| Part I:Multi-omics analysis reveals the evolutionary mechanism of esophageal squamous cell carcinoma and dysplasiaBackground:Esophageal cancer is a malignant tumor with the eighth morbidity and sixth mortality around the world.Esophageal squamous dysplasia is known as the precancerous lesion of esophageal squamous cell carcinoma(ESCC),with a high risk of developing into ESCC,and the risk increases with the levels of dysplasia.Deep research into the carcinogenesis of ESCC helps us to identify molecular evolutionary mechanism,find the transition events and its occurrence time,so that we can reduce the risk and provide clinical strategy.In order to comprehensively observe the mutation spectrum and copy number alterations of esophageal squamous dysplasia compared with ESCC,in recent years,the researchers performed whole exome sequencing(WES)on the precancerous and matched cancerous tissues of the same patients,and conducted preliminary exploration.However,synchronous precancer/cancer pair studies do not always exactly reflect the temporal evolution underlying neoplastic transformation,for example,molecular alterations found in precancerous lesions adjacent to tumor are not comparable to those found in dysplasia which never progressed.Previous studies still lack analysis on samples of the different stages progressed along with time of the same person,which is the ideal way to study cancer development.Besides,conjoint multiple omics analysis about dyspasia and ESCC has not been reported,most of related researches are still limited to the genomic level,and transcriptome is rarely involved.Aim:In order to illustrate the molecular evolutionary mechanism of ESCC and precancerous lesions and provide new ideas and basis for the early diagnosis and treatment of ESCC,we analysed data of WES on temporal samples(samples of different stages of the same person developing with time,including precancerous tissues,cancerous tissues and matched morphology normal tissues)and spacial samples(tumor samples,adjacent precancerous samples and matched morphologically normal tissues of the same person).On this basis,we had further done WES,whole genome sequencing(WGS),transcriptome sequencing and circulating tumor DNA(ctDNA)sequencing of multi-stage samples from different people.Materials and methods:We collected temporal samples from 3 patients,spacial samples from 23 patients,and performed WES.On this basis,WGS,transcriptome sequencing and ctDNA sequencing were performed on multi-stage lesions of ESCC.We validated the findings in the temporal cohort and spacial cohort and further explored the molecular evolutionary mechanism of neoplastic transformation from the multi-omics perspective.Results:The precancerous lesions and cancer tissues in the temporal cohort show continuation of some specific mutant clones,indicating that the precancerous lesions have a continuous evolutionary relationship with the corresponding cancer tissues.Two of the three temporal cohort patients have TP53 mutation in the precancerous lesions,and the mutation of TP53 continues to expanding to cancer.Unlike precancerous tissues,genome doubling(GD)appears in the cancer tissues of the 3 patients in temporal cohort.In the spacial cohort,the number of non-synonymous mutations of driver genes in precursor lesions is similar to that of cancer tissues.While the number of driver genes involved in copy number variations in cancerous lesions is significantly more than that of precancerous lesions.GD exists frequently in precursor lesions of spacial cohort,and its frequency is similar with matched cancer tissues.In multi-stage samples,the number of structural variations(SV)and driver genes involved in SV in precancerous and cancerous tissues is similar.The frequency of GD increased in phase I,but afterwards the frequency dose not increase with clinical stage.Genes related to proliferation and division are mainly highly expressed in phase I,while the expression of genes related to immunity and epithelial-mesenchymal transition gradually increase with clinical stage.Compared with samples without GD,the samples with GD show a more serious immune disorder,in which the fraction of neutrophils increases significantly,and the fraction of M1 type macrophages decreases significantly.Then we focused on the samples with GD and non-GD in phase I,and found that the two have different immune microenvironments.Among the phase I samples with GD,ratios of M2 type macrophages,activated dendritic cells,and resting CD4 memory T cells significantly increase(p values are 0.035,0.033,0.001,wilcox.test),and the proportion of follicular helper T cells and memory B cells decrease significantly(p values are 0.001,0.044,wilcox.test).And compared the precancerous lesions with GD with ESCC in phase T1a without GD,we can still get the similar results.The ctDNA sequencing results show that the detection rate of ctDNA mutation in patients with precancerous lesions(41.7%)is even slightly higher than patients with early cancer(31.6%).Conclusions:1.TP53 mutation is an early driving event of canceration.Compared with cancerous tissues,there is no