ECP 2023 Abstracts

S7 Virchows Archiv (2023) 483 (Suppl 1):S1–S391 13 *Pathology Unit, Department of Woman and Child’s Health and Pub- lic Health Sciences, Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Rome, Italy Background & objectives: Locally advanced cervical cancer (LACC) represents 37% of worldwide cervical cancer. Chemo-radiotherapy fol- lowed by surgery is an alternative to standard treatment. We aimed to assess the prognostic role of the therapeutic response in LACC using Mandard tumour regression score. Methods: We enrolled 244 patients with LACC treated with CT-RT followed by surgery. Histopathologic response was assessed according to Mandard-score system: 1-complete fibrosis; 2- rare residual can- cer cells scattered through the fibrosis; 3- increased number of cancer cells, fibrosis predominating; 4- residual cancer outgrowing fibrosis; 5- no fibrosis. Kaplan–Meier and Cox regression were used for survival analysis. Results: We found a complete pathological response in 118 patients (48.4%), tumour regression score (TRG) 2 in 49 cases (20.1%), TRG3 in 35 cases (14.3%), and 42 (17.2%) were classifies as non-responders (TRG4-5). TRG was significantly associated with OS (p<0.001) and PFS (p<0.001). Main responders (TRG1-2) showed a 92% 5-year overall survival (5y-OS) and a 75% 5-year disease free survival (5y-DFS). Minor or no responders showed a 48% 5y-OS and a 39% 5y-DFS. Kaplan-Meier curves for OS and PFS showed two prognostic groups: TRG1-TRG2 cases had a better outcome than TRG3-TRG4-5 cases. The 2-tiered TRG was independently associated with both DFS and OS in Cox regression analysis. Conclusion: Currently there is not a standard method to classify tumour regression in LACC. We showed that Mandard TRG is strongly associated with OS and PFS. Moreover, 2-tier TRG (TRG1-2 vs TRG3- 4-5) is an independent prognostic factor in post-CT/RT LACC, with potential benefits in defining post-treatment adjuvant therapy. Lastly, 2-tier TRG is an easier method to assess tumour regression overcoming the difficulty of 5-tier TRG in distinguishing regressive fibrosis from the dense fibrous cervical stroma. OFP-02-005 Mutational profiling of recurrent endometrial endometrioid car- cinoma with no specific molecular profile (NSMP)/p53wt using endometrial brush sampling A. Khatib*, T. Revil, L. Jiang, L. Gilbert, B. Tessier-Cloutier *Division of Gynaecologic Oncology, McGill University Health Cen- tre, Montreal, Canada Background & objectives: The NSMP/p53wt endometrial carcinoma (EC) is defined by the exclusion of POLE mutation, mismatch repair protein loss and aberrant p53 protein expression. We set to study clin- icopathological features of NSMP/p53wt EC using a targeted DNA sequencing panel. Methods: Endometrial brush samples were collected at the time of hysterectomy in patients with low-grade (FIGO grade 1-2) and low stage (pT1-2) EC NSMP/p53 between 2015 and 2019. Clinicopathol- ogy review was performed for each case. Mutational profiling assay included low-pass whole genome DNA sequencing, as well as targeted sequencing for 19 common EC genes. Results: We included 180 NSMP/p53 cases, 13 of which recurred. The median age was 60-year-old, and all patients were treated with surgery. The most common site of recurrence was the vaginal apex. Logis- tic regression analyses showed that positive intra-operative cytology (p=0.006), grade 2 (p=0.027), and tumour size (p=0.037) were indepen- dently associated with a higher rate of recurrence. All tumours expressed ER and PR by immunohistochemistry. Mutational analysis, currently available for 99 cases (seven with recurrence), showed that 1q gain was associated with tumour recurrence (p=0.0247), none of which had a PTENmutation. One recurrent tumour had an APC mutation (p=0.0173). No other alterations were significantly associated with recurrence. Conclusion: Our analysis supports that, along high-risk clinicopatho- logic metrics, 1q gain is a helpful marker of more aggressive disease in low-grade and ER positive EC. As we lean further into the molecular classification for the diagnosis of EC, DNA sequencing approaches for NSMP/p53wt subgroup may be necessary for adequate for clini- cal stratification. Sequencing is ongoing to validate our results in the remaining cases. Funding: Henry R. Shibata Cedars Cancer Fellowship/Cedars Cancer Foundation OFP-02-006 Proficiency testing of p53 IHC pattern interpretation in vulvar biopsies shows overcalling of the mutant basal overexpression pat- tern in TP53 wild-type cases M. Kinloch*, C. Parra-Herran, G. Focchi, B. Howitt, S.L. Skala, S. Strickland, J. Watkins, R. Wong, E. Torlakovic, R. Crawford, B. Gilks, L. Anderson, N. Singh, K. Wei, L. Hoang *University of Saskatchewan, Canada Background & objectives: Recently, criteria for p53 immunohisto- chemistry (IHC) interpretation was described in the vulva. However, proficiency of pathologist read out has not been studied. We used a read-out module to assess proficiency variation in a set of gynaecologic pathologists. Methods: Scanned images of p53 IHC on 90 vulvar biopsies (diagno- ses of non-neoplastic dermatoses, high-grade squamous intraepithe- lial lesion, differentiated vulvar intraepithelial neoplasia, verruciform acanthotic vulvar intraepithelial neoplasia) were provided. Scoring completed by pathologists as wild-type or mutational via 6 p53 IHC patterns. Following 45 cases in Module A and given TP53 sequencing data, a second 45-case Module B was completed. Results: Eleven gynaecologic pathologists participated in module A with a mean accuracy of 67.04% (concordant p53 IHC pattern and NGS result), that ranged from 55.56% to 86.67%. Accuracy improved for 75% pathologists from Module A to Module B with a mean of 77.22%. The mean improvement was 10.2%. Null pattern (100% case accuracy) and parabasal/diffuse overexpression patterns (66% case accuracy) were the most accurate in pathologist readout (where ≥9 pathologists accurately predicted the mutation). Remarkably, the basal overexpression pattern had only 20% case accuracy. There were 21 cases where ≥4 pathologists overcalled basal overexpression and 2 cases where all 12 pathologists overcalled a basal overexpression mutant pattern in a wild-type setting. Conclusion: Proficiency readout improved from Module A to Module B with the educational intervention of TP53 sequencing data. How- ever, pathologists must exhibit caution in overcalling the mutant basal overexpression pattern. Given basal overexpression can occur in both wild-type and TP53 mutation settings, pathologists may not be able to recognize a mutant basal overexpression, even with training. Limita- tions of this study are small sample size and pathologist’s blinding to H&E and p16. Funding: Carraresi Foundation and Sumiko Kobayashi Marks Memo- rial OVCARE Research Grants OFP-02-007 Prognostic impact of pathological parameters in molecular sub- groups of endometrial carcinoma A. De Leo*, M. Ruscelli, F. Rosini, A.G. Corradini, T. Maloberti, S. Coluccelli, A. Costantino, C. Ceccarelli, D. de Biase, D. Santini, G. Tallini *University of Bologna, Italy Background & objectives: Molecular characterization of endome- trial carcinoma (EC) has recently included in the ESGO/ESTRO/ESP