AUTHOR=Fang Zeng-Yi , Li Ke-Zhen , Yang Man , Che Yu-Rou , Luo Li-Ping , Wu Zi-Fei , Gao Ming-Quan , Wu Chuan , Luo Cheng , Lai Xin , Zhang Yi-Yao , Wang Mei , Xu Zhu , Li Si-Ming , Liu Jie-Ke , Zhou Peng , Wang Wei-Dong 

TITLE=Integration of MRI-Based Radiomics Features, Clinicopathological Characteristics, and Blood Parameters: A Nomogram Model for Predicting Clinical Outcome in Nasopharyngeal Carcinoma

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.815952

DOI=10.3389/fonc.2022.815952

ISSN=2234-943X

ABSTRACT=Purpose: To set up a nomogram model based on multiparametric magnetic resonance imaging (MRI) radiomic features, clinicopathological and blood parameters to predict the progression-free survival (PFS) for patients with nasopharyngeal carcinoma (NPC).
Methods: 462 patients with pathologically confirmed non-keratinizing carcinoma were recruited in Sichuan Cancer Hospital from 2015 to 2019 and divided into training and validation cohorts in a ratio of 7:3. The least absolute shrinkage and selection operator (LASSO) algorithm was used for radiomics features dimension reduction and screening in the training cohort. Rad-score, age, sex, smoking habit, drinking habit, Ki-67, monocytes (MONO), monocyte ratio (MONO%), and mean corpuscular volume (MCV) were incorporated into the multivariate Cox proportional risk regression to build a multifactorial nomogram model. The concordance index (C-index) and decision curve analysis (DCA) apply to estimate the efficacy of the model.
Results: 9 significant features associated with PFS were selected by LASSO and constructed the rad-score of each patient. Rad-score was verified to be an independent prognostic factor for PFS in NPC. The survival analysis showed that those with lower rad-scores obtained longer PFS in both cohorts (P＜0.05). Compared with the TNM staging system, the multifactorial nomogram got the highest C-index (training cohorts: 0.819 vs. 0.610; validation cohorts: 0.820 vs. 0.602). Moreover, the DCA curve showed that this model could better predict progression within 50% threshold probability.
Conclusion: Nomogram that combined MRI-based radiomics with clinicopathological and blood parameters improved predicting ability of progression in patients with nasopharyngeal carcinoma.