Molecular Classification Of Nasopharyngeal Poorly-Differentiated Squamous Cell Carcinomas Using CDNA Microarrays

Nasopharyngeal carcinoma(NPC) is a malignancy with high incidence in Southern China and South-East Asia. The etiology of NPC is associated with Epstein-Barr Virus (EBV) infection, environmental factors and genetic alterations. However, there exist different histo-pathological classifications with confused nomenclature systems. Clinically, Poorly-differentiated squamous cell carcinoma is present in 98% NPC patients, well-differentiated squamous cell carcinoma and adenocarcinoma are rarely encountered. The current main therapy for NPC is radiotherapy, however, patients with NPC have highly variable clinical courses; although most patients respond initially to radiotherapy, five-year survival rate for NPC patients is about 45%. The clinical heterogeneity within NPC indicates that there are sub-types in nasopharyngeal poorly-differentiated squamous cell carcinoma.Molecular classification of tumor refers to screening several tumor markers for finding unknown human tumor subtypes and/or categorizing "special" tumors into known human ones using relevant molecular techniques cooperated with biological informatics based on gene expression patterns. This up-to-date classification system will facilitate the early diagnosis of tumors, advance the development of individual therapy for rumor patients and improve the prognosis evaluations for therapy. Nowadays, the molecular classification of tumors is becoming one of the most important fields in molecular onco-pathology.cDNA microarray provides a method for monitoring the mRNA expression levels of many thousands of genes simultaneously in primary tumors and cells. It is one of the most important techniques used in molecular classification of tumors.In the experiments of molecular classification of nasopharyngeal poorly-differentiated carcinoma, Human 'Named Genes' GeneFilters (Microarrays GF211, ResGen, USA) were used to study the gene expression profile innasophyrngeal carcinoma and normal nasophryngeal tissue. The Human 'Named Genes' GeneFilters contains approximately 4,000 genes, all of which are of known function. To rule out gene expression alterations because of stromal cell contamination, we confirmed that each tumor specimen used in our study contained >75% cancer cells by analysis of corresponding H&E-stained sections. Each of GF211 was individually hybridized with 32P-dCTP labeled cDNA probe prepared from RNA from 16 nasopharyngeal carcinoma and pooled RNA from 15 normal nasophryngeal tissue respectively. The difference in expression level was calculated by the ratio of intensity. The difference in signal intensity between two microarrays were normalized using Point normalization software from Pathways 4.0.From "global" 4133-gene(except control points, housekeeping genes and ESTs without Accession number) expression profiles, sixty-nine genes were found to be up-regulated in all of sixteen nasopharyngeal carcinoma, including E2F transcription factor, epidermal growth factor receptor pathway substrate 15, mitogen-activatedproteinase kinase kinase kinase 11 (MAPS K11), RAB6A, stress-induced-phosphoproteinl(Hsp70/Hsp90-organizing protein) and so on; Fifteen genes were down-regulated in most of nasopharyngeal carcinomas, including small membrane protein 1, interferon-related developmental regulator 2, et al. Meanwhile, Three hundred and twenty-eight significantly expressed gene were selected by SAM (significant analysis of microarray, SAM) statistical analysis method. Of these 328 SAM-selected genes, twenty genes were found that take part in signal transduction pathways, for example, the five genes involved in MAPK signal transduction pathways were found to be expressed differentially in GF211 microarrays, they might be correlated with the unregulated proliferation of NPC cells, invasive growth and metastasis.And, based on the expression profiling of 4133 genes, sixteen nasopharyngeal poorly-differentiated squamous cell carcinomas were classified into two subtypes using hierarchical clustering : Group one including NPC6, 7, 8, 9, 11, Group two inc...