The American Journal of Pathology

S100A4/Nonmuscle Myosin IIA/p53 Axis Contributes to Aggressive Features in Ovarian High-Grade Serous Carcinoma

S100A4 is a small calcium-binding protein that exerts its biological functions by interacting with nonmuscle myosin IIA (NMIIA) and p53. Although S100A4 promotes metastasis in several tumors, little is known about its involvement in the progression of ovarian high-grade serous carcinomas (HGSCs). Herein, we focused on functional roles of the S100A4/NMIIA/p53 axis in these tumors. In HGSC cell lines harboring mutant p53, knockdown (KD) of S100A4 reduced the expression of several epithelial-mesenchymal transition/cancer stem cell markers and the ALDH1

Anaplastic Lymphoma Kinase Overexpression Is Associated with Aggressive Phenotypic Characteristics of Ovarian High-Grade Serous Carcinoma

Deregulated full-length anaplastic lymphoma kinase (ALK) overexpression has been found in some primary solid tumors, but little is known about its role in ovarian high-grade serous carcinoma (HGSC). The current study focused on the functional roles of ALK in HGSC. Cytoplasmic ALK immunoreactivity without chromosomal rearrangement and gene mutations was significantly higher in HGSC compared with non-HGSC-type ovarian carcinomas, and was significantly associated with several unfavorable clinicopathologic factors and poor prognosis. HGSC cell lines stably overexpressing ALK exhibited increased cell proliferation, enhanced cancer stem cell features, and accelerated cell mobility, whereas these phenotypes were abrogated in ALK-knockdown cells. Expression of the nervous system-associated gene, ELAVL3, and the corresponding protein (commonly known as HuC) was significantly increased in cells overexpressing ALK. Expression of SRY-box transcription factor (Sox)2 and Sox3 (genes associated with the neural progenitor population) increased in ALK-overexpressing but not ALK-knockdown cells. Furthermore, overexpression of Sox2 or Sox3 enhanced both ALK and ELAVL3 promoter activities, suggesting the existence of ALK/Sox/HuC signaling loops. Finally, ALK overexpression was attributed to increased expression of neuroendocrine markers, including synaptophysin, CD56, and B-cell lymphoma 2, in HGSC tissues. These findings suggest that overexpression of full-length ALK may influence the biological behavior of HGSC through cooperation with ELAVL3 and Sox factors, leading to the establishment and maintenance of the aggressive phenotypic characteristics of HGSC.

Deep Learning for Grading Endometrial Cancer

Endometrial cancer is the fourth most common cancer in women in the United States, with a lifetime risk of approximately 2.8%. Precise histologic evaluation and molecular classification of endometrial cancer are important for effective patient management and determining the best treatment options. This study introduces EndoNet, which uses convolutional neural networks for extracting histologic features and a vision transformer for aggregating these features and classifying slides into high- and low-grade cases. The model was trained on 929 digitized hematoxylin and eosin-stained whole-slide images of endometrial cancer from hysterectomy cases at Dartmouth-Health. It classifies these slides into low-grade (endometrioid grades 1 and 2) and high-grade (endometrioid carcinoma International Federation of Gynecology and Obstetrics grade 3, uterine serous carcinoma, or carcinosarcoma) categories. EndoNet was evaluated on an internal test set of 110 patients and an external test set of 100 patients from The Cancer Genome Atlas database. The model achieved a weighted average F1 score of 0.91 (95% CI, 0.86 to 0.95) and an area under the curve of 0.95 (95% CI, 0.89 to 0.99) on the internal test, and 0.86 (95% CI, 0.80 to 0.94) for F1 score and 0.86 (95% CI, 0.75 to 0.93) for area under the curve on the external test. Pending further validation, EndoNet has the potential to support pathologists without the need of manual annotations in classifying the grades of gynecologic pathology tumors.

