1.Establishment of HNSCC metastatic cellular model 2.Study on the aberrant DNA methylation of human HNSCC and its association with clinical implications

• Main Authors: Wei-Li Huang, 黃偉立
• Other Authors: Shou-Yen Kao
• Format: Others
• Language: en_US
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/78229717036844651215

博士 === 國立陽明大學 === 牙醫學系 === 102 === Head and neck squamous cell carcinoma (HNSCC) is the sixth most common type of cancer in the world. Oral squamous cell carcinoma (OSCC), the major type of HNSCC, is the 4th leading malignancy affecting middle-aged men in Taiwan. Investigations on the cause, progression, and prognosis of cancer may improve the prevention and cure of HNSCC. Distant metastasis is the major cause of cancer-related mortality. Prevention or elimination of metastasis may improve the survival of cancer patients; however, the availability of systemic HNSCC cell model with which to investigate the mechanisms of metastasis is limited. This study established several lines of metastatic cells from the high tumorigenic HNSCC cells (SAS_VO3). The SAS_VO3 cells were injected into the tail vein of nude mouse and the pulmonary colonization were observed. The lung tumors were removed and used to generate the metastatic cell lines (SAS_VO3_M). In combination with their low-tumorigenic (SAS) and high-tumorigenic ancestor cell lines, a cell model containing cell lines with varying malignant characteristics was established. Transcriptome analysis revealed distinct signatures among the metastatic cell lines, in comparison to the ancestor cell lines. Signaling of GTPase RhoA and mammalian embryonic stem cell pluripotency were identified in the metastatic cells. Moreover, the capabilities of the cells for migration, invasion, and pulmonary colonization in nude mice were determined. Together, we demonstrated filopodium formation and enhanced invasion activity in metastatic HNSCC cells, rendering this model a powerful tool for the development of diagnostic biomarkers and identification of therapeutic targets for HNSCC. Although tumor metastasis is highly associated with cancer mortality, the prognostic predictor of distant metastasis is limited. The origin of metastatic cells was hypothesized to be exist in the primary tumor. This study aimed to profile the epigenetic modification of HNSCC primary tumors in order to characterize the potential distant metastasis-related features and evaluate the clinical implications of distant metastasis-related DNA methylation. Aberrant DNA methylation of CpG loci in primary tumors was profiled using DNA methylation array. A total of 14 HNSCC tumors and 2 normal oral tissues were included. A cluster of distant metastasis-specific 245 CpG loci was selected and hierarchical diagram revealed differential CpG methylation pattern between tumors with or without distant metastatic activity. Ingenuity Pathway Analysis revealed that distant-metastasis related genes were strongly associated with neuronal disorders and metabolism diseases. The methylation status of p16 (n=107) and brain-derived neurotrophic factor (BDNF) (n=84) of HNSCC tumors was assayed by established methylation-specific PCR. Hypermethylation of either p16 or BDNF were not associated with stage or differentiation of tumors. Hypermethylation of p16 was significantly associated with age (p=0.008) and regional invasion (p=0.013). Hypermethylation of BDNF was not associated with regional invasion (p=0.383), but with distant metastasis of follow-up (p=0.004). Multivariate analysis revealed that distant metastasis was correlated to hypermethylated BDNF (adjusted odds ratio: 5.77, p=0.004). The adjusted odd ratio was 10.37 (p=0.003) when both p16 and BDNF was hypermethylated. Kaplan-Meier analysis showed that hypermethylation of BDNF was significantly associated with poor survival (p=0.013). On the other hand, p16 hypermethylation associated with survival in an age-dependent manner. Combination of hypermethylated p16 and BDNF further distinguished the life span of HNSCC patients (p<0.001). Together, this study shows a distinct DNA-methylation signature of HNSCC tumor with differential metastatic activity. Hypermethylation of p16 and BDNF may be independent predictor of the prognosis of HNSCC.