Objective Pancreatic cancer has the worst prognosis of any gastrointestinal cancer, with the mortality approaching the incidence. Early detection is crucial for improving patient prognosis. We therefore performed a meta-analysis to evaluate and compare the sensitivity and specificity of CA19-9 and CA242 in pancreatic cancer. Methods We searched PubMed, EMBASE and the Cochrane Library for studies that evaluated the diagnostic validity of CA19-9 and CA242 between January 1966 and March 2011. Meta-analysis methods were used to pool sensitivity and specificity and to construct a summary receiver-operating characteristic (SROC) curve. Results A total of 11 studies that included 2316 patients who full filled all of the inclusion criteria were considered for analysis. The pooled sensitivities for CA242 and CA19-9 were 0.719 (95% CI: 0.690–0.746) and 0.803 (95% CI: 0.777–0.826), respectively. The pooled specificities of CA242 and CA19-9 were 0.868 (95% CI: 0.849–0.885) and 0.802 (95% CI, 0.780–0.823), respectively. The diagnostic odds ratio (DOR) estimate was significantly higher for CA242 (16.261) than for CA19-9 (15.637). Conclusion Our meta-analysis showed that CA242 and CA19-9 could play different roles in the diagnosis of pancreatic cancer. Although the sensitivity of CA242 is lower than that of CA19-9, its specificity is greater.