Explainable machine learning model for predicting skeletal muscle loss during surgery and adjuvant chemotherapy in ovarian cancer.

BACKGROUND: Skeletal muscle loss during treatment is associated with poor survival outcomes in patients with ovarian cancer. Although changes in muscle mass can be assessed on computed tomography (CT) scans, this labour-intensive process can impair its utility in clinical practice. This study aimed to develop a machine learning (ML) model to predict muscle loss based on clinical data and to interpret the ML model by applying SHapley Additive exPlanations (SHAP) method. METHODS: This study included the data of 617 patients with ovarian cancer who underwent primary debulking surgery and platinum-based chemotherapy at a tertiary centre between 2010 and 2019. The cohort data were split into training and test sets based on the treatment time. External validation was performed using 140 patients from a different tertiary centre. The skeletal muscle index (SMI) was measured from pre- and post-treatment CT scans, and a decrease in SMI ≥ 5% was defined as muscle loss. We evaluated five ML models to predict muscle loss, and their performance was determined using the area under the receiver operating characteristic curve (AUC) and F1 score. The features for analysis included demographic and disease-specific characteristics and relative changes in body mass index (BMI), albumin, neutrophil-to-lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR). The SHAP method was applied to determine the importance of the features and interpret the ML models. RESULTS: The median (inter-quartile range) age of the cohort was 52 (46-59) years. After treatment, 204 patients (33.1%) experienced muscle loss in the training and test datasets, while 44 (31.4%) patients experienced muscle loss in the external validation dataset. Among the five evaluated ML models, the random forest model achieved the highest AUC (0.856, 95% confidence interval: 0.854-0.859) and F1 score (0.726, 95% confidence interval: 0.722-0.730). In the external validation, the random forest model outperformed all ML models with an AUC of 0.874 and an F1 score of 0.741. The results of the SHAP method showed that the albumin change, BMI change, malignant ascites, NLR change, and PLR change were the most important factors in muscle loss. At the patient level, SHAP force plots demonstrated insightful interpretation of our random forest model to predict muscle loss. CONCLUSIONS: Explainable ML model was developed using clinical data to identify patients experiencing muscle loss after treatment and provide information of feature contribution. Using the SHAP method, clinicians may better understand the contributors to muscle loss and target interventions to counteract muscle loss.

Timosaponin AIII Inhibits Migration and Invasion Abilities in Human Cervical Cancer Cells through Inactivation of p38 MAPK-Mediated uPA Expression In Vitro and In Vivo.

Cervical cancer is one of the most common gynecologic cancers globally that require novel approaches. Timosaponin AIII (TSAIII) is a steroidal saponin that displays beneficial effects in antitumor activities. However, the effect of TSAIII on human cervical cancer remains unknown. In this study, we found that TSAIII showed no influence on cell viability, cytotoxicity, cell cycle distribution and apoptosis induction in human cervical cancer cells. TSAIII was revealed to have a significant inhibitory effect on cell migration and invasion through the downregulation of invasion-related uPA expression and p38 MAPK activation in both human cervical cancer cells and cervical cancer stem cells (CCSCs), indicating that the p38 MAPK-uPA axis mediated the TSAIII-inhibited capacity of cellular migration and invasion. In a synergistic inhibition assay, a TSAIII plus p38 siRNA cotreatment revealed a greater inhibition of uPA expression, migration and invasion in human cervical cancer cells. In an immunodeficient mouse model, TSAIII significantly inhibited lung metastases from human cervical cancer SiHa cells without TSAIII-induced toxicity. These findings first revealed the inhibitory effects of TSAIII on the progression of human cervical cancer through its downregulation of p38 MAPK-uPA axis activation. Therefore, TSAIII might provide a potential strategy for auxiliary therapy in human cervical cancer.

α-Mangostin attenuates stemness and enhances cisplatin-induced cell death in cervical cancer stem-like cells through induction of mitochondrial-mediated apoptosis.

