Comparison between mononucleotide and dinucleotide marker panels in gastric cancer with loss of hMLH1 or hMSH2 expression

Background: DNA mismatch repair deficiency is an important molecular mechanism of genetic instability in gastric cancer, and a high instability at microsatellites is associated with favorable prognosis. We compared mononucleotide and dinucleotide microsatellite instability (MSI) marker panels in 56 paired gastric tumor and normal samples. Methods: The mononucleotide marker panel (mono panel) consisted of 8 markers: BAT25, BAT26, BAT40, BAT-RII, NR21, NR22, NR24 and NR27. The dinucleotide marker panel (di panel) contained D2S123, D5S346, D17S250, D17S261, D17S520, D18S34 and D18S58. The NCI panel was used as reference panel. Results: Among 13 gastric tumors showing no hMLH1 or hMSH2 expression, 8 MSI-H (high) and 5 MSI-L (low) were identified. The analytical sensitivities of the NCI, mono and di panels to detect unstable MSI were 61.5% (8/13), 76.9% (10/13) and 84.6% (11/13), respectively. The size change of allele shift was statistically greater in the mono panel than in the di panel (p = 0.02 by Mann-Whitney U-test). The BAT40 (69.2%, 9/13) and D18S34 (76.9%, 10/13) markers showed high sensitivity for determination of MSI status. Conclusions: To improve the detection rate of MSI in gastric cancer with loss of hMLH1 or hMSH2 expression, the kind of MSI marker may need to be considered more, instead of the repetitive type of marker. Thus, an MSI panel designed with a combination of both BAT40 and D18S34 is suggested for providing more accurate and sensitive MSI analysis in gastric cancer.