Metabolic profile in Mexican young women breast cancer (YWBC) to predict pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC).

Abstract

e12652 Background: YWBC is associated with aggressive biology with a higher mortality rate and recurrence risk. Similar to older patients, pCR after NAC is a surrogate marker of disease-free survival and overall survival in YWBC. However, YWBC who did not achieve pCR did significantly worse. Metabolic profile (MP) has been used for biomarker discovery to monitor chemotherapy response. Few studies focus on using metabolomics to predict pCR in BC. The study aimed to identify MP in YWBC, before NAC to predict response to chemotherapy. Methods: A targeted metabolomics approach was employed to determine the concentration of 11 AAs, free carnitine, and 30 ACCs using electrospray tandem mass spectrometry in plasma of 30 YWBC patients, before NAC to select plasma biomarkers with the most predictive power to identify pCR in YWBC. Patients were classified in responders (R) and non-responders (NR), based on pCR or not-pCR reported by the pathologist department. The differences between R and NR groups were evaluated using an analysis of Random Forest decision trees according to the distribution of the variables. To evaluate the diagnostic power of each biomarker alone, we performed univariate receiver operating characteristics (ROC) analysis on each biomarker to obtain its ROC curve, ROC area under the curve (AUC), and standard error (SE) of the AUC. Results: We observed differences in MP between R and NR YWBC. Ten metabolites contributed most to differentiating between both groups. We developed a model that included two short-chain acyl-carnitines (AC4 and AC5:1) that allowed an AUC of 0.851 (0.767-0.921) and AUC of 0.737 (0.806-0.844) respectively, and ornithine with AUC of 0.723 (0.6-0.626). Conclusions: There are different MP among Mexican YWBC patients classified as R and NR to NAC. To our understanding, this is one of the first studies that evaluate MP in YWBC in Mexico. Further studies are needed to validate a metabolic signature to predict pCR in this patient group.