| 1 Background and Objective Esophageal cancer?EC?is the sixth cause of cancer related deaths worldwide.And EC is the fourth cause of cancer related death in China.China has the highest morbidity and mortality for EC worldwide.More than half of new occurrence and death of EC are occurred in China each year.EC has poor prognosis and the 5-year survival rate is around 20%.While various treatments methods have rapid progress,but the survival rate of EC is at a low level for many years.The EC patients with early stage have long survival time.But clinically more than 90%of EC patients are at advanced stage when they are diagnosed.Advanced ECs have poor survival conditions.But some advanced EC patients have long-term survival of more than ten years,or even more than three decades.How to find out the factors of affecting survival time of EC patients and how to improve the survival rate of EC patients are the mission for researchers and it's one of the attentions of many researchers.On the basis of large sample epidemiological survey,we work on the large-scale survival follow-up.We attempt to identify the factors that influence survival time for EC patients and provide a solid scientific basis for clinical prevention and treatment of EC.The strong difference of regional distribution and significant familial aggregation phenomena suggest that environmental and genetic factors play important role in the pathogenesis of EC.Some studies suggest that the immune defense mechanism may play an important role in the esophageal carcinogenesis.The human major histocompatibility complex?MHC?is an important area for ? ? autoimmune.In our previous study,we found 10 esophageal squamous cell carcinoma?ESCC?susceptible SNP loci and 8 related susceptibility genes by genome-wide association study?GWAS?on Chinese ESCC patients.And also in our previous study we found 3 SNPs in MHC region that related with ESCC,but did not have validation in additional independent samples.Based on the previous study and in order to assess the impact of immune factors on the incidence of EC,we further expand the sample size for GWAS of EC on MHC region.We select the SNPs on the MHC region related with ESCC by GWAS and validate the SNP loci in additional independent samples by using Taq Man.We try to determine the SNP loci on MHC region that associated with high risks of ESCC.2 Materials and methods2.1 The survival analysis for large-sample esophageal cancers2.1.1 Study objects Several hospitals and research institutes were jointed with Henan Key Laboratory,the First Affiliated Hospital of Zhengzhou University,for EC research.We carry out a large-scale follow-up survey for EC.The methods for collect patient data were include field investigation,cancer registration,verification in hospitals and follow-up by telephone,et al.A total of 61012 cases of patients with EC at 40 years spans?1974-2013 year?were collected.The deadline of follow up was July 24,2014.1)The general situation of patients In the 61 012 patients,38 289 were male cases,the minimum age was 15 years old and the oldest was 104 years old,the average age was 59.876 ± 9.480 years;22 723 were female cases,the minimum age was 24 years old and the oldest was 99 years old,the average age was 60.548 ± 9.513 years.2)The pathological diagnosis of research objects We verified the clinical information and pathological data from the hospitals.We recorded detail data of patient that included histological type,gross type,depth of invasion,differentiation,the location of esophageal tumor,lymph node metastasis,pathologic grade,TNM staging,et al.2.1.2 Research Methods1)Data CollectionFirst,all the investigators involved in the work were accepted rigorous training.The patients' information was obtained by using the questionnaire.The contents of questionnaire included general information?name,gender,age,occupation,education level,ethnicity,home address,telephone or phone number of patients or relatives,etc.?,lifestyle behaviors?smoking,drinking history,etc.?,clinical features?family history of cancers,cancer truth telling,the date of visit and treatment hospital,etc.?.Secondly,we collected clinical information and pathological data of patients from the medical record of hospitals.The contents included allergies,HP infection,histological type,gross type,esophageal tumor location,and length of tumor,degrees of differentiation,degree of infiltration,lymph node metastasis,histological grade,TNM staging,treatment and surgical stapling methods.Finally,we obtained survival and prognostic information by followed up patients through the methods of telephone and home visits,etc.All collected data were conducted scientific sorting,many times checking,and updated gradually.2)Survival Analysis The SPSS 21.0 statistical software was used for the analysis.The impacts of univariate element?gender,age,family history of cancers,gross type,depth of invasion,lymph node metastasis,etc.?on survival of patients were analysised by using Kaplan-Meier method.The Log rank P values were calculated.Cox regression model were used for multivariate analysis to comprehensive judgment the prognostic factors.The statistically significant projects by univariate element analysis were intaked to the Cox model.The P values,HR value and 95%confidence intervals?95% CI?were calculated.The standard for test was ? = 0.05.2.2 The GWAS study of esophageal cancer on MHC region2.2.1 Study objects The study samples were from the GWAS for EC of our study group.The case-control experimental design was used for this study.In the screening phase of GWAS,the 1 077 patients with esophageal cancer and 1 733 controls were genotyped by using Illumina 610 chip,and 451 patients with esophageal cancer and 374 controls were genotyped by using the Illumina 660 chip.A total of 3 635 cases,including 1 528 cases of patients?921 male cases,607 female cases,mean age 61 ± 19 