Association between MICA polymorphisms, s-MICA levels, and pancreatic cancer risk in a population-based case-control study.

BACKGROUND: Pancreatic tumor cells may avoid immune surveillance by releasing the transmembrane major histocompatibility complex class I chain-related A (MICA) protein in soluble form (s-MICA). We hypothesized that the presence of the A5.1 polymorphism in the MICA gene, which encodes a truncated MICA protein, is associated with higher s-MICA levels and increased pancreatic cancer risk. METHODS: MICA alleles and s-MICA levels were measured in 121 pancreatic cancer cases and 419 controls. General linear regression with a log transformation assessed geometric means of s-MICA levels across MICA alleles. Unconditional logistic regression was used to calculate the odds ratio (OR) and 95% confidence intervals (CI) for pancreatic cancer associated with MICA alleles. RESULTS: After multivariate adjustment, participants with at least one copy of the A5.1 allele versus no A5.1 allele had 1.35 (95% CI: 1.05-1.74) times greater s-MICA levels (1.65 times higher for cases and 1.28, for controls) and increased risk of pancreatic cancer (OR = 1.91, 95% CI: 1.05-3.48). CONCLUSIONS: Our study suggests higher risk of pancreatic cancer among those with the MICA A5.1 polymorphism, which may be explained by an increase in s-MICA secretion and impaired immune response. IMPACT: These findings provide further evidence at the genetic and molecular level of the important role of MICA in pancreatic cancer development, and may have important implications with regards to pancreatic cancer screening.