Salivary metabolomics with alternative decision tree-based machine learning methods for breast cancer discrimination
The aim of this study is to explore new salivary biomarkers to discriminate breast cancer patients from healthy controls.
Saliva samples were collected after 9 h fasting and were immediately stored at − 80 °C. Capillary electrophoresis and liquid chromatography with mass spectrometry were used to quantify hundreds of hydrophilic metabolites. Conventional statistical analyses and artificial intelligence-based methods were used to assess the discrimination abilities of the quantified metabolites. A multiple logistic regression (MLR) model and an alternative decision tree (ADTree)-based machine learning method were used. The generalization abilities of these mathematical models were validated in various computational tests, such as cross-validation and resampling methods.
One hundred sixty-six unstimulated saliva samples were collected from 101 patients with invasive carcinoma of the breast (IC), 23 patients with ductal carcinoma in situ (DCIS), and 42 healthy controls (C). Of the 260 quantified metabolites, polyamines were significantly elevated in the saliva of patients with breast cancer. Spermine showed the highest area under the receiver operating characteristic curves [0.766; 95% confidence interval (CI) 0.671–0.840, P < 0.0001] to discriminate IC from C. In addition to spermine, polyamines and their acetylated forms were elevated in IC only. Two hundred each of two-fold, five-fold, and ten-fold cross-validation using different random values were conducted and the MLR model had slightly better accuracy. The ADTree with an ensemble approach showed higher accuracy (0.912; 95% CI 0.838–0.961, P < 0.0001). These prediction models also included spermine as a predictive factor.
These data indicated that combinations of salivary metabolomics with the ADTree-based machine learning methods show potential for non-invasive screening of breast cancer.
KeywordsBreast cancer Biomarker Saliva Polyamines Metabolomics Alternative decision tree
We thank Editage (www.editage.jp) for English language editing.
This study was funded by JSPS KAKENHI Grant Numbers 16H05408 and 25461996, and research Grants from the Yamagata Prefectural Government and the City of Tsuruoka.
Compliance with ethical standards
Conflict of interest
The authors declare no competing financial interests. M. Sunamura and M. Sugimoto hold unpaid advisory positions in a commercial organization. No other author declares non-financial competing interests.
All procedures performed in studies involving human participants were in accordance with the ethical standards of the ethics committees of Keio University (No.20120143), Teikyo University (No.15-047-2), and Kitasato University Kitasato Institutional Hospital (No.17006) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent was obtained from all individual participants included in the study.
- 1.International Agency for Research on Cancer. http://gco.iarc.fr/. Accessed 11 Jan 2019
- 2.Cancer Incidence in Five Continents, CI5plus. IARC Cancer Base No.9. Lyon, International Agency for Research on Cancer. http://ci5.iarc.fr. Accessed 11 Jan 2019
- 3.World Health Organization mortality database. http://www.who.int/healthinfo/mortality_data/en/. Accessed 11 Jan 2019
- 5.Habbema JD, van Oortmarssen GJ, van Putten DJ, Lubbe JT, van der Maas PJ (1986) Age-specific reduction in breast cancer mortality by screening: an analysis of the results of the Health Insurance Plan of Greater New York study. J Natl Cancer Inst 77(2):317–320Google Scholar
- 6.OECD Health Statistics https://doi.org/10.1787/health-data-en. Accessed 11 Jan 2019
- 19.Brooks MN, Wang J, Li Y, Zhang R, Elashoff D, Wong DT (2008) Salivary protein factors are elevated in breast cancer patients. Mol Med Rep 1(3):375–378Google Scholar
- 36.Freund Y, Mason L (1999) The alternating decision tree learning algorithm. Icml 1999:124–133Google Scholar
- 41.Elshof LE, Tryfonidis K, Slaets L, van Leeuwen-Stok AE, Skinner VP, Dif N et al (2015) Feasibility of a prospective, randomised, open-label, international muticentre, phase III, non-inferiority trial to assess the saftey of active surveillance for low risk ductal carcinoma in site The LORD study. Eur J Cancer 51(12):1497–1510CrossRefGoogle Scholar