| Background:Colorectal cancer is one of the common malignant tumors of digestive tract,which is a serious threat to human health.The incidence of colorectal cancer accounts for the fourth place of malignant tumors in China,and the mortality rate accounts for the fifth place of malignant tumors.It is estimated that most CRC deaths are due to distant metastasis,and the liver is the most common metastatic site of colorectal cancer.About 15%of CRC patients have liver metastasis at the time of diagnosis,and about 60%of metastatic CRC patients will have liver metastasis.Liver metastasis is the most common form of metastasis of colorectal cancer.The prognosis of colorectal cancer patients with liver metastasis varies greatly.About half of the patients relapse within 3 years after hepatectomy with poor prognosis and short survival time.However,the survival time of some patients can be more than 5 years,or even longer.In order to obtain better clinical treatment results,it is necessary to classify the patient population and distinguish high-risk patients to choose more active anti-tumor therapy,but there is relatively little analysis on the prognostic risk factors of colorectal cancer patients with liver metastasis.In 2011,Robert A Professor Weinberg published a review in Cell magazine,and listed the phenotype of tumor metabolic abnormali ties as one of the top ten characteristics of tumors.Metabolic abnormalities are the risk factors for many tumors,especially gastrointestinal tumors,proposed in recent years.Long non-coding RNA(lncRNA)is a new hot regulatory molecule in basic medical research.It has spatio-temporal and tissue specific expression patterns,participates in a variety of metabolic related signal pathways and further regulates metabolic processes.For example,lncRNA CCAT2(colorectal cancer related transcript 2)regulates cancer metabolism in an allelic specific manner;LncRNAAT 102202 can regulate the expression of HMGCR and participate in cholesterol metabolism;LncRNAHULC promotes the accumulation of triglycerides and cholesterol in cells by inducing methylation of target gene CpG site,increasing the expression of PPARA and ACSL1.Therefore,metabolism related lncRNA has shown great prospects in tumor therapy,and the research around it is also a hot topic in recent years.Tumor is a kind of genetic disease.Whether the clinical manifestation of metabolic abnormalities or the specific biological process regulated by lncRNA,it is realized in the unrestricted cloning process of tumor cells.Some studies have explored the clonal evolution process of metastatic colon cancer,and found that the metastasis process follows the monoclone and polyclonal patterns,and different cloning patterns have obvious correlation with prognosis.Then further in-depth study on the clonal evolution model of the phenotype of liver metastasis of colon cancer can not only describe the evolution process of primary and metastatic lesions,explore their relationship with prognosis,but also summarize the root causes of tumor drug resistance,recurrence and metastasis,then find potential treatment methods.However,there are relatively few studies on the liver metastasis of colon cancer,especially the clonal evolution model without the influence of treatment pressure.There are no studies to analyze the relationship between metabolic abnormalities,clonal evolution and the prognosis of colorectal cancer patients with liver metastasis.In this study,(1)the clinical characteristics of abnormal metabolism such as diabetes,coronary heart disease and blood lipids were analyzed for the first time to explore their effects on the prognosis of these patients.(2)At the level of gene expression,we tried to further verify the results of clinical analysis by studying the relationship between metabolic lncRNA and the prognosis of colon cancer.The genes in the prediction model were detected by RT-PCR with colon cancer tissue samples,and the relationship between the expression level and prognosis was confirmed,then proved by in vitro experiments.(3)Finally,the first part of the study continued to screen the patients who had not received any anti-tumor therapy,multiple liver metastasis,primary liver metastasis and multiple liver metastasis at the same time and had complete follow-up data.The whole exon sequencing technique supported by NGS was used to continue analyze the caracteristics of CRLM,on the level of gene mutation.The heterogeneity of CRLM was shown from the aspects of gene,chromosome variation and subcloning;the clonal evolution map,(phylogenetic tree)was described;the metastasis pathway was speculating,and the biological function of subcloning and its effect on prognosis was analyzed finally.Methods:1.Prognosis analysis of colorectal Liver Metastasis(1)Patients diagnosed with CRLM from 2010 to 2018 at Shandong Cancer Hospital were identified.(2)Clinic characteristics,blood biochemical data and survival data were collected from the electronic medical record.After the preliminary analysis of all the input data,combined with the total number of patients,in order to ensure the sample size of statistics,this study finally makes a follow-up analysis of some of the input parameters.