Differences in outcome between colorectal cancer (CRC) patients (pts) with Lynch syndrome (LS) versus MLH1 hypermethylation (MLH1 HM).

Abstract

1556 Background: In ~15% of CRC tumors mismatch repair deficiency (dMMR) is observed and is usually caused by LS (germline mutations in MLH1, MSH2, MSH6 or PMS2) or sporadic inactivation of MLH1 (hypermethylation [HM] of MLH1 gene promoter). The objective of this study was to retrospectively compare clinical factors and outcomes in pts with dMMR with LS to those without LS. Methods: This cohort includes pts with CRC diagnosed between 5/98 and 5/12 with dMMR on immunohistochemistry (IHC) staining. Prior to 2006 pts came from the Columbus area HNPCC study. After 2006, we included all CRC pts at OSU as universal IHC was initiated. Chi-square was used to compare nominal parameters and log-rank to compare overall survival (OS). Results: A total of 189 pts had dMMR per IHC staining (see characteristics in table). In 36 pts (19.0%) a diagnosis of LS was made after germline testing, in 59 pts (31.2%) a diagnosis of MLH1 HM was made with methylation or BRAF testing, in 65 pts (34.4%) a LS or MLH1 HM diagnosis was assumed based on clinical criteria. Median OS was 159 mos in the LS group and 49 mos in the MLH1MH group (p=0.003) and OS was significantly longer in the LS group across all 4 stages (p=0.002) (see Table). Conclusions: Pts with LS-associated CRC had longer OS than pts with MLH1 HM-associated CRC, even in the setting of a higher incidence of other cancers (44.1% vs. 28%). This is likely related to younger age in the LS group. The MLH1HM group was older at diagnosis and more often female. There are a number of limitations to our study including the potential for survivorship bias and lack of a definitive diagnosis in ~50% of patients despite using clinical criteria to define these pts. [Table: see text]