Abstract
1297
Purpose: Tumor metabolic phenotype can be assessed with integrated image pattern analysis of FDG PET/CT, called radiomics. This study was performed to assess the prognostic value of radiomic PET parameters in head and neck squamous cell carcinoma (HNSCC) patients.
Methods: PET/CT data of 215 patients from HNSCC collection free database in The Cancer Imaging Archive (TCIA) were reviewed. Seventy PET/CT exams for baseline evaluation with same spatial resolution and assessable primary tumor were included. Primary tumors were segmented by adjusted fixed thresholding method. Segmental tumors in PET images were preprocessed using relative resampling of 64 bins. Forty seven PET parameters including conventional parameters and texture parameters were measured. Binary groups of homogeneous imaging phenotypes, clustered by K-means method, were compared for overall survival (OS) and disease free survival (DFS) by log rank test. Individual radiomics features were tested by cox-regression test for OS and DFS, and the most significant feature was tested in multivariate analysis with clinical factors including age, and T, N stages. Two-sided p-values less than 0.05 was regarded as statistically significant.
Results: Median follow up period was 62.1 months. Binary groups with different imaging phenotypes showed significant difference in OS (p=0.036), and borderline difference in DFS (p=0.086). Grey-Level Zone Length Matrix, Gray-Level Non-Uniformity (GLZLM_GLNU) was the most significant prognostic factor for both OS (HR: 0.32, p=0.008) and DFS (hazard ratio [HR] 0.22, p=0.020). Multivariate analysis revealed GLZLM_GLNU as an independent prognostic factor for OS (HR 0.24, p=0.0.008).
Conclusions: PET radiomics based metabolic imaging phenotype can be biomarker for survival prognosis in HNSCC patients, and may assist clinicians for patient risk assessment.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.