RESUMO
INTRODUCTION: Since 2010, the network of rare malignant tumors of the ovary (TMRG) was developed to optimize the management of patients, also allowing a histological second opinion of rare ovarian tumors. The aim of this work was to study the contribution of second opinion to improve histological diagnostic accuracy on ovarian rare malignant tumors included in the TMRG database. MATERIAL AND METHODS: Histological data of patients diagnosed with a rare ovarian tumor included in TMRG network over a one-year period (2018) were collected. Initial diagnoses were compared with second opinion from national gynecological pathologist experts. The modalities of histological second opinion requests were studied, as well as the histological characteristics of the tumors. The discordances were classified as minor (if the modification of histological diagnosis did not change patient management) and major (if the patient management can be modified). RESULTS: Of 1185 included patients, 937 matched the inclusion criteria. Full concordance between primary diagnosis and expert second opinion was reached in 611 cases (65,3%), minor discordance was seen in 114 (12,2%) and major discordance in 209 (22,3%) of cases. In systematic review requested by the network, 26% (n = 137) of cases were reported with a change in histological diagnosis, while the change concerned 44% (n = 186) of cases for a second opinion spontaneously requested by the initial pathologist. The discrepancies concerned all categories of ovarian tumors, with a majority of mucinous tumors (43% of major discordances), followed by stromal and sex-cord tumors (13.8% of major discordances) and clear cell tumors (12,4% of major discordances). CONCLUSION: This analysis confirms the diagnostic difficulty of ovarian tumors, due to their rarity and morphological heterogeneity. French pathologists are aware of these difficulties and spontaneously refer ovarian tumors with unusual histology for a second opinion and collaborate with rare tumor networks for systematic review.
