Abstract
the rest were used as test data set to valid the predicted “AminoIndex”. Results: Plasma concentrations of several amino acids were significantly changed in breast cancer patients compared to control subjects in study data set. Finally, “AminoIndex” for breast cancer composed with six amino acids (Gln, Ala, ABA, Trp, Orn, and Arg) was predicted. To evaluate the performance, the ROC curve was calculated, and this gave an AUC of ROC of 0.832 using the study data. Validation of predicted “AminoIndex” using test data set resulted same discriminating performance (AUC of ROC of 0.822), suggesting the robustness of the predicted classifier. Furthermore, predicted AminoIndex showed notably features. 1. The index could discriminate breast cancer patients in early stages. 2. The index showed higher discrimination performance than those of existing tumor markers especially in stages 0, I, and II patients. 3. The index could equally discriminate breast cancer patients of any histological types. Therefore, predicted “AminoIndex” would be suitable for screening and early detection of breast cancer patients. Conclusion and Perspectives: In this study, we demonstrated that change of plasma amino acid profile would be a helpful tool for early detection of breast cancer patients. “AminoIndex” would be useful to concentrate and inspire the candidates for further survey such as mammography. To evaluate the efficacy of this method, cohort studies are ongoing.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have