News
Article
Author(s):
Early-stage detection of colorectal cancer (CRC) could be achieved using a noninvasive multiomics liquid biopsy test, according to a recent study.
Multiomics liquid biopsy testing could have clinical potential when used as a noninvasive method for detection of early-stage colorectal cancer (CRC), according to a study published in Molecular Cancer.1 The liquid biopsy could act as a first-line screening method for patients before they undergo a colonoscopy.
CRC is the third most common cancer globally and is the second leading cause of cancer death.2 Early detection has been shown to be the most effective way to improve the outcomes for patients, but adherence to colonoscopies is low in the target demographic. Cell-free DNA (cfDNA) has been used in liquid biopsies that can potentially detect early-stage cancer. This study aimed to use Mutation Capsule Plus (MCP) technology to profile multiple genomic features and develop a multiomics assay to detect CRC in its early stages.1
The researchers used both a healthy control group (96 patients) and a group diagnosed with CRC (93 patients) for the training cohort and 89 patients with CRC and 95 healthy individuals for the validation cohort. Patients had stages I to IV CRC, with stage IV making up 2.2% of the population, stage I at 17.2%, stage II at 39.8%, and stage III at 38.7%. Participants were excluded if they had a history of neoadjuvant therapy or a prior history of cancer.
Mutation, methylation, and genome-wide features were profiled for each cfDNA sample using MCP technology. Potential methylation markers were selected for use in determining the targeted sequencing panel for cfDNA. Normal colorectal issues from healthy patients and patients with CRC were compared with hypermethylated sites in tumor tissues and selected for the panel. Using this information from the training cohort, molecular features were profiled and and the most informative biomarkers used to distinguish CRC were selected.
The mutation test was able to detect at least 1 eligible mutation in 41.9% of patients with CRC compared with 9.4% in healthy participants when testing the training cohort. The APC mutation was detected in 19.4%, the TP53 mutation was found in 23.7%, and the KRAS mutation was found in 17.2% of the patients with CRC.
When a logistic regression model was constructed based on these markers to distinguish patients with CRC vs healthy controls, the area under curve (AUC) was between 0.857 (95% CI, 0.804-0.910) for CNV and 0.904 (95% CI, 0.862-0.945) for DNA methylation. An integrated model was constructed and achieved an AUC of 0.993 (95% CI, 0.985-1.000) in the training cohort with a specificity of 94.8% and a sensitivity of 97.8%.
The integrated model was tested in the validation cohort and achieved an AUC of 0.981 (95% CI, 0.965-0.998), with a specificity of 94.7% and a sensitivity of 92.1%. Sensitivity was lowest in stage 1 CRC (80%) and highest in stages III and IV (100% in both); DNA methylation had the best performance with an AUC of 0.926 (95% CI, 0.883-0.968).
There were some limitations to this study. The sample size was limited and could have affected the evaluation of the model, and the study did not include patients with precancerous lesions, which is an indicator for CRC. Future studies will need to include more diverse patient groups to confirm the efficacy of the model.
"We developed a blood-based method for early detection of CRC and demonstrated the screening potential of multiomics cfDNA-based biomarkers," the authors concluded. "With further validation, this multiomics strategy is expected to be implemented in clinical settings as a first-line screening modality prior to colonoscopy."
References