Improved Cervical Lymph Node Characterization among Patients with Head and Neck Squamous Cell Carcinoma Using MR Texture Analysis Compared to Traditional FDG-PET/MR Features Alone

Abstract

Accurate differentiation of benign and malignant cervical lymph nodes is important for prognosis and treatment planning in patients with head and neck squamous cell carcinoma. We evaluated the diagnostic performance of magnetic resonance image (MRI) texture analysis and traditional 18F-deoxyglucose positron emission tomography (FDG-PET) features. This retrospective study included 21 patients with head and neck squamous cell carcinoma. We used texture analysis of MRI and FDG-PET features to evaluate 109 histologically confirmed cervical lymph nodes (41 metastatic, 68 benign). Predictive models were evaluated using area under the curve (AUC). Significant differences were observed between benign and malignant cervical lymph nodes for 36 of 41 texture features (p < 0.05). A combination of 22 MRI texture features discriminated benign and malignant nodal disease with AUC, sensitivity, and specificity of 0.952, 92.7%, and 86.7%, which was comparable to maximum short-axis diameter, lymph node morphology, and maximum standard uptake value (SUVmax). The addition of MRI texture features to traditional FDG-PET features differentiated these groups with the greatest AUC, sensitivity, and specificity (0.989, 97.5%, and 94.1%). The addition of the MRI texture feature to lymph node morphology improved nodal assessment specificity from 70.6% to 88.2% among FDG-PET indeterminate lymph nodes. Texture features are useful for differentiating benign and malignant cervical lymph nodes in patients with head and neck squamous cell carcinoma. Lymph node morphology and SUVmax remain accurate tools. Specificity is improved by the addition of MRI texture features among FDG-PET indeterminate lymph nodes. This approach is useful for differentiating benign and malignant cervical lymph nodes.

Keywords: PET-MRI; cervical lymphadenopathy; machine learning; squamous cell carcinoma; texture analysis.

Figure 1

ROC curves of the best individual texture features for distinguishing benign and malignant cervical lymph nodes in patients with HNSCC. HNSCC, head and neck squamous cell carcinoma; GLCM, Gray-Level Co-Occurrence Matrix; GLZLM, Gray-Level Zone-Length Matrix; GLRLM, Gray-Level Run-Length Matrix; ZLNU, zone-length non-uniformity; RLNU, run-length non-uniformity.

Figure 2

ROC curves of the top combinations for distinguishing benign and malignant cervical lymph nodes in patients with HNSCC. HNSCC, head and neck squamous cell carcinoma; TFs, texture features; SAD, short-axis diameter; SUVmax, maximum standardized uptake value.

Figure 3

Example of a histologically confirmed benign reactive FDG-avid cervical lymph node. FDG-PET/MR evaluation of a patient with HNSCC. (A) attenuation-corrected FDG-PET, (B) fused axial STIR, (C) axial STIR, and (D) segmented axial STIR images. There is a right level 2A FDG-avid lymph node (SUVmax 6.6) that measured 7 mm in short-axis diameter and demonstrated normal morphology. Texture analysis was performed within the visualized region of interest in image (D). This node was histologically confirmed as benign reactive after neck dissection.

Figure 4

Example of a histologically confirmed malignant cervical lymph node. FDG-PET/MR evaluation of a patient with HNSCC. (A) attenuation-corrected FDG-PET, (B) fused axial STIR, (C) axial STIR, and (D) segmented axial STIR images. There is a left level 2B FDG-avid lymph node (SUVmax 9) that measured 7 mm in short-axis diameter and demonstrated abnormal morphology. Texture analysis was performed within the visualized region of interest in image (D). This node was histologically confirmed as malignant after neck dissection. Mild hypermetabolic activity is also visualized in the musculature of the neck, likely related to muscle strain.