Clonal Origin and Lineage Ambiguity in Mixed Neuroendocrine Carcinoma of the Uterine Cervix

Small-cell neuroendocrine carcinoma (SCNEC) of the cervix is a rare disease characterized by a high incidence of mixed tumors with other types of cancer. The mechanism underlying this mixed phenotype is not well understood. This study established a panel of organoid lines from patients with SCNEC of the cervix and ultimately focused on one line, which retained a mixed tumor phenotype, both in vitro and in vivo. Histologically, both organoids and xenograft tumors showed distinct differentiation into either SCNEC or adenocarcinoma in some regions and ambiguous differentiation in others. Tracking single cells indicated the existence of cells with bipotential differentiation toward SCNEC and adenocarcinomas. Single-cell transcriptional analysis identified three distinct clusters: SCNEC-like, adenocarcinoma-like, and a cluster lacking specific differentiation markers. The expression of neuroendocrine markers was enriched in the SCNEC-like cluster but not exclusively. Human papillomavirus 18 E6 was enriched in the SCNEC-like cluster, which showed higher proliferation and lower levels of the p53 pathway. After treatment with anticancer drugs, the expression of adenocarcinoma markers increased, whereas that of SCNEC decreased. Using a reporter system for keratin 19 expression, changes in the differentiation of each cell were shown to be associated with the shift in differentiation induced by drug treatment. These data suggest that mixed SCNEC/cervical tumors have a clonal origin and are characterized by an ambiguous and flexible differentiation state.

Estrogen Inhibits Epithelial Progesterone Receptor–Dependent Progestin Therapy Efficacy in a Mouse Model of Cervical Cancer

Although the uterine cervix responds to the female sex hormone change, the role of progesterone in cervical cancer is poorly understood. It has been shown that medroxyprogesterone acetate (MPA) regresses cervical cancer in the transgenic mouse model expressing human papillomavirus type 16 E6 and E7 oncogenes. As MPA interacts most strongly with progesterone receptor (PR), we reasoned that PR would contribute to MPA-induced regression of cervical cancer. We also hypothesized that estrogen influences the therapeutic activity of MPA because it promotes cervical cancer growth in the same mouse model. The present study showed that the deletion of Pgr in the cervical cancer cells ablated the MPA's therapeutic effect in the human papillomavirus transgenic mouse model. Additionally, estrogen attenuated cancer regression by MPA in the same model system. These observations indicate that MPA can effectively regress cervical cancer only when cancer cells express PR and estrogen levels are low. These results suggest that, if translatable, MPA should be administered when estrogen levels are low in patients with PR-positive cervical cancer.

Ovarian Cancer Detection in Ascites Cytology with Weakly Supervised Model on Nationwide Data Set

Conventional ascitic fluid cytology for detecting ovarian cancer is limited by its low sensitivity. To address this issue, this multicenter study developed patch image (PI)-based fully supervised convolutional neural network (CNN) models and clustering-constrained attention multiple-instance learning (CLAM) algorithms for detecting ovarian cancer using ascitic fluid cytology. Whole-slide images (WSIs), 356 benign and 147 cancer, were collected, from which 14,699 benign and 8025 cancer PIs were extracted. Additionally, 131 WSIs (44 benign and 87 cancer) were used for external validation. Six CNN algorithms were developed for cancer detection using PIs. Subsequently, two CLAM algorithms, single branch (CLAM-SB) and multiple branch (CLAM-MB), were developed. ResNet50 demonstrated the best performance, achieving an accuracy of 0.973. The performance when interpreting internal WSIs was an area under the curve (AUC) of 0.982. CLAM-SB outperformed CLAM-MB with an AUC of 0.944 for internal WSIs. Notably, in the external test, CLAM-SB exhibited superior performance with an AUC of 0.866 compared with ResNet50's AUC of 0.804. Analysis of the heatmap revealed that cases frequently misinterpreted by AI were easily interpreted by humans, and vice versa. Because AI and humans were found to function complementarily, implementing computer-aided diagnosis is expected to significantly enhance diagnostic accuracy and reproducibility. Furthermore, the WSI-based learning in CLAM, eliminating the need for patch-by-patch annotation, offers an advantage over the CNN model.

