Identification of the functional alteration signatures across different cancer types with support vector machine and feature analysis

Cancers are regarded as malignant proliferations of tumor cells present in many tissues and organs, which can severely curtail the quality of human life. The potential of using plasma DNA for cancer detection has been widely recognized, leading to the need of mapping the tissue-of-origin through the identification of somatic mutations. With cutting-edge technologies, such as next-generation sequencing, numerous somatic mutations have been identified, and the mutation signatures have been uncovered across different cancer types. However, somatic mutations are not independent events in carcinogenesis but exert functional effects. In this study, we applied a pan-cancer analysis to five types of cancers: (I) breast cancer (BRCA), (II) colorectal adenocarcinoma (COADREAD), (III) head and neck squamous cell carcinoma (HNSC), (IV) kidney renal clear cell carcinoma (KIRC), and (V) ovarian cancer (OV). Based on the mutated genes of patients suffering from one of the aforementioned cancer types, patients they were encoded into a large number of numerical values based upon the enrichment theory of gene ontology (GO) terms and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. We analyzed these features with the Monte-Carlo Feature Selection (MCFS) method, followed by the incremental feature selection (IFS) method to identify functional alteration features that could be used to build the support vector machine (SVM)-based classifier for distinguishing the five types of cancers. Our results showed that the optimal classifier with the selected 344 features had the highest Matthews correlation coefficient value of 0.523. Sixteen decision rules produced by the MCFS method can yield an overall accuracy of 0.498 for the classification of the five cancer types. Further analysis indicated that some of these features and rules were supported by previous experiments. This study not only presents a new approach to mapping the tissue-of-origin for cancer detection but also unveils the specific functional alterations of each cancer type, providing insight into cancer-specific functional aberrations as potential therapeutic targets. This article is part of a Special Issue entitled: Accelerating Precision Medicine through Genetic and Genomic Big Data Analysis edited by Yudong Cai & Tao Huang.     

The currently declining incidence of cancer of unknown primary

Cancer of unknown primary (CUP) is a heterogeneous group of metastatic cancers for which a primary tumor cannot be identified after a standardized work-up. The biology of CUP has not been fully elucidated and epidemiologic data may be helpful in this regard. The variations in the incidence-rate over time and between countries reflect changes in the risk factors for CUP, incidence trends of the primary tumors that potentially contribute to the burden of CUP and changes in the diagnostic technologies and practice. CUP accounted for 3–5% of cancers in the historical series but its incidence seems to decline in the recent publications. This paper reviews the published cancer-registry studies in order to identify and understand the variations in the incidence-rates of CUP.     

Demographic,social and lifestyle risk factors for cancer registry-notified cancer of unknown primary site (CUP)

BackgroundLittle is known about the risk factors for cancer of unknown primary site (CUP). We examined the demographic, social and lifestyle risk factors for CUP in a prospective cohort of 266,724 people aged 45 years and over in New South Wales, Australia.MethodsBaseline questionnaire data were linked to cancer registration, hospitalisation, emergency department admission, and mortality data. We compared individuals with incident cancer registry-notified CUP (n = 327) to two sets of controls randomly selected (3:1) using incidence density sampling with replacement: (i) incident cancer registry-notified metastatic cancer of known primary site (n = 977) and (ii) general cohort population (n = 981). We used conditional logistic regression to estimate adjusted odds ratios (ORs) and 95% confidence intervals (CIs).ResultsIn a fully adjusted model incorporating self-rated overall health and comorbidity, people diagnosed with CUP were more likely to be older (OR 1.05, 95% CI 1.04–1.07 per year) and more likely to have low educational attainment (OR 1.77, 95% CI 1.24–2.53) than those diagnosed with metastatic cancer of known primary. Similarly, compared to general cohort population controls, people diagnosed with CUP were older (OR 1.10, 95% CI 1.08–1.12 per year), of low educational attainment (OR 1.69, 95% CI 1.08–2.64), and current (OR 3.42, 95% CI 1.81–6.47) or former (OR 1.95, 95% CI 1.33–2.86) smokers.ConclusionThe consistent association with educational attainment suggests low health literacy may play a role in CUP diagnosis. These findings highlight the need to develop strategies to achieve earlier identification of diagnostically challenging malignancies in people with low health literacy.     

