ECP 2023 Abstracts

S94 Virchows Archiv (2023) 483 (Suppl 1):S1–S391 13 difference in median age between two groups (67.8y in HPVIs vs 50y in HPVAs) and HPVIs presenting at higher stage. Most HPVAs were of non-keratinizing growth pattern while most HPVIs were keratiniz- ing. We have also observed differences regarding associated precursor lesions between HPVIs versus HPVAs with HPVI precursor lesions presenting either as undifferentiated basaloid high grade squamous intraepithelial lesion (HSIL-like) or as high grade differentiated keratinizing (similar to differentiated vulvar intraepithelial neoplasia (d-VIN). Among the HPVA SCC cases, none (0%) was associated with LR-HPV while all (100%) were positive for HR-HPV. Conclusion: Most cervical SCCs are associated with HR-HPV how- ever, rare HPVI SCCs tumours can be encountered, with important clinical and histologic differences from their HPVA counterparts. The present study failed to identify SCCs associated with LR-HPV. PS-10-034 In silico methods investigation of molecular changes in adenoid cystic carcinomas M.H. Toper*, S. Sarioglu *Dokuz Eylul University, Turkey Background & objectives: MYB/MYBL1-NFIB rearrangements can be seen frequently in adenoid cystic carcinomas (ACC), along with low frequency mutations in different genes. In this study, we aimed to investigate molecular changes at the genomic/transcriptional level in ACCs by in silico methods. Methods: Open access data in the cBioPortal database were used for DNA alterations. GSE153283-dataset from the NCBI-GEO database was evaluated with the Geo2R online-tool to detect differentially expressed genes (DEG) between normal/tumour tissues. In addition, using these data, gene-set enrichment analysis with GSEA program and functional- enrichment analysis with David tool were performed. Protein-protein association analysis was evaluated in String database/Cytoscape. Results: In cBioPortal, common structural changes (N=239) were seen in NFIB(71%), MYB(58%), MYBL1(20%) genes, and the most frequent sequence mutations were at KDM6A(9.5%), SPEN(9.4%), SLC9C2(8.3%) genes. 387 DEGs were observed by Geo2R-mediated; the highest expression were seen BMP7, TLX1, COL27A1 genes; decreased expression were ETV1, WNT5A, WIF1. GSEA analysis demontrated that E2F targets, Notch pathway, Estrogen-early response, Wnt-βcatenin pathway, G2M-checkpoint, MYC targets-V1. David functional-enrichment analysis has revealed PI3K-Akt pathway, signal transduction, breast cancer, MAPK pathway, and transcriptional mis- regulation. In the protein-protein relationship analysis revealed that the most interacted protein were TP53, MYC, CTNNB1. DEGs interacting with MYB protein were E2F1, TCF3, MYC. Conclusion: In our study, along with MYB/MYBL1 translocation, other genomic/transcriptional changes that affect transcriptional regu- lation, cell cycle and Notch/PI3K/Wnt-β-catenin/MAPK pathways are observed in ACCs, and it provides pioneering information for future studies in terms of pathogenesis or treatment targets. PS-10-035 Prognostic significance and computer aided quantitative analysis of tumour-stroma ratio in high-grade serous ovarian cancer using routine H&E digital pathology slides L. van Wagensveld*, C. Walker, K. Hahn, J. Sanders, R.F. Kruitwagen, M.A. van der Aa, G.S. Sonke, S. Rottenberg, K.K. Van de Vijver, A. Janowczyk, H.M. Horlings *Department of Research and Development, Department of Molecular Pathology, The Netherlands Cancer Institute, The Netherlands, GROW, School for Oncology and Reproduction, Maastricht, The Netherlands Background & objectives: Tumour-stroma ratio (TSR) is prognostic in multiple cancers, but remains understudied in high-grade serous ovarian cancer(HGSOC). We investigated the prognostic significance of TSR in HGSOC and developed a deep-learning based, fully-auto- mated, scoring algorithm(OTSR), and evaluated its relationship with tumour-infiltrating lymphocytes (TILs). Methods: N=360 patients with advanced-stage HGSOC, treated with primary debulking surgery(PDS) or neo-adjuvant chemotherapy(NACT) and interval debulking (IDS), were scored for TSR in both the most invasive(MI) and whole tumour(WT) region of hematoxylin-and-eosin-stained tissue. TSR was scored by two inde- pendent pathologists and quantified with OTSR. Patients were clas- sified as High-TSR (≥50% stroma) or Low-TSR(<50%). TILs were assessed with immunohistochemical staining. Results: In PDS, TSR-high, compared to TSR-low, was significantly associ- ated with papillar growth pattern(60% vs 34%). In NACT it was associated with a lower Mandart-score(score 4&5, 21% vs 57%) and pleural metas- tasis (25% vs 16%). High TSR depicted a significantly shorter overall and progression-free survival(31 versus 45 months; and 38 versus 44) in WT andMI. Grouping TSR and TILs (based on the median) led to three distinct survival groups with good(Low TSR, high TIL), medium(High TSR, high TIL. Or; Low TSR, Low TIL) and poor(High TSR, low TIL) survival, which remained significant for CD8 and CD103 in multivariable analysis. OTSR depicted similar significant results and demonstrated high concord- ance with pathologists (correlation = 0.83). Conclusion: TSR is an independent prognostic biomarker for OS in HGSOC in both whole tumour and most invasive region. High TSR showed a worse prognosis and a higher likelihood of pleural metastasis in case of NACT and may serve as a prognostic parameter in HGSOC. OTSR could provide a cost and time-efficient way of determining TSR, with a high reproducibility and reduced inter-observer variability. This research was funded by the Dutch Cancer Society [IKNL2014-6838]. PS-10-036 ß-catenin and L1CAM expression further stratify endometrial carcinoma patients in no specific molecular profile group and p53 abnormal group respectively H. Yoon*, H. Kim *Department of Pathology, Seoul National Bundang Hospital, Seongnam, Republic of Korea Background & objectives: The Proactive Molecular Risk Classifier for Endometrial Cancer (ProMisE-POLE; MMR-D; p53 wt/NSMP; p53 abn) has consistently been shown to be prognostically significant in endometrial carcinomas (EC). This study investigated the prognostic effect of ß-catenin and L1CAM expression on ProMisE classification. Methods: 240 EC specimens from the Seoul National University Bundang Hospital database were classified according to the ProM- isE classification. ß-catenin and L1CAM were evaluated by immuno- histochemistry on tissue microarray. We defined ß-catenin positive, if ≥10% of tumour cells showed nuclear staining and L1CAM posi- tive, if ≥10% of tumour cells showed membranous staining. Correla- tions between clinicopathological data and survival were calculated. Results: 8.8% showed ß-catenin expression and 12.5% showed L1CAM expression. Both markers were expressed mutually exclusive pattern except for one case. L1CAM expression was associated with non-endometrioid histology and high tumour grade (p<0.001). ß-catenin and L1CAM expres- sion were most frequent in NSMP group (76.2%) and p53 abnormal group (63.3%), respectively. In survival analysis, ß-catenin expressing patients showed shorter progression free survival(PFS) in NSMP group than ß-catenin wild type, whereas L1CAM expressing patients showed longer PFS in p53 abnormal group compared to negative patients (p<0.001). In multivariate analysis, the modified ProMisE classification applied with