| Objective:The diagnostic value of multi-target fecal DNA(mt-s DNA),methylated SEPT9 DNA plasma assay(m SEPT9)and colorectal CT(CTC)for colorectal cancer(CRC)were evaluated separately using general meta-analysis,and the three diagnostic methods were compared by network meta-analysis and the optimal screening method was selected,which provided more reference options for colorectal cancer screening methods.Methods:1.Chinese and English literature related to screening methods for early colorectal cancer were searched in eight databases: China National Knowledge Infrastructure(CNKI),Wanfang Data Knowledge Service Platform(WF),VIP,China Biomedical Literature Service System(CBM),Pubmed,Embase,Web of Science,and Cochrane Library.The limited search period was from January 1992 to December 2022.The search was conducted mainly by computerized search and supplemented by manual search,and the search terms were referred to the MESH subject word list,using a combination of“subject word” + “free word” search term strategy to eliminate free words that would not increase or decrease the volume of literature in order to streamline the search form.The literature diagnosis of “multitarget fecal DNA”,“plasma Septin9 gene methylation”,“colorectal CT”and “colorectal tumor” were all included in the scope of the study,and the reading title,abstract and full text were screened according to the inclusion criteria and exclusion criteria.Finally,the 37 included studies were assessed for quality and the required data were extracted.2.Use a fixed-effect model for mt-s DNA after threshold effect evaluation using Meta-Disc software,and use random-effects models to combine effect sizes for m SEPT9,CTC(threshold≥10 mm)and CTC(threshold≥6 mm);in Stata 14.0,the sensitivity analysis of the data is performed to verify the stability of the data,and if the data is unstable,the cause of instability is found.Subgroup analysis of the diagnosis of non-threshold heterogeneity was performed to identify sources of heterogeneity;The Fagan nomogram was used to analyze the predictive value of diagnosis and prediction value,and the predictive value of four sets of data on CRC was compared and analyzed.Conduct testing for Deeks publication bias;R 4.2.2 software was used to run the ANOVA model to compare and analyze mt-DNA,m SEPT9 and CTC according to the data of“relative sensitivity”,“relative specificity” and “superiority index”.Results:Finally,37 studies were included,including 8 mt-s DNA studies,20 m SEPT9 studies,and 9 CTC studies.1.Results of conventional Meta-analysis: mt-s DNA,m SEPT9,CTC(threshold≥10 mm)and CTC(threshold≥6 mm)diagnosed colorectal cancer with 95% confidence intervals(CI)of 0.93(0.90,0.94),0.76(0.69,0.82),0.89(0.81,0.94),0.77(0.67,0.85),with combined specificity of 0.87(0.79,0.93),0.91(0.87,0.94),0.97(0.94,0.98),0.89(0.85,0.92),and combined positive likelihood ratios of 7.4(4.4,12.5),8.7(5.7,13.5),27.6(14.6,52.2),and 7(5.3,9.1),combined negative likelihood ratios were 0.08(0.06,0.11),0.27(0.20,0.34),0.11(0.06,0.20),and0.25(0.17,0.38),and combined diagnostic ratio DOR were 90(51,159),33(19,57),242(92,636),27(16,46),and combined AUC of 0.96(0.93,0.97),0.91(0.88,0.93),0.98(0.96,0.99),and 0.92(0.89,0.94).The diagnostic predictions were: the post-test probability of detecting colorectal cancer was 45% at a positive likelihood ratio(PLR)of 7 for mt-s DNA;the post-test probability of missing colorectal cancer was 1% at an negative likelihood ratio(NLR)of 0.08.The post-test probability of detecting colorectal cancer was 49% at a PLR of 9 for m SEPT9;the posttest probability of missing colorectal cancer was 3% at an NLR of 0.27.CTC diagnosed colorectal tumors larger than The post-test probability of detecting colorectal tumor was 75% when the PLR of colorectal tumor greater than or equal to 10 mm was 28,and the post-test probability of missing colorectal cancer was 1% when the NLR was 0.11.The post-test probability of detecting colorectal tumor was 44% when the PLR of colorectal tumor greater than or equal to 6 mm was 7,and the post-test probability of missing colorectal cancer was 3% when the NLR was 0.25.The probability of missing colorectal cancer was 3% when NLR was 0.25.2.Mesh meta-analysis: the diagnostic data of mt-s DNA,m SEPT9 and CTC(threshold≥10 mm)diagnostic test data were subjected to reticulated Meta-analysis: the sensitivity and specificity of mt-s DNA,plasma SEPT9 methylation and CTC(threshold≥10 mm)at 95% confidence intervals were0.94(0.89,0.95),0.89(0.81,0.90);0.77(0.72,0.78),0.91(0.87,0.91);0.91(0.84,0.92),0.97(0.94,0.98).Taking mt-s DNA as the comparative study parameter,the relative sensitivity and relative specificity of m SEPT9 were 0.88 and 1.18,and the relative sensitivity and relative specificity of CTC(threshold≥10 mm)were 1.03 and 1.27.mt-s DNA、m SEPT9 and CTC(threshold≥10 mm)superiority index “S”: 5,3,0.3,respectively.Diagnostic test data from mt-s DNA,m SEPT9 and CTC(threshold≥6 mm)were performed for network meta-analysis: the sensitivity,specificity and diagnostic odds ratio of CTC(threshold≥6mm)were 0.81(0.72,0,83),0.92(0.85,0.92)and 37.31(16.82,41.33),respectively,using mt-s DNA as the comparative study parameters,the relative sensitivity and relative specificity of m SEPT9 was 0.88,1.16,CTC(threshold≥6 mm)relative sensitivity and relative specificity was0.92,1.15.mt-s DNA,m SEPT9 and CTC(threshold≥6 mm)superiority index“S”: 5,3,3,respectively.Conclusion:1.For colorectal tumors greater than or equal to 10 mm,the diagnostic efficacy CTC>mt-s DNA>m SEPT9;For colorectal tumors greater than or equal to 6 mm,mt-s DNA>m SEPT9=CTC;2.the diagnostic efficacy of CTC is similar to that of colonoscopy for colorectal tumors above 10 mm,but is inferior to that of colonoscopy for colorectal tumors above 6 mm.3.Of the three diagnostic methods included in the analysis,CTC was the most likely to replace colonoscopy as a non-invasive screening modality,and the accuracy of mt-s DNA and m SEPT9 needs to be improved. |
PDF Full Text Request