Human Papillomavirus and Cystic Node Metastasis in Oropharyngeal Cancer and Cancer of Unknown Primary Origin

The clinical significance of human papillomavirus (HPV) in neck node metastasis from cancer of unknown primary (CUP) is not well established. We aimed to address the relationship of HPV status between node metastasis and the primary tumor, and also the relevance of HPV status regarding radiographically detected cystic node metastasis in head and neck squamous cell carcinoma (HNSCC) and CUP. HPV DNA was examined in 68 matched pairs of node metastasis and primary tumor, and in node metastasis from 27 CUPs. In surgically treated CUPs, p16 was examined immunohistochemically. When tonsillectomy proved occult tonsillar cancer in CUP, HPV DNA and p16 were also examined in the occult primary. Cystic node metastasis on contrast-enhanced computed tomography scans was correlated with the primary site and HPV status in another series of 255 HNSCCs and CUPs with known HPV status. Node metastasis was HPV-positive in 19/37 (51%) oropharyngeal SCCs (OPSCCs) and 10/27 (37%) CUPs, but not in non-OPSCCs. Fluid was collected from cystic node metastasis using fine needle aspiration in two OPSCCs and one CUP, and all fluid collections were HPV-positive. HPV status, including the presence of HPV DNA, genotype, and physical status, as well as the expression pattern of p16 were consistent between node metastasis and primary or occult primary tumor. Occult tonsillar cancer was found more frequently in p16-positive CUP than in p16-negative CUP (odds ratio (OR), 39.0; 95% confidence interval (CI), 1.4–377.8; P = 0.02). Radiographically, cystic node metastasis was specific to OPSCC and CUP, and was associated with HPV positivity relative to necrotic or solid node metastasis (OR, 6.2; 95% CI, 1.2–45.7; P = 0.03). In conclusion, HPV status remains unchanged after metastasis. The occult primary of HPV-positive CUP is most probably localized in the oropharynx. HPV status determined from fine needle aspirates facilitates the diagnosis of cystic node metastasis.     

ND4 mutation in transitional cell carcinoma: Does mitochondrial mutation occur before tumorigenesis?

To investigate how mitochondrial mutation occurs in cancers, we analyzed ND4 mutation in 53 transitional cell carcinomas (TCCs) of the upper urinary tract and the normal counterpart (perirenal soft tissue). Three methods, i.e., DNA sequencing, restriction fragment length polymorphism (RFLP), and denaturing high-performance liquid chromatography (DHPLC), were employed because of their different sensitive of detecting mutation. The results of sequencing and RFLP showed that ND4 mutations were only found in 24.5% (13/53) of tumor. However, 11 of these mutations could also be identified in the normal tissue by DHPLC, indicating that most mitochondrial mutations identified in tumors preexist as minor components, which are too low in quantity to be detected by less sensitive methods such as DNA sequencing. The result suggests that mtDNA mutation occurs before tumorigenesis and become apparent in cancer cells.     

Prevalence of<Emphasis Type="Italic"> BRCA1</Emphasis> and<Emphasis Type="Italic"> BRCA2</Emphasis> mutations among clinic-based African American families with breast cancer

To define the prevalence and relative contributions of BRCA1 and BRCA2 mutations among African American families with breast cancer, we analyzed 28 DNA samples from patients identified through two oncology clinics. The entire coding regions of BRCA1 and BRCA2 were screened by protein truncation test, heteroduplex analysis, or single-stranded conformation polymorphism followed by DNA sequencing of variant bands. Deleterious protein-truncating BRCA1 and BRCA2 mutations were identified in five patients or 18% of the entire cohort. Only 8% (1 of 13) of women with a family history of breast cancer, but no ovarian cancer, had mutations. The mutation rates were higher for women from families with a history of breast cancer and at least one ovarian cancer (three of six, 50%). One woman with a family history of undocumented cancers was also found to carry a deleterious mutation in BRCA2. The spectrum of mutations was unique in that one novel BRCA1 mutation (1625del5) and three novel BRCA2 mutations (1536del4, 6696delTC, and 7795delCT) were identified. No recurrent mutations were identified in this cohort, although one BRCA2 (2816insA) mutation had been previously reported. In addition, two BRCA1 and four BRCA2 missense mutations of unknown significance were identified, one of which was novel. Taken together with our previous report on recurrent mutations seen in unrelated families, we conclude that African Americans have a unique mutation spectrum in BRCA1 and BRCA2 genes, but recurrent mutations are likely to be more widely dispersed and therefore not readily identifiable in this population.     

