Search

SWI/SNF (SWItch/Sucrose Non-Fermentable) complex deficiency has been reported in a wide variety of cancers and is often associated with an undifferentiated phenotype. In the gynecologic tract SWI/SNF-deficient cancers are diagnostically challenging and little is known about their cellular origins. Here we show that undifferentiated endometrial carcinoma (UDEC), SMARCA4-deficient uterine sarcoma (SDUS), and ovarian small cell carcinoma, hypercalcemic type (SCCOHT) harbor distinct DNA methylation signatures despite shared morphology and SWI/SNF inactivation. Our results indicate that the cellular context is an important determinant of the epigenetic landscape, even in the setting of core SWI/SNF deficiency, and therefore methylation profiling may represent a useful diagnostic tool in undifferentiated, SWI/SNF-deficient cancers. Furthermore, applying copy number analyses and group-wise differential methylation analyses including endometrioid endometrial carcinomas and extracranial malignant rhabdoid tumors, we uncover analogous molecular features in SDUS and SCCOHT in contrast to UDEC. These results suggest that SDUS and SCCOHT represent chromosomally stable SWI/SNF-deficient cancers of the gynecologic tract, which are within the broader spectrum of malignant rhabdoid tumors. © 2022 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland., (© 2022 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.)

The color variation of hematoxylin and eosin (H&E)-stained tissues has presented a challenge for applications of artificial intelligence (AI) in digital pathology. Many color normalization algorithms have been developed in recent years in order to reduce the color variation between H&E images. However, previous efforts in benchmarking these algorithms have produced conflicting results and none have sufficiently assessed the efficacy of the various color normalization methods for improving diagnostic performance of AI systems. In this study, we systematically investigated eight color normalization algorithms for AI-based classification of H&E-stained histopathology slides, in the context of using images both from one center and from multiple centers. Our results show that color normalization does not consistently improve classification performance when both training and testing data are from a single center. However, using four multi-center datasets of two cancer types (ovarian and pleural) and objective functions, we show that color normalization can significantly improve the classification accuracy of images from external datasets (ovarian cancer: 0.25 AUC increase, p = 1.6 e-05; pleural cancer: 0.21 AUC increase, p = 1.4 e-10). Furthermore, we introduce a novel augmentation strategy by mixing color-normalized images using three easily accessible algorithms that consistently improves the diagnosis of test images from external centers, even when the individual normalization methods had varied results. We anticipate our study to be a starting point for reliable use of color normalization to improve AI-based, digital pathology-empowered diagnosis of cancers sourced from multiple centers. © 2021 The Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd., (© 2021 The Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.)

TP53 mutations are implicated in the progression of mucinous borderline tumors (MBOT) to mucinous ovarian carcinomas (MOC). Optimized immunohistochemistry (IHC) for TP53 has been established as a proxy for the TP53 mutation status in other ovarian tumor types. We aimed to confirm the ability of TP53 IHC to predict TP53 mutation status in ovarian mucinous tumors and to evaluate the association of TP53 mutation status with survival among patients with MBOT and MOC. Tumor tissue from an initial cohort of 113 women with MBOT/MOC was stained with optimized IHC for TP53 using tissue microarrays (75.2%) or full sections (24.8%) and interpreted using established criteria as normal or abnormal (overexpression, complete absence, or cytoplasmic). Cases were considered concordant if abnormal IHC staining predicted deleterious TP53 mutations. Discordant tissue microarray cases were re-evaluated on full sections and interpretational criteria were refined. The initial cohort was expanded to a total of 165 MBOT and 424 MOC for the examination of the association of survival with TP53 mutation status, assessed either by TP53 IHC and/or sequencing. Initially, 82/113 (72.6%) cases were concordant using the established criteria. Refined criteria for overexpression to account for intratumoral heterogeneity and terminal differentiation improved concordance to 93.8% (106/113). In the expanded cohort, 19.4% (32/165) of MBOT showed evidence for TP53 mutation and this was associated with a higher risk of recurrence, disease-specific death, and all-cause mortality (overall survival: HR = 4.6, 95% CI 1.5-14.3, p = 0.0087). Within MOC, 61.1% (259/424) harbored a TP53 mutation, but this was not associated with survival (overall survival, p = 0.77). TP53 IHC is an accurate proxy for TP53 mutation status with refined interpretation criteria accounting for intratumoral heterogeneity and terminal differentiation in ovarian mucinous tumors. TP53 mutation status is an important biomarker to identify MBOT with a higher risk of mortality.