Molecular classification of endometrial carcinoma applied to endometrial biopsy specimens: Towards early personalized patient management

Abstract

BackgroundThe current risk stratification systems used to guide management of endometrial cancer are based on irreproducible post surgical pathological information, hence the need for more reliable classification systems. Using microarray and sequencing technologies, TCGA recently identified four prognostically significant endometrial carcinoma subtypes, which subsequently proved reproducible using clinically applicable surrogate tests. Using these tests, we sought to determine the level of concordance between endometrial biopsies and subsequent hysterectomy specimens in assessing the molecular classification of endometrial carcinoma. Materials and methodsFifty biopsies with corresponding hysterectomy specimens for endometrial carcinomas were collected. Additionally, 10 cases of biopsy proven atypical hyperplasia/EIN who were found to have endometrial carcinoma on resection were included. IHC for mismatch repair (MMR) proteins (MLH1, PMS2, MSH2 and MSH6) and P53 was performed. Microsatellite instability analysis was performed by PCR and Sanger sequencing was performed to detect mutations in exons 9 and 13 of the POLE gene. The level of concordance for tumor grade, histologic subtype, immunohistochemical and molecular profile in both specimens was determined using Cohen's kappa estimates. ResultsA high level of concordance was achieved for MMR-loss, MSI-high, P53-wild and abnormal types. In contrast, grade and histologic subtype showed only moderate levels of agreement. POLE gene mutation was detected in two patients. For both cases, mutations were detected only in resection specimens. When comparing atypical hyperplasia/EIN with subsequent hysterectomy tumor, the profile was identical to that of endometrial carcinoma. ConclusionIn our cohort of endometrial carcinoma, a high level of concordance was achieved between biopsy and hysterectomy specimens for MMR-loss, MSI-high, P53-wild and abnormal types, superior to that of grade and histologic subtype, providing earlier and more reliable prognostic information to inform management. Similar concordance could not be achieved for POLE mutation, given the low frequency of this mutation in our study.