Cancer stem cells (CSCs) exhibit specific characteristics including decontrolled self-renewal, tumor-initiating, promoting, and metastatic potential, abnormal stemness signaling, and chemotherapy resistance. Thus, targeting CSC is becoming an emerging cancer treatment. α-Mangostin has been shown to have potent and multiple anticancer activities. Accordingly, we hypothesized that α-mangostin may diminish the stemness and proliferation of CSC-like cervical cancer cells. In our results, comparing to the parent cells, CSC-like SiHa and HeLa cells highly expressed CSC marker Sox2, Oct4, Nanog, CK-17, and CD49f. α-Mangostin significantly reduced the cell viability, sphere-forming ability, and expression of the CSC stemness makers of CSC-like cervical cancer cells. Further investigation showed that α-mangostin induced mitochondrial depolarization and mitochondrial apoptosis signaling, including upregulation of Bax, downregulation of Mcl-1 and Bcl-2, and activation of caspase-9/3. Moreover, α-mangostin synergically enhanced the cytotoxicity of cisplatin on CSC-like SiHa cells by promoting mitochondrial apoptosis and inhibiting the expression of CSC markers. Consistent with in vitro findings, in vivo tumor growth assay revealed that α-mangostin administration significantly inhibited the growth of inoculated CSC-like SiHa cells and synergically enhanced the antitumor effect of cisplatin. Our findings indicate that α-mangostin can reduce the stemness and proliferation of CSC-like SiHa and HeLa cells and promote the cytotoxicity of cisplatin, which may attribute to the mitochondrial apoptosis activation. Thus, it suggests that α-mangostin may have clinical potential to improve chemotherapy for cervical cancer by targeting cervical CSC.

Possible surrogate marker for an effective dose-dense chemotherapy in treating ovarian cancer.

OBJECTIVE: To dissect the correlated hematologic markers that reflect the clinical outcome or treatment response in patients receiving dose-dense chemotherapy with a combination of platinum (cisplatin or carboplatin) and paclitaxel. MATERIALS AND METHODS: From 2009 to 2014, we enrolled 55 ovarian cancer patients (total 67 courses) including first-line, persistent, platinum-sensitive, or platinum-resistant disease in MacKay Memorial Hospital, Taipei, Taiwan. Weekly pretreatment complete blood counts and calculated ratios [platelet/neutrophil ratio, neutrophil/lymphocyte ratio (NLR), platelet/lymphocyte ratio (PLR), neutrophil/monocyte ratio, platelet/monocyte ratio, lymphocyte/monocyte ratio] during dose-dense chemotherapy were collected. By grouping these hematologic biomarkers into three different response subgroups (responsive, stable, and nonresponsive) according to CA125 trend, the data were analyzed using one-way analysis of variance, and using post hoc-Tukey test for comparing each other. A p value < 0.05 was considered to be statistically significant. RESULTS: Absolute counts of lymphocytes and platelets, PLR, platelet/neutrophil ratio, platelet/monocyte ratio (all p < 0.001), and NLR (p=0.013) had statistically significant differences. Moreover, using box-and-whisker plot, absolute count of lymphocyte, PLR, and NLR showed most remarkable discrepancy in responsive, stable, and nonresponsive patients. Subgroup analysis for primary, platinum-sensitive, and platinum-resistant patients further revealed that PLR and NLR were significantly correlated to the outcome of dose-dense chemotherapy. CONCLUSION: Lower PLR or lower NLR had better treatment response for dose-dense chemotherapy and are possible markers for representing treatment response in dose-dense chemotherapy. For a clinician, this is useful for timing when to switch to another chemotherapy regimen.

Paraneoplastic cerebellar degeneration in a patient with ovarian cancer.

OBJECTIVE: Herein, we report a case of ovarian serous carcinoma with paraneoplastic cerebellar degeneration. CASE REPORT: A 44-year-old female presented to our hospital with dizziness, slurred speech, and ataxic gait. Brain magnetic resonance imaging was normal. A lumbar puncture revealed a normal cell count in the cerebrospinal fluid, but slightly elevated protein. Her serum cancer antigen -125 level was high (2126.4 U/mL), and abdominal computed tomography disclosed a pelvic mass measuring 11 cm in diameter. Exploratory laparotomy was then performed, and a frozen section of the tumor revealed serous carcinoma. CONCLUSION: According to the surgical findings and pathological report, The International Federation of Gynecology and Obstetrics (FIGO) Stage IIIC, Grade 3, serous-type ovarian cancer was diagnosed. Due to the abovementioned symptoms and signs, we performed a serial test to document the presence of anti-Yo antibody in this patient.