An audit of cancer of unknown primary notifications: A cautionary tale for population health research using cancer registry data

Background: Cancer of unknown primary (CUP) is a common cancer yet little is known about the reliability of incidence data. Methods: We audited 574 CUP (C80.9) diagnoses (median age 81 years) registered by the New South Wales (NSW) Central Cancer Registry (2004–2007) in a cohort of Australian Government Department of Veterans’ Affairs clients. The registry did not clarify diagnoses with notifiers during this period due to interpretation of privacy legislation. For the audit, current registry practice was applied by seeking additional information from CUP notifiers and reclassifying diagnoses as necessary. In addition, clinicopathological characteristics were extracted from notifications. Fisher's exact test and Student's t-test were used to compare the demographic and clinicopathological characteristics of the CUP subgroups. Age/sex-standardised CUP incidence rates and 95% confidence intervals were calculated, standardised to the 2001 Australian population. Results: 172 (30.0%) cases were reclassified to a known primary site, mostly cutaneous, and nine (1.6%) were found to be non-malignant diagnoses. After the audit the age/sex-standardised CUP incidence rates decreased from 26.0 (95% CI 21.2–30.8) to 15.9 (95% CI 12.5–19.3) per 100,000 person-years. Of the 393 remaining CUP cases, 202 (51%) were registered on the basis of a clinical diagnosis (46 by death certificate only) and 191 (49%) by pathological diagnosis (79 by cytology alone). Compared to cases with a pathological diagnosis, cases with a clinical diagnosis were older (85.6 vs. 82.0 years, p < 0.001), and the reported number and location of metastases differed (p < 0.001); metastatic sites were more likely to be unspecified for clinical diagnoses (36.1% vs. 4.2%). Conclusions: Cancer registry processes can markedly influence CUP incidence. Future population-based CUP research should take this into account, and consider stratification by basis of diagnosis due to differences in patient and tumour characteristics.     

Ice-COLD-PCR enables rapid amplification and robust enrichment for low-abundance unknown DNA mutations

Identifying low-abundance mutations within wild-type DNA is important in several fields of medicine, including cancer, prenatal diagnosis and infectious diseases. However, utilizing the clinical and diagnostic potential of rare mutations is limited by sensitivity of the molecular techniques employed, especially when the type and position of mutations are unknown. We have developed a novel platform that incorporates a synthetic reference sequence within a polymerase chain reaction (PCR) reaction, designed to enhance amplification of unknown mutant sequences during COLD-PCR (CO-amplification at Lower Denaturation temperature). This new platform enables an Improved and Complete Enrichment (ice-COLD-PCR) for all mutation types and eliminates shortcomings of previous formats of COLD-PCR. We evaluated ice-COLD-PCR enrichment in regions of TP53 in serially diluted mutant and wild-type DNA mixtures. Conventional-PCR, COLD-PCR and ice-COLD-PCR amplicons were run in parallel and sequenced to determine final mutation abundance for a range of mutations representing all possible single base changes. Amplification by ice-COLD-PCR enriched all mutation types and allowed identification of mutation abundances down to 1%, and 0.1% by Sanger sequencing or pyrosequencing, respectively, surpassing the capabilities of other forms of PCR. Ice-COLD-PCR will help elucidate the clinical significance of low-abundance mutations and our understanding of cancer origin, evolution, recurrence-risk and treatment diagnostics.     

Chromatin accessibility of primary human cancers ties regional mutational processes and signatures with tissues of origin

