Search

Mo Tang,1 Lihao Gao,2 Bin He,1 Yufei Yang1 1Oncology Department, Xiyuan Hospital of China Academy of Chinese Medical Sciences, Beijing, People’s Republic of China; 2Smart City Business Unit, Baidu Inc., Beijing, People’s Republic of ChinaCorrespondence: Yufei YangXiyuan Hospital of China Academy of Chinese Medical Sciences, No. 1 Xiyuan Caochang Road, Haidian District, Beijing, 100091, People’s Republic of ChinaEmail yyf93@vip.sina.comLihao GaoSmart City Business Unit, Baidu Inc., No. 51 Dezhen Road, Haidian District, Beijing, 100091, People’s Republic of ChinaEmail gaolihao@baidu.comPurpose: The present study aimed to develop prognostic prediction models based on machine learning (ML) for non-metastatic colon cancer (CRC), which can provide a precise quantitative risk assessment and serve as an assistive method for treatment strategy development. The possibility of improving prediction accuracy using nonlinear methods compared to linear methods was investigated.Patients and Methods: A cancer-specific survival (CSS) model constructed using logistic regression, extreme gradient boosting (XGBoost), and random forest algorithms was trained on the Surveillance, Epidemiology, and End Results datasets for 15,254 patients with non-metastatic CRC (split into training [70%] and internal validation [30%] datasets) and externally validated with an outpatient cohort of 311 cases from Xiyuan Hospital in China. A Chinese cohort was also used to develop recurrence and metastasis (R&M) models for CRC patients. The experiments for each model were performed 100 times to obtain average scores and 95% confidence intervals. The model performance was evaluated using the area under the receiver operating characteristic curve (AUC) values.Results: The XGBoost approach showed the highest AUC values of 0.86 (0.84– 0.88), 0.82 (0.81– 0.83), and 0.81 (0.79– 0.82) for one-, three-, and five-year CSS cohorts, respectively, along with a relatively high generalization ability. The XGBoost approach also performed best for the R&M model, with the AUC values of 0.71 (0.64– 0.79), 0.79 (0.74– 0.86), and 0.89 (0.82– 0.95) for one-, three-, and five-year R&M cohorts, respectively. The rankings of predictor importance for the CSS and R&M models were different, and the higher model accuracy was associated with more prognostic predictors.Conclusion: Three different ML algorithms for developing prognostic prediction models for non-metastatic CRC were compared. The predictive performance results showed that the nonlinear XGBoost approach performed best, suggesting that it can be used for quantifying the prognostic risk. It was also demonstrated that the model performance can be improved when more prognostic predictors are considered.Keywords: colon cancer, machine learning, extreme gradient boosting, prognostic prediction models more...

Background CD109 was involved in the tumorigenesis and progression of various cancers via TGF-β1 signalling and STAT3 activation. As CD109 is strongly expressed in cervical squamous cell carcinoma, this study was conducted to investigate its functional characteristics in cervical cancer. Methods CD109 expression was examined by immunohistochemistry (IHC) with cervical tissue microarray. The effects of CD109 expression were examined on migration, cell proliferation, spheroid formation and soft-agar colony-formation assay. Meanwhile, cervical cancer cell lines with high CD109 expression were chosen for the functional study using siRNA knockdown and CRISPR/Cas9 knockout. Results IHC demonstrated an upregulation of CD109 in the cell membrane of cervical squamous cell carcinoma. CD109( + ) cells isolated by flow-cytometric sorting displayed enhanced migration, cell proliferation, sphere-forming and anchorage-independent cell growth ability. In contrast, silencing of CD109 expression could reverse the in vitro and in vivo tumorigenic and aggressive properties. Furthermore, CD109 induced EGFR-mediated STAT3 phosphorylation known to be responsible for cell migration, proliferation and maintenance of CSC phenotype. Conclusion Abundant CD109( + ) populations in cervical cancer cells potentially contributed to carcinogenesis and aggressiveness, whereas silencing of CD109 expression could reverse those properties. CD109 mediates cervical tumorigenicity and aggressiveness via CD109/EGFR/STAT3 signalling. more...

Background Ovarian cancer is characterised by frequent recurrence due to persistent presence of residual cancer stem cells (CSCs). Here, we identify and characterise tumour subsets from ascites-derived tumour cells with stemness, metastasis and metabolic switch properties and to delineate the involvement of pyruvate dehydrogenase kinase 4 (PDK4) in such process. Methods Ovarian cancer cells/cell lines derived from ascites were used for tumourspheres/ALDH+CD44+ subset isolation. The functional roles and downstream signalling of PDK4 were explored. Its association with clinical outcome of ovarian cancer was analysed. Results We demonstrated enhanced CSC characteristics of tumour cells derived from ovarian cancer ascites, concomitant with ALDH and CD44 subset enrichment and high PDK4 expression, compared to primary tumours. We further showed tumourspheres/ALDH+CD44+ subsets from ascites-derived tumour cells/cell lines with CSC properties and enhanced glycolysis. Clinically, PDK4 expression was correlated with aggressive features. Notably, blockade of PDK4 in tumourspheres/ALDH+CD44+ subsets led to inhibition of CSC characteristics, glycolysis and activation of STAT3/AKT/NF-κB/IL-8 (signal transducer and activator of transcription 3/protein kinases B/nuclear factor-κB/interleukin-8) signalling. Conversely, overexpression of PDK4 in ALDH−CD44– subsets exerted the opposite effects. Conclusion Ascites-derived ALDH+CD44+ tumour cell subsets endow stemness, metastatic and metabolic switch properties via PDK4-mediated STAT3/AKT/NF-κB/IL-8 signalling, suggesting PDK4 as a viable therapeutic molecular target for ovarian cancer management. more...

Supplementary Material 1. The comparison of baseline characteristics between participants in and out of the study

Supplemental material, sj-doc-2-ict-10.1177_15347354211058169 for Oral Chinese Patent Medicine Combined With Oxaliplatin-Based Chemotherapy Regimen for the Treatment of Colorectal Cancer: A Network Meta-Analysis by Mo Tang, Bin He, Jiawei Zhai and Lei Wang in Integrative Cancer Therapies more...

Supplemental material, sj-doc-5-ict-10.1177_15347354211058169 for Oral Chinese Patent Medicine Combined With Oxaliplatin-Based Chemotherapy Regimen for the Treatment of Colorectal Cancer: A Network Meta-Analysis by Mo Tang, Bin He, Jiawei Zhai and Lei Wang in Integrative Cancer Therapies more...

Supplemental material, sj-doc-4-ict-10.1177_15347354211058169 for Oral Chinese Patent Medicine Combined With Oxaliplatin-Based Chemotherapy Regimen for the Treatment of Colorectal Cancer: A Network Meta-Analysis by Mo Tang, Bin He, Jiawei Zhai and Lei Wang in Integrative Cancer Therapies more...

Supplemental material, sj-doc-1-ict-10.1177_15347354211058169 for Oral Chinese Patent Medicine Combined With Oxaliplatin-Based Chemotherapy Regimen for the Treatment of Colorectal Cancer: A Network Meta-Analysis by Mo Tang, Bin He, Jiawei Zhai and Lei Wang in Integrative Cancer Therapies more...