scholarly journals Comparisons of forecasting for Survival Outcome for Head and Neck Squamous Cell Carcinoma by using Six Machine Learning Models Based on Multi-Omics

2021 ◽  
Author(s):  
Liying Mo ◽  
Yuangang Su ◽  
Jianhui Yuan ◽  
Zhiwei Xiao ◽  
Ziyan Zhang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Head And Neck ◽  
Squamous Cell ◽  
Survival Outcome ◽  
Learning Models ◽  
Machine Learning Methods ◽  
Machine Learning Models ◽  
Better Than

Abstract Background: Machine learning methods showed excellent predictive ability in a wide range of fields. For the survival of head and neck squamous cell carcinoma (HNSC), its multi-omics influence is crucial. This study attempts to establish a variety of machine learning multi-omics models to predict the survival of HNSC and find the most suitable machine learning prediction method. Results: For omics of HNSC, the results of the six models all showed that the performance of multi-omics was better than each single-omic alone. Results were presented which showed that the BN model played a good prediction performance (area under the curve [AUC] 0.8250) in HNSC multi-omics data. The other machine learning models RF (AUC = 0.8002), NN (AUC = 0.7200), and GLM (AUC = 0.7145) also showed high predictive performance except for DT(AUC = 0.5149) and SVM(AUC = 0.6981). And the results of a vitro qPCR were consistent with the Random forest algorithm. Conclusion: Machine learning methods could better forecast the survival outcome of HNSC. Meanwhile, this study found that the Bayesian network was the most superior. Moreover, the forecast result of multi-omics was better than single-omic alone in HNSC.

scholarly journals Machine Learning Incorporating Host Factors for Predicting Survival in Head and Neck Squamous Cell Carcinoma Patients

Cancers ◽  
2021 ◽  
Vol 13 (18) ◽  
pp. 4559
Author(s):  
Han Yu ◽  
Sung Jun Ma ◽  
Mark Farrugia ◽  
Austin J. Iovoli ◽  
Kimberly E. Wooten ◽  
...  
Keyword(s):  
Machine Learning ◽  
Overall Survival ◽  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Head And Neck ◽  
Risk Group ◽  
Host Factors ◽  
Learning Models ◽  
Random Survival Forest ◽  
Machine Learning Models

Prognostication for cancer patients is integral for patient counseling and treatment planning, yet providing accurate prediction can be challenging using existing patient-specific clinical indicators and host factors. In this work, we evaluated common machine learning models in predicting head and neck squamous cell carcinoma (HNSCC) patients’ overall survival based on demographic, clinical features and host factors. We found random survival forest had best performance among the models evaluated, which achieved a C-index of 0.729 and AUROC of 0.792 in predicting two-year overall survival. In addition, we verified that host factors are independently predictive of HNSCC overall survival, which improved the C-index by a margin of 0.026 and the AUROC by 0.034. Due to the strong correlation among host factors, we showed that proper dimension reduction is an important step before their incorporation into the machine learning models, which provides a host factor score reflecting the patients’ nutrition and inflammation status. The score by itself showed excellent discriminating capacity with the high-risk group having a hazard ratio of 3.76 (1.93–7.32, p < 0.0001) over the low-risk group. The hazard ratios were further improved to 7.41 (3.66–14.98, p < 0.0001) by the random survival forest model after including demographic and clinical features.

Machine learning analysis of DNA methylation profiles distinguishes primary lung squamous cell carcinomas from head and neck metastases

2019 ◽  
Vol 11 (509) ◽  
pp. eaaw8513 ◽  
Author(s):  
Philipp Jurmeister ◽  
Michael Bockmayr ◽  
Philipp Seegerer ◽  
Teresa Bockmayr ◽  
Denise Treue ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Dna Methylation ◽  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Head And Neck ◽  
Squamous Cell ◽  
Pulmonary Metastases ◽  
Normal Lung ◽  
Support Vector

Head and neck squamous cell carcinoma (HNSC) patients are at risk of suffering from both pulmonary metastases or a second squamous cell carcinoma of the lung (LUSC). Differentiating pulmonary metastases from primary lung cancers is of high clinical importance, but not possible in most cases with current diagnostics. To address this, we performed DNA methylation profiling of primary tumors and trained three different machine learning methods to distinguish metastatic HNSC from primary LUSC. We developed an artificial neural network that correctly classified 96.4% of the cases in a validation cohort of 279 patients with HNSC and LUSC as well as normal lung controls, outperforming support vector machines (95.7%) and random forests (87.8%). Prediction accuracies of more than 99% were achieved for 92.1% (neural network), 90% (support vector machine), and 43% (random forest) of these cases by applying thresholds to the resulting probability scores and excluding samples with low confidence. As independent clinical validation of the approach, we analyzed a series of 51 patients with a history of HNSC and a second lung tumor, demonstrating the correct classifications based on clinicopathological properties. In summary, our approach may facilitate the reliable diagnostic differentiation of pulmonary metastases of HNSC from primary LUSC to guide therapeutic decisions.

scholarly journals Concurrent Chemoradiotherapy With Weekly Low-Dose Cisplatin for Japanese Patients With Head and Neck Squamous Cell Carcinoma

