Protocol and registration
Herein the protocol will be developed with adherence to Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) 2020 guideline and the review protocol will be submitted to PROSPERO review registry for registration
Eligibility Criteria
Inclusion
All original research articles reporting on the effect of epigenetic signatures, epigenetics and field cancerization on the carcinogenesis and recurrence of gastrointestinal cancers (esophageal cancer, gastric cancer, colonic cancer, colorectal cancer and hepatobiliary cancer) in adults will be included. All cross sectional observational studies, cohort studies, case-control studies and randomized clinical trials published from January 2013 to December 2022 without language limitation will be included.
Exclusion
All articles reporting on animal subjects and pediatric population will be excluded; all articles with missing data will be excluded.
Information Sources
The literature search will be executed through several electronic databases including EMBASE, PubMed, Scopus, Web of Science, Cochraine, Global index medicus , semantic scholar and google scholar .
Search and search strategy
The following key strings will be used in various permutations in combination with Boolean operators and Filters “Epigenetic” [MeSH Terms], “Epigenetics”[MeSH Terms], “Epigenetic process”[MeSH Terms] , “Gene Expression signature”[MeSH Terms], “Field defect”, “Field cancerization” , “Carcinogenesis”, “Tumorigenesis”[MeSH Terms], “Locoregional neoplasm recurrence”[MeSH Terms], “Cancer of gastrointestinal trasct”[MeSH Terms] , “Cancer of esophagus”[MeSH Terms], “Gastric cancer”[MeSH Terms], “Colonic Cancer”[MeSH Terms], “Colorectal cancer” [MeSH Terms], “ Cancer of pancreas”[MeSH Terms] , “Cholangiocarcinoma”[MeSH Terms]
Search Strategy
Epigenetics[MeSH Terms] OR Epigenetic process [MeSH Terms] OR Gene expression signature[ MeSH Terms] OR Field Defect OR Field Cancerization AND Carcinogenesis OR Tumorigenesis [MeSH Terms] OR Locoregional neoplasm recurrence [MeSH Terms] AND Cancer of gastrointestinal tract [MeSH Terms] OR Cancer of esophagus [MeSH Terms] OR Gastric Cancer [MeSH Terms] OR Colonic Cancer [MeSH Terms] OR Colorectal Cancer [MeSH Terms] OR Cancer of Pancreas [MeSH Terms] OR Cholangiocarcinoma [MeSH Terms]
The sample of PubMed Search Strategy is annexed for Figures 1 and 2
Study selection
Three reviewers (VK, EM, and AA) will do the primary screening of all text articles collected from all electronic databases by search strategy independently; the screening process will be blinded. The three reviewers will also evaluate and record the selected articles for full text screening based on the stipulated eligibility criteria and all conflict dissolution will be mitigated upon the mutual consensus or adjudication from the fourth party author (TS). The Rayyan or Covidence and RevMan software will be employed for screening and data synthesis .The screened results will be presented in PRISMA flow diagram shown in the annex figure 3 and the reason for exclusion of full text articles will be justified within the exclusion box of the diagram.
Data extraction
After the full text screening the three reviewers (VK, EM and AA) will use the piloted Microsoft, excel checklist to extract the data items for data synthesis all counterfeits encountered will be resolved by consensus or adjudication from the fourth party reviewer (TS)
Data Items
The data entities such as inclusion and exclusion criteria , study design, number and location of study area , sample size , epidemiological characteristics of study participants , study period , year of study publication , histological type of cancer , grade and stage.
Exposure gestures such as epigenetic signatures, signature pathways, epigenetics , field cancerization status and the outcome indices including carcinogenesis or tumorigenesis , metachronous status and locoregional recurrence of cancers.
Risk of bias (ROB) in individual studies
The ROB will be evaluated using the Modified Newcastle Ottawa Scale (NOS) for observational studies (15,16) and Cochran ROB tool Version 2 for RCT studies (17).The reviewers ( VK , EM and AA ) will assess the risk of bias and include only articles with NOS score above 4 and Low ROB based on D1-D5 for RCTs for Meta-analysis.
Measurement of effect
For RCTs, the Hazard Ratio (HR) for evaluation of the effect of intervention to the outcome will be employed and for dichotomous outcome in non-RCTs, the Relative Risk (RR) for evaluation of the effect of exposure to the outcome of interest will be analyzed. For continuous outcome, the Mean Difference (MD) and Standard Mean Difference (SMD) will be analyzed to evaluate the effect size across the groups of participants.
Data Synthesis
If optimal data extraction will be attained the Meta-analysis will be conducted based on Per Protocol (PP) analysis. The data reporting on epigenetic signatures and pathways, epigenetic phenotypes and field cancerization will be synthesized differently and the Random Effect Model (REM) across the study participants will be reported. For Dichotomous, outcomes the pooled RR at 95% CI will be determined and the MD or SMD at 95% CI will be reported for continuous variables.
The data consistency will be determined from the articles by evaluating the clinical heterogeneity, methodological heterogeneity, statistical heterogeneity and bias of publication. Heterogeneity will be analyzed and reported using the funnel plot, Chi-test and I2 - test, if minimal consistency will be identified the meta-analysis will be aborted and instead the qualitative analysis with a scoping report will be implored.
For the optimal data suitable for Meta-analysis, the analysis will be employed using RevMan Software if the data extracted is limited for Meta-analysis synthesis, the scoping review will be conducted using the synthesis without meta-analysis (SWiM) protocol and the results will be presented in summary tables.
Risk of bias across studies
The publication bias will be analyzed and reported using a funnel plot for more than 10 articles combined for meta-analysis.
Subgroup Analysis
If a significance inconsistency is encountered, advanced analysis will be performed for specific subgroup determinants and outcomes
Epigenetic signature pathways, Epigenetic variations, Field Cancerization, Location and Histological types of gastrointestinal cancers
Sensitivity Analysis
The sensitivity analysis will be implored for evaluating the robustness of the synthesized data, the crude sensitivity of data results will be adjusted by eliminating the articles with high ROB, and missing data and any contradicting definition of determinants and outcome indices.
Strength of body of evidence
Three reviewers (VK , EM and AA) will critically appraise and estimated the certainty applicability of the reported evidence in this review using the Grading of Recommendation , Assessment , Development and Evaluation (GRADE) scheme , the strength of evidence will be rated according to for quality levels classified as high , moderate , low and very low LOE (18,19).
Ethical consideration and Dissemination
There is no involvement of human subject participation in this review thus giving no effect to ethical implication .The evidence report of this review will be disseminated on scientific conferences and will be published to a reputable journal of gastroenterology oncology. This review will be registered on PROSPERO protocol registry.