ECP 2023 Abstracts

Vol.:(0123456789) 13 Virchows Archiv (2023) 483 (Suppl 1):S1–S391 https://doi.org/10.1007/s00428-023-03602-w ABSTRACTS 35 th European Congress of Pathology – Abstracts Published online: 2 September 2023 © Springer-Verlag GmbH Germany, part of Springer Nature 2023 Oral Free Paper Sessions OFP-01 | Oral Free Paper Session Digestive Diseases Pathology – GI OFP-01-001 Results of the “Uniform Noting for International application of the tumour-stroma ratio as easy diagnostic tool” (UNITED): a multi- centre prospective validation study M. Polack*, A.G. Roodvoets, E. Meershoek-Klein Kranenbarg, H.A. Gelderblom, V. Terpstra, G. Petrushevska, G. Gasljevic, S. Kjær- Frifeldt, E.M. de Cuba, N. Bulkmans, G.R. Vink, R. Al Dieri, R.A. Tollenaar, H.J. van Krieken, W.E. Mesker *Department of Surgery, Leiden University Medical Center, Leiden, The Netherlands Background & objectives: As advised by the UICC and CAP, the prospective multicentre UNITED study was initiated to validate the tumour-stroma ratio (TSR) as prognostic independent parameter, pre- dicting patient-related outcomes for stage II-III colon cancer (CC) patients and subsequently aid in personalized treatment. Methods: The UNITED study was enrolled in 27 participating centres in 12 countries worldwide. The effect of TSR, categorized as stroma- high (>50%) or stroma-low (≤50%) through standardized histopathol- ogy microscopic assessment by certified pathologists, was evaluated on disease-free survival (DFS) as primary endpoint for a 3-year median follow-up period. Secondary endpoints were response to adjuvant chemotherapy and overall survival (OS). Results: A total of 1537 patients were included, of which 1388 were deemed eligible and operated between 2015-2021. DFS was signifi- cantly shorter (p<0.001) in stroma-high CC patients (n=428) than in stroma-low CC patients (n=960). In multivariate analysis, the TSR was an independent prognosticator for DFS (p=0.001; hazard ratio 1.498, 95% confidence interval 1.180-1.903) as well. As secondary outcomes, stroma-high stage II-high risk and stage III CC patients also notably had a worse DFS despite receiving adjuvant chemotherapy (p<0.001). For OS, although the follow-up period was relatively short, already a slight trend towards worse OS in stroma-high CC patients was visible (p=0.102). Conclusion: The multicentre UNITED study hereby unequivocally validates the TSR as an independent, prognostic factor for DFS in stage II-III CC patients, proving that stroma-high CC patients have a worse survival. Furthermore, these patients were also observed to respond worse to adjuvant chemotherapy. As the TSR can thus aid in shared decision-making and personalized treatment, future implementation in the tumour-node-metastasis-classification is aimed. This work was supported by grants from the Dutch Cancer Society (KWF Kankerbestrijding; project 10174) and the Stichting Fonds Oncologie Holland, as well as the Bollenstreekfonds, Lisse, the Netherlands. These funders had no role in study design, data collec- tion and analysis, nor in the decision to publish, nor in the preparation of the manuscript. OFP-01-002 Multi-site multi reader study on artificial intelligence-assisted pri- mary diagnosis of gastric biopsies M. Rodriguez-Justo*, G.Y. Lauwers, A.L. Booth, A.P. Levine, W. Jervis, K. Kantekure, L. Aherfi, E. Lanteri, L. Apple, A. Arat, R. Miku- linsky, T. Amit, M. Vecsler, M. Grinwald, J. Sandbank *Cancer Institute, University College London, United Kingdom Background & objectives: This study aimed to clinically validate the use of an artificial intelligence (AI)-based solution by pathologists for reviewing and reporting gastric biopsies. Methods: A two-arm prospective reader study comparing the perfor- mance of pathologists supported by AI with pathologists reviewing digital slides was performed at multiple sites (varied staining proto- cols & different scanners). Both arms were compared to ground truth (GT) established by a consensus of 2 GI-pathologists. Rates of major discrepancies between each arm and GT, as determined by an adjudi- cating-pathologist, were compared. Results: 6 pathologists reported on 235 cases (426 H&E-slides), each case being reported twice, once in each study arm. Pathologists first reviewed only slides and IHC after request, while the AI results were rendered on H&E-slides only. The AI solution demonstrated high performance for the detection of gastric-neoplasia (Carcinoma/HGD/ HG-Lymphoma):AUC of 0.98 (95% CI: 0.967,0.994), sensitivity 96%, specificity 90%, NPV 100% and PPV 94%. High performance was demonstrated for H Pylori detection: AUC of 0.93 (95% CI: 0.88,0.97) sensitivity 91%, specific- ity 80%. Pathologists’ feedback showed the AI solution is user friendly (92.5%), adds confidence to the cases review (80%) and a majority (83%) would prefer continuing to work with the system. Conclusion: This multi-site multi-reader study reports high accuracy for the detection of gastric neoplasia by the AI solution. The AI solu- tion performed accurately and equally well with slides issued from different staining platforms and scanners. Thus, AI solutions have the potential to be a significant helping tool for pathologists in various clinical decision-making in routine pathology practice, enhancing the quality and reproducibility of diagnoses. OFP-01-003 Quantitative and qualitative analyses of exosomes derived from cancer-associated fibroblasts (CAFs) according to the desmoplastic reaction (DR) patterns in colorectal cancer K. Tashiro*, K. Okamoto, S. Mochizuki, E. Shinto, Y. Kajiwara, M. Yamadera, H. Abe, Y. Sota, K. Kikuya, T. Kawauchi, K. Aihara, H. Ueno *Department of Surgery, National Defense Medical College, Japan