Mutational spectrum of 35 Lynch syndrome argentinean families: more to say about Lynch syndrome in South America

MARINA ANTELO; MARIANO GOLUBICKI; DANIELA MILITO; STELLA MARIS HIRMAS; LUIS PEPE; MARCELA CARBALLIDO; GUILLERMO MENDEZ; SOLEDAD ISEAS; MARIO BARUGEL; MARIA AMELIA BARTELLINI; ANA CABANNE; DANIEL CISTERNA; ENRIQUE ROCA

Florencia

Congreso; International society for gastrointestinal hereditary tumours?InSiGH; 2017

InSIGHT

Introduction: although only a minority of the total cases, Lynch syndrome (LS) represents around 150 cases of colorectal cancer (CRC) per year in Argentina. Nevertheless, we still lack a unified Argentinean Registry of Hereditary Colorectal Cancer. Recently published data about LS in South America revealed 128 mismatch repair genes (MMRg) variants in 243 South American families with suspected LS, being 98 classified as deleterious. Of these, 48 families were argentinean. We here report the mutational spectrum of 31 additional MMRg variants in 35 families of our Registry, located in the Hospital of Gastroenterology ?Dr. C. B. Udaondo?, a public metropolitan hospital in Buenos Aires, Argentina. Methods: MMR status of families with Amsterdam criteria and/or Bethesda guidelines registered in our Hospital was assessed, and those with MMR deficiency were followed by direct sequencing +/- Multiplex ligation-dependent probe amplification (MLPA) of MMRg. Those with deleterious variants in a MMRg were classified as LS. Mutation nomenclature was in accordance with the Human Genome Variation Society guidelines. Results: in total, 35 families harbored 31 MMRg variants that we classified as LS: 52% affected MLH1, 36% affected MSH2, 10% affected PMS2 and 2% MSH6. Missense and frameshit mutations were the most common alterations (35 % and 26%, respectively), followed by nonsense mutations (19%), splice site mutations (10%), large deletions (6%), and inframe deletions (4%). Of the 31 disease predisposing MMRg mutations, 18 (58%) were reported in the InSIGHT variant classifications database (IVCD). Of these 18, 15 were classified as LS predisposing variants, 2 were classified as variant of uncertain significance (VUS) (1 of which was predicted pathogenic by the in silico analysis with Mutation Taster-ISAMT-, and the other one was predicted as a polymorphism, but it was present in a patient with CRC, pancreas and prostate cancer who belonged to a family with Amsterdam II criteria). The last variant was classified as not pathogenic, but it was associated with LS in several reports, it was predicted as pathogenic with ISAMT, and it was highly suspicious of LS fulfilling Bethesda 1, 2, 3 and 4 criteria. The remaining 13 (42%) identified MMRg variants were classified as VUS because they are not reported in the IVCD, but we considered them as disease-predisposing since 2 were large deletions and the remaining 11 were all predicted pathogenic by ISAMT and not present in two control population databases (Exome Aggregation Consortium and Exome Variant Server). Conclusions: lining up with the last report in 2016 of LS in South America, we believe that our data represents valuable information to continue deepening the genetic CRC profile of our understudied population, with a mixed and intriguing ancestry.