Wang Xiao-Ling, Qin Shuo, Liu Li, Jia Yu-Qi, Wang Ying, Zhou Yu-Jie, Ramos-Molina Bruno, Valladares-Ayerbes Manuel, Eidens Moritz, Liu Zhi-Zhen
Department of Clinical Laboratory, Shanxi Hospital of Traditional Chinese Medicine, Taiyuan, China.
Academy of Medical Sciences, Shanxi Medical University, Taiyuan, China.
J Gastrointest Oncol. 2025 Jun 30;16(3):965-977. doi: 10.21037/jgo-2025-319. Epub 2025 Jun 27.
Colorectal cancer (CRC) is a common malignant tumor of the digestive tract. Screening of high risk populations and early diagnosis of CRC are of great importance. Tissue factor pathway inhibitor 2 () methylation is an important indicator for screening CRC. However, the sensitivity and specificity of methylation vary significantly across different studies, and its diagnostic performance is susceptible to the influence of detection methods and sample types. This study set out to evaluate the diagnostic value of methylation in blood and stool samples for CRC.
A search strategy based on the Population, Intervention, Comparison, Outcomes and Study (PICOS) principle was employed to retrieve relevant literatures from the PubMed, Web of Science, and EMBASE databases focused on the diagnosis of CRC patients with methylation. The literature must include the following details: the types of tissue samples tested, detailed sample sizes of both control and case groups, comprehensive diagnostic parameters, and specific methodologies for methylation detection. The quality of the included articles was evaluated using the guidelines for the Quality Assessment of Diagnostic Accuracy Studies (QUADAS). Deeks' funnel plot was used for the assessment of the publication bias. Data were extracted from the studies, and the random-effects model was used for the meta-analysis.
In total, 14 diagnostic test accuracy studies, comprising 3,330 subjects (1,983 CRC patients and 1,347 non-CRC controls), were included in the meta-analysis. The meta-analysis demonstrated a non-significant publication bias (P=0.29). Meta-regression further identified country, detection methodology, and sample type as significant contributors to heterogeneity. The combined sensitivity and specificity of in diagnosing CRC were 0.83 [95% confidence interval (CI): 0.72-0.91] and 0.96 (95% CI: 0.93-0.97), respectively. While the combined positive likelihood ratio (PLR) was 19.2 (95% CI: 11.1-33.5), the combined negative likelihood ratio (NLR) was 0.18 (95% CI: 0.10-0.31), the diagnostic odds ratio (DOR) was 109 (95% CI: 45-261), and the area under the summary receiver operating characteristic curve (SROC) for the included studies was 0.97 (95% CI: 0.95-0.98).
Our findings showed that methylation in blood and stool samples could be a potential biomarker for the detection of CRC. methylation demonstrates potential as a non-invasive screening tool and dynamic monitoring biomarker for CRC, offering a complementary or alternative approach to invasive procedures such as colonoscopy. This is particularly applicable for preliminary risk stratification in populations with low screening adherence or high-risk profiles.
结直肠癌（CRC）是一种常见的消化道恶性肿瘤。对高危人群进行筛查以及CRC的早期诊断至关重要。组织因子途径抑制剂2（）甲基化是筛查CRC的一项重要指标。然而，不同研究中甲基化的敏感性和特异性差异显著，其诊断性能易受检测方法和样本类型的影响。本研究旨在评估血液和粪便样本中甲基化对CRC的诊断价值。
采用基于人群、干预、对照、结局和研究（PICOS）原则的检索策略，从PubMed、Web of Science和EMBASE数据库中检索聚焦于甲基化诊断CRC患者的相关文献。文献必须包含以下详细信息：检测的组织样本类型、对照组和病例组的详细样本量、综合诊断参数以及甲基化检测的具体方法。使用诊断准确性研究质量评估指南（QUADAS）对纳入文章的质量进行评估。采用Deeks漏斗图评估发表偏倚。从研究中提取数据，并使用随机效应模型进行荟萃分析。
荟萃分析共纳入14项诊断试验准确性研究，包括3330名受试者（1983例CRC患者和1347例非CRC对照）。荟萃分析显示发表偏倚不显著（P = 0.29）。荟萃回归进一步确定国家、检测方法和样本类型是异质性的重要影响因素。甲基化诊断CRC的合并敏感性和特异性分别为0.83 [95%置信区间（CI）：0.72 - 0.91]和0.96（95% CI：0.93 - 0.97）。合并阳性似然比（PLR）为19.2（95% CI：11.1 - 33.5），合并阴性似然比（NLR）为0.18（95% CI：0.10 - 0.31），诊断比值比（DOR）为109（95% CI：45 - 261），纳入研究的汇总受试者工作特征曲线（SROC）下面积为0.97（95% CI：0.95 - 0.98）。
我们的研究结果表明，血液和粪便样本中的甲基化可能是检测CRC的潜在生物标志物。甲基化作为一种非侵入性筛查工具和CRC的动态监测生物标志物具有潜力，为结肠镜检查等侵入性程序提供了一种补充或替代方法。这对于筛查依从性低或高危人群的初步风险分层尤其适用。
