Abstract
The polyunsaturated fatty acid (PUFA) elongase, ELOVL5, is upregulated in breast cancer (BC) vs. adjacent normal tissue. We performed a comprehensive lipid metabolomic analysis of serum using high-resolution accurate mass spectrometry from two case-control studies that included non-BC, BC subjects pre-surgery, and BC subjects one-month post-surgery to determine if the metabolic signatures of over-active fatty acid elongation and other lipid changes could be detected in BC vs. non-BC subjects: study 1 (n = 48: non-BC, n = 69: pre-surgery BC); study 2 (blinded validation: n = 121: non-BC, n = 62: pre-surgery BC, n = 31: one month post-surgery). The ratio of the ELOVL5 precursor, linoleic acid (18:2) to a non-ELOVL5 precursor, oleic acid (18:1) was evaluated in multiple lipid pools (phosphatidylethanolamine (PtdEtn), phosphatidylcholine (PtdCho), lyso-PtdCho, and free fatty acids). This ratio was lower in pre-surgery BC subjects in all pools in both studies (p < 0.001). At one-month post-surgery, the 18:2/18:1 ratios increased vs. pre-surgery and were no longer different from non-BC subjects (p > 0.05 expect for lyso-PtdCho). In contrast to the elongation biomarkers, docosahexaenoic acid (22:6n-3) containing ethanolamine plasmalogen (EtnPls) species were observed to be further decreased in BC subjects one-month post-surgery vs. pre-surgery levels (p < 0.001). These results are consistent with the hypothesis that ELOVL5 is upregulated in BC tissue, which would result in the selective depletion of 18:2 vs. 18:1 containing lipid species. Surgical removal of the tumor removes the overactive ELOVL5 effect on serum lipids. In contrast, the low EtnPls levels do not appear to be caused by BC tumor activity and may be pre-existent and a possible risk factor for BC. These results indicate that it may be possible to screen for both breast cancer risk and breast cancer activity using a simple blood test.