(3)The effect of prognostic factors on overall survival(OS)were analyzed by univariate and multivariate Cox hazard analysis.2.Study on metabolism related-lncRNAs as potential signature in prognosis of colon cancer(1)In the current study,the transcriptome data and clinical data of colon cancer was downloaded from The Cancer Genome Atlas(TCGA).Metabolism-related gene sets were downloaded from the Molecular Signatures Database(MSigDB).Differential lncRNAs related to metabolism was obtained by performing the correlations between differential expression profile of metabolic genes and lncRNAs.(2)To construct a prognostic model of CC based on metabolism-related lncRNAs,we divided patients,whose clinical data were available,into a training set and a validation set at a ratio of 7:3.The prognostic metabolism related-lncRNA signature was established using the training set by univariate and multivariate Cox regression analysis,and the validation set was used to test the capacity of the prognostic model.(3)All patients included in the study were divided into high and low risk groups using the prognostic model,the correlation between risk score and clinicopathological characteristics of the two groups were further analyzed.GSEA pathway enrichment analysis was investigated for the high-and low-risk groups.(4)GO and KEGG analysis was carried out using the gene set correlated with the identified lncRNA,to observe which metabolism-related biological processes are closely related to the lncRNAs in this study.(5)The four lncRNAs in the risk model of this study were further explored,paired paraffin-embedded colon cancer tissues and normal paracancerous tissues were identified,and samples were screened by concentrations for further PCR testing,then prognostic analysis of patients was performed by relative expression.(6)Finally,in vitro cellular assays were performed to verify the effect of the screened genes on cell proliferation and migration ability.3.Study on the clonal evolution of liver metastatic colon cancer(1)Screening after the expansion of the research objects in the first part,to collecte patients who were diagnosed with primary tumors and multiple metastases simultaneously and treatment-naive before operation.Collect of postoperative tissue wax blocks including primary colon tumors and multiple liver metastases.(2)Whole exome sequencing was performed on all samples using the next-generation sequencing platform.(3)A variety of biological information software was used to analyze the mutations of the primary tumor and metastasis at the SNV and CNV aspects,showing the specific manifestations of tumor heterogeneity.(4)Intra-and intertumor heterogeneity were calculated by the computational method PyClone.LICHeE(Lineage Inference for Cancer Heterogeneity and Evolution)was implied to reconstruct the cancer phylogeny trees and infer the pathways of colon cancer liver metastases.(5)To perform biological function analysis of subclones in the phylogenetic tree to show the biological roles of subclones with different characteristics.(6)After analysis of the mutated genes contained in the clusters identified by PyClone,8 patients were divided into 2 groups,and the prognosis analysis was implied.Results:1.Prognosis analysis of colorectal liver metastasis(1)A total of 503 CRLM patients were included in this study,among the patients in the primary cohort,median age was 58 years,ranging from 27 to 81 years.There were 332(66.0%)male and 171(34.0%)female patients.(2)The liver metastasis detected interval from the primary tumor diagnosis;liver metastasis number;with or without dyslipidemia;liver metastasis occurred with or after primary tumor diagnosis and gender were associated with outcomes in univariate analysis only.(3)Female,liver metastasis number≥3,liver metastasis detected interval from the primary tumor diagnosis>2,with dyslipidemia,prealbumin≤0.225 and liver metastasis occurred after primary tumor diagnosis were associated with poor prognosis.(4)Multivariate analysis included the three prognostic significance factors,including liver metastasis number,prealbumin and dyslipidemia.(5)Risks associated with combinations of risk factors were calculated from each multivariable model High-risk patients exhibited a poorer prognosis than low-risk patients did.The liver metastasis detected interval from the primary tumor diagnosis was still not a prognostic factor in the dyslipidemia cohort.2.Study on metabolism related-lncRNAs as potential signature in prognosis of colon cancer(1)We identified four lncRNAs were finally identified to build a signature,which was verified the effectiveness by the TCGA validation set.The risk score=(-0.064×expression value of AC124067.4)+(0.291×expression value of AL161729.4)+(0.478 × expression value of AP001469.3)+(0.126 × expression value of PCAT6).(2)The median risk score is the threshold,according to which patients in the training and validation sets are divided into high-risk and low-risk groups.The high-risk group had poorer OS(p<0.001).Applying the predictive model to predict 1-,3-,and 5-year survival rates,the area under the curve(AUC)was 0.701,0.708,and 0.646 in the training set and 0.615,0.7,and 0.652 in the validation set,respectively.The PCA analysis showed that the risk model could clearly distinguish samples from high and low risk groups.(3)The multivariate Cox regression analysis showed that the risk score,age of diagnosis and T stage were independent prognostic factor for CC patients.The C index of the nomogram is 0.762(95%CI=0.735-0.789),which predicts well.(4)Phosphatidylglycerol is the major classes of glycerophospholipids comprise,whose metabolic process upregulated in high-risk group.