Incorporating Inflammatory Markers and Clinical Indicators into a Predictive Model of Single Small Hepatocellular Carcinoma Recurrence After Primary Locoregional Treatments

Abstract

Wenying Qiao,1– 3,* Yiqi Xiong,1,* Kang Li,4 Ronghua Jin,2,3 Yonghong Zhang1,5 1Interventional Therapy Center for Oncology, Beijing You’an Hospital, Capital Medical University, Beijing, 100069, People’s Republic of China; 2National Center for Infectious Diseases, Beijing Ditan Hospital, Capital Medical University, Beijing, 100015, People’s Republic of China; 3Changping Laboratory, Beijing, 102206, People’s Republic of China; 4Research Center for Biomedical Resources, Beijing You’an Hospital, Capital Medical University, Beijing, 100069, People’s Republic of China; 5Beijing Research Center for Respiratory Infectious Diseases, Beijing, 100006, People’s Republic of China*These authors contributed equally to this workCorrespondence: Yonghong Zhang, Interventional Therapy Center for Oncology, Beijing You’an Hospital, Capital Medical University, 8 Xitoutiao, Youanmenwai Street, Fengtai District, Beijing, People’s Republic of China, Tel [phone], Email [email protected] Ronghua Jin, National Center for Infectious Diseases, Beijing Ditan Hospital, Capital Medical University, 8 Jingshundong Street, Chaoyang District, Beijing, People’s Republic of China, Tel [phone], Email [email protected]: We explored the role of tumor size and number in the prognosis of HCC patients who underwent ablation and created a nomogram based on machine learning to predict the recurrence.Patients and Methods: A total of 990 HCC patients who underwent transcatheter arterial chemoembolization (TACE) combined ablation at Beijing Youan Hospital from January 2014 to December 2021 were prospectively enrolled, including 478 patients with single small HCC (S-S), 209 patients with single large (≥ 30mm) HCC (S-L), 182 patients with multiple small HCC (M-S), and 121 patients with multiple large HCC (M-L). S-S patients were randomized in a 7:3 ratio into the training cohort (N=334) and the validation cohort (N=144). Lasso-Cox regression analysis was carried out to identify independent risk factors, which were used to construct a nomogram. The performance of the nomogram was evaluated by C-index, receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA) curves. Patients in the training and validation cohorts were divided into low-risk, intermediate-risk, and high-risk groups based on the risk scores of the nomogram.Results: The median recurrence-free survival (mRFS) in S-S patients was significantly longer than the S-L, M-S, and S-L patients (P< 0.0001). The content of the nomogram includes age, monocyte-to-lymphocyte (MLR), gamma-glutamyl transferase-to-lymphocyte (GLR), International normalized ratio (INR), and Erythrocyte (RBC). The C-index (0.704 and 0.71) and 1-, 3-, and 5-year AUCs (0.726, 0.800, 0.780, and 0.752, 0.761, 0.760) of the training and validation cohorts proved the excellent predictive performance of the nomogram. Calibration curves the DCA curves showed that the nomogram had good consistency and clinical utility. There were apparent variances in RFS between the low-risk, intermediate-risk, and high-risk groups (P< 0.0001).Conclusion: S-S patients who underwent ablation had the best prognosis. The nomogram developed and validated in the study had good predictive ability for S-S patients.Keywords: hepatocellular carcinoma, HCC, transcatheter arterial chemoembolization, TACE, ablation, nomogram, recurrence

Keywords

• hepatocellular carcinoma
• hcc
• transcatheter arterial chemoembolization
• tace
• ablation
• nomogram
• recurrence