Somatic mutations in cancer genomes are associated with DNA replication timing (RT) and chromatin accessibility (CA), however these observations are based on normal tissues and cell lines while primary cancer epigenomes remain uncharacterised. Here we use machine learning to model megabase-scale mutation burden in 2,500 whole cancer genomes and 17 cancer types via a compendium of 900 CA and RT profiles covering primary cancers, normal tissues, and cell lines. CA profiles of primary cancers, rather than those of normal tissues, are most predictive of regional mutagenesis in most cancer types. Feature prioritisation shows that the epigenomes of matching cancer types and organ systems are often the strongest predictors of regional mutation burden, highlighting disease-specific associations of mutational processes. The genomic distributions of mutational signatures are also shaped by the epigenomes of matched cancer and tissue types, with SBS5/40, carcinogenic and unknown signatures most accurately predicted by our models. In contrast, fewer associations of RT and regional mutagenesis are found. Lastly, the models highlight genomic regions with overrepresented mutations that dramatically exceed epigenome-derived expectations and show a pan-cancer convergence to genes and pathways involved in development and oncogenesis, indicating the potential of this approach for coding and non-coding driver discovery. The association of regional mutational processes with the epigenomes of primary cancers suggests that the landscape of passenger mutations is predominantly shaped by the epigenomes of cancer cells after oncogenic transformation.     

Descriptive epidemiology of cancer of unknown primary site in Scotland, 1961–2010

BackgroundCancers of unknown primary site (CUP) pose problems for diagnosis, treatment, and accurate prediction of prognosis. However, there are limited published data describing the epidemiology of this disease entity. Our aim was to describe the epidemiology of CUP in Scotland.MethodsAnonymised data, covering the period 1961–2010, were extracted from the Scottish Cancer Registry database, based on the following ICD-10 diagnostic codes: C26.0, C26.8, C26.9, C39, and C76–C80. Age-standardised incidence rates were calculated by direct standardisation to the World Standard Population. Estimates of observed survival were calculated by the Kaplan–Meier method.ResultsBetween 1961 and 2010, there were 50,941 registrations of CUP, representing 3.9% of all registrations of invasive cancers. Age-standardised rates increased to a peak in the early to mid-1990s, followed by a steeper decrease in rates. During 2001–2010, age-standardised rates of CUP were higher in the most compared with the least deprived fifth of the population. Observed survival was marginally higher in patients diagnosed during 2001–2010 (median 5.6 weeks) compared with those diagnosed in the previous two decades. During the most recent decade, survival decreased with age at diagnosis, and was higher in patients with squamous cell carcinoma and with lymph node metastases.ConclusionPatterns of CUP in Scotland are largely consistent with those reported from the few other countries that have published data. However, in comparing studies, it is important to note that there is heterogeneity in terms of definition of CUP, as well as calendar period of diagnosis or death. Variation in the definition of CUP between different epidemiological studies suggests that there would be merit in seeking international agreement on guidelines for the registration of CUP as well as a standard grouping of diagnostic codes for analysis.     

Screening of the c-kit gene missense mutation in invasive ductal carcinoma of breast among north Indian population

Breast cancer is one of the most frequently diagnosed cancers and the leading cause of cancer deaths among females across the world, accounting for 23 % (1.38 million) of total new cancer cases and 14 % (0.45 million) of the total cancer deaths in 2008. c-kit is expressed in mast cell growth factor, cellular migration, proliferation, melanoblasts, haematopoietic progenitors and germ cells. We have designed our study with aim to explore the c-kit gene mutations in invasive ductal carcinoma (IDC) breast. To ascertain the range of mutations in exon 11, 13 and 17 of c-kit gene in 53 cases of IDC breast, we carried out PCR-SSCP followed by DNA sequencing. The mutation frequency of c-kit gene in exon 11, 13 and 17 were 9.43 % (5/53), 1.88 % (1/53) and 3.77 % (2/53), respectively. During our mutational analysis, we have detected five missense mutations in exon 11 (Pro551Leu, Glu562Val, Leu576Phe, His580Tyr and Phe584Leu), one missense mutation in exon 13 (Ser639Pro) and two missense mutations in exon 17 (Arg796Gly and Asn822Ser). It seems that c-kit mutations might participate in breast cancer pathogenesis and may be utilized as predictive marker, since the loss of c-kit positivity is generally linked with different types of breast cancer. Further molecular studies are necessary to validate the association of c-kit gene mutation in IDC breast pathogenesis.     

p16基因甲基化状态与散发性大肠癌的相关性研究

为探讨p1 6基因甲基化状态与散发性大肠癌发生发展的关系 ,用甲基化特异性的聚合酶链反应 (methylati omspecificPCR ,MSP)结合测序检测散发性大肠癌及相应癌旁组织p1 6基因甲基化状态。研究发现p1 6基因在散发性大肠癌中甲基化率为 2 8 9% (1 3 4 5 ) ,有 8例癌及癌旁组织都发生了甲基化 ;有淋巴结及远处转移的甲基化率为5 0 % (8 1 6 ) ,高于无转移的甲基化率 2 0 8(5 2 4 ) (P <0 0 5 )。p1 6基因高甲基化是散发性大肠癌中常见的分子改变之一 ,大肠癌中p1 6基因高甲基化可能发生在癌变早期并与大肠癌的恶性进展有相关性     