Artificial Intelligence-Based Histopathological Subtyping of High-Grade Serous Ovarian Cancer

Four subtypes of ovarian high-grade serous carcinoma (HGSC) have previously been identified, each with different prognoses and drug sensitivities. However, the accuracy of classification depended on the assessor's experience. This study aimed to develop a universal algorithm for HGSC-subtype classification using deep learning techniques. An artificial intelligence (AI)-based classification algorithm, which replicates the consensus diagnosis of pathologists, was formulated to analyze the morphological patterns and tumor-infiltrating lymphocyte counts for each tile extracted from whole slide images of ovarian HGSC available in The Cancer Genome Atlas (TCGA) data set. The accuracy of the algorithm was determined using the validation set from the Japanese Gynecologic Oncology Group 3022A1 (JGOG3022A1) and Kindai and Kyoto University (Kindai/Kyoto) cohorts. The algorithm classified the four HGSC-subtypes with mean accuracies of 0.933, 0.910, and 0.862 for the TCGA, JGOG3022A1, and Kindai/Kyoto cohorts, respectively. To compare mesenchymal transition (MT) with non-MT groups, overall survival analysis was performed in the TCGA data set. The AI-based prediction of HGSC-subtype classification in TCGA cases showed that the MT group had a worse prognosis than the non-MT group (P = 0.017). Furthermore, Cox proportional hazard regression analysis identified AI-based MT subtype classification prediction as a contributing factor along with residual disease after surgery, stage, and age. In conclusion, a robust AI-based HGSC-subtype classification algorithm was established using virtual slides of ovarian HGSC.

A Deep Learning–Based Assessment Pipeline for Intraepithelial and Stromal Tumor-Infiltrating Lymphocytes in High-Grade Serous Ovarian Carcinoma

Tumor-infiltrating lymphocytes (TILs) are associated with improved survival in patients with epithelial ovarian cancer. However, TIL evaluation has not been used in routine clinical practice because of reproducibility issues. The current study developed two convolutional neural network models to detect TILs and to determine their spatial location in whole slide images, and established a spatial assessment pipeline to objectively quantify intraepithelial and stromal TILs in patients with high-grade serous ovarian carcinoma. The predictions of the established models showed a significant positive correlation with the number of CD8

Multi-Class Segmentation Network Based on Tumor Tissue in Endometrial Cancer Pathology Images

Endometrial cancer has the second highest incidence of malignant tumors in the female reproductive system. Accurate and efficient analysis of endometrial cancer pathology images is one of the important research components of computer-aided diagnosis. However, endometrial cancer pathology images have challenges such as smaller solid tumors, lesion areas varying in morphology, and difficulty distinguishing solid and nonsolid tumors, which would affect the accuracy of subsequent pathologic analyses. An Endometrial Cancer Multi-class Transformer Network (ECMTrans-net) is therefore proposed herein to improve the segmentation accuracy of endometrial cancer pathology images. An ECM-Attention module can sequentially infer attention maps along three separate dimensions (channel, local spatial, and global spatial) and multiply the attention maps and the input feature map for adaptive feature refinement. This approach may solve the problems of the small size of solid tumors and similar characteristics of solid tumors to nonsolid tumors and further improve the accuracy of segmentation of solid tumors. In addition, an ECM-Transformer module is proposed, which can fuse multi-class feature information and dynamically adjust the receptive field, solving the issue of complex tumor features. Experiments on the Solid Tumor Endometrial Cancer Pathological (ST-ECP) data set found that performance of the ECMTrans-net was superior to state-of-the-art image segmentation methods, and the average values of accuracy, Mean Intersection over Union, precision, and Dice coefficients were 0.952, 0.927, 0.931, and 0.901, respectively.

Two Distinct Tumorigenic Processes in Endometrial Endometrioid Adenocarcinoma

Endometrial endometrioid adenocarcinoma (EEA) is conventionally considered to be a single pathologic entity that develops through a hyperplasia-carcinoma sequence under the influence of estrogen. Previously, another EEA subtype was described and proposed to arise directly from normal endometrium. These conventional and de novo subtypes are designated groups 1 and 2, respectively. To identify the molecular mechanisms of these distinct tumorigenic processes, we conducted comprehensive integrated analyses of genomic data with hormonal status for group 1 paired carcinoma and hyperplasia and group 2 carcinoma samples. Although group 1 carcinomas mostly exhibited genomically stable characteristics and the activation of estrogen signaling, group 2 EEAs showed enriched hypermutator and CpG island methylator phenotypes. Pairwise comparisons of hyperplasia and carcinoma, along with time-course analyses of the hyperplasia-carcinoma sequence, revealed the acquisition of driver mutations in the evolutionary process of hyperplasia but not in the transition from hyperplasia to carcinoma. The current study confirms the existence of two different histopathologic programs during EEA development that harbor distinct molecular bases and demonstrates the biological relevance of these differential tumorigenic processes.

