ECP 2023 Abstracts

S9 Virchows Archiv (2023) 483 (Suppl 1):S1–S391 13 moderate/strong intensity and one (with SPECCL1::NTRK3 fusion) was negative. Eleven of 461 (2.4%) cases without NTRK rearrange- ments also exhibited pan-TRK expression: 3 low-grade and 7 high- grade endometrial stromal sarcomas and 1 undifferentiated uterine sarcoma. Molecular confirmation of the absence of NTRK rearrange- ments was possible in 9 tumours; molecular testing was not successful in the other 2 cases. Of the 9 pan-TRK positive neoplasms without NTRK-rearrangements, 7 exhibited focal/multifocal (<70%) cytoplas- mic staining with weak/moderate positivity. None exhibited strong and diffuse staining. Conclusion: Even though pan-TRK immunohistochemistry is not entirely sensitive or specific for NTRK-rearranged sarcomas, these neoplasms tend to exhibit diffuse staining with moderate/strong inten- sity, unlike the positive staining in neoplasms without NTRK-rear- rangements. Pan-TRK should be performed in uterine monomorphic spindle neoplasms negative for smooth muscle markers and hormone receptors and variably positive for CD34 and S100. The diagnosis of an NTRK-rearranged sarcoma requires molecular confirmation but pan- TRK immunohistochemistry is a useful screening tool to direct which cases require molecular testing OFP-02-011 Oestrogen receptor expression in endometrial adenocarcinomas of no specific molecular profile shows a bimodal distribution B. Gilks*, L. Hoang, B. Howitt, T. Longacre, J. McAlpine, A. Jamieson, N. Singh, J. Pors *Canada Background & objectives: Decreased Oestrogen receptor (ER) expression is a poor prognostic marker in endometrial carcinomas of no specific molecular profile (NSMP), but the optimal cut-off for low ER (high-risk) versus high ER (low-risk) expression has not been defined. Methods: Endometrial biopsy specimens from 120 cases of endome- trial carcinoma of NSMP molecular subtype were stained for ER and Allred score was assigned (sum of staining intensity score, with 0, 1=weak, 2=moderate and 3=strong staining, and distribution score, with 0=negative, 1=<1% of cells staining, 2=1-10%, 3=11-33%, 4=34-66% and 5=67-100%). Histotype and grade was available for every case. Clinical follow-up data pending. Results: The Allred score distribution is bimodal with a small peak at 0 and larger peak for scores 7/8. Twelve tumours had an Allred score of 0-3, including four endometrioid carcinomas, four clear cell carcino- mas, and one each of: mesonephric-like adenocarcinoma, gastric-type adenocarcinoma, carcinosarcoma and endometrioid carcinoma with yolk sac differentiation. One mesonephric-like adenocarcinoma had an Allred score of 4 and the two tumours with an Allred score of 5 were a mesonephric-like adenocarcinoma and an endometrioid carcinoma. Five tumours had an Allred score of 6: 4 endometrioid carcinomas and 1 mixed clear cell and endometrioid carcinoma. 99% of tumours with an Allred score of 7 or 8 (99/100) were endometrioid. Conclusion: The distribution of ER expression in NSMP endometrial carcinoma is bimodal, with most (>80%) showing high level ER expres- sion (Allred score 7 or 8) when staining is performed in well-fixed endo- metrial biopsy specimens. All non-endometrioid carcinomas and a few endometrioid carcinomas had lower-level ER expression (Allred score of 6 or less). These results suggest an Allred score cut-off of 6 sepa- rates high from low ER expressing NSMP endometrial carcinomas, with potential implications for risk stratification and clinical management. Funding: Terry Fox Research Institute OFP-02-012 Cytology evaluation of neuroblastic tumours E. Dvidenko*, F. Santos, M.M. Lemos *Instituto Português de Oncologia de Lisboa Francisco Gentil, Portugal Background & objectives: Neuroblastic tumours (NTs) arise from neural crest-derived tissues and are divided in four categories by International Neuroblastoma Pathology Committee (INPC). We aim to describe our cytological experience with NTs as a useful tool for initial diagnostic approach and molecular characterization. Methods: NTs diagnosed in our institution from 2010 to 2022 with pre-treatment cytological and histological first diagnosis were reviewed. Air-dried touch preparations from fine-needle aspira- tions and biopsy imprints were available. Flow cytometry, seg- mental chromosomal alterations analysis and MYC amplification by fluorescence in situ hybridization (FISH) were performed in both smears and fresh tissue. Cytological and histological findings were correlated. Results: Twenty-nine NTs were included. Mean patient age was 38.7 months (2 – 132 months). Fifteen were females (51,7%). Adre- nal gland was the main primary site (55,2%). A diagnosis of NT was achieved in 22 smears (75,9%): 14 poorly differentiated NTs, 4 undifferentiated NTs and 4 were reported as unclassifiable NTs. The remaining 7 cases were cytologically diagnosed as small round cell tumours. Undifferentiated and poorly differentiated NTs were concordant with final histological diagnosis. Molecular studies and FISH results were available in 25 cases (86.2%): 12 had segmental chromosomal aberrations; 4 had MYC amplification; none had ALK gene alterations. DNA index was feasible in 23 cases (79.3%) with seven diploid tumours. Conclusion: Cytology, a minimally invasive technique, reliably yields material for diagnosis and molecular studies of NTs. Nevertheless, due to tumour heterogeneity, definitive cytological evaluation and tumour subtyping according to INPC is not widely recommended. Our study confirms the suitability of cytological samples for molecular charac- terization, but also demonstrates that a rapid diagnosis and subtype assessment of non-treated NTs is possible in these specimens, pro- viding they are adequately handled and evaluated by an experienced cytopathologist. OFP-02-013 Application of the newWHO reporting system for lung cytopathology on bronchial wash samples: a cytohistological correlation study from a tertiary hospital J. Vaz Silva*, L. Leça, C. Lobo, N. Coimbra, A.L. Cunha, P. Monteiro *PathologyDepartment, Portuguese Oncology Institute of Porto (IPO-Porto), School of Medicine, University of Minho, Braga, Portugal Background & objectives: Bronchial wash(BW) is a minimally inva- sive method to sample mucosal surfaces of the bronchial tree to assist in the diagnosis of infection and malignancy. Several studies report a wide range of sensitivity for malignancy detection, with different fac- tors influencing sensitivity. Methods: Recently, WHO introduced a new reporting system for lung cytopathology. Our aim was to assess the new system accuracy for BW samples, through cytohistological correlation, while characterizing BW performance in lung diagnosis. We retrieved all BW from 2021-2022 with an available histological diagnosis. Cytology reports were divided into negative- and positive- tiers (including suspicious/atypical categories). BW-performance parameters were calculated following histological correlation. Results: We assessed a total of 440 cytological and 440 histological samples. Each biopsy/histologic diagnosis correlated with a BW per- formed either simultaneously or prior. BW performance analysis demonstrated a sensitivity of 37,9% and a speci- ficity of 93,3%; positive predictive value of 95,6% and a negative predictive value of 28,5%. Overall diagnostic accuracy was 49,5%. Sensitivity was slightly higher in patients with a diagnosis of squamous cell carcinoma than in patients with adenocarcinoma diagnosis.