Aysun Alci

Prediction of Clavien Dindo Classification ≥ Grade III Complications After Epithelial Ovarian Cancer Surgery Using Machine Learning Methods

Background and Objectives: Ovarian cancer surgery requires multiple radical resections with a high risk of complications. The aim of this single-centre, retrospective study was to determine the best method for predicting Clavien–Dindo grade ≥ III complications using machine learning techniques. Material and Methods: The study included 179 patients who underwent surgery at the gynaecological oncology department of Antalya Training and Research Hospital between January 2015 and December 2020. The data were randomly split into training set n = 134 (75%) and test set n = 45 (25%). We used 49 predictors to develop the best algorithm. Mean absolute error, root mean squared error, correlation coefficients, Mathew’s correlation coefficient, and F1 score were used to determine the best performing algorithm. Cohens’ kappa value was evaluated to analyse the consistency of the model with real data. The relationship between these predicted values and the actual values were then summarised using a confusion matrix. True positive (TP) rate, False positive (FP) rate, precision, recall, and Area under the curve (AUC) values were evaluated to demonstrate clinical usability and classification skills. Results: 139 patients (77.65%) had no morbidity or grade I-II CDC morbidity, while 40 patients (22.35%) had grade III or higher CDC morbidity. BayesNet was found to be the most effective prediction model. No dominant parameter was observed in the Bayesian net importance matrix plot. The true positive (TP) rate was 76%, false positive (FP) rate was 15.6%, recall rate (sensitivity) was 76.9%, and overall accuracy was 82.2% A receiver operating characteristic (ROC) analysis was performed to estimate CDC grade ≥ III. AUC was 0.863 with a statistical significance of p < 0.001, indicating a high degree of accuracy. Conclusions: The Bayesian network model achieved the highest accuracy compared to all other models in predicting CDC Grade ≥ III complications following epithelial ovarian cancer surgery.

Evaluation of systemic inflammation- and nutrition-based indices in the prediction of HPV persistence

Persistent high-risk human papillomavirus (HPV) infection is the primary etiological factor in cervical cancer, with HPV16 and HPV18 posing the greatest oncogenic risk. Although systemic inflammation and nutritional indices such as the Hemoglobin, Albumin, Lymphocyte, and Platelet (HALP) score and Prognostic Nutritional Index (PNI) have prognostic value in various malignancies, their role in predicting HPV persistence remains unclear. This study aimed to evaluate the predictive value of HALP and PNI scores for one-year HPV persistence and additionally assessed other inflammatory markers, including neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and C-reactive protein (CRP). This retrospective study included 470 HPV-positive women aged 31-67 years who were followed for at least one year between January 2021 and March 2025. Participants were categorized into Group N (HPV clearance, n = 271) and Group P (HPV persistence, n = 199) according to one-year HPV results. Baseline demographic, clinical, histopathological, and laboratory data were collected. HALP and PNI scores were calculated from hemoglobin, albumin, lymphocyte, and platelet counts, and inflammatory markers including NLR, PLR, and CRP were evaluated. Group comparisons were performed using appropriate statistical tests, and predictive performance was assessed using receiver operating characteristic (ROC) analysis. Continuous variables were standardized using Z-score transformation before multivariate logistic regression to minimize scale differences among predictors. Within a year, 271 (57.7%) of the 470 women achieved viral clearance, whereas 199 (42.3%) had persistent infection. No substantial differences were observed in demographic or histological parameters. Vaccination after diagnosis did not affect clearance rates. The median HALP and PNI scores were similar between groups and provided no discriminatory ability [HALP AUC = 0.475 (95% CI: 0.422-0.528); PNI AUC = 0.487 (95% CI: 0.434-0.540)], as did PLR [AUC = 0.513 (95% CI: 0.460-0.566)] and CRP [AUC = 0.462 (95% CI: 0.409-0.515)]. In contrast, NLR was significantly higher in the persistence group (p < 0.001) and demonstrated modest discriminatory ability [AUC = 0.623 (95% CI: 0.573-0.673)]. HPV16 and HPV18 positivity was strongly associated with persistence. When the researchers evaluated the subjects' systemic inflammatory and nutritional status, they found that HALP, PNI, PLR, and CRP did not provide effective predictive value for one-year HPV persistence. Conversely, NLR exhibited a promising capacity to function as a straightforward systemic inflammatory marker. The presence of HPV16 and HPV18 was found to be significantly associated with the persistence of the infection, thereby corroborating the established immune evasion mechanisms associated with these types. These findings underscore the necessity for prognostic markers that integrate both local mucosal immune responses and systemic inflammatory pathways to enhance risk stratification and management of HPV-related disease.