2021 ◽  
Vol 15 ◽  
pp. 117955492110484
Author(s):  
Shogo Shinohara ◽  
Shinji Takebayashi ◽  
Kiyomi Hamaguchi ◽  
Tetsuhiko Michida ◽  
Yota Tobe ◽  
...  
Keyword(s):  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Head And Neck ◽  
Squamous Cell ◽  
Low Dose ◽  
Concurrent Chemoradiotherapy ◽  
Age Distribution ◽  
Survival Outcome ◽  
Neck Squamous Cell Carcinoma ◽  
High Dose

Background: Concurrent chemoradiotherapy (CCRT) with tri-weekly high-dose cisplatin (HDC) is considered the standard regimen. However, due to significant toxicity, various weekly low-dose schedules have been increasingly used. We investigated the tolerability and survival of patients with head and neck squamous cell carcinoma (HNSCC) who underwent CCRT with low-dose weekly cisplatin (LDC) for Japanese population. Methods: A retrospective review was conducted among patients with HNSCC who were treated with CCRT/LDC in our institute. Ninety-five patients who met the criteria were enrolled in this study. We evaluated the cycle and cumulative cisplatin dose, completion rate of radiotherapy, adverse events, and survival outcome. Results: The mean cycles and cumulative cisplatin dose were 4.7 cycles and 187 mg/m2. All patients completed planned dose of radiation without prolonged breaks. Leucopoenia was the most frequent dose-limiting factor and 44% patients developed grade 3 or 4 toxicity. The 2-year overall survival and recurrence-free survival were 93% and 74%, respectively. The significant differences of survival outcomes between the patients with total cisplatin dose (⩾200 mg and <200 mg) or among age distribution (35-55, 56-75, and ⩾76) were not observed. Conclusions: Concurrent chemoradiotherapy/LDC can be safely administered with acceptable toxicity and survival outcome even if the patients with higher age, lower eGFR, and so on.

scholarly journals Site-Specific Variation in Radiomic Features of Head and Neck Squamous Cell Carcinoma and Its Impact on Machine Learning Models

Cancers ◽  
2021 ◽  
Vol 13 (15) ◽  
pp. 3723
Author(s):  
Xiaoyang Liu ◽  
Farhad Maleki ◽  
Nikesh Muthukrishnan ◽  
Katie Ovens ◽  
Shao Hui Huang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Head And Neck ◽  
Squamous Cell ◽  
Model Performance ◽  
Tumor Location ◽  
Nodal Metastasis ◽  
P Value ◽  
Learning Models ◽  
Hpv Status ◽  
Machine Learning Models

Current radiomic studies of head and neck squamous cell carcinomas (HNSCC) are typically based on datasets combining tumors from different locations, assuming that the radiomic features are similar based on histopathologic characteristics. However, molecular pathogenesis and treatment in HNSCC substantially vary across different tumor sites. It is not known if a statistical difference exists between radiomic features from different tumor sites and how they affect machine learning model performance in endpoint prediction. To answer these questions, we extracted radiomic features from contrast-enhanced neck computed tomography scans (CTs) of 605 patients with HNSCC originating from the oral cavity, oropharynx, and hypopharynx/larynx. The difference in radiomic features of tumors from these sites was assessed using statistical analyses and Random Forest classifiers on the radiomic features with 10-fold cross-validation to predict tumor sites, nodal metastasis, and HPV status. We found statistically significant differences (p-value ≤ 0.05) between the radiomic features of HNSCC depending on tumor location. We also observed that differences in quantitative features among HNSCC from different locations impact the performance of machine learning models. This suggests that radiomic features may reveal biologic heterogeneity complementary to current gold standard histopathologic evaluation. We recommend considering tumor site in radiomic studies of HNSCC.

scholarly journals Fifteen mRNA-lncRNA expression-based signature predicted the survival of late-staged head and neck squamous cell carcinoma

2020 ◽  
Vol 40 (7) ◽  
Author(s):  
He Ren ◽  
Huaping Li ◽  
Ping Li ◽  
Yuhui Xu ◽  
Gang Liu ◽  
...  
Keyword(s):  
Gene Expression ◽  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Head And Neck ◽  
Median Survival ◽  
Squamous Cell ◽  
Gene Set Enrichment Analysis ◽  
The Other ◽  
Neck Squamous Cell Carcinoma ◽  
Better Than

Abstract Background: Gene expression is necessary for regulation in almost all biological processes, at the same time, it is related to the prognosis for head and neck squamous cell carcinoma (HNSCC). The prognosis of late-staged HNSCC is important because of its guiding significance on the therapy strategies. Methods: In this work, we analyzed the relationship between gene expression and HNSCC in The Cancer Genome Atlas (TCGA) cohort, and optimized the panel with random forest survival analysis. Subsequently, a Cox multivariate regression-based model was developed to predict the clinical outcome of HNSCC. The performance of the model was assayed in the training cohort and validated in another three independent cohorts (GSE41614, E-TABM-302, E-MTAB-1328). The underlying pathways significantly associated with the model were identified. According to the results, patients of low-score group (median survival months: 27.4, 95% CI: 18.2–43) had a significant poor survival than those of high-score group (median survival months: 69.4, 95% CI: 58.7–72.1, P=2.7e-5), and the observation was repeatable in the other validation cohorts. Further analysis revealed that the model performed better than the other clinical indicators and is independent of these indicators. Results: Comparison revealed that the model performed better than existing models for late HNSCC prognosis. Gene set enrichment analysis (GSEA) elucidated that the model was significantly associated with various cell processes and pathways.

Sign in / Sign up

Export Citation Format

Share Document