(5)Lipid metabolism ranked first in all the metabolic processes involved in the four lncRNA.(6)In paraffin-embedded tissues,the expression of AC124067.4,AP001469.3 and PCAT6 was significantly higher in tumor tissues than in normal tissues adjacent to the cancer,and the prognosis of patients with high expression of AC 124067.4 was significantly better than that of patients with low expression,while the prognosis of patients with high expression of AL161729.4 was worse.(7)In vitro experiments initially demonstrated that AC 124067.4 could inhibit the proliferation,clone formation and migration ability of colorectal cancer cells.3.Study on the clonal evolution of liver metastatic colon cancer(1)A total of 30 samples from 8 colon cancer patients,including matched two to four metastases,underwent next-generation whole-exome DAN sequencing.(2)The mutations differed significantly not only between primary and metastases but also among the multiple metastases in all the eight patients.(3)Heat map analysis of the correlation between mutation characteristics and COSMIC characteristics showed that all samples were clustered on Signature1,but differences were reflected in Signature3,Signature5,Signature6,Signature15 and Signature16.(4)The identification and analysis of high-frequency CNVs revealed that the distribution of high-frequency copy number amplification and deletion in primary colon cancer and liver metastases was different.(5)Intratumor heterogeneity was demonstrated by calculating the MATH value for each sample.It was found that the number and distribution of subclones calculated for each patient’s primary tumor and each metastatic tumor were different.(6)Using Pyclone to calculate intratumor and intertumor heterogeneity,not only demonstrated intratumor heterogeneity,but also found obvious intertumor heterogeneity.(7)Three different models of phylogenetic tree were identified,including parallel evolution,linear evolution and branching evolution.The liver metastases can be seeded from primary or other metastasis.(8)Pathway and process enrichment analysis showed that the transition and ancestor subclons exerted their different functional effects during metastatic evolution.Intrahepatic specific subclones were also significantly enriched in cholesterol transport and efflux,bile secretion and other pathways.(9)Pathway and process enrichment analysis has been carried out on the clustered mutation list in the result of PyClone,two cohorts was identified,and their survival were significantly different.Conclusion:1.There is heterogeneity in liver metastases from colorectal cancer.Dyslipidemia,the number of liver metastases and the level of prealbumin are independent prognostic risk factors for CRLM.At the clinicopathological level,it was found that abnormal lipid metabolism may affect the prognosis of colorectal cancer liver metastases,which will prompt further research.2.At the gene expression level,we proved that abnormal lipid metabolism affects the prognosis of colon cancer.A signature consisting of 4 metabolism-related lncRNAs was developed to predict the prognosis of colon cancer,which combined with several features that are more readily available in the clinic may help to select high-risk patients who need more aggressive treatment or intervention.AC 124067.4 can inhibit the proliferation and migration of colorectal cancer cells,and its potential mechanism will continue to be explored in the future,which will eventually provide a reference for the clinical treatment of colorectal cancer.3.At the level of gene mutation,this study comprehensively described the significant intratumoral and intertumoral heterogeneity of colon cancer liver metastases;demonstrated the clonal evolution process and metastasis pathway of liver metastases;found that intrahepatic specific subclones were involved in cholesterol transport and efflux,bile efflux,bile The significant enrichment of the secretory pathway suggests that abnormal lipid metabolism may play an important role in the colonization and growth of subclones in the liver;after clustering and grouping by cell cluster function,the prognosis of different patient cohorts is significantly different.Originality and Significance:1.First,it was proposed and proved that abnormal lipid metabolism has a significant impact on the prognosis of colorectal cancer patients with liver metastases.2.For the first time,metabolism-related lncRNAs were used to construct a colon cancer prognostic signature.Only four lncRNAs were used in combination with the tumor stage,age,and T stage that were easily obtained in the clinic to induce a nomogram for predicting the survival rate.The C-index of the nomogram was 0.762,which predicted excellent results.3.For the first time,the results of whole-exome sequencing of tissue paraffin specimens without any anti-tumor treatment before surgery,multiple liver metastases,primary tumors and multiple liver metastases were resectioned at the same time were used to analyze the heterogeneity of colon cancer liver metastases,and to analyze the heterogeneity of liver metastases from colon cancer.The phylogenetic tree has shown and the results of functional clustering of cell clusters were used for prognostic analysis.It is hoped that these results can continue to be validated in larger samples,ultimately advancing the development of new treatments and further improving patient outcomes. |
PDF Full Text Request