significant difference in the number of driver gene mutations in precancerous lesions.But the copy number variations in cancer tissues involve significantly more driver genes than precancerous tissues.The number of SV and frequency spectrum of driver genes involved in SV in precancerous lesions and cancer tissues are similar.2.In the process of canceration,the peak expression of genes related to cell proliferation and division occurs in the early stage of cancer,and the expression of genes related to immunity and epithelial-mesenchymal transition gradually increase with clinical stage3.GD may be a transformation event in the carcinogenesis of ESCC4.GD tends to occur relatively early in the development of cancer,and may be related to errors in cell proliferation and division5.Compared with samples without GD,samples with GD have a more malignant immune microenvironment,and GD related changes in the immune microenvironment have already occurred in precancerous lesions with GD,suggesting that precancerous lesions with GD already have the potential for malignant transformation6.The detection rate of ctDNA mutations in patients with precancerous lesions is not lower than patients with early cancer,which indicates that in the precancerous stage,abnormal cells carrying mutations have already entered the blood.Part ?:Effect of PD-1 inhibitor on immune status of peripheral blood in cancer patientsBackground:One of the important mechanisms of tumor immune escape is the immune suppression caused by immune checkpoint PD-1/PD-L1 in tumor microenvironment.PD-1/PD-L1 inhibitor is a major milestone in the history of tumor immunotherapy.However,there are still some problems in the clinical application of PD-1 inhibitors,such as some side effects.In addition to tumor microenvironment,the overall immune status of the host also plays a great role in anti-tumor process.However,the effect of PD-1 inhibitors on the overall immune status of cancer patients still remains unclear.Objective:In order to reveal the effect of PD-1 inhibitors on the systemic immune status of cancer patients,strengthen the understanding of the efficacy of PD-1 inhibitors and guide the use of immunotherapy,we conducted this study.Materials and methods:We enrolled 7 peripheral blood samples of cancer patients before and after PD-1 inhibitors treatment,and performed peripheral blood mononuclear cell transcriptome sequencing and TCR immune repertoire sequencing.Results:After PD-1 inhibitor treatment,the main components of the expression profile of peripheral blood mononuclear cells show similar changes.After medication,the up-regulated genes in peripheral blood mononuclear cells are mainly enriched in lymphocyte-mediated immunity,NK cell-mediated immunity,IL-2 secretion,positive regulation of immune response,cell killing,Th1 type immune response and other biological pathways.Down-regulated genes are mainly enriched in biological pathways such as humoral immune response,complement activation,neutrophil activation,B cell-mediated immunity,and anti-bacterial defense response.Analysis of the TCR repertoire show that the sum reads detected in samples before medication are between 178,115 and 590,122,and the sum reads detected in samples after medication are between 98388 and 795039.The number of unique clonotypes detected in the peripheral blood TCR repertoire before treatment are 63533?184859,and the unique clonotypes detected after treatment are 26838?149867.No significant difference exsists in the number of unique clonotypes in TCR repertoire before and after medication.After treatment,the proportion of small clones in the TCR library tends to decrease,and the proportion of highly expanded clones tends to increase,but the p values are not significant.The length of the CDR3 region of the repertoire and the distribution of gene use are similar before and after treatment.The most commonly used V genes are TRBV20-1 and TRBV5-1.The most commonly used J genes are TRBJ2-1 and TRBJ2-7.The cumulative frequency of the top 100 clones before treatment is between 0.16 and 0.53,the cumulative frequency of the top 100 clones after treatment is between 0.16 and 0.68,and the cumulative frequency of the top 100 clones of 4 patients is higher after medication.After treatment,the diversity of TCR repertoire in peripheral blood is significantly reduced(P=0.031,Wilcoxon test,Paired=TRUE).Conclusions:1.PD-1 inhibitors activate the T cell immune response in the peripheral blood of cancer patients,and inhibit the humoral immune response and neutrophil-mediated inflammatory response.2.PD-1 inhibitors have no significant effect on the number of unique clone types,the amino acid length of the CDR3 region,and the preference for V and J genes in the peripheral blood TCR immune repertoire.3.After PD-1 inhibitors were administered,the top 100 clones of the peripheral blood TCR immune repertoire expand in more than half of the patients.4.The diversity of the peripheral blood TCR repertoire is significantly reduced after PD-1 inhibitors were administered,suggesting that some specific high-frequency clones have expanded. |
PDF Full Text Request