ECP 2022 Abstract Book

Virchows Archiv (2022) 481 (Suppl 1):S1–S364 13 1. EBUS-TBNA (n=415): C1=8.7%, C2=54.7%, C3=2.4%, C4=1.4%, C5=32.8%. 2. Bronchial brushings (n=96): C1=1.0%, C2=49.0%, C3=8.3%, C4=3.1%, C5=38.5%. 3. Bronchial washings (n=486): C1=3.1%, C2=75.7%, C3=8.4%, C4=2.3%, C5=10.5%. 4. Bronchial lavages (n=429): C1=1.9%, C2=97.0%, C3=0.5%, C4=0.2%, C5=0.5%. 5. Sputum (n=7): C1=14.3%, C2=85.7%, C3=0.0%, C4=0.0%, C5=0.0%. Conclusion: Coding systems are widely utilised in other areas of cytopathology, such as in thyroid and salivary gland specimens. The consistency and standardisation of reporting achieved from such systems provides greater clarity for the treating physician and facilitates clear communication of essential information. We have been using a classification system analogous to the pro- posed new respiratory cytology classification system for approxi- mately 10 years. This data provides indicative rates of cytological outcomes in a tertiary referral lung oncology centre. OFP-02-013 A retrospective report-based review of 4,155 consecutive patients undergoing minimally invasive thoracic lymph node sampling M. Bonert*, S. Sayeda, H. Begum, J. Agzarian, J. Cutz, C. Finley, A. Naqvi *Pathology, McMaster University, Canada Background & objectives: Endoscopic bronchial ultrasound (EBUS) and endoscopic ultrasound (EUS) are minimally invasive strategies to assess thoracic lymph nodes. This project examined the pathology of patients from a regional thoracic surgery cen- tre that underwent EBUS/EUS and (if applicable) lung cancer resection. Methods: All EBUS/EUS specimens accessioned 2011-2020 were retrieved and linked via anonymized patient identifier to lung cancer resections with synoptic reports. Cases were auto- matically classified and grouped into mutually exclusive catego- ries (malignant(MAL), suspicious(SUSP), insufficient(INSUF), benign(BEN)), based on the most recent specimens using a previously validated program. Data was stratified by location (station(ST)7, ST4R, ST4L, all other N2 ST(STN2others), all N1 ST(STN1all)). Results: The cohort contained 4,155 patients with 10,922 EBUS/ EUS specimens. Patients had 1 to 13 specimens and median number of specimens per patient was three. Patients with station MAL/number of patients(% MAL) was: ST7 750/3,407(22%), ST4R 816/2,844(29%), ST4L 293/1,504(20%), STN2others 283/668(42%), STN1all 381/1,264(30%). Overall, N2 MAL sta- tus/number of patients(%) was 1,399/3,992(35%) and multiple N2 stations/STN2others were involved in 604 of 1,399(43%) patients. 829(20%) patients went for lung cancer resection; 47 had two sur- geries and two had three surgeries. Nodal stage by most recent surgical specimen was 20 (patients) pNX, 563 pN0, 175 pN1 and 71 pN2. 469 of 4,155 patients (11%) had granulomatous inflam- mation in 838 EBUS/EUS specimens. Conclusion: EBUS/EUS is used for lung cancer staging, sarcoido- sis and other indications. Patients with N2 positivity frequently have multiple positive stations. The low pN2 rate on resection in our environment confirms EBUS/EUS is effective at selecting lung cancer patients for surgery. The assessment of pathologic cate- gorizations can provide insights in thoracic surgery and tumour biology. OFP-02-014 The Paris System for reporting urinary cytology: 5 years of institutional experience in a tertiary hospital M. Pinho*, R. Moiteiro da Cruz, J. Boavida, R. Luís *Department of Pathology - Hospital de Santa Maria, CHULN; Faculdade de Medicina da Universidade de Lisboa, Portugal Background & objectives: The Paris System for Reporting Urinary Cytology (TPS) standardizes the triage criteria for high- grade urothelial carcinoma (HGUC) in urine specimens. Our aim was to assess the TPS accuracy through cytohistological correlation, stratified by both cytological category and histological grade. Methods: The reports of urinary tract histological diagnoses and pertaining retrospective cytological specimens, examined in 2016-2021, were retrieved from our Department software data- base; non-satisfactory/non-diagnostic samples were unaccounted for. The TPS categories were divided into negative and positive tiers (suspicious and HGUC combined); the risk of malignancy (ROM) and performance parameters were then calculated, follow- ing histological correlation. Results: A total of 1261 cytological and 638 histological samples were assessed. Each biopsy/surgical diagnosis correlated with an average of 2 urinary (voided/instrumented) specimens (ranging from 1 to 12), sampled either simultaneously or up to 52 months before- hand. The ROM ranged from 17.4% in negative samples to 71.5% in positive categories; the performance analysis revealed a sensitiv- ity of 46.7%, a specificity of 91.8%, a positive predictive value of 78.4% and a negative predictive value of 72.7%. Atypical urothelial cells were independently analysed and revealed mixed results. Conclusion: We present a heterogeneous series, encompassing the learning curve following TPS implementation and clustering both “de novo” diagnoses and urothelial neoplasm follow-ups. The large volume of these samples impacts its diagnostic yield and hence a low sensitivity was expected, contrasting with a high specificity that translates the atypia of even scarce shed tumour cells. On the other hand, although pathologists are not blinded to clinical information and previous diagnosis, histological false negatives due to undersampling must not be overlooked. OFP-02-015 Fine needle aspiration biopsy of orbital masses: a review of 36 cases H. Domanski*, J. Köster *Department of Genetics and Pathology, Division of Laboratory Medicine Region Skåne, Lund, Sweden Background & objectives: Fine needle aspiration biopsy (FNA) is a useful tool to triage orbital lesions in cases where a diagnosis cannot be made by clinical and imaging findings alone. FNAB can provide crucial information before a surgical and oncologic procedure is undertaken. Methods: Over a period of 5 years, 2017-2021, 36 patients under- went FNAB of orbital masses. Twenty-nine lesions were sampled by FNAB using a transcutaneous or transconjunctival approach and 7 patients underwent CT guided FNAB with the assistance of a cytopathologist. Thirteen patients had a prior history of malignan- cies. Flow cytometry was performed in 16 cases and immunocyto- chemical examination in one. Results: Diagnostic material was obtained in 28 lesions. Malignant haematolymphoid lesions represented most of the malignancies: 6 primary and 4 recurrences of non-Hodgkins lymphomas and acute myeloid leukaemia. There were 8 metastatic malignancies: 3 breast, 1 lung, 1 adrenal and 2 neuroendocrine carcinomas, and 1 metastasis of glioblastoma. S10