years?and 2 107 controls?1 052 male,1 055 female,with an average age of 31 ± 15 years?.In validation phase,the Taq Man technology were used for verification on another separate 2 026 cases and 2 384 controls.A total of 4 410 cases for validation,including 2 026 cases of patients?1256 male,770 female,mean age 60 ± 9 years?and 2 384 controls?1 198 male,1 186 female,with an average age of 50 ± 11 years?.The patients were diagnosed ESCC by histopathology and the controls were confirmed no early EC and other upper gastrointestinal tumors by upper gastrointestinal endoscopy.2.2.2 Research Methods1)GWAS screening The genetical deflection populations were excluded by using the method of principal component analysis?PCA?.The DNA of matched cases and controls were gone through whole genome SNP scan analysis by using Illumina 610 and 660 chips.The criteria for SNP loci exclusion:?1?the call rate was less than 95% in the cases of patients and the controls;?2?the minor allele frequency?MAF?was less than 1% in controls;?3?the P value of HardyWeinberg equilibrium test?HWE?was less than 10-7 in controls.The SNPs in the MHC region that potential associated with high risk of ESCC were selected after removing unqualified SNPs.2)Taq Man verification The selected SNPs from screening were screened at the second time.The selection criteria of SNP for Taq Man validation were:?1?the MAF was more than 0.05 in the cases and controls;?2?the P value of HWE was ? 0.001 in the controls;?3?the P value of GWAS analysis?Cochran-Armitage trend test?less than 10-4 in cases and controls.The final locked SNP site that may be relevant with high-risk of ESCC were selected by using these standards.The selected SNPs were validated in another separate large sample size of ESCC and normal controls by using Taq Man fluorescent quantitative genotyping technology.3)Statistical analysis The call rate,MAF,and HWE of the SNPs in the MHC region were calculated.After quality control the data were analyzed by using Plink1.03 software.The P values,OR ? ? and 95%CI were calculated by using Cochran-Armitage trend test.The standard for test was ? = 0.05.3 Results3.1 The results of survival analysis on large sample of esophageal cancer3.1.1 The impact of general status on survival time of esophageal cancer patientsThe survival time of female patients were longer than male?Log rank P = 0.000?.The age at diagnosis had a significant effect on the survival time of patients with EC.The survival time of patients were progressive shortening with increasing age at diagnosis?Log rank P = 0.000?.The nationality had no impact on the survival time of patients with EC?Log rank P = 0.268?.The education level had a significant effect on survival time of patients with EC?Log rank P= 0.000?.The occupation had impact on the survival time of patients with EC?Log rank P =0.000?.3.1.2 The impact of behavior lifestyle on survival time of patients with esophageal cancerHistory of smoking had a significant effect on the survival time of patients with EC.The survival time of patients with smoking history were significantly shorter than those without a history of smoking?Log rank P = 0.000?.History of alcohol had significantly impact on survival time of patients with EC.The survival time of patients with history of alcohol were significantly shorter than patients with no history of drinking?Log rank P = 0.000?.3.1.3 The effect of high and low risk areas on survival time of esophageal cancer patientsThe high and low incidence area for EC had significantly impact on survival time of EC patients.The survival time of high incidence patients were significantly longer than the patients in middle or low risk areas?Log rank P = 0.000?.3.1.4 The effect of the clinical features on survival time of patients with esophageal cancerThe survival time of patients with positive family history for cancer were significantly longer than patients with negative family history?Log rank P = 0.000?.The blood type had no significant effect on survival time of patients with EC?Log rank P = 0.996?.The history of allergy had no significant effect on survival time of patients with EC?Log rank P = 0.148?.The single-double eyelid had significant effect on survival time of patients with EC?Log rank P = 0.003?.The HP infection had no effect on survival time of patients with EC?Log rank P= 0.328?.There were no difference of survival time between HBV infection positive and negative patients with EC?Log rank P = 0.102?.The survival time of HCV positive patient were significantly longer than HCV negative patients?Log rank P = 0.001?.The survival time of HIV-positive patient were significantly longer than the HIV negative patients?Log rank P= 0.003?.3.1.5 The effect of pathologic factors on the survival time of patients with esophageal cancerThe survival time of patients with esophageal squamous cell carcinoma and carcinosarcoma were longer than esophageal adenocarcinoma.The patients with esophageal small cell carcinoma had shortest survival time?Log rank P = 0.000?.There had statistically significant difference on survival time of patients with different tumor location,the survival time of patients with middle-lower location EC were longer than the upper section.The patients with cervical EC had shortest survival time?Log rank P = 0.000?.The length of tumor had a significant effect on survival time of EC patients.With the increasing of tumor length,the survival time of EC patients were progressively decreased?Log rank P = 0.000?.The width of tumor had significant effect on survival time of EC patients.With the width of tumor growing,the survival time of patient progressively decreased?Log rank P = 0.000?.The thickness of tumor had a marked impact on survival time of patients with EC.With the growing of tumor thickness,the survival time of EC patient progressively decreased?Log rank P = 0.000?.The overall survival time