Ovarian cancer is a leading cause of deaths among gynecological cancers, and a method to detect early-stage epithelial ovarian cancer (EOC) is urgently needed. We aimed to develop an artificial intelligence (AI)-based comprehensive serum glycopeptide spectra analysis (CSGSA-AI) method in combination with convolutional neural network (CNN) to detect aberrant glycans in serum samples of patients with EOC. We converted serum glycopeptide expression patterns into two-dimensional (2D) barcodes to let CNN learn and distinguish between EOC and non-EOC. CNN was trained using 60% samples and validated using 40% samples. We observed that principal component analysis-based alignment of glycopeptides to generate 2D barcodes significantly increased the diagnostic accuracy (88%) of the method. When CNN was trained with 2D barcodes colored on the basis of serum levels of CA125 and HE4, a diagnostic accuracy of 95% was achieved. We believe that this simple and low-cost method will increase the detection of EOC.

Shigeta S, et al. J Gynecol Oncol. 2017 Aug;28(6):e76. https://doi.org/10.3802/jgo.2017.28.e76

Ikeda M, et al. J Gynecol Oncol. 2021 May;32:e68. https://doi.org/10.3802/jgo.2021.32.e68

With increased screening, more patients with precancerous or early cervical cancer are now being identified. Age at pregnancy, and thus number of patients requiring fertility preservation, have also increased, resulting in more diagnostic and therapeutic cervical conization (conization) procedures. We here investigated the pathological and clinical characteristics of patients undergoing conization, with a focus on age. The objectives of our study were to identify the risk factors potentially involved in cervical intraepithelial neoplasia (CIN) recurrence or persistence, additional treatment after conization, the effects of conization on pregnancy, and the actual status of conization in Japan. A “Subcommittee for Investigation of Cervical Conization” within the Gynecologic Oncology Committee in the Japan Society of Obstetrics and Gynecology investigated pathological and clinical characteristics of conization at 205 institutions in Japan. We analyzed pathological and clinical characteristics according to age ≤50 and >50 years. Patients aged 20–40 years accounted for 12904 (87%) of the 14,832 study patients (median: 37 years, range: 16–88 years). However, 1838 (12.4%) were aged >50 years. The commonest post-operative diagnosis was CIN grade 3 in all age groups. Rates of invasive cancer, post-operative detection of more advanced lesions, positive surgical margins, additional treatment, and recurrence were significantly higher in patients aged ≥50 years than those aged <50 years (all p < 0.01), whereas rates of post-operative complications did not differ significantly between age groups. The relationship between cerclage and the incidence of amniorrhexis or premature birth did not differ significantly in any age group. Post-conization management of patients aged ≥50 years requires considering the high rates of detection of more advanced lesions post-operatively, positive surgical margins, and recurrence. Cerclage should not be performed post-conization without careful consideration.

Objectives: To conduct a comprehensive glycopeptide spectra analysis of serum between cancer and non-cancer patients to identify early biomarkers of epithelial ovarian cancer (EOC). Methods: Approximately 30,000 glycopeptide peaks were detected from the digested serum glycoproteins of 39 EOC patients (23 early-stage, 16 advanced-stage) and 45 non-cancer patients (27 leiomyoma and ovarian cyst cases, 18 endometrioma cases) by liquid chromatography mass spectrometry (LC–MS). The differential glycopeptide peak spectra were analyzed to distinguish between cancer and non-cancer groups by employing multivariate analysis including principal component analysis (PCA), orthogonal partial least squares discriminant analysis (OPLS-DA) and heat maps. Results: Examined spectral peaks were filtered down to 2281 serum quantitative glycopeptide signatures for differentiation between ovarian cancer and controls using multivariate analysis. The OPLS-DA model using cross-validation parameters R2 and Q2 and score plots of the serum samples significantly differentiated the EOC group from the non-cancer control group. In addition, women with early-stage clear cell carcinoma and endometriomas were clearly distinguished from each other by OPLS-DA as well as by PCA and heat maps. Conclusions: Our study demonstrates the potential of comprehensive serum glycoprotein analysis as a useful tool for ovarian cancer detection.

Ikeda M, et al. J Gynecol Oncol. 2020 Oct;32:e44. https://doi.org/10.3802/jgo.2021.32.e44