Investigators wanted to analyze the diagnostic performance of methylation-sensitive restriction enzyme digestion followed by sequencing using cell-free DNA and explore the cancer signal origin of the cancer using DNN analyses for colorectal and lung cancers.
Methylation-sensitive restriction enzyme digestion followed by sequencing (MRE-Seq) was found to be an advantageous method for identifying global hypomethylation patterns in liquid biopsy samples and precisely diagnosing colorectal and lung cancers, as well as determining the cancer signal origin (CSO) of the cancer using deep neural network (DNN) analysis, according to Scientific Reports.
Methylation patterns in cell-free DNA (cfDNA) have come about as a hopeful genomic feature for identifying the presence of cancer and detecting its origin. The aim of this study was to analyze the diagnostic performance of MRE-Seq using cfDNA and explore the cancer signal origin (CSO) of the cancer using DNN analyses on liquid biopsy of colorectal and lung cancers. Liquid biopsy has shown potential for early cancer detection and diagnosis.
First, the researchers created a selective MRE-Seq method with a DNN learning-based prediction model using demethylated-sequence-depth patterns from 63,266 CpG sites using SacII enzyme digestion. A total of 191 patients with stage I-IV cancers (95 lung cancers and 96 colorectal cancers) and 126 participants who did not have cancer were enrolled in this study.
The study showed an area under the receiver operating characteristic curve (AUC) of 0.978 with a sensitivity of 78.1% for colorectal cancer and an AUC of 0.956 with a sensitivity of 66.3% for lung cancer; the combined specificity was 99.2%. Sensitivities for stages I-IV for colorectal cancer ranged from 76.2% to 83.3%, while for lung cancer, sensitivities ranged from 44.4% to 78.9%, again at a combined specificity of 99.2%. The CSO model’s true-positive rates were 94.4% for colorectal cancers and 89.9% for lung cancers.
Methylated DNA immunoprecipitation coupled with high-throughput sequencing was used as an affinity purification–based method, with AUCs of 0.978, 0.918, and 0.971 for acute myeloid leukemia, pancreatic cancer, and lung cancer, respectively. However, the overall CSO prediction accuracy was less than 70%, which isn’t high enough for it to become a practical tool for early multicancer screening.
The researchers determined that the MRE-seq method was more suitable for diagnosing early cancer in liquid biopsy using a small amount of cfDNA in regular clinical testing.
The general accuracy of the liquid biopsy method was observed to be lower for lung vs colorectal cancer, probably because of the greater diversity of histological subtypes and larger differences in DNA methylation patterns in lung cancer compared with colorectal cancer. But the accuracy of lung cancer identification is anticipated to improve with the use of an adequate number of lung adenocarcinoma samples in the training set.
Additionally, it may be possible to create more accurate and specific liquid biopsy methods for cancer diagnosis by analyzing the methylation patterns of cancer and benign disease samples. By including samples with both, the study might give a more realistic assessment of the accuracy and reliability of the liquid biopsy testing.
“Our proposed MRE-seq performance is comparable to the previous bisulfite sequencing method in accuracy of the cancer detection and classification of CSO with use of [a] relatively smaller amount of blood from a single tube collection and [requires] lower sequencing depth coverage compared to the whole genome bisulfite sequencing analysis,” said the researchers.
These features enhance the practicality for routine clinical adoption by decreasing the requirement of patient blood samples and decreasing testing cost.
“Additional multicenter studies are needed to validate the results with the independent test samples and confirm the utility of the method in different populations,” emphasized the researchers.
This study has some limitations. First, each case study was undertaken at just 1 center, which may present selection bias and curb the generalizability of the findings. Additionally, no follow-up was conducted for the healthy controls, which might have led to misclassification bias if certain people developed cancer following the study.
“The study highlights the potential of liquid biopsy methods for improving cancer diagnosis and detection, suggesting that further development and validation of these methods could have important implications for improving cancer survival and quality of life,” concluded the study authors.
Kwon HJ, Shin SH, Kim HH, et al. Advances in methylation analysis of liquid biopsy in early cancer detection of colorectal and lung cancer. Sci Re. Published online August 19, 2023. doi.org:10.1038/s41598-023-40611-w