had significant differences among different gross types of patients with EC?Log rank P = 0.000?.The survival time had significant differences among different degree of differentiation of patients with EC.The worse of differentiation degree,the shorter of survival time?Log rank P = 0.000?.The degree of infiltration was an important factor for survival time in patients with EC.With the gradual deepening of T staging,the survival time of EC patient progressively reduced?Log rank P = 0.000?.The survival time of EC patients without lymph node metastasis was significantly longer than patients with lymph node metastasis?Log rank P = 0.000?.With increased number of positive lymph node metastasis,the survival time of EC patients gradually reduced?Log rank P = 0.000?.The total number of lymph node dissection had a significant effect on survival time of patients with EC?Log rank P = 0.000?.With the gradual rise of pathological grade,the survival time of EC patient progressively decreased?Log rank P = 0.000?.The survival time of patients with early stage cancer was significantly longer than the medium-term patients.The patients with advanced stage had shortest survival time?Log rank P = 0.000?.3.1.6 The influence of the treatment factors on survival time of patients with esophageal cancerThe methods of treatment had significantly impact on survival time of patients with EC?Log rank P = 0.000?.The surgical stapling method had significant influence on the survival time of EC patients.The patients with method of cervical anastomosis had longer survival time than below the arch anastomosis.The method of under the arch anastomosis had shortest survival time?Log rank P = 0.000?.There were no difference in survival time on EC patients with different ways for thoracotomy surgery?Log rank P = 0.655?.The survival time of EC patients with cancer truth telling was significantly longer than patients with no cancer truth telling?Log rank P = 0.000?.The survival time of EC patients with different treatment period had statistical difference?Log rank P = 0.000?.3.1.7 The results of Cox regressionWomen was protective factors for the prognosis of patients with EC?p = 0.000,HR = 0.834,95% CI = 0.799-0.870?.The age at diagnosis was impact factors for prognostic of EC patients.The risk of death in patients ?70 years of age were increased significantly compared with <40 year-old patient?P = 0.000,HR = 1.856,95% CI = 1.487-2.317?.The tumor location was impact factors for prognostic in patients with EC.The patients with lower location of EC had significantly lower risk of death than the patients with cervical location of EC?P = 0.006,HR = 0.609,95% CI = 0.426-0.869?.The degree of differentiation was impact factors for prognostic of patients with EC.The patients with undifferentiated EC had significantly higher risk of death than patients with high-differentiated carcinoma?P = 0.002,HR = 1.503,95% CI = 1.155-1.957?.The degree of infiltration was an independent prognostic factor for patients with EC?P= 0.000,HR=1.372,95%CI=1.323-1.423?.The lymph node metastasis was risk factors for prognostic in patients with EC.The EC patients without lymph node metastasis had the significantly lower risk of death than patients with lymph node metastasis?P = 0.000,HR = 0.679,95% CI = 0.626-0.737?.The pathological grade was prognostic factors in patients with EC.The risk of death of patients with stage IV of EC were increased significantly compared with 0 stage patients?P = 0.007,HR = 3.169,95% CI = 1.361-7.378?.The treatment period was impact factor for prognostic in patients with EC.The risk of death of EC patients with treated in the last five years?2009-2013year?was significantly increased than patients treatment in earlier ten years?1974-1983year??P =0.000,HR = 4.145,95% CI = 2.911-5.902?.3.2 The results of GWAS on MHC region for esophageal cancer3.2.1 The Screening results of GWAS on MHC region for esophageal cancerWe obtained genetically matched 1 528 cases of patients with EC and 1 056 controls after removing deviated samples by PCA for 1 528 cases and 2 107 controls.The DNA of matched cases and controls were gone through genome-wide SNP scanning and 6 252 SNP loci in the MHC region were detected.The 2 533 SNP loci were selected for the next step analysis after Quality control.3.2.2 The results of Taq Man validationWe locked five SNP loci in the MHC region may be associated with high risk of ESCC after removed the failed SNPs by SNP select standard for the validation.The five SNPs were:rs17533090(PGWAS = 1E-05),rs35399661(PGWAS = 6E-06),rs1536501(PGWAS = 9E-04),rs911178(PGWAS = 6E-04)and rs6901869(PGWAS = 3E-05).After Taq Man verification on 2026 cases and 2 384 controls,finally,we got one significant SNP loci: rs911178?P replication =1.41E-22?located in the 6p22.1,in the location of upstream 35-kb of SCAN domain containing 3?SCAND3?.4 Conclusions4.1 The gender,age at diagnosis,esophageal cancer location,tumor differentiation,the degree of infiltration,lymph node metastasis,histological grade,and the treatment period are the independent impact factors for prognostic in patients with EC.The survival time of EC patients are related with history of smoking,history of alcohol,high and low incidence area for EC,family history for cancer,histological type,the length of tumor,the width of tumor,the thickness of tumor,the gross type,the number of positive lymph node metastasis,clinical stage,the methods of treatment,and the surgical stapling method.Survival analysis can provide strong basis for prevention and treatment of EC.4.2 The rs911178?SCAND3 gene?in MHC region is significantly correlated with the high risk of ESCC.This study provides more understanding of the role of MHC region for the pathogenesis of ESCC,and also provides important clues for the establishment of the tools and methods at the screening for high risk population. |
PDF Full Text Request