Yuting Wang

AI-Based Identification Method for Cervical Transformation Zone Within Digital Colposcopy: Development and Multicenter Validation Study

Background In low- and middle-income countries, cervical cancer remains a leading cause of death and morbidity for women. Early detection and treatment of precancerous lesions are critical in cervical cancer prevention, and colposcopy is a primary diagnostic tool for identifying cervical lesions and guiding biopsies. The transformation zone (TZ) is where a stratified squamous epithelium develops from the metaplasia of simple columnar epithelium and is the most common site of precancerous lesions. However, inexperienced colposcopists may find it challenging to accurately identify the type and location of the TZ during a colposcopy examination. Objective This study aims to present an artificial intelligence (AI) method for identifying the TZ to enhance colposcopy examination and evaluate its potential clinical application. Methods The study retrospectively collected data from 3616 women who underwent colposcopy at 6 tertiary hospitals in China between 2019 and 2021. A dataset from 4 hospitals was collected for model conduction. An independent dataset was collected from the other 2 geographic hospitals to validate model performance. There is no overlap between the training and validation datasets. Anonymized digital records, including each colposcopy image, baseline clinical characteristics, colposcopic findings, and pathological outcomes, were collected. The classification model was proposed as a lightweight neural network with multiscale feature enhancement capabilities and designed to classify the 3 types of TZ. The pretrained FastSAM model was first implemented to identify the location of the new squamocolumnar junction for segmenting the TZ. Overall accuracy, average precision, and recall were evaluated for the classification and segmentation models. The classification performance on the external validation was assessed by sensitivity and specificity. Results The optimal TZ classification model performed with 83.97% classification accuracy on the test set, which achieved average precision of 91.84%, 89.06%, and 95.62% for types 1, 2, and 3, respectively. The recall and mean average precision of the TZ segmentation model were 0.78 and 0.75, respectively. The proposed model demonstrated outstanding performance in predicting 3 types of the TZ, achieving the sensitivity with 95% CIs for TZ1, TZ2, and TZ3 of 0.78 (0.74-0.81), 0.81 (0.78-0.82), and 0.8 (0.74-0.87), respectively, with specificity with 95% CIs of 0.94 (0.92-0.96), 0.83 (0.81-0.86), and 0.91 (0.89-0.92), based on a comprehensive external dataset of 1335 cases from 2 of the 6 hospitals. Conclusions Our proposed AI-based identification system classified the type of cervical TZs and delineated their location on multicenter, colposcopic, high-resolution images. The findings of this study have shown its potential to predict TZ types and specific regions accurately. It was developed as a valuable assistant to encourage precise colposcopic examination in clinical practice.

Serum lipid profiling analysis and potential marker discovery for ovarian cancer based on liquid chromatography–Mass spectrometry

Low early diagnosis rate and unclear pathogenesis are the primary reasons for the high mortality of epithelial ovarian cancer (EOC). Lipidomics is a powerful tool for marker discovery and mechanism explanation. Hence, a ultra high-performance liquid chromatography-mass spectrometry based non-targeted lipidomics analysis was performed to acquire lipid profiling of 153 serum samples including healthy control (HC, n = 50), benign ovarian tumor (BOT, n = 41), and EOC (n = 62) to reveal lipid disturbance, then differential lipids were verified in another sample set including 187 sera. Significant lipid disturbance occurred in BOT and EOC, fatty acid, lyso-phosphatidylcholine, and lyso-phosphatidylethanolamine were observed to be increased in BOT and EOC subjects, while phosphatidylcoline, ether phosphatidylcoline (PC-O), ether phosphatidylethanolamine (PE-O), and sphingomyelin significantly decreased. Compared with BOT, PC-Os and PE-Os presented a greater reduction in EOC, and serum ceramide increased only in EOC. Moreover, potential markers consisting of 4 lipids were defined and validated for EOC diagnosis. High areas under the curve (0.854∼0.865 and 0.903∼0.923 for distinguishing EOC and early EOC from non-cancer, respectively) as well as good specificity and sensitivity were obtained. This study not only revealed the characteristics of lipid metabolism in EOC, but also provided a potential marker pattern for aiding EOC diagnosis.

Glutaminase Inhibitors Induce Thiol-Mediated Oxidative Stress and Radiosensitization in Treatment-Resistant Cervical Cancers

Abstract The purpose of this study was to determine if radiation (RT)-resistant cervical cancers are dependent upon glutamine metabolism driven by activation of the PI3K pathway and test whether PI3K pathway mutation predicts radiosensitization by inhibition of glutamine metabolism. Cervical cancer cell lines with and without PI3K pathway mutations, including SiHa and SiHa PTEN−/− cells engineered by CRISPR/Cas9, were used for mechanistic studies performed in vitro in the presence and absence of glutamine starvation and the glutaminase inhibitor, telaglenastat (CB-839). These studies included cell survival, proliferation, quantification of oxidative stress parameters, metabolic tracing with stable isotope-labeled substrates, metabolic rescue, and combination studies with L-buthionine sulfoximine (BSO), auranofin (AUR), and RT. In vivo studies of telaglenastat ± RT were performed using CaSki and SiHa xenografts grown in immune-compromised mice. PI3K-activated cervical cancer cells were selectively sensitive to glutamine deprivation through a mechanism that included thiol-mediated oxidative stress. Telaglenastat treatment decreased total glutathione pools, increased the percent glutathione disulfide, and caused clonogenic cell killing that was reversed by treatment with the thiol antioxidant, N-acetylcysteine. Telaglenastat also sensitized cells to killing by glutathione depletion with BSO, thioredoxin reductase inhibition with AUR, and RT. Glutamine-dependent PI3K-activated cervical cancer xenografts were sensitive to telaglenastat monotherapy, and telaglenastat selectively radiosensitized cervical cancer cells in vitro and in vivo. These novel preclinical data support the utility of telaglenastat for glutamine-dependent radioresistant cervical cancers and demonstrate that PI3K pathway mutations may be used as a predictive biomarker for telaglenastat sensitivity.