Abstract 834: Plasma MICA and pancreatic cancer

Abstract

Abstract Background: We hypothesized that major histocompatibility complex class I chain-related protein (MICA) may be useful for detecting early pancreatic cancer and guiding immunotherapy. MICA, which is up-regulated on the tumor cell surface, is a major ligand for natural killer (NK) cell receptors - NKG2D. The binding of transmembrane MICA to NKG2D facilitates the recognition of tumors by NK cells, resulting in tumor elimination. However, many tumors release MICA into circulation in the soluble form and thus escape immune surveillance. Several small studies showed that soluble MICA levels (s-MICA) are higher in pancreatic cancer patients than healthy people. The goal of this project was to define whether s-MICA concentration is associated with pancreatic cancer in a population-based case-control study in Minnesota (1994-98, mean age 66 years). Methods: Cases ≥20 years (n = 163) were ascertained from hospitals. Controls (n = 542) were randomly selected from the general population and frequency matched to cases by age (within 5 y), sex and race. All subjects were interviewed in-person regarding demographic information, lifestyle, and medical history. For this analysis, only Caucasians (96% of all participants) were included. S-MICA plasma concentrations were measured using a commercially available Luminex kit (R&D Systems, MN); level of detection was 2 pg/ml. Unconditional logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (CI) for pancreatic cancer in relation to s-MICA concentration. S-MICA was modeled as 4 categories after adjustment for age, sex, smoking status, pack-years and diabetes. S-MICA concentration ≤2 pg/ml was assigned to the null category, whereas all other participants were divided into tertiles. The first tertile was the reference category since the number of cases in the null category was small (N = 20). Results: Among controls, s-MICA concentrations were positively associated with older age (≥70 y) and diabetes. The ORs for pancreatic cancer were 1.17 (0.71-1.94) and 1.82 (1.12-2.97) in the second and third tertiles, respectively, compared to the first tertile (p-trend = 0.02), while ORs for the null category were 1.05 (0.57-1.92). For the 3rd versus 1st tertile, s-MICA was significantly associated with pancreatic cancer in the following groups: men, OR = 2.46 (1.31-4.64; p-trend = 0.007); people younger than 70 y, 2.42 (1.26-4.64; p-trend = 0.007), and in those without diabetes, 1.91 (1.15-3.19; p-trend = 0.02). Notably, among cases with known death dates (N = 107), s-MICA concentrations were significantly higher for those who died in the first three months of follow-up, suggesting a more virulent or later stage cancer for those with higher concentrations. Conclusions: Our study supports the hypothesis that plasma s-MICA concentration is positively associated with increased odds of pancreatic cancer and suggests that s-MICA concentration may be a promising biomarker for early detection of pancreatic cancer. Citation Format: Anna E. Prizment, Heather H. Nelson, Bharat Thyagarajan, Amy M. Linabery, Kristin E. Anderson. Plasma MICA and pancreatic cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 834. doi:10.1158/1538-7445.AM2015-834