| Summary: | 碩士 === 國立交通大學 === 生物科技學系 === 103 === Head and neck squamous cell carcinoma (HNSCC) is the sixth leading cancer by incidence worldwide. About 75% of HNSCC patients were diagnosed as later stage and even metastasis due to lack of diagnostic biomarkers. In addition, recent studies have revealed that the molecular classification of HNSCC remains unclear and HNSCC lacks of biomarkers for early diagnosis and clinical outcome. Therefore, identification of a set of genes (as biomarkers) for diagnosis, prognosis, and molecular classification of HNSCC is an emergency task to develop therapy and clarify the mechanisms of HNSCC.
To address these issues, we have collected gene expressions of 235 HNSCC clinical samples, contained both tumor and corresponding normal tissues, from three different sources. We found 289 genes which are consistently and significantly high expression in most tumor samples. Moreover, we proposed an integrated method to select 88 HNSCC-related genes (called HNC88) based on quantifying gene expression behavior and influence of regulation in biological networks during carcinogenesis. In HNC88, we examined the expression levels of eight genes by immunohistochemical (IHC) staining. The results show that three, MMP9, NCF2, and IFI30, of eight genes are strong positive in tumor invasion front. In the center of tumor mass, these three genes also represented positive but not in surrounding stromal cells, inflammatory cells and normal epithelial cells.
To further analyze and validate HNC88, we collected RNA-Seq data (522 tumor and 44 corresponding normal tissues) and clinical information of HNSCC patients from the cancer genome atlas (TCGA) database as an independent testing set. Among these 566 samples, we selected 112 tumor samples with human papilloma virus (HPV) states (+/-) and 44 normal samples from TCGA database, and then clustered the gene expression profiles by using HNC88. These results show that HNC88 is not only able to distinguish normal, HPV(+), and HPV(-) samples, but also divide HPV(-) samples into two sub-groups (i.e. HPV1(-) and HPV2(-)). In HPV1(-) samples, we found that gene expressions of most genes involved in the immune system and metabolism of oxygen are significantly higher than HPV2(-) samples. In addition, patients with ≥80% up-regulated genes in HNC88 exhibited a relatively poor prognosis than the other patients in five-year survival (Log-Rank p=0.012). We believe that HNC88 a set of effective biomarkers for distinguishing sub-groups of HNSCC, predicting prognosis, and providing the clues to develop target therapy of HNSCC. Furthermore, our method can be a general framework for identifying a set of genes for the other cancers.
|
|---|
