and aromatase inhibitors). RESULTS: Mean (SD) age at diagnosis was 62.3 (11.8) years. Over a median follow-up of 8.6 years, 2,876 patients died; 1,080 of breast cancer. Patients with low HDL-C (≤ 45 vs. > 45 mg/dL) had higher breast cancer-specific mortality (HR, 1.77; 95% CI, 1.53-2.05), as did those with elevated fasting glucose (> 99 vs. 60-99 mg/dL) (HR, 1.19; 95% CI, 1.03-1.37). Elevated levels of triglycerides and LDL-C were not associated with breast cancer-specific mortality. CONCLUSIONS: High fasting glucose and low HDL-C evaluated over time after cancer diagnosis were associated with higher breast cancer mortality independent of cancer treatments and changes in other metabolic risk factors. Future studies should address whether pharmacologic or lifestyle treatment of glucose and lipids after breast cancer diagnosis can optimize survival outcomes.