Publications by authors named "Ngoc-An Trinh Le"

Article Synopsis
  • * This study analyzed blood samples from 159 CRC patients and 158 healthy individuals, using a deep neural network to classify based on DNA fragment length and methylation profiles.
  • * The SPOT-MAS model showed high accuracy with a sensitivity of 96.8% and specificity of 97%, along with strong external validation results, indicating its potential for effective early-stage CRC detection.
View Article and Find Full Text PDF

Identification of tumor-derived mutation (TDM) in liquid biopsies (LB), especially in early-stage patients, faces several challenges, including low variant-allele frequencies, interference by white blood cell (WBC)-derived mutations (WDM), benign somatic mutations and tumor heterogeneity. Here, we addressed the above-mentioned challenges in a cohort of 50 nonmetastatic colorectal cancer patients, via a workflow involving parallel sequencing of paired WBC- and tumor-gDNA. After excluding potential false positive mutations, we detected at least one TDM in LB of 56% (28/50) of patients, with the majority showing low-patient coverage, except for one TDM mapped to that recurred in 30% (15/30) of patients.

View Article and Find Full Text PDF

The identification and quantification of actionable mutations are critical for guiding targeted therapy and monitoring drug response in colorectal cancer. Liquid biopsy (LB) based on plasma cell-free DNA analysis has emerged as a noninvasive approach with many clinical advantages over conventional tissue sampling. Here, we developed a LB protocol using ultra-deep massive parallel sequencing and validated its clinical performance for detection and quantification of actionable mutations in three major driver genes ( and ).

View Article and Find Full Text PDF