Health-related predictors of cancer registry-notified cancer of unknown primary site (CUP)

BackgroundThe relationship between comorbid disease and health service use and risk of cancer of unknown primary site (CUP) is uncertain.MethodsA prospective cohort of 266,724 people aged 45 years and over in New South Wales, Australia. Baseline questionnaire data were linked to cancer registration, health service records 4–27 months prior to diagnosis, and mortality data. We compared individuals with incident registry-notified CUP (n = 327; 90% C80) to two sets of randomly selected controls (3:1): (i) incident metastatic cancer of known primary site (n = 977) and (ii) general cohort population (n = 981). We used conditional logistic regression to estimate adjusted odds ratios (ORs) and 95% confidence intervals (CIs).ResultsIn fully adjusted models incorporating sociodemographic and lifestyle factors, people with cancer registry-notified CUP were more likely to have fair compared with excellent self-rated overall health (OR 1.78, 95% CI 1.01–3.14) and less likely to self-report anxiety (OR 0.48, 95% CI 0.24−0.97) than those registered with metastatic cancer of known primary. Compared to general cohort population controls, people registered with CUP were more likely to have poor rather than excellent self-rated overall health (OR 6.22, 95% CI 1.35–28.6), less likely to self-report anxiety (OR 0.28, 95% CI 0.12−0.63), and more likely to have a history of diabetes (OR 1.89, 95% CI 1.15–3.10) or cancer (OR 1.62, 95% CI 1.03–2.57). Neither tertiary nor community-based health service use independently predicted CUP risk.ConclusionLow self-rated health may be a flag for undiagnosed cancer, and an investigation of its clinical utility in primary care appears warranted.     

Somatic Variations in Cervical Cancers in Indian Patients

There are very few reports that describe the mutational landscape of cervical cancer, one of the leading cancers in Indian women. The aim of the present study was to investigate the somatic mutations that occur in cervical cancer. Whole exome sequencing of 10 treatment naïve tumour biopsies with matched blood samples, from a cohort of Indian patients with locally advanced disease, was performed. The data revealed missense mutations across 1282 genes, out of 1831 genes harbouring somatic mutations. These missense mutations (nonsynonymous + stop-gained) when compared with pre-existing mutations in the COSMIC database showed that 272 mutations in 250 genes were already reported although from cancers other than cervical cancer. More than 1000 novel somatic variations were obtained in matched tumour samples. Pathways / genes that are frequently mutated in various other cancers were found to be mutated in cervical cancers. A significant enrichment of somatic mutations in the MAPK pathway was observed, some of which could be potentially targetable. This is the first report of whole exome sequencing of well annotated cervical cancer samples from Indian women and helps identify trends in mutation profiles that are found in an Indian cohort of cervical cancer.     

Somatic Mutations,Viral Integration and Epigenetic Modification in the Evolution of Hepatitis B Virus-Induced Hepatocellular Carcinoma

Liver cancer in men is the second leading cause of cancer death and hepatocellular carcinoma (HCC) accounts for 70%-85% of the total liver cancer worldwide. Chronic infection with hepatitis B virus (HBV) is the major cause of HCC. Chronic, intermittently active inflammation provides “fertile field” for “mutation, selection, and adaptation” of HBV and the infected hepatocytes, a long-term evolutionary process during HBV-induced carcinogenesis. HBV mutations, which are positively selected by insufficient immunity, can promote and predict the occurrence of HCC. Recently, advanced sequencing technologies including whole genome sequencing, exome sequencing, and RNA sequencing provide opportunities to better under-stand the insight of how somatic mutations, structure variations, HBV integrations, and epigenetic modifications contribute to HCC development. Genomic variations of HCC caused by various etiological factors may be different, but the common driver mutations are important to elucidate the HCC evolutionary process. Genome-wide analyses of HBV integrations are helpful in clarifying the targeted genes of HBV in carcinogenesis and disease progression. RNA sequencing can identify key molecules whose expressions are epigenetically modified during HCC evolution. In this review, we summarized the current findings of next generation sequencings for HBV-HCC and proposed a theory framework of Cancer Evolution and Development based on the current knowledge of HBV-induced HCC to characterize and interpret evolutionary mechanisms of HCC and possible other cancers. Understanding the key viral and genomic variations involved in HCC evolution is essential for generating effective diagnostic, prognostic, and predictive biomarkers as well as therapeutic targets for the interventions of HBV-HCC.     

