Machine Learning-Assisted Evaluation of Circulating DNA Quantitative Analysis for Cancer Screening.

While the utility of circulating cell-free DNA (cfDNA) in cancer screening and early detection have recently been investigated by testing genetic and epigenetic alterations, here, an original approach by examining cfDNA quantitative and structural features is developed. First, the potential of cfDNA quantitative and structural parameters is independently demonstrated in cell culture, murine, and human plasma models. Subsequently, these variables are evaluated in a large retrospective cohort of 289 healthy individuals and 983 patients with various cancer types; after age resampling, this evaluation is done independently and the variables are combined using a machine learning approach. Implementation of a decision tree prediction model for the detection and classification of healthy and cancer patients shows unprecedented performance for 0, I, and II colorectal cancer stages (specificity, 0.89 and sensitivity, 0.72). Consequently, the methodological proof of concept of using both quantitative and structural biomarkers, and classification with a machine learning method are highlighted, as an efficient strategy for cancer screening. It is foreseen that the classification rate may even be improved by the addition of such biomarkers to fragmentomics, methylation, or the detection of genetic alterations. The optimization of such a multianalyte strategy with this machine learning method is therefore warranted.

Quantifying circulating cell-free DNA in humans.

To our knowledge, this is the first comprehensive study on the influence of several pre-analytical and demographic parameters that could be a source of variability in the quantification of nuclear and mitochondrial circulating DNA (NcirDNA and McirDNA). We report data from a total of 222 subjects, 104 healthy individuals and 118 metastatic colorectal cancer (mCRC) patients. Approximately 50,000 and 3,000-fold more mitochondrial than nuclear genome copies were found in the plasma of healthy individuals and mCRC patients, respectively. In healthy individuals, NcirDNA concentration was statistically influenced by age (p = 0.009) and gender (p = 0.048). Multivariate analysis with logistic regression specified that age over 47 years-old was predictive to have higher NcirDNA concentration (OR = 2.41; p = 0.033). McirDNA concentration was independent of age and gender in healthy individuals. In mCRC patients, NcirDNA and McirDNA levels were independent of age, gender, delay between food intake and blood collection, and plasma aspect, either with univariate or multivariate analysis. Nonetheless, ad hoc study suggested that menopause and blood collection time might have tendency to influence cirDNA quantification. In addition, high significant statistical differences were found between mCRC patients and healthy individuals for NcirDNA (p < 0.0001), McirDNA (p < 0.0001) and McirDNA/NcirDNA ratio (p < 0.0001). NcirDNA and McirDNA levels do not vary in the same way with regards to cancer vs healthy status, pre-analytical and demographic factors.