A multidimensional nomogram combining overall stage, dose volume histogram parameters and radiomics to predict progression-free survival in patients with locoregionally advanced nasopharyngeal carcinoma. (November 2019)

• Record Type: Journal Article
• Authors: Yang, Kaixuan; Tian, Jiangfang; Zhang, Bin; Li, Mei; Xie, Wenji; Zou, Yating; Tan, Qiaoyue; Liu, Lihui; Zhu, Jinbing; Shou, Arthur; Li, Guangjun
• Is Part Of: Oral oncology. Volume 98(2019)
• Page Start: 85
• Page End: 91
• Publication Date: 2019-11
• Subjects: Locoregionally advanced nasopharyngeal carcinoma -- Radiomics -- Intensity modulated radiotherapy -- Dose volume histogram parameters -- Lymph node -- Nomogram; Mouth -- Cancer -- Periodicals; Mouth -- Tumors -- Periodicals; Mouth Diseases -- Periodicals; Mouth Neoplasms -- Periodicals; Bouche -- Cancer -- Périodiques; Bouche -- Tumeurs -- Périodiques; Tumeurs -- Périodiques; Electronic journals; 616.9943105
• DOI: 10.1016/j.oraloncology.2019.09.022
• Languages: English
• ISSNs: 1368-8375

Abstract

Highlights: Radiomic signatures of regional lymph node (RSnd), Dose Volume Histogram signature reflecting planning score (PS), and TNM stage were independent predictors for PFS. The nomogram integrating the RSnd, PS and TNM stage can accurately predict 3-year and 5-year PFS. Subgroup analysis showed Epstein–Barr virus DNA status improved the predictive power of the nomogram. The nomogram can successfully classify patients into low- and high-risk groups, with significantly different PFS. Abstract: Objectives: To develop a multidimensional nomogram for predicting the progression-free survival (PFS) in patients with locoregionally advanced nasopharyngeal carcinoma (NPC) (stage III-IVa). Materials and methods: A total of 224 patients with locoregionally advanced NPC (training cohort, n = 149; validation cohort, n = 75) were retrospectively included. We extracted 260 radiomic features from the primary tumor and lymph nodes on the axial contrast-enhanced T1 weighted and T2 weighted MRI. Radiomic signatures of the gross tumor volume (RSnx) and lymph node (RSnd), Dose Volume Histogram (DVH) signature reflecting planning score (PS), and clinical characteristics were included as potential predictors of PFS. The least absolute shrinkage and selection operator (LASSO) regression were applied for feature selection and data dimension reduction. A nomogram was developed by incorporating the selected predictors. The C-index and calibration curve were used to assess discrimination and calibration power of the nomogram, respectively. Results: RSnd, PS, and tumor-node-metastasis (TNM) stage were the independent predictors for PFS (all p < 0.05). The nomogram integrating the three factors achieved a C-index of 0.811 (95% CI: 0.74–0.882) in the validation cohort for predicting PFS, which outperformed than that of the TNM stage alone (C-index, 0.613, 95% CI: 0.532–0.694). Subgroup analysis showed Epstein–Barr virus (EBV) DNA status improved the predictive accuracy of the nomogram (C-index, 0.86, 95% CI: 0.787–0.933). Conclusions: The multidimensional nomogram incorporating RSnd, PS, and TNM stage showed high performance for predicting PFS in patients with locoregionally advanced NPC.