INTRA-TUMOR HETEROGENEITY DETERMINATION BASED ON GENOMIC AND PROTEOMIC DATA

Erdoğan E. O., Turna Ö. C.

International Başkent Congress, Ankara, Türkiye, 16 - 17 Temmuz 2021, ss.224-234, (Tam Metin Bildiri)

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Ankara
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.224-234
  • İstanbul Üniversitesi-Cerrahpaşa Adresli: Evet

Özet

Cancer is a disease related to uncontrolled cell proliferation in a tissue or organ. It stems from molecular alterations within the cells leading the intracellular mechanisms to deviate from its normal functioning. Cancerous cells may undergo consecutive molecular alterations. Thus, several types of cancerous cell groups emerge within the same tumor. Intra-Tumor Heterogeneity (ITH) refers to the distinct groups of cells that a single tumor comprises. ITH is associated with numerous prognostic factors including survival, risk of metastasis and so forth. Therefore, it is essential to determine ITH to draw inferences about disease prognosis. Next Generation Sequencing (NGS), which is a massively parallel sequencing technology, allows researchers to focus on ITH by providing large datasets. Hitherto, the determination of ITH based on protein data has not been extensively studied. This study proposes a novel approach by utilizing Reverse-Phase Protein Arrays (RPPA) data for the purpose of establishing a prognostic biomarker that explains survival as the most crucial ITH-associated feature. Since the proteins regulate the intracellular activity, under- or over-synthesis of the proteins may disrupt the intracellular mechanisms. Therefore, Protein Aberrancy Index is calculated to reflect how aberrantly a protein is produced. Utilizing Protein Aberrancy Index in the survival analysis yields meaningful results. In this scope, Cox proportional hazards model is developed by using Gene Expression Aberrancy Index, Protein Aberrancy Index, CNV and DNA mutation data. The datasets are provided by TCGA project including 33 distinct tumor types and more than 5000 samples. Each sample has gene expression, RPPA, CNV, DNA mutation data along with a clinical data. Pan-cancer survival analysis results show that RPPA is significantly associated with the survival in both univariate and multivariate model. RPPA is also strongly associated with survival in numerous distinct cancers such as COAD, GBM, KIRC, LGG, LUSC, THCA.