A Complex Interplay between Notch Effectors and β-Catenin Signaling in Morular Differentiation of Endometrial Carcinoma Cells

Notch signaling contributes to tissue development and homeostasis, but little is known about its role in morular differentiation of endometrial carcinoma (Em Ca) cells. The current study focused on crosstalk between Notch and β-catenin signaling in Em Ca with morules. Promoters of hairy and enhancer of split 1 (Hes1) and mastermind-like 2 (MAML2) were activated by Notch intracellular domain 1 but not β-catenin, and a positive feedback loop between Hes1 and MAML2 was observed. Immunoreactivities for nuclear β-catenin, Hes1, and MAML2, as well as the interaction between β-catenin and Hes1 or MAML2, were significantly higher in morular lesions compared with surrounding carcinoma in Em Ca. Inhibition of glycogen synthase kinase-3β (GSK-3β) increased expression of total nuclear and cytoplasmic GSK-3β and its phosphorylated forms, as well as Notch intracellular domain 1, Hes1, and active β-catenin. GSK-3β inhibition also decreased proliferation and migration, consistent with the response of cells stably overexpressing Hes1. Finally, the nuclear/cytoplasmic GSK-3β score was significantly higher in morules compared with surrounding carcinoma in Em Ca, and it was positively correlated with nuclear β-catenin, Hes1, and MAML2 scores. This complex interplay between Notch effectors and β-catenin signaling through GSK-3β inhibition contributes to the establishment and maintenance of β-catenin-mediated morular differentiation, which is, in turn, associated with reduced proliferation and inhibition of migration in Em Ca.

A Histopathologic Image Analysis for the Classification of Endocervical Adenocarcinoma Silva Patterns Depend on Weakly Supervised Deep Learning

Twenty-five percent of cervical cancers are classified as endocervical adenocarcinomas (EACs), which comprise a highly heterogeneous group of tumors. A histopathologic risk stratification system known as the Silva pattern system was developed based on morphology. However, accurately classifying such patterns can be challenging. The study objective was to develop a deep learning pipeline (Silva3-AI) that automatically analyzes whole slide image-based histopathologic images and identifies Silva patterns with high accuracy. Initially, a total of 202 patients with EACs and histopathologic slides were obtained from Qilu Hospital of Shandong University for developing and internally testing the Silva3-AI model. Subsequently, an additional 161 patients and slides were collected from seven other medical centers for independent testing. The Silva3-AI model was developed using a vision transformer and recurrent neural network architecture, utilizing multi-magnification patches, and its performance was evaluated based on a class-specific area under the receiver-operating characteristic curve. Silva3-AI achieved a class-specific area under the receiver-operating characteristic curve of 0.947 for Silva A, 0.908 for Silva B, and 0.947 for Silva C on the independent test set. Notably, the performance of Silva3-AI was consistent with that of professional pathologists with 10 years' diagnostic experience. Furthermore, the visualization of prediction heatmaps facilitated the identification of tumor microenvironment heterogeneity, which is known to contribute to variations in Silva patterns.

The Origin of Ovarian Cancer Species and Precancerous Landscape

Unlike other human cancers, in which all primary tumors arise de novo, ovarian epithelial cancers are primarily imported from either endometrial or fallopian tube epithelium. The prevailing paradigm in the genesis of high-grade serous carcinoma (HGSC), the most common ovarian cancer, posits to its development in fallopian tubes through stepwise tumor progression. Recent progress has been made not only in gathering terabytes of omics data but also in detailing the histologic-molecular correlations required for looking into, and making sense of, the tissue origin of HGSC. This emerging paradigm is changing many facets of ovarian cancer research and routine gynecology practice. The precancerous landscape in fallopian tubes contains multiple concurrent precursor lesions, including serous tubal intraepithelial carcinoma (STIC), with genetic heterogeneity providing a platform for HGSC evolution. Mathematical models imply that a prolonged time (decades) elapses from the development of a TP53 mutation, the earliest known molecular alteration, to an STIC, followed by a shorter span (6 years) for progression to an HGSC. Genetic predisposition accelerates the trajectory. This timeline may allow for the early diagnosis of HGSC and STIC, followed by intent-to-cure surgery. This review discusses the recent advances in this tubal paradigm and its biological and clinical implications, alongside the promise and challenge of studying STIC and other precancerous lesions of HGSC.