Relationship Between FGF/FGF4/FGF19 Co-amplification In Hepatocellular Carcinoma Tissues And Patients’ Clinical Phenotype And Prognosis

Purpose:The aim of this study was to investigate the distribution of FGF3/FGF4/FGF19 coamplification in Chinese patients with hepatocellular carcinoma（HCC）,to explore the relationship between the occurrence of FGF3/FGF4/FGF19coamplification and clinicopathological characteristics and prognosis in HCC patients,and then to provide new ideas for clinical management of HCC with different genetic status.Method:From April 1,2016 to December 30,2019,this study collected a total of 253tissue specimens,which came from patients who received radical hepatectomy for HCC and consented to accept the next-generation sequencing in the Affiliated Hospital of Qingdao university.This study used the QIAamp^?DNA FFPE Tissue Kit（Qiagen,Hilden,Germany）to extract DNA from FFPE tumors and paired paracancerous tissues,respectively,and performed sequencing and genomic analysis with Yuansu 450^TM(Origi Med^?,Shanghai,China).The genes in the samples were captured and sequenced by the Nova Seq 6000（Illuminaintegrated,CA,USA）sequencing platform at an average depth of 1000×.Single nucleotide variants（SNVs）,insertion/deletion,copy number variations（CNVs）,gene rearrangements and fusions were the main genetic variants of our concern.At the same time,clinicopathological data and variant gene information of patients were collected and categorized.Disease-free survival（DFS）,survival time after relapse（SAR）,and overall survival（OS）were considered as the endpoints of this study.Resectable recurrent HCC（r-HCC）patients with well-preserved liver function（Child-Pugh A）and with indocyanine green（ICG-R15）<30%underwent hepatectomy again,unresectable patients with Child-Pugh A or B liver function,multiple intra-or extrahepatic tumors,and with large vessel invasion were treated with transarterial chemoembolization（TACE）combined with targeted therapy;those who progressed on targeted therapy or those who cannot tolerate it were treated with immunotherapy;patients who were unwilling or unable to receive surgical treatment or systemic therapy（advanced age,poor economic condition,etc.）were treated with TACE alone.All clinical data and follow-up information of the patients were obtained through His system in our hospital and telephone callbacks.Differences in the distribution of variant events（FGF3/FGF4/FGF19 coamplification）between groups of categorical variables were tested by Pearsonχ²test or,if not applicable,by continuous correctedχ²test or Fisher’s exact probability method.The cumulative recurrence rate and cumulative survival probability were estimated by the life table method,DFS,SAR,and OS were calculated by the Kaplan-Meiers method and compared with the log-rank test.Independent risk factors affecting DFS after radical resection in HCC patients were identified by multivariable Cox proportional hazards regression model based on factors with P<0.1 in the univariate analysis.All the data analysis in this study was performed by SPSS 24.0 statistical software（IBM Corp,NY,USA）.The two-tailed P<0.05 was considered to be a statistically significant difference.Results:Alterations in FGF3,FGF4 and FGF19 were observed in 30 of 253 cancer tissues.Independent variant of FGF19（four with gene amplification,two with point mutations:c.398 T>C,C.214 C>T）was observed in 6 cases,independent variant of FGF4（SNV,A:T>C:G）was observed in one case,no independent variant of FGF3was observed,common variant of FGF3,FGF4,and FGF19（CNV,All three are gene amplification）was observed in 23 cases.The incidence of the above co-variant was about9.09%（23/253）.FGF3/FGF4/FGF19 coamplification occurred mainly in the chromosome 11q13.3 region and was usually accompanied by amplification of the