Dataset Information

Using Texture Analysis to Determine Human Papillomavirus Status of Oropharyngeal Squamous Cell Carcinomas on CT.

ABSTRACT:

Background and purpose

Human papillomavirus-associated oropharyngeal squamous cell carcinoma is increasing in prevalence and typically occurs in younger patients than human papillomavirus-negative squamous cell carcinoma. While imaging features of human papillomavirus-positive versus human papillomavirus-negative squamous cell carcinoma nodal metastases have been described, characteristics distinguishing human papillomavirus-positive from human papillomavirus-negative primary squamous cell carcinomas have not been well established. The purpose of this project was to evaluate the use of texture features to distinguish human papillomavirus-positive and human papillomavirus-negative primary oropharyngeal squamous cell carcinoma.

Materials and methods

Following institutional review board approval, 40 patients with primary oropharyngeal squamous cell carcinoma and known human papillomavirus status who underwent contrast-enhanced CT between December 2009 and October 2013 were included in this study. Segmentation of the primary lesion was manually performed with a semiautomated graphical-user interface. Following segmentation, an in-house-developed texture analysis program extracted 42 texture features from each segmented volume. A t test was used to evaluate differences in texture parameters between human papillomavirus-positive and human papillomavirus-negative squamous cell carcinomas.

Results

Of the 40 included patients, 29 had human papillomavirus-positive oropharyngeal squamous cell carcinoma and 11 had human papillomavirus-negative oropharyngeal squamous cell carcinoma. Significant differences were seen in the histogram parameters median (P = .006) and entropy (P = .016) and squamous cell carcinoma entropy (P = .043).

Conclusions

There are statistically significant differences in some texture features between human papillomavirus-positive and human papillomavirus-negative oropharyngeal tumors. Texture analysis may be considered an adjunct to the evaluation of human papillomavirus status and characterization of squamous cell carcinoma.

SUBMITTER: Buch K

PROVIDER: S-EPMC7965289 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectiveTo develop and test a machine learning model for classifying human papillomavirus (HPV) status of patients with oropharyngeal squamous cell carcinoma (OPSCC) using 18F-fluorodeoxyglucose (18F-FDG) PET-derived parameters in derived parameters and an appropriate combination of machine learning methods in patients with OPSCC.Materials and methodsThis retrospective study enrolled 126 patients (118 male; mean age, 60 years) with newly diagnosed, pathologically confirmed OPSCC, that underwent 18F-FDG PET-computed tomography (CT) between January 2012 and February 2020. Patients were randomly assigned to training and internal validation sets in a 7:3 ratio. An external test set of 19 patients (16 male; mean age, 65.3 years) was recruited sequentially from two other tertiary hospitals. Model 1 used only PET parameters, Model 2 used only clinical features, and Model 3 used both PET and clinical parameters. Multiple feature transforms, feature selection, oversampling, and training models are all investigated. The external test set was used to test the three models that performed best in the internal validation set. The values for area under the receiver operating characteristic curve (AUC) were compared between models.ResultsIn the external test set, ExtraTrees-based Model 3, which uses two PET-derived parameters and three clinical features, with a combination of MinMaxScaler, mutual information selection, and adaptive synthetic sampling approach, showed the best performance (AUC = 0.78; 95% confidence interval, 0.46-1). Model 3 outperformed Model 1 using PET parameters alone (AUC = 0.48, p = 0.047) and Model 2 using clinical parameters alone (AUC = 0.52, p = 0.142) in predicting HPV status.ConclusionUsing oversampling and mutual information selection, an ExtraTree-based HPV status classifier was developed by combining metabolic parameters derived from 18F-FDG PET/CT and clinical parameters in OPSCC, which exhibited higher performance than the models using either PET or clinical parameters alone.

Project description:Importance:Clinical trials that deintensify treatment for patients with suspected human papillomavirus (HPV)-positive oropharyngeal squamous cell carcinoma (OPSCC) use p16 expression to identify HPV-mediated tumors and guide treatment. While p16 staining has a strong correlation with good outcomes, approximately 12% of p16-positive patients have recurrent disease. Biomarkers that reveal tumor-specific characteristics, such as nodal involvement, may change therapy decisions. Objective:To assess whether if a tumor-specific genetic signature exists for node-negative vs node-positive HPV 16-positive/p16-positive OPSCCs. Design, Setting, and Participants:This was a retrospective cohort study with randomized case selection for p16 OPSCCs undertaken at a university-based, tertiary care cancer center. Samples were collected from patients with p16-positive OPSCC. A total of 21 HPV 16/p16-positive tumors were used in this study. Main Outcomes and Measures:Gene expression profiles of node-negative vs node-positive tumor samples were evaluated using a differential expression analysis approach and the sensitivity and specificity of a molecular signature was determined. Results:Among the 21 patients in the study (3 women, 18 men; mean [SD] age, 54.6 [9.6] years), 6 had node-negative disease and 15 had node-positive disease. Using differential expression analysis, we found 146 genes that were significantly different in patients with node-negative disease vs those with node-positive disease, of which 15 genes were used to create a genetic signature that could distinguish node-negative-like from node-positive-like disease. The resultant molecular signature has a sensitivity of 88.2% (95% CI, 63.6%-98.5%) and specificity of 85.7% (95% CI, 42.1%-99.6%). The positive likelihood ratio of this signature was 6.1 (95% CI, 1.0-38.2) and the negative likelihood ratio was 0.1 (95% CI, 0.04-0.5). Given this population's prevalence of node-positive disease of 70.8%, the positive- and negative-predicative values for this gene signature were 93.7% (95% CI, 70.8%-98.9%) and 75.0% (95% CI, 44.1%-92.0%), respectively. In addition, we developed a gene signature using agnostic, machine learning software that identified a 40-gene profile that predicts node-negative disease from node-positive disease (area under the curve, 0.93; 95% CI, 0.63-1.00). Conclusions and Relevance:Many HPV-16 and p16-positive tumors are treated as "lower-risk," but they do not have similar genetic compositions at the biological level. The identification of subgroups with unique expression patterns, such as those with nodal metastases, may guide physicians toward alternative or more aggressive therapies. In our study, unguided clustering suggested that that the larger biological characteristics of a tumor could be a better prognostic biomarker.