ECP 2023 Abstracts

S168 Virchows Archiv (2023) 483 (Suppl 1):S1–S391 13 *University of Pisa, Italy Background & objectives: Neoadjuvant therapy has become the standard of care for HER2-positive breast cancer. However, only half of the patients achieve a complete pathological response. Our study aimed to investigate the correlation between HER2 spatial distribution and treatment response. Methods: Preoperatory biopsies from 101 HER2-positive/3+ IHC- score invasive breast cancer patients undergoing neoadjuvant chem- otherapy were retrieved. Four groups of HER2 distribution (named Pisa Prediction Groups – PPG) were defined: 1 - homogeneous; 2a - big clusters; 2b - small clusters; 2c - mosaic pattern. The study was approved by the local Ethical Committee. Statistical analysis was per- formed using GraphPad Prism Software. Results: Within the 64 cases of PPG 1, 75% achieved pCR; instead, among the 37 cases of the heterogeneous groups (PPGs 2a+2b+2c), only 43% achieved pCR: in detail, the pCR rate was 54% for PPG 2a, 50% for PPG 2b and 35% for PPG 2c. Interestingly, no statistically significant difference was found between PPG 1 and PPG 2a (p-value = 0.1768); instead, PPG 2c resulted different from PPG 1 (p-value = 0.0023), as well as PPGs 2c+2b from PPG 1 (p-value = 0.0022). Conclusion: This study confirms the existence of considerable heterogene- ity in the expression of HER2 protein within breast cancer. Intriguingly, the results demonstrate the need to integrate the percentage of HER2-positive cells with the assessment of their distribution within the tumour. Pattern analysis of HER2 spatial distribution could help identify breast cancer patients who will not achieve pCR. This could represent a powerful deci- sion tool in terms of escalation treatment approaches. E-PS-02-062 The added value of SOX10 immunohistochemistry to other breast markers in identifying Cytokeratin 5-positive triple negative breast cancers as of mammary origin A. Sejben*, A. Voros, A. Golan, T. Zombori, G. Cserni *University of Szeged, Hungary Background & objectives: Triple-negative breast cancer (TNBC) rep- resents a specific group that lacks the expression of hormone receptors and might also lack the expression other breast markers; when this occurs, proving the breast origin of a metastasis is a challenging task. Methods: In the present study, we assessed the added value of SOX10 immunohistochemistry to known GATA3, mammaglobin (MG), GCDFP15 (growth cystic disease fluid protein 15), and NYBR1 statuses in a series of CK5-positive primary TNBCs. Tissue microarrays were made from the formalin-fixed and paraffin-embedded blocks of 120 TNBCs, and 3-4-mm- thick sections were immunostained for SOX10. Results: In our cohort, SOX10 positivity was seen in 82/119 cases, 61, 74, 76, and 82 all of which were GATA3, MG, GCDFP15, and NY-BR-1 negative, respectively. The cut-off for a positive reaction was at least 10% of tumour cells staining. Of the SOX10 negative cases, 12 stained with at least another breast marker. Nevertheless, 25/119 (21%) cases remained negative with all markers assessed. Conclusion: SOX10 proved to be the most commonly positive breast marker in our CK5 expressing TNBCs, but the other markers also had some additive value to SOX10. E-PS-02-063 Examination of tumour regression grading systems in breast can- cer patients who received neoadjuvant therapy A. Sejben*, R. Kószó, Z. Kahán, G. Cserni, T. Zombori *University of Szeged, Hungary Background & objectives: Neoadjuvant therapy is a common form of treatment in locally advanced breast cancer (LABC) patients. Besides some guidelines for grading regression, a standardized general scheme is not yet available. Methods: The aim of our study was to compare the prognostic impact of different regression grading systems, namely the TR/NR, Chevallier, Sataloff, Denkert-Sinn, Miller-Payne, NSABP-B18, Residual Disease in Breast and Nodes and Residual Cancer Burden (RCB) on disease- free (DFS) and overall survival (OS). Results: Data of 746 breast cancer patients treated in neoadjuvant set- ting between 1999 and 2019 have been included. The DFS and OS estimates of patients with complete pathological regression and residual in situ carcinoma have been significantly more favourable than those having partial regression or no signs of regression (pDFS<0.001, pOS < 0.001). Significant differences were found between DFS estimates of classes with partial regression and without regression defined by RCB. Concerning DFS estimates, the RCB classification (p = 0.019), while regarding OS data the y-stage (p = 0.011) and the nodal status (ypN; p = 0.045) were significant prognosticators by multivariate Cox regression. Conclusion: Regression grading systems help the evaluation of regres- sion in LABC patients treated with neoadjuvant therapy. Of the sev- eral grading systems compared, the RCB classification makes the best distinction between the outcomes of the different classes, therefore we recommend the inclusion of RCB into the histopathological findings. E-PS-02-064 During the COVID-19 lockdown younger breast cancer patients and those with more unfavourable prognostic parameters had the opportunity to be treated S. Sevim*, H.K. Kahraman, E.D. Serbes, S. Dizbay Sak *Ankara University Faculty of Medicine, Pathology Department, Turkey Background & objectives: The COVID-19 pandemic caused by the SARS-CoV-2, has affected the routine pathology workflow, as well as many applications in medicine. In the current study, it was aimed to investigate the changes in breast pathology caused by the pandemic. Methods: Breast specimens examined in our laboratory were divided into three groups, each for 16-month periods, as pre-COVID-19 (pre- COV), lockdown period (LP), and post-COVID-19 (postCOV). Macro- scopic, histopathological, immunohistochemical, molecular features, and neoadjuvant therapy status were analysed to reveal the changes in these periods. In comparative statistical analyzes of different periods in terms of clinicopathological parameters, p<0.05 was considered significant. Results: A total of 2758 breast specimens were examined. Number of breast specimens decreased during LP (n=718), compared to pre- COV (n=969) and postCOV (n=1071) (p=0.000). In LP, non-tumoral pathologies and benign tumours were decreased compared to pre- COV, whereas malignancies (n=365, 56.2% vs n=406, 41.9%) were increased (p=0.000). During LP, patient age was younger (49.46 vs 51.55, p=0.004), Ki67 index was higher (p=0.04) and LVI was more frequent (p=0.031) in breast carcinomas, compared to preCOV. In post- COV, frequency of tru-cut biopsy and mammoplasty specimens and benign tumours increased. Tumour size (p=0.038) and frequency of PR positivity (p=0.027) decreased compared to LP. HER2 positivity by IHC/SISH in postCOV was higher than LP (p=0.040). Conclusion: The COVID-19 pandemic has adversely affected breast carcinoma patients. It can be speculated that, older patients could not access to the hospital and breast physicians in the LP, due to age restric- tions. Patients with unfavourable prognostic parameters like higher Ki67 and LVI were operated more in LP, probably due to the more prominent clinical symptoms/findings indirectly related to these param- eters. Less essential operations such as mammoplasties and benign resections have been postponed. E-PS-02-065 Diabetic mastopathy: two cases report and review of the literature I. Sfiniadakis*, E. Sfiniadaki, D. Vasileiou-Dervisoglou, P. Sfiniada- kis, D. Sfiniadaki, K. Sfiniadakis *Athens Naval Hospital, Greece