ECP 2022 Abstract Book

Virchows Archiv (2022) 481 (Suppl 1):S1–S364 13 antibody (clone EPR20330, Rabbit Monoclonal, Abcam) on auto- mated system (Ventana Benchmark Ultra), according to the manu- facturer’s instructions. Samples were scored blindly according to both the percentage of positive cells and the intensity of expression (H-score). The study was approved by the local Ethical Committee. Results: The study included a total of 250 melanocytic tumours of which: 50 high grade and 50 low grade dysplastic nevi; 50 nevi with architectural disorder and minimal cytological atypia; 50 melanomas and 50 common nevi as controls. The histopathological diagnosis was reviewed collegially by four expert dermatopathologists according to the 2018 WHO classification of skin tumours. Written consent was obtained from each participant. Statistical analysis was performed using ANOVA analysis and Kruskal-Wallis test. As expected, PRAME immunoreactivity was markedly different between melanomas (diffuse and strong staining) and common nevi (weak and focal). Instead, among dysplastic nevi PRAME protein displayed different staining patterns in the epidermal and dermal portions. Conclusion: This work represents the first large study on PRAME expression in dysplastic nevi in a real-world setting. Our results suggest that immunohistochemical analysis for PRAME expression may aid in the differential diagnosis between low grade and high grade dysplastic nevi as well as between high grade dysplastic nevus and melanoma, in addition to and as a complement to the WHO morphological criteria used in routine practice. OFP-07-012 Somatostatin receptor type 2a (SSTR2a) immunohistochemical expression pattern in Merkel cell carcinoma- a pilot study A. Sykaras*, C. Vourlakou, C. Kouvidou *Department of Pathology, Evangelismos General Hospital, Ath- ens, Greece Background & objectives: Merkel Cell Carcinoma (MCC) is an aggressive, high-grade cutaneous neuroendocrine carcinoma. Recent data suggest that MCC patients may benefit from therapies targeting somatostatin receptors. Our aim was to evaluate the expression pattern of somatostatin receptor type 2a (SSTR2a) in MCC. Methods: SSTR2a immunohistochemistry was performed in ten MCC specimens (five primary and five metastatic to lymph nodes). SSTR2a expression was evaluated by using a scoring system pro- posed from Volante and colleagues. Specifically, score 0 describes no staining and score 1+ cytoplasmic staining whereas membrane expression corresponds to score 2+ (staining in <50% of tumour cells) or score 3+ (>50% of cells). Results: All primary MCC were SSTR2a positive with membrane staining in 4/5 cases and focal cytoplasmic staining (score 1+) in one case. Membrane SSTR2a staining was scored as 2+ in two cases and 3+ in the other two. Score 2+ cases displayed heteroge- neous-focal membrane staining of moderate intensity with partial or circumferential pattern whereas score 3+ cases were character- ized by a circumferential, strong and diffuse membrane expression of SSTR2a. In contrast with primary MCC, 3/5 metastatic MCC were SSTR2a negative (score 0) with endothelial and follicular dendritic cells serving as internal positive controls. The other two metastatic MCC displayed focal cytoplasmic SSTR2a staining (score 1+) of weak to moderate intensity. Conclusion: SSTR2a is expressed in 6/10 MCC cases, with mem- brane staining that is considered as clinically important in 4/10. Although preliminary, our data are in agreement with the few imag- ing and immunohistochemical studies that previously assessed the SSTR2a expression in MCC. Given that SSTR2a is typically expressed in well-differentiated neuroendocrine tumours (NETs), the positive SSTR2a staining in MCC is a surprising finding. SSTR2a may represent a potential target for imaging and therapies with somatostatin analogues, in a similar fashion with NETs. OFP-08 | Joint Oral Free Paper Session Paediatric and Perina- tal Pathology / Autopsy Pathology OFP-08-001 Digital pathology approach for prognostisation of neuroblastoma N. Tari, T. Werber, T. Micsik* *Semmelweis University, Hungary Background & objectives: Neuroblastoma (NB) diagnosis needs precise percentages of differentiating/proliferating cells and mitosis-karyorrhexis index (MKI), which are hard to assess manually and objectively. Our aim was to test the benefit of using digital pathological evaluation to harbour these features, especially for prognostisation. Methods: Our retrospective study was performed on 41 NB-cases from Semmelweis University, Budapest, Hungary. Hematoxylin- Eosin and Ki-67 immunostained slides were digitalized by Panno- ramic 1000 Scanner (3DHistech, Budapest, Hungary). MKI, tumour differentiation, Ki-67 proliferation index were digitally calculated with Quant Center algorithms of the same vendor. Statistical cor- relations, ROC analysis, t-tests and graphic visualisation were per- formed with Microsoft Excel and XLSTAT programs. Results: Our pilot study incorporated 5 undifferentiated, 28 poorly differentiated and 7 differentiating NBs. Average age at diagnosis was 25,54±38,39 months, with slightly more males (23:18). 5-year overall survival was 80%, disease-free survival was 65%. Manual and digital MKI values corre- lated well (r=0,78, p<0,05) and delivered prognostic value. Manual and digital Ki-67 values correlated well (r1=0,63, r2=0,56, r3=0,91, p1,2,3<0,05), and proved to be prognostic in different aspects. Tumour cells’ biometric data visualisa- tion by violin-plots showed differences among differentiation classes, offering promising ways for digital classification. Combining the various digitally assessed features into the Neuroblastoma Digital Pathology Index we reached a clas- sification accuracy of 94% in non-high-risk and 88% in high- risk NB-patients. Conclusion: Our epidemiological findings were similar to lit- erature data. Digital MKI counting is accessible and delivers prognostic data. Ki-67 proliferation index’s own prognostic value falls behind some classic prognostic factors such as MKI, though it adds valuable tools for NB prognostisation through complex formulas, especially by automatic digital evaluation on hot-spots or whole slides. Our Neuroblastoma Digital Pathology Index’s classification accuracy is promising, just like the visual repre- sentation of tumour cells with violin-plots which might show differentiation related distribution patterns. OFP-08-002 Wilms’ tumour 1 expression in eutopic and ectopic decidua: a correlation with the pregnancy status J. Pacheco*, F. Rosa, E. Dvindenko, R. Barros, A. Costa Braga *Centro Hospitalar Universitário de São João, Portugal Background & objectives: Wilms’ tumour 1 gene (WT1) is reported to be overexpressed in decidual cells during preg- nancy, however, case series are lacking. Our aim is to evaluate WT1 expression in eutopic and ectopic decidua and correlate it with pregnancy status. S31