PT - JOURNAL ARTICLE AU - Soyoun Rachel Kim AU - Alicia Tone AU - Raymond Kim AU - Matthew Cesari AU - Blaise Clarke AU - Lua Eiriksson AU - Tae Hart AU - Melyssa Aronson AU - Spring Holter AU - Alice Lytwyn AU - Manjula Maganti AU - Leslie Oldfield AU - Steven Gallinger AU - Marcus Q Bernardini AU - Amit M Oza AU - Bojana Djordjevic AU - Jordan Lerner-Ellis AU - Emily Van de Laar AU - Danielle Vicus AU - Trevor J Pugh AU - Aaron Pollett AU - Sarah Elizabeth Ferguson TI - Tumor site discordance in mismatch repair deficiency in synchronous endometrial and ovarian cancers AID - 10.1136/ijgc-2020-001927 DP - 2020 Oct 18 TA - International Journal of Gynecologic Cancer PG - ijgc-2020-001927 4099 - http://ijgc.bmj.com/content/early/2020/10/15/ijgc-2020-001927.short 4100 - http://ijgc.bmj.com/content/early/2020/10/15/ijgc-2020-001927.full AB - Objectives For synchronous endometrial and ovarian cancers, most centers rely on mismatch repair testing of the endometrial cancer to identify Lynch syndrome, and neglect the ovarian tumor site completely. We examined the mismatch repair immunohistochemistry and microsatellite instability results from the endometrium and ovary to assess discordance between the tumor sites and between tests.Methods 30 women with newly diagnosed synchronous endometrial and ovarian cancer were prospectively recruited from three cancer centers in Ontario, Canada. Both tumor sites were assessed for mismatch repair deficiency by immunohistochemistry and microsatellite instability test; discordance in results between tumor sites and discordance between test results at each site was examined. Cases with discordant results had tumors sequenced with a targeted panel in order to reconcile the findings. All women underwent mismatch repair gene germline testing.Results Of 30 patients, 11 (37%) were mismatch repair deficient or microsatellite instable at either tumor site, with 5 (17%) testing positive for Lynch syndrome. Mismatch repair immunohistochemistry expression was discordant between endometrial and ovarian tumor sites in 2 of 27 patients (7%) while microsatellite instability results were discordant in 2 of 25 patients (8%). Relying on immunohistochemistry or microsatellite instability alone on the endometrial tumor would have missed one and three cases of Lynch syndrome, respectively. One patient with Lynch syndrome with a PMS2 pathogenic variant was not detected by either immunohistochemistry or microsatellite instability testing. The rate of discordance between immunohistochemistry and microsatellite instability test was 3.8% in the ovary and 12% in the endometrium.Conclusions There was discordance in immunohistochemistry and microsatellite instability results between tumor sites and between tests within each site. Endometrial tumor testing with mismatch repair immunohistochemistry performed well, but missed one case of Lynch syndrome. Given the high incidence of Lynch syndrome (17%), consideration may be given to germline testing in all patients with synchronous endometrial and ovarian cancers.