Radiomic biomarkers of locoregional recurrence: prognostic insights from oral cavity squamous cell carcinoma preoperative CT scans

Xiao Ling ¹, Gregory S Alexander ², Jason Molitoris ¹

⁰Department of Radiation Oncology, University of Maryland School of Medicine, Baltimore, MD, United States.

Frontiers in Oncology +

|

May 8, 2024

View abstract on PubMed

Summary

This study identified two radiomics imaging biomarkers, Large Dependence Emphasis and Long Run Emphasis, to predict locoregional recurrence in Oral Cavity Squamous Cell Carcinoma patients. These biomarkers show promise for personalized treatment and improved survival outcomes.

Area Of Science

Oncology
Radiology
Medical Imaging

Background

Oral Cavity Squamous Cell Carcinoma (OSCC) poses a significant challenge due to the risk of locoregional recurrence (LR).
Accurate preoperative prediction of LR is crucial for tailoring treatment strategies and improving patient outcomes.

Purpose Of The Study

To identify and validate CT-based radiomic imaging biomarkers for predicting locoregional recurrence in OSCC patients.
To assess the efficacy of these biomarkers in distinguishing between patients with and without LR.

Main Methods

Extraction of 1,092 radiomic features from pre-treatment CT scans of 78 OSCC patients.
Application of feature selection algorithms and Logistic Regression Modeling (LRM) to identify significant imaging biomarkers.
Validation of selected biomarkers using Receiver Operating Characteristic (ROC) curve analysis for prediction accuracy.

Main Results

Two radiomics biomarkers, Large Dependence Emphasis (LDE) and Long Run Emphasis (LRE), were identified as significant predictors of LR.
These biomarkers demonstrated high sensitivity (0.82) and specificity (1.00) in distinguishing patients with and without LR.
Patients with higher LRE or LDE values showed significantly improved 3-year recurrence-free survival rates.

Conclusions

CT-based radiomics biomarkers, specifically LDE and LRE, can effectively predict locoregional recurrence in OSCC.
These imaging biomarkers offer a non-invasive method for risk stratification and personalized treatment planning.
The findings contribute to advancing the use of radiomics in oncology for improved patient management and survival.

Keywords:

CT biomarker classification logistic regression oral cavity squamous cell carcinoma outcome prediction radiomics recurrence