Aim: The potential of biomarkers in detecting early cholangiocarcinoma (CCA) is facilitated by examining CCA-associated proteins from primary studies. One such protein is mucin 5AC (MUC5AC) but inconsistency of reported associations between its expression/serum levels and CCA prompts a meta-analysis to obtain more precise estimates.
Methods: A literature search yielded 17 included articles where multiple data in some raised the number of studies to 22. We calculated pooled odds ratios (OR) and 95% confidence intervals from negative and positive readings of MUC5AC levels. Data were subgrouped by ethnicity, detection method, sample source, and cancer type.
Results: Outcome in the overall analysis was non-significant but those in the subgroups were. Thus, significant associations (P < 0.001) indicating high MUC5AC levels were found in three subgroups: (i) Thai (OR 8.32) and (ii) serum (OR 4.52). Heterogeneity of these two outcomes (I2 = 90-93%) was erased with outlier treatment (I2 = 0%) which also modulated the pooled effects (OR 2.48-2.59). (iii) Immunoblot (OR 2.61) had low initial heterogeneity (I2 = 2%). Robustness and significant tests for interaction (Pinteraction = 0.01-0.02) improved MUC5AC associations with CCA in the Thai population.
Conclusions: Our pooled effect findings target the biomarker potential of MUC5AC to the Thai population.
Keywords: Biomarker; Cholangiocarcinoma; MUC5AC; Meta-analysis.