Mismatch Repair Deficient Cancer Diagnostic Aspects in Colorectal Cancer and the Role of Urological Cancer in Lynch Syndrome

Abstract: Mismatch repair (MMR) deficiency is carcinogenic and can either have somatic/sporadic causes (i.e. epigenetic silencing or somatic inactivation) or hereditary causes (Lynch syndrome due to a germline mutation in one of the MMR genes ¬- MLH1, MSH2, MSH6, PMS2). The identification of MMR defective colon cancer is clinically relevant for diagnostic, prognostic and potentially also for treatment-predictive purposes. The aims of this thesis were to validate the application of the MMR index for the prediction of MMR deficiency in colon cancer (study I), to investigate heterogeneous MMR protein expression in colorectal cancer (study II) and to study urological cancers in Lynch Syndrome (studies III-V). The most common cancers in Lynch syndrome are colorectal cancer and endometrial cancer, but also other tumour types are linked to the syndrome. We investigated the roles of bladder cancer (study III), prostate cancer (study IV) and renal cell cancer (study V) in Lynch syndrome. Study I confirmed that the MMR index is easy to apply and identifies MMR defective colon cancers with high sensitivity (93%) and specificity (76%). The MMR index evaluates features such as expanding growth pattern, lack of dirty necrosis, mucinous differentiation and presence of tumour-infiltrating lymphocytes. Study II showed that heterogeneous (retained/lost) MMR protein expression occurred in three distinct patterns, i.e. as intraglandular, clonal and compartmental protein expression. These patterns co-existed in 9/14 tumours and correlated to differences in the MMR status. Attention to this phenomenon is recommended to prevent false-positive or false-negative evaluations of MMR protein immunostaining. Studies III-V linked urinary bladder cancer, prostate cancer and renal cell cancer to Lynch syndrome families. The cancers frequently showed MMR defects in line with the underlying disease-predisposing mutation, indicating that these cancer types should be considered part of the Lynch syndrome tumour spectrum. Urothelial cancer development was predominantly linked to MSH2 mutations. These findings should be considered in risk estimates and surveillance recommendations.