Implementation of Universal Colorectal Cancer Screening for Lynch Syndrome in Hispanics Living in Puerto Rico.

OBJECTIVE: Colorectal cancer is the leading cause of cancer death in Puerto Rico and third among Hispanics in the USA. Up to 2-4% of colorectal cancer cases are a result of Lynch syndrome (LS), a hereditary cancer syndrome caused by a germline mutation in at least one of the DNA mismatch repair genes. The objective of this study was to determine the prevalence of LS in colorectal tumors during the first 15-months after the implementation of universal tumor-based screening for LS in Puerto Rico. METHODS: A total of 317 colorectal tumors were evaluated in a large private pathology laboratory from September 2014 to December 2015. Clinical characteristics were obtained from the pathology reports. Unadjusted and adjusted logistic regression models were used to estimate the magnitude of association (odds ratio [OR] with 95% confidence intervals [CI]) between absent MMR protein expression and patient characteristics. RESULTS: Most cases (93.4%) were analyzed by immunohistochemistry; 11.8% (35 of 296) had deficient mismatch repair protein expression. While 29 of the 317 cases were subjected to PCR-based microsatellite instability analysis of which 10.3% (3 of 317) had microsatellite instability. In total, 11.0% of the tumors were reported MMR deficient. These tumors were more likely from females and more likely localized in the proximal colon compared to those with proficient MMR expression. CONCLUSIONS: Our data is consistent with the results from other studies including US Hispanics, where approximately 10% of Hispanic individuals with colorectal cancer have microsatellite instability. Our results support universal tumor-based screening for LS among Hispanics in accordance with National Comprehensive Cancer Network guidelines.

Evaluation of MLH1 variants of unclear significance.

Inactivating mutations in the MLH1 gene cause the cancer predisposition Lynch syndrome, but for small coding genetic variants it is mostly unclear if they are inactivating or not. Nine such MLH1 variants have been identified in South American colorectal cancer (CRC) patients (p.Tyr97Asp, p.His112Gln, p.Pro141Ala, p.Arg265Pro, p.Asn338Ser, p.Ile501del, p.Arg575Lys, p.Lys618del, p.Leu676Pro), and evidence of pathogenicity or neutrality was not available for the majority of these variants. We therefore performed biochemical laboratory testing of the variant proteins and compared the results to protein in silico predictions on structure and conservation. Additionally, we collected all available clinical information of the families to come to a conclusion concerning their pathogenic potential and facilitate clinical diagnosis in the affected families. We provide evidence that four of the alterations are causative for Lynch syndrome, four are likely neutral and one shows compromised activity which can currently not be classified with respect to its pathogenic potential. The work demonstrates that biochemical testing, corroborated by congruent evolutionary and structural information, can serve to reliably classify uncertain variants when other data are insufficient.

Characterization of germline mutations of MLH1 and MSH2 in unrelated south American suspected Lynch syndrome individuals.

Lynch syndrome (LS) is an autosomal dominant syndrome that predisposes individuals to development of cancers early in life. These cancers are mainly the following: colorectal, endometrial, ovarian, small intestine, stomach and urinary tract cancers. LS is caused by germline mutations in DNA mismatch repair genes (MMR), mostly MLH1 and MSH2, which are responsible for more than 85% of known germline mutations. To search for germline mutations in MLH1 and MSH2 genes in 123 unrelated South American suspected LS patients (Bethesda or Amsterdam Criteria) DNA was obtained from peripheral blood, and PCR was performed followed by direct sequencing in both directions of all exons and intron-exon junctions regions of the MLH1 and MSH2 genes. MLH1 or MSH2 pathogenic mutations were found in 28.45% (34/123) of the individuals, where 25/57 (43.85%) fulfilled Amsterdam I, II and 9/66 (13.63%) the Bethesda criteria. The mutations found in both genes were as follows: nonsense (35.3%), frameshift (26.47%), splicing (23.52%), and missense (9%). Thirteen alterations (35.14%) were described for the first time. The data reported in this study add new information about MLH1 and MSH2 gene mutations and contribute to better characterize LS in Brazil, Uruguay and Argentina. The high rate of novel mutations demonstrates the importance of defining MLH1 and MSH2 mutations in distinct LS populations.