Nomograms combining computed tomography-based body composition changes with clinical prognostic factors to predict survival in locally advanced cervical cancer patients

Article type: Research Article

Authors: Fu, Baoyue^{a; 1} | Wei, Longyu^{a; 1} | Wang, Chuanbin^b | Xiong, Baizhu^a | Bo, Juan^b | Jiang, Xueyan^a | Zhang, Yu^a | Jia, Haodong^b | Dong, Jiangning^{a; b; *}

Affiliations: [a] Bengbu Medical College, Bengbu, Anhui, China | [b] Department of Radiology, the First Affiliated Hospital of University of Science and Technology of China, Hefei, Anhui, China

Correspondence: [*] Corresponding author: Jiangning Dong. E-mail: dongjn@ustc.edu.cn.

Note: [1] These authors contributed equally to this study.

Abstract: OBJECTIVE:To explore the value of body composition changes (BCC) measured by quantitative computed tomography (QCT) for evaluating the survival of patients with locally advanced cervical cancer (LACC) underwent concurrent chemoradiotherapy (CCRT), nomograms combined BCC with clinical prognostic factors (CPF) were constructed to predict overall survival (OS) and progression-free survival (PFS). METHODS:Eighty-eight patients with LACC were retrospectively selected. All patients underwent QCT scans before and after CCRT, bone mineral density (BMD), subcutaneous fat area (SFA), visceral fat area (VFA), total fat area (TFA), paravertebral muscle area (PMA) were measured from two sets of computed tomography (CT) images, and change rates of these were calculated. RESULTS:Multivariate Cox regression analysis showed ΔBMD, ΔSFA, SCC-Ag, LNM were independent factors for OS (HR = 3.560, 5.870, 2.702, 2.499, respectively, all P < 0.05); ΔPMA, SCC-Ag, LNM were independent factors for PFS (HR = 2.915, 4.291, 2.902, respectively, all P < 0.05). Prognostic models of BCC combined with CPF had the highest predictive performance, and the area under the curve (AUC) for OS and PFS were 0.837, 0.846, respectively. The concordance index (C-index) of nomograms for OS and PFS were 0.834, 0.799, respectively. Calibration curves showed good agreement between the nomograms’ predictive and actual OS and PFS, decision curve analysis (DCA) showed good clinical benefit of nomograms. CONCLUSION:CT-based body composition changes and CPF (SCC-Ag, LNM) were associated with survival in patients with LACC. The prognostic nomograms combined BCC with CPF were able to predict the OS and PFS in patients with LACC reliably.

Keywords: Locally advanced cervical cancer, quantitative computer tomography-based body composition, survival

DOI: 10.3233/XST-230212

Journal: Journal of X-Ray Science and Technology, vol. 32, no. 2, pp. 427-441, 2024