A toolkit for a modern gynecologic oncology tissue bank.

OBJECTIVES: Tissue banking procedures have evolved to keep pace with precision medicine, technology, emerging understanding of racial disparities, and regulatory requirements. However, there is little published guidance regarding strategies to create and maintain a successful biorepository. Our objective is to describe the infrastructure and protocols used by our Gynecologic Oncology Tissue Bank. METHODS: Our Tissue Bank was founded in 1992. In August 2022, internal funding was used to modernize the Tissue Bank. We hired three full-time employees, implemented universal screening of patients treated by gynecologic oncology faculty, updated consenting protocols, and standardized communication with providers. Tumor tissue, blood derivatives, ascites, and pleural fluid were collected from eligible, consenting patients and processed. Patient-derived cell lines and organoids were generated. For quality control purposes, one formalin-fixed, paraffin-embedded (FFPE) sample per tissue site was analyzed by a board-certified pathologist. All samples were labeled and tracked in an OpenSpecimen collection protocol and clinically annotated in a secure database. RESULTS: From August 2022 to October 2023, 227 patients (83% white, 15% Black, 1% Asian) were enrolled and 4249 specimens were collected. Adherent cell lines were generated from 15 patients with ovarian cancer and cell suspensions for organoid generation were collected from 46 patients with ovarian cancer. A recharge center was established to self-sustain the Tissue Bank. Samples have been shared with academic and commercial collaborators. CONCLUSIONS: Our Tissue Bank has enrolled a large number of diverse patients, collected numerous specimen types, and collaborated widely. The procedures described here provide guidance for other institutions establishing similar resources.

Selective breast/gynecologic pathology fellowship training in the United States: Experience of program directors.

Published data on combined breast and gynecologic [breast/gyn] surgical pathology fellowship training programs are limited. Our study aimed to survey the landscape of such fellowships in the United States (US), including specific information about their characteristics and the educational activities therein. Using web searches, we identified programs offering combined breast/gyn surgical pathology fellowship training. We developed a 26-item questionnaire asking program directors to report on the characteristics of their fellowship training structure. The search revealed 25 academic based programs offering one-year combined breast/gyn fellowship training, predominantly located (40 %) in the Northeast area. The following data was obtained: 44 % of the programs were accredited by the ACGME, 82 % required >19 weeks of breast and gyn service, and 69.6 % accepted the common application, 54.5 % of programs require completion of a research project for graduation. An annual average of 3000 breast and 3000 gyn cases appears to be the usual volume of cases. Interestingly, only 36 % of the program directors are graduates of a combined breast/gyn fellowship program. In conclusion, we present the most comprehensive and up-to-date census of combined breast/gyn pathology fellowships in the US. Our study provides valuable information on the current state of combined breast/gyn pathology fellowship training. The information will be helpful to current and prospective trainees, as well as program leaders.

End-to-end deep learning method for predicting hormonal treatment response in women with atypical endometrial hyperplasia or endometrial cancer.

Purpose: Endometrial cancer (EC) is the most common gynecologic malignancy in the United States, and atypical endometrial hyperplasia (AEH) is considered a high-risk precursor to EC. Hormone therapies and hysterectomy are practical treatment options for AEH and early-stage EC. Some patients prefer hormone therapies for reasons such as fertility preservation or being poor surgical candidates. However, accurate prediction of an individual patient's response to hormonal treatment would allow for personalized and potentially improved recommendations for these conditions. This study aims to explore the feasibility of using deep learning models on whole slide images (WSI) of endometrial tissue samples to predict the patient's response to hormonal treatment. Approach: We curated a clinical WSI dataset of 112 patients from two clinical sites. An expert pathologist annotated these images by outlining AEH/EC regions. We developed an end-to-end machine learning model with mixed supervision. The model is based on image patches extracted from pathologist-annotated AEH/EC regions. Either an unsupervised deep learning architecture (Autoencoder or ResNet50), or non-deep learning (radiomics feature extraction) is used to embed the images into a low-dimensional space, followed by fully connected layers for binary prediction, which was trained with binary responder/non-responder labels established by pathologists. We used stratified sampling to partition the dataset into a development set and a test set for internal validation of the performance of our models. Results: The autoencoder model yielded an AUROC of 0.80 with 95% CI [0.63, 0.95] on the independent test set for the task of predicting a patient with AEH/EC as a responder vs non-responder to hormonal treatment. Conclusions: These findings demonstrate the potential of using mixed supervised machine learning models on WSIs for predicting the response to hormonal treatment in AEH/EC patients.

Malignant Phyllodes Tumor: Imaging Features With Histopathologic Correlation.

Phyllodes tumors (PT) are rare fibroepithelial lesions of the breast that commonly present as rapidly enlarging, palpable masses. Phyllodes tumors may be classified as benign, borderline, or malignant on the basis of histopathologic analysis. Although malignant PT cannot be distinguished from benign PT on the basis of imaging findings alone, studies suggest that malignant PT tend to be larger and irregular in shape, and they are less likely to have circumscribed margins. If biopsy results are indeterminate, excisional biopsy should be performed. Malignant PT can be difficult to distinguish histologically from sarcomas and spindle cell metaplastic breast carcinoma; the distinction is important for prognosis and treatment. Malignant PT are treated surgically with wide local excision, without a clear role for adjuvant radiation or chemotherapy in most cases. Nearly one-third of malignant PT recur locally, usually within a few years after initial diagnosis. Distant metastatic disease is rare, and the five-year overall survival rate of malignant PT is close to 80%. The purpose of this article is to review the clinical presentation, imaging appearance, histopathology, and management of malignant PT.