Predictive models for stage and risk classification in head and neck squamous cell carcinoma (HNSCC)

Abstract

Machine learning techniques are increasingly used in the analysis of high throughput genome sequencing data to better understand the disease process and design of therapeutic modalities. In the current study, we have applied state of the art machine learning (ML) algorithms (Random Forest (RF), Support Vector Machine Radial Kernel (svmR), Adaptive Boost (AdaBoost), averaged Neural Network (avNNet), and Gradient Boosting Machine (GBM)) to stratify the HNSCC patients in early and late clinical stages (TNM) and to predict the risk using miRNAs expression profiles. A six miRNA signature was identified that can stratify patients in the early and late stages. The mean accuracy, sensitivity, specificity, and area under the curve (AUC) was found to be 0.84, 0.87, 0.78, and 0.82, respectively indicating the robust performance of the generated model. The prognostic signature of eight miRNAs was identified using LASSO (least absolute shrinkage and selection operator) penalized regression. These miRNAs were found to be significantly associated with overall survival of the patients. The pathway and functional enrichment analysis of the identified biomarkers revealed their involvement in important cancer pathways such as GP6 signalling, Wnt signalling, p53 signalling, granulocyte adhesion, and dipedesis. To the best of our knowledge, this is the first such study and we hope that these signature miRNAs will be useful for the risk stratification of patients and the design of therapeutic modalities.

FOXD1 Regulates the Sensitivity of Cetuximab by Regulating the Expression of EGFR in Head and Neck Squamous Cell Cancer

“…Their dysfunction was closely related to many major diseases [ 12 ]. The forkhead box (FOX) family, composed of proteins that share related winged helix-turn-helix DNA binding motifs, belongs to the “winged helix” superfamily [ 13 ]. FOX genes are widely present in the evolution of vertebrates and invertebrates, involved in many molecular cascades and regulation of biological functions, such as embryonic development, cell cycle regulation, early morning metabolism control, and signal pathway transduction [ 14 ].…”

A Combined Long Noncoding RNA Signature as a Candidate Prognostic Biomarker for Ovarian Cancer

“…With the rapid development of high-throughput technologies such as gene chip and RNA sequencing, gene analysis has become a powerful tool for screening molecular biomarkers of tumour prognosis prediction ( 24 ). Recently, several studies demonstrated that the robustness of several biomarker combinations is better than the robustness of a single biomarker ( 25 ).…”

Genetic variants in autophagy-related gene ATG2B predict the prognosis of colorectal cancer patients receiving chemotherapy

“…Firstly, we extracted SNPs within 2 kb up-and down-stream regions of 8 differentially expressed autophagy-related genes using the Han Chinese in Beijing (CHB) data from the 1000 Genomes Project (March 2012) based on these selection conditions: (a) minor allele frequency (MAF) in population ≥ 0.05, (b) Hardy-Weinberg equilibrium (HWE) ≥ 0.05, (c) call rate > 95%. Secondly, SNPinfo Web Server (24), HaploReg (25), and RegulomeDB (26) were used to predict functional SNPs. SNPs were not included when the RegulomeDB score > 6.…”

“…Statistical methods with univariate Cox regression analysis were utilized to assess associations between the clinical characteristics and OS of CRC patients (24). The results were corrected by the FDR, which was employed to mitigate against false-positive results.…”

“…The results were corrected by the FDR, which was employed to mitigate against false-positive results. The associations between targeted SNPs and CRC prognosis were assessed depending on multivariate and univariate COX regression models (24). Logistic regression models were employed to identify the effects of SNPs on DCR.…”