Initial CT findings in early tongue and oral floor cancer as predictors of late neck metastasis

Magnetic resonance imaging-based radiomics and deep learning models for predicting lymph node metastasis of squamous cell carcinoma of the tongue

OBJECTIVE

This study aimed to establish a combined method of radiomics and deep learning (DL) in magnetic resonance imaging (MRI) to predict lymph node metastasis (LNM) preoperatively in patients with squamous cell carcinoma of the tongue.

STUDY DESIGN

In total, MR images of 196 patients with lingual squamous cell carcinoma were divided into training (n = 156) and test (n = 40) cohorts. Radiomics and DL features were extracted from MR images and selected to construct machine learning models. A DL radiomics nomogram was established via multivariate logistic regression by incorporating the radiomics signature, the DL signature, and MRI-reported LN status.

RESULTS

Nine radiomics and 3 DL features were selected. In the radiomics test cohort, the multilayer perceptron model performed best with an area under the receiver operating characteristic curve (AUC) of 0.747, but in the DL cohort, the best model (logistic regression) performed less well (AUC = 0.655). The DL radiomics nomogram showed good calibration and performance with an AUC of 0.934 (outstanding discrimination ability) in the training cohort and 0.757 (acceptable discrimination ability) in the test cohort. The decision curve analysis demonstrated that the nomogram could offer more net benefit than a single radiomics or DL signature.

CONCLUSION

The DL radiomics nomogram exhibited promising performance in predicting LNM, which facilitates personalized treatment of tongue cancer.

Immunohistochemical analysis of EGFR and hyaluronan in tongue cancer and the development of regional recurrence in patients initially diagnosed N0

OBJECTIVE

To investigate whether the extent of expression of hyaluronan (HA) and epidermal growth factor receptor (EGFR) in squamous cell carcinoma of the mobile tongue can predict the risk of cervical metastasis and survival.

STUDY DESIGN

Retrospective histopathologic study.

METHODS

Surgical specimens from 64 patients who had undergone surgery for squamous cell carcinoma of the mobile tongue were assessed using immunohistochemistry to investigate the expression of HA and EGFR in the primary tumours, and the data were then correlated to cervical metastasis and survival.

RESULTS

There was a significant correlation between the intensity of HA staining and patient survival (p .024), and a weak correlation between the staining proportion of EGFR and the risk for regional recurrence (AUC 66).

CONCLUSIONS

This study indicates that immunoscoring using HA could be used to provide prognostic tools for tongue cancer, and that it might be of interest to study the prognostic properties of EGFR further concerning the risk for regional recurrence after the primary treatment.

Contrast-enhanced CT and MRI for detecting neck metastasis of oral cancer: comparison between analyses performed by oral and medical radiologists

OBJECTIVES

The aim of the study was to verify the concordance of contrast-enhanced CT (CECT) and MRI evaluation among four radiologists in detecting metastatic cervical lymph nodes of oral cancer patients.

METHODS

Ten patients underwent clinical and imaging examinations (CECT and MRI). Four radiologists, two oral and maxillofacial radiologists (OMRs) and two medical radiologists (MRs), independently analysed the images twice. Cohen's kappa index and Wilcoxon signed-rank test were used to verify the concordance between all analyses.

RESULTS

Regarding the interobserver agreement, the OMRs presented excellent kappa values for determining the regional lymph nodes (N-stage) in both CECT and MRI. The MRs presented moderate agreement for CECT evaluation at the first reading, but no concordance was found for the other analyses. When each imaging modality was analysed separately, kappa values were higher between all examiners. Greater variability was demonstrated between N-stage evaluation using different examinations. All radiologists were able to identify a greater number of metastatic lymph nodes in CECT than in MRI, except one MR, but no significant difference was found for all readers. The differences between the number of metastatic lymph nodes among all radiologists were not statistically significant. Moderate intraobserver agreement was observed for CECT and MRI evaluation, except for one MR.

CONCLUSIONS

The differences found between the N-stage performed by OMRs and MRs support the necessity of a multidisciplinary approach in the imaging evaluation of metastatic nodes. Further studies are necessary to confirm which imaging modality should be employed when evaluating neck areas.

Recent advances in oral oncology 2008; squamous cell carcinoma imaging, treatment, prognostication and treatment outcomes

This paper provides a synopsis of the main papers on diagnosis, imaging, treatment, prognostication and treatment outcomes in patients with oral and oropharyngeal squamous cell carcinoma (OSCC) and head and neck SCC (HNSCC) published in 2008 in Oral Oncology - an international interdisciplinary journal which publishes high quality original research, clinical trials and review articles, and all other scientific articles relating to the aetiopathogenesis, epidemiology, prevention, clinical features, diagnosis, treatment and management of patients with neoplasms in the head and neck, and orofacial disease in patients with malignant disease.