Machine learning enables detection of early-stage colorectal cancer by whole-genome sequencing of plasma cell-free DNA

• Main Authors: Nathan Wan, David Weinberg, Tzu-Yu Liu, Katherine Niehaus, Eric A. Ariazi, Daniel Delubac, Ajay Kannan, Brandon White, Mitch Bailey, Marvin Bertin, Nathan Boley, Derek Bowen, James Cregg, Adam M. Drake, Riley Ennis, Signe Fransen, Erik Gafni, Loren Hansen, Yaping Liu, Gabriel L. Otte, Jennifer Pecson, Brandon Rice, Gabriel E. Sanderson, Aarushi Sharma, John St. John, Catherina Tang, Abraham Tzou, Leilani Young, Girish Putcha, Imran S. Haque
• Format: Article
• Language: English
• Published: BMC 2019-08-01
• Series: BMC Cancer
• Subjects: Cell-free DNA; Colorectal cancer; Screening; Whole-genome sequencing; Early-stage cancer
• Online Access: http://link.springer.com/article/10.1186/s12885-019-6003-8

Abstract Background Blood-based methods using cell-free DNA (cfDNA) are under development as an alternative to existing screening tests. However, early-stage detection of cancer using tumor-derived cfDNA has proven challenging because of the small proportion of cfDNA derived from tumor tissue in early-stage disease. A machine learning approach to discover signatures in cfDNA, potentially reflective of both tumor and non-tumor contributions, may represent a promising direction for the early detection of cancer. Methods Whole-genome sequencing was performed on cfDNA extracted from plasma samples (N = 546 colorectal cancer and 271 non-cancer controls). Reads aligning to protein-coding gene bodies were extracted, and read counts were normalized. cfDNA tumor fraction was estimated using IchorCNA. Machine learning models were trained using k-fold cross-validation and confounder-based cross-validations to assess generalization performance. Results In a colorectal cancer cohort heavily weighted towards early-stage cancer (80% stage I/II), we achieved a mean AUC of 0.92 (95% CI 0.91–0.93) with a mean sensitivity of 85% (95% CI 83–86%) at 85% specificity. Sensitivity generally increased with tumor stage and increasing tumor fraction. Stratification by age, sequencing batch, and institution demonstrated the impact of these confounders and provided a more accurate assessment of generalization performance. Conclusions A machine learning approach using cfDNA achieved high sensitivity and specificity in a large, predominantly early-stage, colorectal cancer cohort. The possibility of systematic technical and institution-specific biases warrants similar confounder analyses in other studies. Prospective validation of this machine learning method and evaluation of a multi-analyte approach are underway.