Does HPV‐18 co‐infection increase the risk of cervical pathology in individuals with HPV‐16?

Abstract Objective We aimed to investigate differences between HPV‐16 mono‐ and HPV‐16/18 co‐infections in terms of cervical dysplasia and invasive cancer. Methods This multicentre, retrospective study spanned from December 2017 to December 2020, involving women who visited gynaecological oncology clinics for colposcopy with either HPV‐16 or HPV‐16/18 positivity. A total of 736 patients, 670 in Group 1 (HPV‐16 positivity) and 66 in Group 2 (HPV‐16/18 positivity), were compared for the presence of CIN2+ lesions detected by colposcopic biopsy or endocervical curettage (ECC). Exclusions included hysterectomized patients, those with prior gynaecological cancers, and patients with HPV positivity other than types 16 and 18. Results Among the included patients, 42.4% had a diagnosis of CIN2+ lesions. The cytology results demonstrated abnormal findings in 45.3% in Group 1 and 42.2% in Group 2, with no significant difference between the groups. ECC revealed CIN2+ lesion in 49 (8.7%) patients in group 1, while only 1 (1.7%) patient had CIN2+ lesion in group 2. There was no difference between 2 groups in terms of ECC result ( p  = 0.052). In group 1, 289 (43.1%) patients had CIN2+ lesion, while 23 (34.8%) patients had CIN2+ lesions in group 2. There was no difference between group 1 and 2 in terms of diagnosis of CIN2+ lesions ( p  = 0.19). Conclusion This multicentre retrospective study found no significant differences between HPV‐16 mono‐ and HPV‐16/18 co‐infections regarding cervical pathologies. Larger studies are needed to validate and further explore these findings.

Defining the relationship between ovarian adult granulosa cell tumors and synchronous endometrial pathology: Does ovarian tumor size correlate with endometrial cancer?

Abstract Objective The main feature of adult granulosa cell tumors (AGCT) is their capacity to secrete hormones, with nearly all of them capable of synthesizing oestradiol. The primary goal of this study is to identify synchronized endometrial pathologies, particularly endometrial cancer, in AGCT patients who had undergone a hysterectomy. Materials and Methods The study cohort comprised retrospectively of 316 AGCT patients from 10 tertiary gynecological oncology centers. AGCT surgery consisted of bilateral salpingo‐oophorectomy, hysterectomy, peritoneal cytology, omentectomy, and the excision of any suspicious lesion. The median tumor size value was used to define the relationship between tumor size and endometrial cancer. The relationship between each value and endometrial cancer was evaluated. Results Endometrial intraepithelial neoplasia, or hyperplasia with complex atypia, was detected in 7.3% of patients, and endometrial cancer in 3.1% of patients. Age, menopausal status, tumor size, International Federation of Gynecology and Obstetrics stage, ascites, and CA‐125 level were not statistically significant factors to predict endometrial cancer. There was no endometrial cancer under the age of 40, and 97.8% of women diagnosed with endometrial hyperplasia were over the age of 40. During the menopausal period, the endometrial cancer risk was 4.5%. Developing endometrial cancer increased to 12.1% from 3.2% when the size of the tumor was &gt;150 mm in menopausal patients ( p  = 0.036). Conclusion Endometrial hyperplasia, or cancer, occurs in approximately 30% of AGCT patients. Patients diagnosed with AGCT, especially those older than 40 years, should be evaluated for endometrial pathologies. There may be a relationship between tumor size and endometrial cancer, especially in menopausal patients.