CCND1（χ²=43.481,P<0.000）.The variants of FGF3,FGF4,and FGF19 were not detected in the corresponding 253 paracancerous tissues.The top ten variant genes in our HCC samples included TP 53（56.13%）,TERT（42.69%）,CTNNB 1（24.11%）,AXIN 1（15.42%）,ARID 1A（11.46%）,RB 1（11.07%）,TSC 2（10.67%）,ATM（9.09%）,CCND 1（4.74%）,MYC（3.95%）.Our study found that the co-amplification frequency of FGF3/FGF4/FGF19 was significantly higher in patients with CD 10（-）,TP 53 mutation and CCND 1 amplification（P<0.05）.However,there was no significant difference in the distribution of FGF3/FGF4/FGF19 coamplification among the group of age,sex,blood type,BMI,antivirals,smoking,alcoholism,hypertension,diabetes,hepatitis B,hepatocirrhosis,portal hypertension,preoperative ascites,family history of malignancy,Child-Pugh classification,ALT,AST,STB,GGT,ALP,AFP,PNI,CNLC stage,number of tumor,maximum tumor diameter,degree of tumor differentiation（Edmondson classification）,MVI,PVTT,satellite lesions,tumor necrosis,TERT mutation,CTNNB1mutation,AXIN1 mutation and TMB（P>0.05）.The differences in DFS and OS between patients with FGF3/FGF4/FGF19 coamplification and patients with wild-type FGF3/FGF4/FGF19 were estimated by the Kaplan-Meier method,and the results indicated that patients in the group with FGF3/FGF4/FGF19 coamplification had a significantly shorter median DFS than those in the group without FGF3/FGF4/FGF19coamplification（10.5 months vs.43.6 months,P=0.003）,and that the 1-,2-,and 3-year cumulative OS rates in the group with FGF3/FGF4/FGF19 coamplification were89.1%,76.2%,and 68.6%,respectively,which were significantly lower than those in the group without FGF3/FGF4/FGF19 coamplification（98.5%,92.0%,and 84.6%,respectively;P=0.02）.Based on the factors affecting DFS of HCC patients in univariate analysis（P<0.1）,multivariate Cox proportional hazards regression model suggested that multiple tumor（HR=2.457,95%CI=1.441-4.188,P=0.001）and coamplification of FGF3/FGF4/FGF19（HR=2.114,95%CI=1.028-4.347,P=0.042）were independent risk factors that could predict poor DFS after surgical radical resection in HCC patients.For patients in the group of FGF3/FGF4/FGF19 coamplification,serum AFP≥400 ng/ml,multiple tumors,no antiviral therapy with nucleoside analogs,and preoperative ALP>125 U/L were important factors affecting their poor postoperative DFS（P<0.05）.The median follow-up time of this study was 26.6 months（range,6.0 months-64.4 months）.During the follow-up,91 cases recurred,with a recurrence rate of 62.6%（91/243）.The median SAR for these 91 HCC patients was 34.9months（95%CI:20.05 months-49.75 months）.R-HCC Patients with FGF3/FGF4/FGF19 coamplification responded better to TACE combined with targeted therapy than r-HCC patients without FGF3/FGF4/FGF19 coamplification（P=0.025）,whereas r-HCC patients with FGF3/FGF4/FGF19 coamplification responded worse than r-HCC patients without FGF3/FGF4/FGF19 coamplification to TACE alone treatment（P=0.000）or immunotherapy（P=0.025）.Conclusion:FGF3/FGF4/FGF19 coamplification was more common in HCC tumor tissues with CD 10（-）,TP 53 mutation and CCND 1 amplification,while this study did not find a correlation between the amplification status of FGF3/FGF4/FGF19 and any clinical parameters of patients.The amplification status of FGF3/FGF4/FGF19 in HCC tumor tissue was strongly associated with DFS and OS of HCC patients after surgical radical resection.R-HCC patients with FGF3/FGF4/FGF19 coamplification responded better to TACE combined with targeted therapy than r-HCC patients without FGF3/FGF4/FGF19 coamplification,whereas r-HCC patients with FGF3/FGF4/FGF19 coamplification responded worse than r-HCC patients without FGF3/FGF4/FGF19 coamplification to TACE alone treatment or immunotherapy,and our findings may provide new insights into clinical management strategies for HCC.