Abstract: Objective To investigate  the value of plasma cell⁃free DNA (cfDNA) fragmentomics in the prognostic analysis of hepatocellular carcinoma (HCC). Methods A total of 50 HCC patients admitted to the Department of Hepatobiliary Surgery, the First Affiliated Hospital of Fourth Military Medical University from June 2016 to September 2016 were selected as the subjects of study. Preoperative blood of HCC patients was collected for cfDNA extraction, library construction, and cfDNA sequencing. The cfDNA fragmentomic characteristics of HCC patients were analyzed, including fragment size distribution and terminal base motifs. The clinical data of patients combined with cfDNA fragmentomic data was performed by Cox regression analysis to screen independent factors affecting the prognosis of HCC patients, and a nomogram model was constructed to predict the overall survival rate of HCC patients. Results The cfDNA fragment sizes were mainly in the range of  150-200 bp, and the peak was at 167 bp; and continuous peak waves with an interval of about 10 bp appeared between 100-200 bp. Comparing the end motifs of total fragments, the differences were statistically significant (all P<0.001), except for the differences between the mean percentages of  A⁃end and G⁃end, and T⁃end and C⁃end (all P>0.05). Meanwhile, the differences between the mean proportion of each end motifs in the short fragments and the long fragments were also statistically significant (all P<0.001). The tumor differentiation, BCLC stage and fragment size score were independent factors affecting the prognosis of HCC patients(all P<0.05). Based on the above three factors to construct the nomogram model, the area under the receiver operating characteristic (ROC) curve for predicting  the 1⁃ and 3⁃year overall survival rate of HCC patients was 0.836 and 0.840, respectively. Conclusions  The combination of cfDNA fragmentomics and clinical indicators have great potential in the prognostic assessment of HCC. 

Key words: Hepatocellular carcinoma, Cell?free DNA, Fragmentomics, Prognosis