Dataset Information

Histopathological image and gene expression pattern analysis for predicting molecular features and prognosis of head and neck squamous cell carcinoma.

ABSTRACT:

Background

Histopathological image features offer a quantitative measurement of cellular morphology, and probably help for better diagnosis and prognosis in head and neck squamous cell carcinoma (HNSCC).

Methods

We first used histopathological image features and machine-learning algorithms to predict molecular features of 212 HNSCC patients from The Cancer Genome Atlas (TCGA). Next, we divided TCGA-HNSCC cohort into training set (n = 149) and test set (n = 63), and obtained tissue microarrays as an external validation set (n = 126). We identified the gene expression profile correlated to image features by bioinformatics analysis.

Results

Histopathological image features combined with random forest may predict five somatic mutations, transcriptional subtypes, and methylation subtypes, with area under curve (AUC) ranging from 0.828 to 0.968. The prediction model based on image features could predict overall survival, with 5-year AUC of 0.831, 0.782, and 0.751 in training, test, and validation sets. We next established an integrative prognostic model of image features and gene expressions, which obtained better performance in training set (5-year AUC = 0.860) and test set (5-year AUC = 0.826). According to histopathological transcriptomics risk score (HTRS) generated by the model, high-risk and low-risk patients had different survival in training set (HR = 4.09, p < 0.001) and test set (HR=3.08, p = 0.019). Multivariate analysis suggested that HTRS was an independent predictor in training set (HR = 5.17, p < 0.001). The nomogram combining HTRS and clinical factors had higher net benefit than conventional clinical evaluation.

Conclusions

Histopathological image features provided a promising approach to predict mutations, molecular subtypes, and prognosis of HNSCC. The integration of image features and gene expression data had potential for improving prognosis prediction in HNSCC.

SUBMITTER: Chen L

PROVIDER: S-EPMC8267162 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundImmune-related genes (IRGs) were linked to the prognosis of head and neck squamous cell carcinoma (HNSCC). This study aimed to identify the effects of an immune-related gene signature (IRGS) that can predict the of HNSCC prognosis.MethodsThe expression data of 770 HNSCC patients from the TCGA database and the GEO database were used. To explore a predictive model, the Cox proportional hazards model was applied. The Kaplan-Meier survival analysis, as well as univariate and multivariate analyses were performed to evaluate the independent predictive value of IRGS. To explore biological functions of IRGS, enrichment analyses and pathway annotation for differentially expressed genes (DEGs) in different immune groups were applied, as well as the immune infiltration.ResultsA prognostic signature comprising 27 IRGs was generated. IRGS significantly stratified HNSCC patients into high and low immune risk groups in regard to overall survival in the training cohort (HR = 3.69, 95% CI 2.73-4.98, P < 0.001). Likewise, IRGS could be linked to the prognosis of HNSCC in patients of the validation cohort (HR = 1.84, 95% CI 1.21-2.81, P < 0.01). Even after adjusting for TNM stage, IRGS was maintained as an independent predictor in the multivariate analysis (HR = 3.62, 95% CI 2.58-5.09, P < 0.001), and in the validation cohort (HR = 1.73, 95% CI 1.12-2.67, P = 0.014). The IFN-α response, the IFN-γ response, IL-2/STAT5 signaling, and IL-6/JAK/STAT3 signaling were all negatively correlated with the immune risk (P < 0.01). Immune infiltration of the high-risk group was significantly lower than that of the low-risk group (P < 0.01). Most notably, the infiltration of CD8 T cells, memory-activated CD4 T cells, and regulatory T cells was strongly upregulated in the low immune risk groups, while memory resting CD4 T cell infiltration was downregulated (P < 0.01).ConclusionOur analysis provides a comprehensive prognosis of the immune microenvironments and outcomes for different individuals. Further studies are needed to evaluate the clinical application of this signature.

Project description:Head and neck squamous cell carcinoma (HNSCC) is a common malignancy worldwide with a poor prognosis. DNA methylation is an epigenetic modification that plays a critical role in the etiology and pathogenesis of HNSCC. The current study aimed to develop a predictive methylation signature based on bioinformatics analysis to improve the prognosis and optimize therapeutic outcome in HNSCC. Clinical information and methylation sequencing data of patients with HNSCC were downloaded from The Cancer Genome Atlas database. The R package was used to identify differentially methylated genes (DMGs) between HNSCC and adjacent normal tissues. We identified 22 DMGs associated with 246 differentially methylated sites. Patients with HNSCC were classified into training and test groups. Cox regression analysis was used to build a risk score formula based on the five methylation sites (cg26428455, cg13754259, cg17421709, cg19229344, and cg11668749) in the training group. The Kaplan-Meier survival curves showed that the overall survival (OS) rates were significantly different between the high- and low-risk groups sorted by the signature in the training group (median: 1.38 vs. 1.57 years, log-rank test, p < 0.001). The predictive power was then validated in the test group (median: 1.34 vs. 1.75 years, log-rank test, p < 0.001). The area under the receiver operating characteristic curve (area under the curve) based on the signature for predicting the 5-year survival rates, was 0.7 in the training and 0.73 in test groups, respectively. The results of multivariate Cox regression analysis showed that the riskscore (RS) signature based on the five methylation sites was an independent prognostic tool for OS prediction in patients. In addition, a predictive nomogram model that incorporated the RS signature and patient clinical information was developed. The innovative methylation signature-based model developed in our study represents a robust prognostic tool for guiding clinical therapy and predicting the OS in patients with HNSCC.

Project description:BackgroundFerrogenesis was strongly associated with tumorigenesis and development, and activating the ferrogenic process was a novel regimen in treating cancer, especially conventional treatment-resistant cancers. The purpose of the article was to construct a ferroptosis-related long noncoding RNAs (FRlncRNAs) signature, regardless of expression levels to effectively predict prognosis and immunotherapeutic response for head and neck squamous cell carcinoma (HNSCC).MethodsThe RNA-seq data for HNSCC and corresponding clinical information were obtained in the TCGA database, and ferroptosis-related genes (FRGs) were extracted in the ferroptosis database. On this basis, differentially expressed FRlncRNAs (DEFRlncRNAs) pairs were identified through coexpression analysis, differential expression analysis, and a fresh pairing algorithm. Then, a risk assessment model was established with univariate Cox, LASSO, and multivariate Cox regression analysis. Finally, we evaluated the model from various aspects, including survival status, clinicopathological characteristics, infiltration status of immune cells, immune functions, chemotherapeutic sensitivity, immune checkpoint inhibitors (ICIs)-related molecules, and N6-methyladenosine (m6A) mRNA status.ResultWe established a signature of 11-DEFRlncRNA pairs related to the prognosis of HNSCC that had AUC values above 0.75 in the one-, three-, and five-year ROC curves, underscoring the high susceptibility and specifiability of predicting HNSCC prognosis. Survival rates were remarkably higher for the low-risk patients than for the high-risk patients, and the signature was significantly correlated with survival, clinical, T, and N stages. Finally, immune cell infiltration status, immune functions, chemotherapeutic sensitivity, and expression levels of ICIs-related and m6A-related molecules were statistically different among different groups.ConclusionOur study established a novel lncRNA signature, which is independent of specific expression levels, could predict patient prognosis, and might have promising clinical applications in HNCSS.