Abstract 32P
Background
Endometrial cancer (EC) is the most prevalent amongst gynaecological malignancies, and microsatellite instability (MSI), has been reported in up to 30% of cases. This is mainly caused by alteration of DNA mismatch repair (MMR) genes and can be explored by loss of MMR proteins expression by immunohistochemistry (IHC), or alternatively, by MSI status.
Methods
Firstly, 276 early stage low grade (G1-G2) EC patients were classified by IHC by evaluation of tissue microarrays. Based on the expression of MLH1, PMS2, MSH2 and MSH6, samples were classified as MMR-proficient (MMR-p; all four proteins retained expression) or deficient (MMR-d; one or more proteins showed loss of expression). Secondly, the Idylla™ MSI Test (Idylla™), was used to retest DNA from all MMR-d cases and a random selection of an approximately equal number of MMR-p cases. Non-informative cases by IHC were also retested by Idylla™.
Results
A total of 276 EC were analysed by IHC; 30 (11%) samples were not initially informative, 35 (13%) samples were classified as MMR-d, and 211 (77%) as MMR-p. Idylla™ was able to rescue 29/30 (97%) of the non-informative IHC samples. 23/29 (79%) were determined as Microsatellite Stable (MSS) and 6/29 (21%) as MSI-High (MSI-H). In the MMR-d group, 16 out of 35 samples were equally classified, but the other 19 were classified as MSS, showing a discrete sensitivity compared to IHC (45.71%). Within the 44 MMR-p randomly selected samples, all were identically classified by Idylla™ except one, that was classified as MSI-H by Idylla™, showing high specificity compared to IHC (97.72%). Overall, there is a concordance of 74.68% between IHC and Idylla™.
Conclusions
This preliminary data show that the Idylla™ MSI Test is able to rescue the majority of samples which are classified as non-informative using IHC and that the Idylla™ MSI Test seems to be highly correlated with IHC in the identification of MMR-p EC cases, but a substantial lower concordance rate has been observed for the MMR-d EC cases. Further analyses have to be performed to explain the observed discordances of these MMR-d cases.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
Hospital Universitario la Paz-IdiPAZ.
Funding
Biocartis NV.
Disclosure
M. Mendiola: Non-remunerated activity/ies, research collaboration: Biocartis; Honoraria (self), speaker: AstraZeneca; Honoraria (self), speaker: Tesaro. A. Gallego Martínez: Advisory/Consultancy: Pharmamar; Travel/Accommodation/Expenses: Pharmamar; Speaker Bureau/Expert testimony: Clovis; Speaker Bureau/Expert testimony: AstraZeneca; Speaker Bureau/Expert testimony: MSD; Travel/Accommodation/Expenses: MSD; Travel/Accommodation/Expenses: Tesaro (GSK); Travel/Accommodation/Expenses: Pierre-Fabre; Non-remunerated activity/ies, Non remunerated membership: SEOM; Non-remunerated activity/ies, Non remunerated membership: GEICO; Non-remunerated activity/ies, Non remunerated membership: European Orgnisation Treatment Throphoblastic disease; Non-remunerated activity/ies, Non remunerated membership: ESMO. D. Hardisson: Non-remunerated activity/ies, Principal investigator, Molecular Pathology and Therapeutic Targets Lab: IdiPAZ. A. Redondo: Honoraria (self), Research grant/Funding (self), speaker and advisory role: Roche; Honoraria (self), Research grant/Funding (institution), speaker and advisory role: Pharmamar; Honoraria (self), speaker and advisory role: AstraZeneca; Honoraria (self), speaker and advisory role: GSK; Honoraria (self), speaker and advisory role: Clovis; Honoraria (self), speaker: Novartis; Honoraria (self), Advisory role: Amgen; Research grant/Funding (institution): Roche; Research grant/Funding (institution): Eisai; Non-remunerated activity/ies, Member of executive board and secretary: GEICO; Non-remunerated activity/ies, PI of phase II Cecilia study sponsored by Roche: Roche. All other authors have declared no conflicts of interest.