Early colorectal cancer detection with a spectroscopic liquid biopsy.

Abstract

e15627 Background: Earlier diagnosis and treatment of colorectal cancer (CRC) maximizes the opportunity to combat or control disease progression. The average 5-year survival rate after diagnosis decreases from 91% in early-stage CRC, to as low as 15% for stage IV CRC. Furthermore, rapid detection and removal of pre-cancerous adenomas – e.g., advanced adenomas (AA) – can significantly improve survival rates of affected patients. However, current stool-based tests have inadequate AA sensitivities, such as FIT (24%) and FIT-DNA (42%). Liquid biopsies have great potential to supplement FIT tests and improve diagnostic pathways, but current blood tests based on tumor-derived biomarkers also have limited sensitivity for AA detection. Methods: The Dxcover® Cancer Liquid Biopsy is based on Fourier-transform infrared (FTIR) spectroscopy applied to serum from a standard blood sample. The spectral data are collected and analyzed using pattern recognition and machine learning algorithms to detect disease-specific signatures. We initially examined test performance for the detection of multiple cancer types. Additionally, we have analyzed a retrospective cohort of samples comprising 100 CRC, 99 AA removed by surgical resection and 97 colonoscopy screening controls. Results: The CRC classifier from the discovery multi-cancer dataset resulted in 74% sensitivity with 91% specificity when differentiating CRC and non-cancer, which surpasses the targets set by the Centers for Medicare & Medicaid Services (CMS) for coverage of CRC tests. Receiver operating characteristic (ROC) analysis reported an area under the curve (AUC) value of 0.91. The machine learning algorithms can be fine-tuned to maximize either sensitivity or specificity depending on the requirements of different patient pathways. When tuned for higher sensitivity, the model produced 97% sensitivity (49% specificity), and when tailored for greater specificity (97%) the sensitivity was 47%. We have now progressed these findings to examine the ability of the technology to differentiate patients with CRC, AA and colonoscopy controls. Furthermore, we assess whether combining spectral data with additional clinical information, such as biomarker data and patient demographics, can enhance test performance. Conclusions: A rapid blood test that is sensitive to AA and early-stage CRC could improve patient prognosis and reduce cancer burden. With further development, this liquid biopsy could have a profound impact on the earlier detection of CRC.