A Therapeutic Application of Systematic Review with Meta-analysis: Epigenetic Analysis Tools and Cancer’s Therapy Assessment

Epigenetics will produce a vast array of novel biomarkers to help refine cancer therapies assessment and drug response of the patients; they can serve as markers used to predict clinical outcome and probability of relapse after chemotherapies.

A Prisma flow chart revealed that; a total of 1302 articles were retrieved, 151 full-text were assessed for eligibility where 47 studies with low to moderate quality were found to be eligible and involved in systematic review as qualitative phase, on the other hands, only 19 studies with a moderate quality; fit the meta-analysis criteria with a total patients of 6,086 from more than 25 countries used in the quantitative study phase.

A systematic review summarized that, among the five epigenetic tools used only two modifications; DNA methylation and histone modifications were used to assess the response outcomes of cancer therapies. The highest reputation epigenetic modification was found to be DNA methylation that associated with either better or bad outcomes depending on the related gene which can be oncogenes or tumor suppresser genes.

Both the fixed and random meta-analysis models of the studies proved that, the epigenetic DNA methylation have a tendency to predict the outcomes (HR= 0.603, CI= 0.462- 0.788) in different cancer types treated with chemotherapeutic agents, surgery and radiotherapy. But, high heterogeneity and publication bias were reported. Sub-groups analysis was performed to explore the reasons of heterogeneity and appear to be due to clinical and methodological variability among the studies.

The first subgroup; methylation of oncogenes or DNA repair genes showed better outcomes prediction of cancer patients (HR= 0.5, CI= 0.45-0.55), with low publication bias and heterogeneity (Q= 25.24, and I2 = 40.58%) among the studies. The second subgroup; tumor suppresser genes methylation associated with worse outcomes (HR= 2.07, CI= 1.77-2.43), with no heterogeneity (Q= 1.73, and I2 = 0.00) among the studies.

Our analysis revealed that, MGMT gene-promoter methylation was a strong outcomes predictor (HR=0.528, CI= 0.493-0.566) of cancer patients particularly the use of alkylating chemotherapeutic agents with or without radiotherapy, moreover, correlation of MGMT with overall survival revealed that, the MGMT promoter methylation was a good predictive biomarkers of cancer therapies. These findings were with low heterogeneity (Q= 28.35, and I2= 43.57) and no publication bias was reported.

The three sensitivity analysis techniques; statistical models, one study removal and study design for MGMT-promoter methylation as a good predictive biomarker indicated that, the pooled effect was not differ upon the change of the statistical models (fixed or random) of meta-analysis, or one study removal analysis as well as the study design change.

Huge articles with different designs were scattered around 13% of the world countries that represent more than 6000 patients with low to moderate quality, regarding the use of epigenetic tools in cancer therapies assessment.

Application of meta-analysis confirmed the systematic review findings regarding the DNA-methylation as a best epigenetic tool with high heterogeneity and publication bias, due to clinical and methodological variability among the studies.

Finally, we conclude that, MGMT gene methylation can be considered as a good predictive biomarker of cancer therapies and associated with improved overall survival, these findings were not depend on one study and not changed with any modification in statistical models and study designs, also the unpublished or un retrieve studies in our review do not change the result obtained.