Abstract 4610: Inflammation-associated microsatellite alterations in colorectal cancers (CRCs): Definition, assay, prognostic value, and pathologic significance

Abstract

Abstract Background & Aims. Deficiency in the DNA mismatch repair (MMR) gene product MSH3 results in low levels of microsatellite instability (MSI-L) at dinucleotide loci and elevated microsatellite instability at selected tetranucleotide repeats (EMAST), but not at mononucleotide loci in CRC. We previously demonstrated that the pro-inflammatory cytokine interleukin 6 (IL6) and/or hydrogen peroxide triggers translocation of MSH3 from the nucleus to cytoplasm, resulting in MSI-L/EMAST (L/E). This is consistent with observed correlation between intraepithelial and nests of inflammatory cells within malignant epithelium that are the likely origin of IL6 and oxidative stress with the occurrence of L/E in CRCs, informing that L/E is a consequence of exposure to inflammation. Microbes such as Fusobacterium nucleatum may play a role in CRC initiation and progression, and may trigger inflammation. Clinically, L/E is associated with poor patient prognosis and CRC recurrence and/or metastasis. Threshold values for determining L/E range in the literature from 1 to 3 polymorphic markers frameshifted when compared to normal tissue. Our aim was to assess an optimal threshold value for determining L/E in CRCs. Materials & Methods. For MSI-H, MSI-L and EMAST status, we developed a multiplex PCR-based assay consisting of 2 mono-, 5 di-, and 7 tetra-nucleotide microsatellite markers. MSI-H was defined when >2/7 mono- and/or di-nucleotide repeats exhibited frameshift mutations (FSM), and MSI-L when <2/7 mono- and/or di-nucleotide markers exhibited FSM. EMAST was defined when 1/7 (relaxed criteria) or >2/7 (stringent criteria) tetranucleotide markers showed FSM. Stage II/III CRCs (132 cases) from Korea were analyzed for the association between L/E and recurrence-free survival (RFS) using relaxed and stringent criteria for EMAST. A second CRC cohort (308 cases) was analyzed for MSI/EMAST and F. nucleatum infection. Results. When EMAST was defined by relaxed criteria, P value for the association between L/E and RFS of 132 CRC patients was 0.415 while P value was 0.0038 when EMAST defined by stringent criteria was used. Similarly, F. nucleatum infection was associated with L/E by univariate analysis (P=0.027) but not by multivariate analysis (P=0.08) when EMAST was defined by relaxed criteria for 308 CRC cases. In contrast, this became significant in both uni- (P=0.014) and multi-variate analysis (P=0.024) when EMAST was defined by stringent criteria. Conclusions. A clear distinction for pathological significance is detected between non-MSI-H, L/E CRCs (associated with inflammation) and non-L/E CRCs when EMAST is defined by <2/7 EMAST markers. This threshold may better define EMAST for these observed inflammation-associated microsatellite alterations. Citation Format: Minoru Koi, Yoshiki Okita, Stephanie Tseng-Rogenski, Koji Munakata, Chan Choi, Hyeong-Rok Kim, Erika Koeppe, Elena Stoffel, Joseph A. Galanko, Nikki McCoy, Temitope Keku, Takahito Kitajima, Takeshi Nagasaka, Yuji Toiyama, John M. Carethers. Inflammation-associated microsatellite alterations in colorectal cancers (CRCs): Definition, assay, prognostic value, and pathologic significance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 4610.