384 Quantifying the Giants: A Novel Immunohistochemical and Digital Pathology Approach to Polyploid Giant Cancer Cell Detection and Prognostic Stratification in High-Grade Ovarian Carcinoma

Abstract Introduction/Objective Polyploid giant cancer cells (PGCCs) are large, mono- or multinucleated tumor cells implicated in cancer initiation, metastasis, and therapy resistance. First described in 1914, PGCCs have been observed across various solid tumors, including breast, prostate, and ovarian cancers. They are believed to arise through endoreplication in response to environmental stressors such as hypoxia, inflammation, and anticancer therapies. Despite their clinical relevance, PGCCs remain underrecognized due to their morphological heterogeneity and the absence of standardized detection methods. Their identification is further complicated by the lack of computational tools capable of reliably detecting and characterizing them in tumor samples. In ovarian cancer, particularly high-grade serous ovarian carcinoma (HGSOC), PGCCs may contribute to treatment failure and poor prognosis. This study aimed to address this gap by integrating immunohistochemistry (IHC) with digital pathology to develop a reproducible method for PGCC detection. Using membrane-targeted markers, we sought to delineate tumor cell boundaries and identify PGCCs in formalin-fixed paraffin-embedded (FFPE) samples. The study also evaluated the association between PGCC burden and clinical, histopathological, and molecular features in ovarian cancer. By establishing a standardized detection method, this research aims to enhance our understanding of PGCCs and their potential prognostic value in ovarian cancer management. Methods/Case Report This study evaluated four membrane markers—EpCAM, Na+/K+ ATPase, cadherins, and PMCA1—via immunohistochemistry (IHC) in 40 aggressive tumor cases, including high-grade serous ovarian carcinoma (HGSOC), glioblastoma, triple-negative breast cancer, and pancreatic adenocarcinoma. Marker performance was assessed based on specificity, intensity, uniformity, reproducibility, and compatibility with other IHC reagents. EpCAM was selected for its superior membrane delineation and applied to 26 FFPE ovarian cancer samples collected between 2015 and 2022. Slides were digitized using the Aperio ScanScope CS System and analyzed with Aperio ImageScope software. PGCCs were identified and scored (0–3) using both digital pathology and conventional light microscopy. EpCAM staining intensity was evaluated visually and digitally, and differences were statistically analyzed using the Wilcoxon signed-rank test. Clinical data (age, stage, treatment type), histological parameters (mitotic index, morphology), and molecular findings (BRCA, TP53, ATM, CHEK2) were collected. TP53 mutations were assessed via next-generation sequencing (NGS) in 25 patients. Treatment sequence (primary debulking vs. neoadjuvant chemotherapy) and survival outcomes were also analyzed. The study aimed to correlate PGCC burden with clinical and molecular features to assess its prognostic significance. All procedures were conducted under institutional review board approval and followed standard pathology protocols. Results All four membrane markers delineated tumor cell membranes effectively, enabling PGCC identification. EpCAM, Na+/K+ ATPase, and PMCA1 outperformed cadherins in staining quality. EpCAM was selected for further analysis due to its superior performance. In the 26 HGSOC cases, EpCAM staining revealed significant intra- and intertumoral heterogeneity. PGCCs were more frequently identified using digital pathology than visual assessment, with statistically significant differences in scoring (Wilcoxon signed-rank test, p < 0.05). Patients with low PGCC scores (0–1) had a longer median overall survival (62 months) compared to those with high scores (2–3; 39 months), though this difference did not reach statistical significance (Mann–Whitney U test: p = 0.066). At the time of analysis, 50% of low-score patients were alive versus 17% in the high-score group (Fisher’s Exact Test: p = 0.149). All sequenced tumors harbored TP53 mutations, including frameshifts, missense substitutions, and deletions, with strong concordance between mutation type and p53 IHC patterns. BRCA mutations were present in ∼18% of cases; ATM and CHEK2 mutations were less frequent. Patients who underwent primary debulking surgery showed a trend toward improved survival compared to those receiving neoadjuvant chemotherapy, although this was not statistically significant (p = 0.096). Conclusion This study establishes a reproducible method for detecting polyploid giant cancer cells (PGCCs) in high-grade serous ovarian carcinoma using EpCAM-based immunohistochemistry and digital pathology. Digital analysis proved more sensitive than visual assessment, enabling consistent identification of PGCCs and revealing a trend toward poorer survival in patients with higher PGCC scores. Although not statistically significant, this trend suggests a potential prognostic role for PGCCs. The universal presence of TP53 mutations and the observed variability in treatment outcomes underscore the biological and clinical complexity of HGSOC. These findings support further investigation into PGCCs as prognostic biomarkers and therapeutic targets in ovarian cancer.