Founder mutations in the BRCA1 gene in Polish families with breast-ovarian cancer

We have undertaken a hospital-based study, to identify possible BRCA1 and BRCA2 founder mutations in the Polish population. The study group consisted of 66 Polish families with cancer who have at least three related females affected with breast or ovarian cancer and who had cancer diagnosed, in at least one of the three affected females, at age <50 years. A total of 26 families had both breast and ovarian cancers, 4 families had ovarian cancers only, and 36 families had breast cancers only. Genomic DNA was prepared from the peripheral blood leukocytes of at least one affected woman from each family. The entire coding region of BRCA1 and BRCA2 was screened for the presence of germline mutations, by use of SSCP followed by direct sequencing of observed variants. Mutations were found in 35 (53%) of the 66 families studied. All but one of the mutations were detected within the BRCA1 gene. BRCA1 abnormalities were identified in all four families with ovarian cancer only, in 67% of 27 families with both breast and ovarian cancer, and in 34% of 35 families with breast cancer only. The single family with a BRCA2 mutation had the breast-ovarian cancer syndrome. Seven distinct mutations were identified; five of these occurred in two or more families. In total, recurrent mutations were found in 33 (94%) of the 35 families with detected mutations. Three BRCA1 abnormalities-5382insC, C61G, and 4153delA-accounted for 51%, 20%, and 11% of the identified mutations, respectively.     

A high proportion of mutations in the BRCA1 gene in German breast/ovarian cancer families with clustering of mutations in the 3′ third of the gene

We have analyzed 61 German breast and breast/ovarian cancer families for BRCA1 mutations using single-strand conformation polymorphism analysis (SSCP) followed by sequencing. Forty-seven of the families had at least three cases (at least two under 60 years) and 14 families had only two cases of breast/ovarian cancer (at least one under 50 years). Twenty-eight families were breast/ovarian and 33 were breast cancer-only families. Eighteen mutations in BRCA1 were detected in 11/28 breast/ovarian cancer families and 7/33 breast cancer families and none in the families with only two cases. We identified 17 truncation mutations (8 frameshift, 7 nonsense and 2 splice variants) and one missense mutation. Seven of these are novel and two, the 5382insC and 5622C→T mutations, occurred in two apparently unrelated families. The genotype of the two families with the 5382insC mutation is compatible with the rare haplotype segregating with the 5382insC mutation in different populations, further supporting its European origin. One unclassified missense alteration, R841W, was found in one family but did not segregate with the disease, suggesting that it is more likely a polymorphism. We also report and discuss the sequence of several new unclassified single-nucleotide changes first identified by SSCP. Of the 18 mutations, 13 occurred in the 3′ third of the gene (end of exon 11–24) and ovarian cancers were found in eight of these families. Received: 5 February 1998 / Accepted: 7 April 1998     

RNA mutations of prox1 detected in human esophageal cancer cells by the shifted termination assay

We have recently reported a novel finding that a candidate tumor suppressor gene prox1 suffered adenosine-to-inosine (A-to-I) RNA mutation without genomic mutation in a subset of human cancer cells and lost its function. Hence, screening of mutations in both cDNA and genomic DNA could be important in the analysis of causes for cancers. Here, we applied a sensitive, accurate, and simple method, called shifted termination assay (STA) for detection of an A-to-I RNA mutation (R334G) in prox1. We prepared PCR-amplified samples containing the target base of RNA mutation from cDNAs and genomic DNAs of various cell lines and clinical samples, to demonstrate that the STA method can be used to identify not only genomic mutations but also RNA mutations more effectively compared to sequencing. By means of STA, we found prox1 R334G RNA mutations but not genomic DNA mutations in 4 of 8 cases of esophageal cancers. This method can help us to detect RNA mutation effectively and progress research of a potential oncogenic principle.