Immunohistochemistry, molecular biology and clinical scoring for the detection of muir-torre syndrome in cutaneous sebaceous tumors: which strategy?

Background Sebaceous neoplasms (SNs) always raise the possibility of an association with Muir-Torre syndrome (MTS) and permit to screen internal malignancies, colorectal and endometrial carcinomas, before they become symptomatic. Immunohistochemistry (IHC), molecular biology and clinical examination are different approaches for detection of MTS. We conducted a retrospective analysis of non-selected SNs in order to determine the optimal tools to implement for MTS screening. Methods Deficient MMR phenotype (dMMR) was determined by either IHC using antibodies directed to four mismatch repair (MMR) antigens on tissue-microarray or molecular biology using pentaplex PCR. The Mayo-Clinic-Risk score of MTS was calculated from medical records. Sensibility and specificity of each test for the detection of MTS was determined. Results We included 107 patients, 8 with multiple SNs, for a total of 123 SNs (43 sebaceous adenomas, 19 sebaceomas and 61 sebaceous carcinomas (SC)). Loss of at least one MMR protein was observed in 70.7% of tumors while 48% had a microsatellite instable phenotype. Concordance between both techniques was 92.9%, with a 0.85 Cohen's kappa coefficient. Nineteen patients (20.2%) had a ≥2 points Mayo-Clinic-Risk-Score, one having a pMMR SC. Among the 13 patients with confirmed MTS, 2 had a low Mayo-Clinic-risk score (1 point). IHC had the highest sensitivity for MTS screening (100%) with a specificity of 34.1% while a >2 points Mayo-Clinic-Risk-Score had a lower sensitivity (92%) but a higher specificity (89%). Conclusion To detect MTS in SNs patients, first line Mayo-Clinic-risk-score followed by IHC appears the most accurate strategy with lower cost for society. This strategy should be adapted to the medico-economic resources of each country.  .