Prognostic Value of Baseline Radiomic Features of ¹⁸F-FDG PET in Patients with Diffuse Large B-Cell Lymphoma

• 發表在: Diagnostics
• Main Authors: Kun-Han Lue, Yi-Feng Wu, Hsin-Hon Lin, Tsung-Cheng Hsieh, Shu-Hsin Liu, Sheng-Chieh Chan, Yu-Hung Chen
• 格式: Article
• 語言: 英语
• 出版: MDPI AG 2020-12-01
• 主題: ¹⁸F-FDG; PET; radiomics; prognosis; diffuse large B-cell lymphoma
• 在線閱讀: https://www.mdpi.com/2075-4418/11/1/36

This study investigates whether baseline ¹⁸F-FDG PET radiomic features can predict survival outcomes in patients with diffuse large B-cell lymphoma (DLBCL). We retrospectively enrolled 83 patients diagnosed with DLBCL who underwent ¹⁸F-FDG PET scans before treatment. The patients were divided into the training cohort (<i>n</i> = 58) and the validation cohort (<i>n</i> = 25). Eighty radiomic features were extracted from the PET images for each patient. Least absolute shrinkage and selection operator regression were used to reduce the dimensionality within radiomic features. Cox proportional hazards model was used to determine the prognostic factors for progression-free survival (PFS) and overall survival (OS). A prognostic stratification model was built in the training cohort and validated in the validation cohort using Kaplan–Meier survival analysis. In the training cohort, run length non-uniformity (RLN), extracted from a gray level run length matrix (GLRLM), was independently associated with PFS (hazard ratio (HR) = 15.7, <i>p</i> = 0.007) and OS (HR = 8.64, <i>p</i> = 0.040). The International Prognostic Index was an independent prognostic factor for OS (HR = 2.63, <i>p</i> = 0.049). A prognostic stratification model was devised based on both risk factors, which allowed identification of three risk groups for PFS and OS in the training (<i>p</i> < 0.001 and <i>p</i> < 0.001) and validation (<i>p</i> < 0.001 and <i>p</i> = 0.020) cohorts. Our results indicate that the baseline ¹⁸F-FDG PET radiomic feature, RLN_GLRLM, is an independent prognostic factor for survival outcomes. Furthermore, we propose a prognostic stratification model that may enable tailored therapeutic strategies for patients with DLBCL.